Professional Documents
Culture Documents
THESE DE DOCTORAT
présentée par
Bersissa Kumsa
MOLECULAR INVESTIGATION
OF ARTHROPODS AND VECTOR-
BORNE BACTERIA FROM ETHIOPIA
Membres du Jury :
1
2
Acknowledgements
First and foremost I would like to express my sincere
gratitude to my PhD supervisor, Prof. Didier Raoult,
the Director of the laboratory and renowned scientist of
Clinical microbiology, for his continuous guidance,
suggestions and constructive comments throughout my
research. I also would like to thank him very much for
giving me a very important chance to do a PhD degree,
for his financial support (AP+HM) of my PhD research
at his laboratory.
I am very pleased and thankful to my PhD supervisor
Prof. Philippe Parola for his constant suggestions, lots
of good ideas, rigorous guidance, very professional
comments, great advice and mentor for my PhD
research.
I am greatly thankful to my 3rd PhD supervisor Dr.
Cristina Scolvoccshi for her constant suggestions, lots
of critically important comments and great advice
during my PhD research.
I would like to thank the reviewers of my PhD thesis
Prof. Sarah Bonnet and Prof Marthy Pierre for their
high quality professional advices and detailed review
during the preparation of the thesis. Their sincere
suggestions were critically important to improve the
quality of this PhD thesis.
3
4
My heartfelt thanks go to the technicins of URMITE
laboratory particularily Annick Bernard, Laurence
Thomas, Veronique Brice, Stephanie Junoy, Marielle
Bedotto, Anne-Marie Gottrau and Jean-Michel
Berenger for their technical support during the whole of
my PhD research in the Laboratory.
Also my deepest heartfelt thanks go to the secretaries of
URMITE especially MS. Francine Simula and Valerie
Filosa for their very kind and honest help regarding
very important issues like accommodation and VISA
arrangements throughout my 3 years stay in Marseille
for PhD research at the laboratory.
I take this opportunity to thank my wife Tadelech Fite
and son Barecha Bersissa for their great love, moral
support and patience throughout my long absence from
the lovely home.
Lastly but not least my great thanks goes to my beloved
mother Dinkitu Abetu Morka, for her moral support,
great love and thinking about me for the entire of her
life.
5
TABLE OF CONTENTS
Acknowledgements 3
INTRODUCTION 13
2. REVIEW 19
2.1. Article 1 21
Negelected Vector-borne Bacterial Zoonoses in East
Africa. Kumsa, B., Raoult, D., Parola, P.,
To be submitted to Acta Tropica
3. TICKS 121
3.1. Article 2 127
Spotted fever group rickettsiae in ixodid ticks in
Oromia, Ethiopia. Kumsa, B., Socolovschi, C., Raoult,
D., Parola, P., 2014. Ticks Tick Borne Dis
3.2. Article 3 137
Occurrence and genotyping of Coxiella burnetii in
ixodid ticks in Oromia, Ethiopia. Kumsa, B.,
Socolovschi, C., Almeras, L., Raoult, D., Parola, P., To
be submitted to Plos Neglected tropical Diseases.
3.3. Article 4 165
New Borrelia species detected in ixodid ticks in
Oromia, Ethiopia. Kumsa, B., Socolovschi, C., Raoult,
R., Parola, P., To be submitted to Ticks Tick Borne Dis.
3.4. Article 5 193
Morphological, MALDI TOF Mass spectrometry and
molecular identification of ixodid tick species in
Oromia, Ethiopia. Kumsa, B., Almeras, L., Yussuf, A.,
Mediannikov, O., Raoult, D., Parola, P., To be
submitted to Veterinary Parasitollogy.
6
4. FLEAS 237
4.1. Article 6 215
Molecular detection of Rickettsia felis and Bartonella
henselae in dog and cat fleas in central Oromia,
Ethiopia. Kumsa, B., Parola, P., Raoult, D.,
Socolovschi, C., Am. J. Trop. Med. Hyg., 90(3), 2014,
pp. 457–462
5. LICE AND SHEEP KED 249
5.1. Article 7 227
Molecular detection of Acinetobacter species in the lice
and keds of domestic animals in Oromia Regional State,
Ethiopia. Kumsa, B., Socolovschi, C., Raoult, D.,
Parola, P., PLoS ONE 7(12): e52377.
doi:10.1371/journal.pone.0052377.
5.2. Article 8 241
Bartonella melophagi in Melophagus ovinus (sheep
ked) collected from sheep in northern Oromia, Ethiopia.
Kumsa, B., Raoult, D., Parola, P., Socolovschi, C.,
Comparative Immunology, Microbiology and Infectious
Diseases 37 (2014) 69– 76.
6. CONCLUSIONS 249
6.1. Conclusions and perspectives 251
Bibliography 281
7
Résumé
Les arthropodes comme les tiques, les puces, les poux et les moutons ked
se trouvent dans toutes les régions du monde et sont plus fréquents dans
les pays tropicaux. Les vecteurs arthropodes hématophages comme les
tiques, les puces, les poux et les moutons ked transmettent les bactéries
qui sont généralement associés aux maladies fébriles chez l'homme. Les
informations sur les bactéries transmises par les arthropodes est rare en
Ethiopie. Ainsi, l'objectif principal de cette thèse est d'accroître les
connaissances sur les bactéries à transmission vectorielle en Ethiopie par
l'exploration de leur présence dans plusieurs types d'arthropodes y compris
les tiques, les puces, les poux et les moutons ked prélevés sur des animaux
dans plusieurs départements du pays. En outre, nous avons fait une
expérience sur les nouveaux outils pour identifier les tiques par MALDI-TOF
MS protéines profilage et des méthodes moléculaires.
Notre étude visant à explorer les bactéries dans les ixodidae prélevés sur
des animaux domestiques en Éthiopie a révélé une prévalence globale
de 6% (46/767) des rickettsies de SFG, 3,8% (29/767) ADN de Borrelia et
6,4% (54/842) de C. burnetii dans différentes espèces de tiques. Africae
Rickettsia a été détecté dans Amblyommma variegatum, Am. gemma,
Am. cohaerens, Amblyomma spp. larves, nymphes Amblyomma,
Rhipicephalus (Boophilus) decoloratus et Rh. (Bo.) Nymphes de
decoloratu. La prévalence de R. africae était plus élevée dans le
département centrale et orientale que dans l'ouest. R. aeschlimannii a été
détecté dans Hyalomma rufipes marginatum et Hy. truncatum. Notre
étude est la première à détecter R. massiliae dans Rhipicephalus
Praetextatus en Ethiopie. Borrelia spp. a été détecté dans Am. cohaerens,
Am. variegatum, Am. gemma, les larves Amblyomma et les nymphes
Amblyomma. La nouvelle Borrelia spp. détectée dans les tiques
Amblyomma clustérise entre le groupe de la fièvre récurrente et le groupe
Borrelia de la maladie de Lyme alors que la Borrelia sp. détectée dans
Rhipicephalus tiques clustérise avec B. theileri / B. lonestari. Coxiella
burnetii a été détecté dans Am. gemma, Rh. pulchellus et dans d'autres
espèces de tiques, principalement en zone Borana en Oromia. Les 18
génotypes de tiques des départements du sud-est et les 20 génotypes de
tiques dans les départements centraux ont été déterminés par le typage
par séquençage multilocus (MST) de C. burneti. Après la création d'une
base de données de références fiables des spectres obtenus à partir des
jambes sur un total de 41 spécimens de tiques suivie par un test à
l'aveugle avec un total de 44 spécimens de tiques, nous avons identifié
11 espèces de tiques ixodidae par la méthode MALDI-TOF-MS. Les
méthodes d'identification morphologiques des tiques et MALDI-TOF ont
8
été confirmées par séquençage du gène de l'ARNr 12S d’où une
identification moléculaire des différentes espèces de tiques.
L'étude pour étudier les bactéries dans 303 puces prélevés sur des chiens
et des chats domestiques en Ethiopie qui ont été identifiés comme étant
morphologiquement Ctenocephalides felis felis, Ctenocephalides canis,
Pulex irritans et Echidnophaga gallinacé montré Rickettsia felis dans 21%
des puces, principalement dans Ctenocephalides felis, avec une
semblable prévalence dans les puces de chiens et de chats. Notre étude
est la première à signaler Bartonella henselae en C. felis de l'Ethiopie.
Notre étude sur les bactéries des poux et des moutons ked (Melophagus
ovinus) a révélé la présence d'Acinetobacter spp. dans M. ovinus,
Heterodoxus spiniger, Bovicola ovis et Linognathus vituli. La séquence du
gène rpoB partiel a révélé la présence de A. soli, A. lowffii, A. Pitti et
3 nouveaux Acinetobacter spp. dans les poux et Keds. L'identification
moléculaire des poux à l'aide d'une analyse du gène 18S a confirmé les
méthodes morphologiques d'identification des poux. Bartonella
melophagi a été identifié par une PCR standard, suivi par un séquençage
du fragment de la gltA et gène rpoB chez M. ovinus.
La présente étude a contribué à accroître les connaissances actuelles en
Ethiopie en donnant la répartition géographique des rickettsies du groupe
de la fièvre boutonneuse avec les espèces de tiques associées, la
première preuve moléculaire de détection de R. massiliae dans
Rhipicephalus Praetextatus, de B. henselae dans C. felis, B . melophagi
dans M. ovinus, nouvelle Borrelia spp. dans des tiques Amblyomma,
nouveau Acinetobacter spp. dans des poux et KED et première preuve de
génotypes de C. burnetii chez les tiques en Ethiopie et a introduit un
nouveau procédé d'identification des tiques par la technique MALDI TOF
MS.
Dans l'ensemble, nos résultats alerte les médecins en charge des patients
avec fièvre d'étiologie inconnue en Ethiopie et ceux qui se soucient de
voyageurs en provenance de l'Ethiopie à prendre en compte la présence
de plusieurs espèces zoonotiques à transmission vectorielle de bactéries, y
compris SFG rickettsies, C. burnetii, R. felis, B. henselae et B. melophagi
comme agents pathogènes potentiels. Cependant, les études futures sont
nécessaires pour traiter l'isolement, la culture et les génotypes de ces
bactéries provenant de différentes sources et de déterminer l'importance
en santé publique de ces bactéries et la compétence vectorielle des
arthropodes en Ethiopie.
9
Abstract
Arthropods such as ticks, fleas, lice and sheep ked are found in
every part of the world and are more common in tropical countries.
Haematophagous arthropod vectors such as ticks, fleas, lice and
sheep ked transmit bacteria that are usually associated with febrile
illnesses in humans. Information on vector-borne bacteria
transmitted by arthropods is scarce in Ethiopia. Thus, the main
objective of this thesis was to increase the present understanding on
vector-borne bacteria in Ethiopia by exploring their presence in
several types of arthropods including ticks, fleas, lice and sheep ked
collected from domestic animals in several districts in the country. In
addition, we made an experiment on emerging tools to identify
ticks by MALDI-TOF MS protein profiling and molecular methods.
Our study to explore bacteria in ixodid ticks collected from
domestic animals in Ethiopia revealed an overall prevalence of 6%
(46/767) SFG rickettsiae, 3.8% (29/767) Borrelia DNA and 6.4%
(54/842) C. burnetii in different tick species. Rickettsia africae was
detected in Amblyommma variegatum, Am. gemma, Am.
cohaerens, Amblyomma spp. larvae, Amblyomma nymphs,
Rhipicephalus (Boophilus) decoloratus and Rh. (Bo.) decoloratu
nymphs. The prevalence of R. africae was higher in the central and
eastern than in the western districts. R. aeschlimannii was detected
in Hyalomma marginatum rufipes and Hy. truncatum. Our study is
the first to detect R. massiliae in Rhipicephalus praetextatus in
Ethiopia. Borrelia spp. was detected in Am. cohaerens, Am.
variegatum, Am. gemma, Amblyomma larvae and Amblyomma
nymphs. The new Borrelia spp. detected in Amblyomma ticks cluster
in between the recurrent fever and the Lyme disease group
borreliae whereas the Borrelia sp. detected in Rhipicephalus ticks
clusters with B. theileri/B. lonestari. Coxiella burnetii was detected in
Am. gemma, Rh. pulchellus and in other tick species mainly in
Borana zone in Oromia. Genotype 18 in ticks from south-eastern
districts and genotype 20 in ticks from central districts were
determined by multi-spacer sequence typing (MST) of C. burneti.
After creating a reliable reference database of spectra obtained
from the legs of a total of 41 tick specimens followed by blind test
with a total of 44 tick specimens we correctly identified 11 ixodid
ticks species by MALDI-TOF-MS method. The morphological and
10
MALDI-TOF tick identification methods were confirmed by
sequencing of the 12S rRNA gene molecular identification of tick
species.
The study to investigate bacteria in 303 fleas collected from
domestic dogs and cats in Ethiopia that were morphologically
identified as Ctenocephalides felis felis, Ctenocephalides canis,
Pulex irritans and Echidnophaga gallinacean showed Rickettsia felis
in 21% of fleas, mainly in Ctenocephalides felis, with a similar
prevalence in fleas from dogs and cats. Our study is the first to
report Bartonella henselae in C. felis from Ethiopia.
The study to investigate bacteria in lice and sheep ked
(Melophagus ovinus) revealed Acinetobacter spp. in M. ovinus,
Heterodoxus spiniger, Bovicola ovis and Linognathus vituli. Partial
rpoB gene sequence revealed A. soli, A. lowffii, A. pitti and 3 new
Acinetobacter spp. in the lice and keds. Molecular identification of
lice using an 18S rRNA gene analysis confirmed the morphological
methods of lice identification. Bartonella melophagi was identified
by standard PCR followed by sequencing of fragments of the gltA
and rpoB genes in M. ovinus.
The present study contributed to increase the current knowledge in
Ethiopia by providing updated geographical distribution of spotted
fever group rickettsiae with associated tick species, provided first
molecular evidence of R. massiliae in Rhipicephalus praetextatus,
first report of B. henselae in C. felis, B. melophagi in M. ovinus, new
Borrelia spp. in Amblyomma ticks, new Acinetobacter spp. in lice
and ked of animals and first report of C. burnetii genotypes in ticks
in Ethiopia and introduces new method of tick identification by
MALDI TOF MS technique.
Overall, our findings alert physicians managing patients with fever of
unknown aetiology in Ethiopia and those who care for travellers
from Ethiopia to consider the presence of several vector-borne
zoonotic species of bacteria including SFG rickettsiae, C. burnetii,
R. felis, B. henselae and B. melophagi as potential causative agents.
However, future studies are needed to address the isolation, culture
and genotypes of these bacteria from different sources and
determine the public health significance of these bacteria and
vector competence of arthropods in Ethiopia.
11
12
INTRODUCTION
13
14
Introduction
lice and sheep ked are found in every part of the world and
15
creatures (17). Arthropods such as ticks, fleas, lice and sheep
(17).
16
like relapsing fever, trench fever and epidemic fever
17
investigators of the present study to explore the diversity of
18
2. REVIEW
19
20
2.1. Article 1
21
22
NEGLECTED VECTOR-BORNE
23
24
ABSTRACT
Vector-borne bacterial diseases cause significant morbidity
and mortality in both humans and animals around the world
and affect the global economy. They are transmitted during
the blood feeding activities of hematophagous arthropod
vectors such as ticks, fleas, lice and flies. Febrile illness is
one of the most common clinical symptoms associated with
vector-borne zoonotic bacterial infection and is the major
cause of morbidity and mortality in east African countries
but too often, tests are not available to determine the causes,
leading to misdiagnosis and inappropriate treatment. Due to
the widespread occurrence of malaria in east African
countries there is great confusion between malaria and
malaria mimicking treatable vector-borne bacterial infections
associated with febrile illnesses. Tick-borne bacterial
diseases (spotted fever rickettsiosis, typhus group
rickettsiosis, tick-borne relapsing fever borreliosis,
bartenolosis and Q fever), louse-borne bacterial diseases
(epidemic relapsing fever, trench fever, epidemic typhus and
Acinetobacter baumannii) and flea-borne bacterial zoonoses
(flea-borne spotted fever, murine typhus, cat scratch disease
and plague) are common in east African countries but poorly
investigated. Even though recent investigations indicate
25
escalating problems associated with arthropod vectors and
vector-borne diseases across other parts of the world
relatively little attention has been given to arthropod
transmitted zoonotic bacterial diseases in east African
countries. Thus the main objective of this review is to update
our current understanding on the geographical distribution,
epidemiology, clinical aspects, public health issues,
treatment and show research gaps on arthropod transmitted
zoonotic bacterial diseases in east African countries so as to
provide better understanding and draw the attention of
researchers, clinicians, microbiologists, scientific and
medical community working in this part of the world.
26
TABLE OF CONTENTS
1. INTRODUCTION
Africa
Animals
12. CONCLUSION
27
28
1. INTRODUCTION
29
the world due to expanding vector populations, global
warming, increased contacts between humans, animals and
vectors due to enhanced international commerce, increased
and more rapid global transport, human and animal
population dynamics, and emerging drug resistance in both
vectors and pathogens (Dantas-Torres et al., 2012; De la
Fuente and Estrada-Pena, 2012).
East Africa as defined by the United Nations comprises
countries including Djibouti, Eritrea, Ethiopia, Somalia,
South Sudan, Tanzania, Kenya, Uganda, Rwanda, Burundi,
Zambia, Zimbabwe, Malawi, Mozambique, Madagascar,
Mauritius, Comoros, Seychelles, Mayotte and Reunion
(Fig. 1). East Africa is characterized by very huge human,
domestic and wild animal populations and agriculture is the
backbone of the economy of the countries hence the daily
activities of the farming community favours close contact
between humans, animals and their ectoparasites (Cutler et
al., 2010a). The region is known for diverse vegetation and
climates very suitable and conducive for reproduction and
maintenance of all stages of arthropod vectors. Despite these
global facts, little information on vector-borne zoonotic
bacterial diseases is emanating in general from Africa and in
particular very scanty information from east African
30
countries (Mediannikov et al., 2013b; Parola, 2011).
Moreover, poor facility of health institutions that usually
lack resources for rigorous diagnostic tests to determine the
underlying cause of febrile illnesses is additional problem.
As a result misdiagnoses and mistreatment of common
diseases in east Africa is common due to great confusion
between malaria and malaria mimicking treatable vector-
borne bacterial infections that causes febrile illnesses in
infected patients (Mediannikov et al., 2013b). Recently,
there is growing number of studies reporting death of
humans in east African countries from vector-borne zoonotic
bacterial diseases due to misdiagnosis and mistreatment for
malaria and other mimicking diseases (Rutherford et al.,
2004). Most vector-borne zoonotic bacterial diseases are
commonly manifested by nonspecific febrile illnesses that
are difficult to diagnosis clinically, thus their clinical signs
and symptoms overlap with those of widely distributed
diseases such as malaria in east Africa, which may lead to
inaccurate diagnoses (Mediannikov et al., 2013b; Parola and
Raoult, 2001; Raoult and Roux, 1999). The majority of
recent research and reviews on vector-borne zoonotic
bacterial diseases focus and concentrate on USA, Europe,
and Asian countries, south, west and North Africa (Reif and
31
Macaluso, 2009). However, comparatively little attention has
been given to arthropod transmitted zoonotic bacteria in east
African countries (Cutler, 2010). Thus, the main objective of
this review is to update our present understanding on the
geographical distribution, epidemiology, clinical aspects,
public health issues, treatment and to show research gaps on
arthropod transmitted zoonotic bacterial diseases in east
African countries.
32
2.0. MAJOR ARTHROPOD VECTORS OF
ZOONOTIC BACTERIA IN EAST AFRICA
33
Fenollar, 2014). Close contact with domestic animals and
their ectoparasites is reported as the important risk of
infection with many vector-borne zoonotic bacteria for
humans (Angelakis and Raoult, 2010). The most common
arthropod vectors known to transmit zoonotic bacteria to
humans in east African countries are briefly described as
follows:
2.1. Ticks
Ticks are one of the most important arthropods responsible
for health problems in both humans and animals (Nicholson
et al., 2010; Parola and Raoult, 2001). They are vectors of
several agents of emerging and re-emerging infectious
diseases of medical and veterinary importance (Baneth,
2014). Ticks are second only after mosquitoes as vectors of
human infectious diseases (Jongejan and Uilenberg, 2004;
Parola and Raoult, 2001). Ticks are vectors and reservoirs of
zoonotic bacteria such as Borrelia spp., spotted fever and
typhus group Rickettsia spp., Ehrlichia spp., Coxiella
burnetii, Tularaemia spp. and Bartonella spp. which cause
emerging zoonoses in humans (Irwin and Jefferies, 2004;
Otranto et al., 2009; Pfaffle et al., 2013). Recent reports
show the growing health risks associated with ticks and tick-
34
borne diseases in both humans and animals (Dantas-Torres et
al., 2012; De la Fuente and Estrada-Pena, 2012; Piesman and
Eisen, 2008). To date, almost 900 tick species have been
described. These are divided into three families: the
Argasidae or soft ticks (191 species), the Ixodidae or hard
ticks (701 species) and the Nuttalliellidae consisting of only
one species, Nuttalliella namaqua (Barker and Murrell,
2004). The generic taxonomy of argasid ticks is currently
highly controversial with different authorities listing between
four and 10 valid genera. The genera Argas, Ornithodoros
and Carios, if recognized, are all of medical importance for
humans. For example, various relapsing fever Borrelia spp.
are transmitted by a variety of Ornithodoros spp. (Estrada-
Pena and Jongejan, 1999; Mediannikov and Fenollar, 2014).
The Ixodidae consist of 13 current genera of which the most
important for humans are Amblyomma, Dermacentor,
Haemaphysalis, Hyalomma, Ixodes and Rhipicephalus
(Estrada-Pena and Jongejan, 1999; Parola and Raoult, 2001).
Ticks can be identified to the family, genus and species level
with use of numerous taxonomic keys that are available from
different regions of the world (Hoogstraal, 1956; Walker,
2003a). However, identification of larvae and nymphs of tick
is difficult and may require entomological skills. Molecular
35
methods such a mitochondrial markers (mitochondrial 12S
and 16S ribosomal DNAs, cytochrome oxidase subunit 1
[COX1], cytochrome b) are usually easy to amplify and
sequence, but the degree of intraspecies difference is rarely
enough to distinguish the phylogenetically close species but
are important to improve identification of ticks (Beati and
Keirans, 2001; Beati et al., 2012; Mediannikov and Fenollar,
2014). Nuclear markers (18S, 28S, internal transcribed
spacers 1 and 2) have also been used, but only entire
mitochondrial genome seems to be enough for phylogeneitc
studies (Burger et al., 2014). In addition to the technical and
logistical drawbacks of PCR assays, this approach is further
limited by the availability of gene sequences in genetic
databases. However, currently there is no PCR assay that can
distinguish tick species, and ideal PCR primer pairs that can
amplify the relevant gene fragments are not available but are
expected to be used more widely in the future particularly for
the differentiation of closely related species. Protein
profiling by matrix-assisted laser desorption ionization–time
of flight mass spectrometry (MALDI-TOF MS) is now
commonly used as routine identification method of
microorganisms in clinical microbiology (Karger et al.,
2012). There is growing reports on the use of MALDI-TOF
36
MS for the differentiation of arthropods from different
countries of the world. Recently, MALDI-TOF MS study has
been used to identify ticks in Germany (Karger et al., 2012)
and France (Yssouf et al., 2013) and is considered as
promising alternative method for the identification of ticks.
Even though molecular and MALDI-TOF MS studies on
ticks of east African countries are not available, the species
of ticks reported by using taxonomic keys and their
associated zoonotic bacterial diseases is compiled on table 1.
2.2. Lice
Lice are also one of the very important ectoparasites of
humans and animals worldwide. Lice are small, wingless
insects, obligate permanent ectoparasites and can‟t survive
more than few days off their hosts and are highly host
specific (Barker, 1994). Lice belong to the order Phthiraptera
which is divided into two major groups: Anoplura (the
hematophagous sucking lice of placental mammals) and
Mallophaga (the chewing or biting lice of birds, marsupials
and placental mammals) (Raoult and Roux, 1999). Although,
the Mallophaga group is likely paraphyletic, it is widely
agreed that sucking and chewing lice originated from a
common non-parasitic ancestral group closely related to the
order Psocoptera (book lice and bark lice) (Barker, 1994).
37
Humans can be infested with two types of lice (Anoplura):
Pediculus on the head (Pediculus capitis) and/or body
(Pediculus humanus) and Pthirus (in the pubic area Pthirus
pubis) (Raoult and Roux, 1999). Both anopluran and
mallophgan lice infest domestic farm animals and
companion animals; however, birds are infested only by
mallophagan lice species. Pediculus humanus (human body
louse) is the well established vector and known to transmit
Rickettsia prowazekii, Borrelia recurrentis, and Bartonella
quintana the causative agents of epidemic typhus, relapsing
fever, and trench fever, respectively (Badiaga and Brouqui,
2012). Recently Borrelia recurrentis and Bartonella
quintana are reported from Pediculus capitis in some
countries of the world including east African countries
(Badiaga and Brouqui, 2012; Boutellis et al., 2013).
Furthermore, Acinetobacter baumannii DNA have been
reported from both head and body lice in east African
countries (Badiaga and Brouqui, 2012; Gillespie et al.,
2009).
In domestic animals pediculosis (dermatitis due to lice) is
more common in cattle than any other species of domestic
animals (Kumsa et al., 2012a; Kumsa et al., 2012b). For
instance, Bovicola (Damalina) bovis) (chewing lice),
38
Linognathus vituli (the long-nosed cattle louse), Solenopotes
capillatus (the little blue cattle louse), Haematopinus
eurysternus (the short-nosed cattle louse), Haematopinus
quadripertusus (the cattle tail louse in tropics) and
Haematopinus tuberculatus (the buffalo louse) are the
species of lice known to infest cattle in different countries of
the world (Kumsa et al., 2012b). Likewise, 4 species of lice
on sheep (Kumsa et al., 2012a), 3 species on goats, 2 species
on equines, 1 species on pigs, 3 species in dogs and 1 species
on cats are documented on domestic animals around the
world (Kumsa et al., 2012a; Kumsa et al., 2012b). There is
little information on the role of lice of domestic animals as
vectors of zoonotic bacterial diseases from east African
countries. The only available information on vector-borne
bacteria in lice of domestic animals is the recent report of
Acinetobacter species in lice of cattle, sheep and dogs in
Ethiopia (Kumsa et al., 2012b). Details on the common
species of lice on humans, domestic animals and the
associated zoonotic bacteria is presented on table 2.
39
2.3. Fleas
Fleas (order Siphonaptera) are holometabolous obligate
blood-feeding ectoparasites of birds and mammals; the
immature stages (egg, larva and pupa) are found in burrows
or nests, whereas adult fleas are usually found on animal
hosts (Dobler and Pfeffer, 2011). Siphonaptera (fleas)
currently comprising 246 genera and approximately 2575
described taxa. Fleas are laterally compressed, wingless
insects that range from 1 to 10 mm in length (Traversa,
2013). They usually have small and shield shaped head.
Fleas are found in all continents including Antarctica and
they inhabit wide range of habitats and hosts, from equatorial
deserts through tropical rainforests to the arctic tundra
(Traversa, 2013).
Fleas are important vectors and reservoirs of several
emerging or re-emerging infectious diseases of great medical
and veterinary importance around the world (Dobler and
Pfeffer, 2011). The discovery that fleas were capable of
transmitting Yersinia pestis, the causative agent of plague
and later Rickettsia typhi, the causative agent of murine
typhus, stimulated a frenzy of flea studies in the early 20th
century (Eisen et al., 2012). Fleas have played a historic role
in human plagues, including the „Black Death‟ (bubonic
40
plague), which is transmitted by rodent fleas and is estimated
to have caused the deaths of a third of the world‟s population
during the Middle Ages (Eisen and Gage, 2012). Also
Ctenocephalides felis is a known vector for Bartonella
henselae, Bartonella clarridgeiae, and Rickettsia felis which
cause cat scratch disease, endocarditis and cat flea typhus in
humans, respectively. Xenosylla cheopis is the principal
vector of Rickettsia typhi the causative agent of murine
typhus and Yersinia pestis the causative agent of plague in
humans (Eisen and Gage, 2012).
Fleas are normally considered as ectoparasites of dogs and
cats, however, they infest and readily feed on other species
of domestic animals and humans (Dobler and Pfeffer, 2011).
Ctenocephalides felis felis (cat flea), Ctenocephalides canis
(dog flea), Pulex irritans (human flea), Echidnophaga
gallinacea (sticktight poultry flea) and Xenopsylla cheopis
(rat flea) are the most commonly reported species on dogs
and cats worldwide (Dobler and Pfeffer, 2011). Tunga
penetrans (called jigger or sand flea) is the species found
commonly in humans and also reported on dogs in all east
African countries (Franck et al., 2003). The females of the T.
penetrans penetrate the skin of the host to the basal layer of
the corium where they feed on blood and tissue exudates
41
produced by the host‟s inflammatory response. Only female
T. penetrans fleas cause tungiasis. The infestation of the skin
may cause severe damage by inflammatory response or by
bacterial super-infection (Mazigo et al., 2012). Domestic
dogs and cats play a peculiar role as bridging hosts for fleas
of different wild animals, domestic animals and humans, as
they will come into contact with different animals during
their hunting behaviour and therefore acquire the fleas of
different animals (Dobler and Pfeffer, 2011). Several flea
species are known to feed on humans and domestic animals
in east African countries (Kumsa and Mekonnen, 2011).
42
3. TICK-BORNE ZOONOTIC BACTERIA
43
(Parola et al., 2013; Raoult and Roux, 1997) as demonstrated
for some species. This favours transovarial and transstadial
transmission in some tick species and to transmit rickettsiae
to vertebrate hosts during blood feeding (Parola et al., 2013).
SFG rickettsiae are mainly associated with ticks, have an
optimal growth temperature of 32 ◦C, a guanosine plus
cytosine (G+C) content between 32% and 33%, can
polymerize actin and thereby enter the nuclei of host cells
and cause spotted fever in humans (Merhej and Raoult,
2011; Parola et al., 2013). In humans, spotted fever
rickettsiae induces symptoms that generally include fever,
headache, myalgia, rash, local lymphadenopathy, and an
eschar at the site of the tick bite, which might be useful in
diagnosis (Raoult and Roux, 1997).
The public health significance of rickettsial diseases among
indigenous east African populations is generally
underdiagnosed and frequently confused with malaria
(Mediannikov et al., 2013b; Parola, 2011). For instance,
infection with Rickettsia africae, the most common SFG
rickettsae in sub-Saharan Africa, is usually detected in
European and North American travellers returning home
from Africa than native nations (Parola et al., 2013). From
east African countries so far six Rickettsia species
44
pathogenic for human have been reported in ticks and /or
humans presented as follows.
45
R. africae has been detected in Rhipicephalus species ticks
but usually with very lower prevalence (Parola et al., 2013;
Raoult and Roux, 1997).
In east African countries R. africae has been detected in
5 Amblyomma species including in 15.8% Am. variegatum
(Macaluso et al., 2003) and in Am. variegatum,
Am. hebraeum and Am. gemma ticks (Mutai et al., 2013) in
Kenya, in 67% (26/39) (Lorusso et al., 2013) and 97.1%
(Nakao et al., 2013) Am. variegatum in Uganda, 1/13 Am.
variegatum Burundi (Parola et al., 2001), in 100% (10/10)
Am. variegatum collected from cattle in Juba in South Sudan
(Morita et al., 2004), in Amblyomma lepidum ticks (20%) in
Djibouti, in 10.17% Am . lepidum and ½ Am. variegatum
(Mura et al., 2008) and in 30.2% (16/53) Am. variegatum,
28.6% (12/42) Am. gemma , 0.8% (1/119) Am. cohaerens in
Ethiopia (Kumsa et al., 2014b) and in 90% (60/67) of
Amblyomma variegatum in the Union of the Comoros
(Yssouf et al., 2014). R. africae has also been detected in
Rhipicephalus species ticks with low prevalence in some east
African counties: in 0.7% (1/139) Rh. (Bo.) decoloratus and
25% (1/4) nymphs of Rh. (Bo.) decoloratus in Ethiopia
(Kumsa et al., 2014b) and in Rh. appendiculatus,
Rh. pulchellus and Rh. annulatus in Kenya (Macaluso et al.,
46
2003; Mutai et al., 2013) and in 1% (1/92) Rhipicephalus
appendiculatus and in 2.7% (8/296) of Rhipicephalus
(Boophilus) microplus in the Union of the Comoros (Yssouf
et al., 2014). However, the role of Rhipicephalus species
ticks in the epidemiology of ATBF is not yet determined and
it is suggested that the lower prevalence reflects the bacteria
is acquired during cofeeding with Amblyomma ticks or from
bacteraemic animals (Parola et al., 2013).
In humans SFG Rickettsia spp. was confirmed in 36/450
(8%) in hospitalized febrile patients in Tanzania (Prabhu et
al., 2011) and 2.9% (3/102) SFG Rickettsia DNA was
detected in dried thin blood smears prepared to test malaria
in children with febrile illnesses at Soddo Christian Hospital
in Wolaitta Soddo in Ethiopia (Aarsland et al., 2012), are the
available reports from east African countries. Despite the
very high abundance and prevalence of Amblyomma spp. and
other ticks and the highly suitable climates in east African
countries little information is available on R. africae
infection in humans (Parola and Raoult, 2001). However,
comparatively greater numbers of studies have been made in
west, central, south and North African countries on
R. africae in both ticks and humans than in eastern African
countries due to the involvement and construction of modern
47
laboratories by many European countries like France, UK
and Germany (Parola et al., 2013).
48
R. conorii was reported (Okabayashi et al., 1999). In human
patients the typical form of the disease is manifested by
clinical symptoms like high fever, flulike symptoms, a local
necrotic inflammation with a black crust called an eschar
(the “tache noire”) at the tick bite site and a maculopapular
rash (Parola et al., 2005a).
49
2005a). The first disease in human was described in two
patients returning from Morocco to France showing MSF-
like illness. Hyalomma spp. ticks are considered as the main
vectors and reservoirs and transmit both transstadially and
transovarially to their offspring. It has been detected in
Hy. marginatum rufipes, Hy. marginatum marginatum,
Hy. aegyptium, Hy. dromedarii Hy. impeltatum and
Hy. truncatum mostly in north and western African countries
in Niger, Mali, Algeria, Egypt and Senegal (Parola et al.,
2013). To date, this bacterium has been detected in 8 sub-
Saharan African countries. In east African countries
R. aeschlimannii has been detected in Hyalomma spp. in
Kenya (Mutai et al., 2013), Sudan (Morita et al., 2004),
Zimbabwe and in 45.4% (5/11) Hyalomma marginatum
rufipes and 2.2% (1/46) Hy. truncatum in Ethiopia (Kumsa
et al., 2014). So far R. aeschlimannii has not reported in
humans from east African countries.
50
usually associated with the hard ticks of the genus
Rhipicephalus ticks (Rhipicephalus mushamae,
Rhipicephalus lunulatus, Rhipicephalus sulcatus and
Rhipicephalus guilhoni) mainly in the central, western and
north African countries in Central African Republic, Mali,
Algeria, Morocco. The first human case of infection with
R. massiliae was confirmed in 2005, 20 years after the
isolate was obtained from a 45-year-old male human patient
in Palermo (Italy) in 1985 (Vitale et al., 2006). R. massiliae
has been detected mainly in North and West African
countries predominantly in Rhipicephalus spp. including
8.1% (2/37) in Rh. muhsamae in Mali, 8.2% (5/61) in
Rh. senegalensis in Guinea and in 22.4% (11/49) in Rh.
guilhoni in Senegal (Parola et al., 2013). In east African
countries R. massiliae has been detected in Hy. truncatum in
Kenya (Mutai et al., 2013) and in 1.9% (1/52)
Rh. praetextatus in Ethiopia (Kumsa et al., 2014b). There is
no report on R. massiliae from humans in east African
countries.
51
subspecies of R. sibirica first isolated in Beijing in China
from Hyalomma asiaticum ticks collected in Mongolia in
1991 (Parola et al., 2005b). R. sibirica mongolitimonae was
detected in Hy. truncatum (Parola et al., 2001). The first
human case of this Rickettsia was reported in South Africa
after PCR amplification from a biopsied eschar. Human
patients develop lymphangitis, fever, maculopapular rash,
engorged lymph nodes, and multiple eschars. R. sibirica
subsp. mongolitimonae has been detected in 14% of
H. truncatum ticks in Senegal. The only report in east
African countries is the recent detection in Am. gemma ticks
in Kenya (Mutai et al., 2013). There is no report on
R. sibirica subsp. mongolitimonae from humans from in east
African countries.
52
Likewise, a similar species has been detected in European
hard tick Haemaphysalis sulcata for the name Rickettsia
hoogstraalii was proposed. Phylogenetic studies of this
rickettsia clusters within the spotted fever group of
rickettsiae at the periphery of this group at location close to
R. akari and R. felis (Cutler et al., 2012).
53
from cattle and other domestic animals from different
districts in Ethiopia.
On the other hand, investigators in Mexico (Medina-Sanchez
et al., 2005) detected R. prowazekii in Amblyomma ticks and
obtained an isolate of R. prowazekii from one of the ticks.
This finding again raises a question on natural reservoir of
epidemic typhus rickettsiae involving ixodid ticks and
challenges our current understanding that the human latent
infection-reactivation-louse-human cycle or the flying
squirrel-louse and/or flea-flying squirrel cycle is generally
believed to play important role in the origin and evolution of
R. prowazekii.
54
The tick vector, Ornithodoros moubata, feeds for a short
time only (usually less than half an hour), then returns to the
earth floor or walls of the house. Humans are believed to be
the only natural reservoir for B. duttonii, unlike the situation
for B. crocidurae in West Africa, where rodents are
reservoirs.
Recently, B. duttonii has been detected in the peripheral
blood collected from both chickens and pigs in Tanzania that
implicate the possibility of zoonotic importance of this
Borrelia species (Cutler et al., 2010b). In human patients it
causes symptoms like fever accompanied in more than 90%
of cases by tachycardia, headache, myalgia, arthralgia,
conjunctivitis, hepatomegaly and splenomegaly, along with
orange urine in a few cases (Cutler et al., 2006).
Recently, a new Borrelia species distant from both relapsing
fever group and Lyme borreliae was detected in 7.3% of the
Amblyomma cohaerens collected from cattle in southwest
parts of Ethiopia (Mediannikov et al., 2013a). Similarly, new
Borrelia species has been recently detected in 8/119 (6.7%)
Amblyomma cohaerens, 1/42 (2.4%) Amblyomma gemma,
3/53(5.7%) Ammblyomma variegatum, 5/22(22.7%)
Amblyomma larvae and 3/60(5%) Amblyomma nymphs
(Kumsa et al., 2014b) that cluster at intermediate position in
55
between the recurrent fever and the Lyme disease borreliae
group using 16S genus-specific qPCR. These Amblyomma
ticks were collected from domestic animals in different parts
of Ethiopia. However, the pathogenecity for humans of these
newly detected Borrelia species in Amblyomma ticks in
Ethiopia remains to be determined in the future.
3.4. Bartonellosis
Bartonella species are intracellular arthropod-borne small
hemotropic facultative Gram-negative bacilli or coccobacilli
belonging to the alpha 2 subgroup of Proteobacteria, family
Bartonellaceae (Billeter et al., 2008). Bartonella are aerobic
bacilli, oxidative negative, fastidious, slow growing and
highly adapted to their vertebrate hosts (Anderson and
Neuman, 1997; Breitschwerdt and Kordick, 2000; Guptill,
2010). The bacteria infect the erythrocytes and endothelial
cells of their mammalian hosts. In humans Bartonella spp.
are implicated in causing several emerging and remerging
diseases manifested by a wide variety of clinical signs
(Anderson and Neuman, 1997; Breitschwerdt and Kordick,
2000).
Many different species of haematophagous arthropods
including fleas, lice, sandflies and ticks are implicated to
56
mediate the transmission of Bartonella spp. between
reservoir hosts and humans (Anderson and Neuman, 1997;
Angelakis and Raoult, 2014). To date, the genus Bartonella
comprises more than 30 different species and subspecies that
infect a wide variety of mammalian hosts. In several
European countries, USA and Asia a number of tick species
(Ixodes spp., Dermacentor spp., Haemaphysalis spp. and
Rhipicephalus sanguineus,) have been shown to be positive
for Bartonella spp. based mainly on PCR and very rarely by
culture (Angelakis et al., 2010; Angelakis and Raoult, 2014).
On the contrary, few previous studies made to detect
Bartonella spp. in ixodid ticks from African countries
reported the absence of these pathogens (Kumsa et al.,
2014b; Mediannikov et al., 2013a). Review of the recent
literature also shows the presence of contradicting views on
the transmission of Bartonella spp. by ticks. For instance
Telford III and Wormser (Telford, III and Wormser, 2010)
concluded that tick transmission of any Bartonella spp. to
either animals or humans has not been established, however,
Angelakis et al. (Angelakis et al., 2010; Cotte et al., 2008;
Reis et al., 2011) (2010) reported the vector competence and
transmission of Bartonella species by different tick species.
As most Bartonella species are transmitted by lice and fleas
57
brief descriptions of some species are given on the respective
sections in this review.
3.5. Q fever
Coxiella burnetii, the causative agents of Q fever, is obligate
intracellular, small pleomorphic, nonmotile, rod (0.2–0.4mm
wide, 0.4–1.0mm long), Gram-negative bacteria belonging to
the genus Coxiella, the family Coxiellaceae, the order
Legionellales, the class Gammaproteobacteria, and the
phylum Proteobacteria.icks are one of the most important
arthropods known to act as reservoirs of C. burnetii. So far
C. burnetii was reported in more than 40 different tick
species in 12 genera in several countries of the world (Cutler
et al., 2007; Maurin and Raoult, 1999). Ticks acquire
infection during feeding on blood meal from infected
animals and can transmit to their offspring both
transovarially and transstadially (Arricau-Bouvery and
Rodolakis, 2005; Maurin and Raoult, 1999). Ticks play
important role in the epidemiology of Q fever by
contaminating the environment by excreting C. burnetii via
their faeces, saliva and coxal fluids. However, sufficient data
that shows transmission of C. burnetii to humans by blood
feeding ticks is not yet available (Porter et al., 2011).
58
In humans inhalation of aerosolized C. burnetii from the
contaminated environment is considered as the main means
of infection (Maurin and Raoult, 1999). In addition,
consumption of raw milk and milk-products, transplacental,
intradermal inoculation, blood transfusion, contact with
urine, faces and semen of infected animals (especially
ruminants) are also regarded as possible routes of infection
(Maurin and Raoult, 1999). Excretion of C. burnetii by
domestic ruminants is implicated as major cause of
environmental contamination and the source for human
infection. Infection in humans is often asymptomatic, but it
can manifest as an acute disease, such as pneumonia,
hepatitis or appear as self-limited flu-like syndrome (Cutler
et al., 2007). Q fever can also appear in a chronic form,
mainly involving endocarditis and vascular infection, as well
as hepatitis and chronic infection after pregnancy (Maurin
and Raoult, 1999). Generally the number of studies on
Q fever from African countries are lower than from the other
continents, in particular the number of reports on ticks,
animals and humans from east African countries are very
lower than North, south and western African countries
(Vanderburg et al., 2014).
59
From east African countries seroprevalence of antibodies to
C. burnetii in humans was reported in Tanzania (Prabhu et
al., 2011), in Zimbabwe (Kelly et al., 1993) , in Zambia
(Okabayashi et al., 1999), in Somalia (Botros et al., 1995)
and in Kenya (Knobel et al., 2013). In domestic ruminants
seroprevalence of antibodies to C. burnetii was reported in
Malawi (Staley et al., 1989), Tanzania (Hummel, 1976),
Zimbabwe (Kelly et al., 1993), Zambia (Qiu et al., 2013),
Ethiopia (Gumi et al., 2013) and in Kenya (Knobel et al.,
2013). In east African countries C. burnetii was detected by
molecular method in different species of ticks in Kenya
Knobel et al. (Knobel et al., 2013) and in Ethiopia in 5.3%
Amblyomma variegatum and 10.8% Hyalomma truncatum
ticks (Philip et al., 1966a) and 28.6% (14/49) Amblyomma
gemma, 25% (31/124) Rhipicephalus pulchellus, 7.1 %
(1/14) Hyalomma marignatum rufipes, 3.2 % (2/62)
Amblyomma variegatum, 3.1% (4/128) Amblyomma
cohaerens, 1.6% (1/63) Rhipicephalus praetextatus and
0.6% (1/153) Rhipicephalus (Booophilus) decoloratus
(Kumsa et al., 2014). As it can clearly be seen on Fig 2 very
fewer studies were made in east African countries than in the
North, west, middle and South African countries on
C. burnetii infection in both in humans and animals.
60
4. LOUSE-BORNE ZOONOTIC BACTERIA
4.1. Borrelia recurrentis
Borrelia recurrentis, transmitted by Pediculus humanus
corporis (human body lice), is the causative agent of
epidemic relapsing fever, that affected several million people
worldwide during the first half of the 20th century,
particularly during the world wars (Badiaga and Brouqui,
2012). Humans are the only known reservoir and host of
B. recurrentis. It was initially described in Ireland and was
one of the first louse-borne infectious diseases identified by
microscopy. Lice are infected upon ingestion of infected
blood meal from human hosts (Badiaga and Brouqui, 2012).
Then B. recurrentis passes from the gut to the coelomic
cavity and multiplies in the hemolymph. B. recurrentis don‟t
disseminate in tissues and hence don‟t found in the saliva or
the feces of lice. In lice infection last for the rest of entire
lifespan but there is no transovarial transmission of
B. recurrentis hence can‟t transmit borreliae to its progeny.
In humans the only possibility of infection by B. recurrentis
is by contamination when louse is usually crushed between
nails of fingers that releases and spreads B. recurrentis
which is highly infective bacterium and capable of
61
penetrating intact human mucosa and skin (Badiaga and
Brouqui, 2012; Raoult and Roux, 1999).
In infected humans clinical manifestations start abruptly with
a high-grade fever, pain, anorexia, dry coughs, and fatigue.
Complications include skin and mucosal haemorrhaging;
neurological, liver and renal involvement; and spleenic
rupture. Jaundice is a diagnostic clue that suggests relapsing
fever among louse-borne diseases (Raoult and Roux, 1999).
The disease is predominantly characterized by a series of
relapses that are less severe and shorter in duration, and
occur at intervals of 7 or 10 days. Common diagnostic
methods include the detection of Borrelia in Giemsa-stained
blood films and PCR assays (Brouqui et al., 2004; Raoult
and Roux, 1999). The death rate varies from 10% to 40% in
untreated patients, and from 2% to 4% in patients treated
with antibiotics such as doxycycline or erythromycin.
Louse-borne relapsing fever caused by B. recurrentis has
disappeared from many regions around the world (Raoult
and Roux, 1999). On the contrary this disease is still a major
public health problem in populations living in poor-hygiene
conditions where body louse infestations are prevalent in
east African countries including Ethiopia (Almaviva et al.,
1993; Borgnolo et al., 1993; Yimer et al., 2014a; Yimer et
62
al., 2014b) and the neighbouring countries like Sudan (de et
al., 1995), Eritrea, and Somalia. In the highland parts of
Ethiopia louse-borne relapsing fever is listed as the seventh
most common reason for hospital admission (2.5% of total
3777 cases) and the fifth most common cause of death (0.9%
of the total 42 cases and is associated with substantial illness
and death (Yimer et al., 2014a; Yimer et al., 2014b).
Recently, B. recurrentis DNA was detected in 23% of head
lice from patients with louse-borne relapsing fever in
Ethiopia, however, the role of head lice in the transmission
of this bacterium was not determined (Boutellis et al., 2013).
63
especially in Russia and on the Eastern, Central and Western
European fronts (Badiaga and Brouqui, 2012). The incidence
of trench fever dramatically decreased after World War II.
However, in the early 1990s, B. quintana was recognized as
a major re-emerging infectious disease in urban homeless
populations of developed countries in both Europe and the
United States who have poor living conditions (Raoult and
Roux, 1999). Even though, the natural reservoir of
B. quintana has not yet been fully established, infection with
B. quintana in lice lasts for the entire lifespan (Raoult and
Roux, 1997; Raoult and Roux, 1999). B. quintana multiplies
widely in louse intestines and passes with louse faeces.
Humans are infected by contamination of faeces containing
B. quintana from infected lice. B. quintana can survive for
long periods in the louse feces and can remain infectious for
up to 1 year (Badiaga and Brouqui, 2012; Raoult and Roux,
1999).
After an incubation period of 15 to 25 days infected persons
show signs like acute onset of a high-grade fever, headache,
dizziness, and characteristic pain in the legs. The first fever
episode can last for between 2 and 4 days. Occasionally, the
primary episode is followed by a relapse every 4-5 days
(Badiaga and Brouqui, 2012; Raoult and Roux, 1999). The
64
term “quintan fever” refers to the recurring 5-day attacks.
B. quintana causes trench fever, bacillary angiomatosis,
endocarditis, chronic bacteremia, and chronic
lymphadenopathy (Badiaga and Brouqui, 2012; Raoult and
Roux, 1999). Although trench fever often results in
prolonged disability, no deaths have been reported.
In east African countries B. quintana is commonly detected
in human body lice in Ethiopia, Zimbabwe, Burundi and
Rwanda (Fournier et al., 2002; Raoult et al., 1997; Roux and
Raoult, 1999). B. quintana was detected in head lice in
Ethiopia, however, the role of head lice was not determined
in the transmission of this bacterium (Sangare et al., 2014).
65
been suggested as a possible reservoir host of this bacterium.
Infected humans retain some rickettsiae for the rest of their
lives and under certain stressful conditions, they may relapse
in a relatively a milder bacteremic form illness called Brill–
Zinsser disease (Gillespie et al., 2009). These bacteraemic
humans are considered as the source of infection for lice and
to start a new epidemic. In untreated cases the mortality rate
of epidemic typhus varies from 0.7% to 60%. After an
incubation period of 10–14 days, patients usually experience
1–3 days of malaise associated with fever and multiple
painful symptoms, which are then followed by the
development of rashes (20–40% of patients), neurological
manifestations (80%), respiratory manifestations (38–70%),
and shock (7%) (Raoult and Roux, 1999).
Lice are infected upon ingestion of blood meal containing
rickettsiae from infected persons and then the bacteria infect
the midgut epithelial cells of the louse and multiply rapidly
(Badiaga and Brouqui, 2012; Raoult and Roux, 1999).
Excessive multiplication and growth of R. prowazekii causes
enlargement and eventually bursting of infected epithelial
cells releasing the rickettsiae into the gut lumen. Very large
quantities of rickettsiae are discharged with the feces of lice
and can remain infective for up to 100 days (Badiaga and
66
Brouqui, 2012). All stages of the louse life cycle can acquire
infection. Infection with R. prowazekii causes death of the
louse within 1 week due to the ruptured epithelial cells are
not replaced and this causes the blood to pass through the
intestine and the louse becomes red. Infected red lice die
shortly thereafter thus giving the named “red louse disease
Typhus (Raoult and Roux, 1999).
Ethiopia and Burundi are considered the endemic foci
countries of this disease in Africa and in east African
countries R. prowazekii was detected in human body lice in
Ethiopia, Burundi and Rwanda (Fournier et al., 2002; Raoult
et al., 1997).
67
associated with severe community-acquired infections. They
have the capacity to survive for prolonged periods under a
wide range of environmental conditions. Acinetobacter
baumannii is described as the most common species of the
genus frequently associated with outbreaks and has been
repeatedly reported to develop a high level of resistance
against all of the available classes of antimicrobial drugs in
many parts of the world (Badiaga and Brouqui, 2012). In
east African countries Acinetobacter baumannii was detected
in 54 head (47%) and 77 body (71%) lice in Ethiopia
(Kempf et al., 2012). Although A. baumannii infections are
not confirmed to be transmitted by body lice, the role of
human lice in the transmission of this bacterium remains to
be seen in the future. Furthermore, other Acinetobacter
species such as A. pitti in Heterodoxus spiniger of dogs and
A. soli DNA in Linognathus vituli of cattle were detected in
animal lice in Ethiopia (Kumsa et al., 2012b). A. pittii is
reported as the second most commonly isolated
Acinetobacter sp. after A. baumannii in human patients
around the world. More recently other less known species
such as A. soli have been regularly associated with serious
infections and considered as emerging pathogens. For
instance A. soli outbreak as a cause of infection in neonatal
68
intensive health care unit in Korea and nosocomial
bloodstream infection due to A. pittii in United States were
reported (Kumsa et al., 2012b).
69
vomiting, and diarrhoea, as well as solitary, black crusted
skin lesions, muscle pain, in some cases local
lymphadenopathy and a characteristic inoculation eschar at
the site of the flea bite (Parola, 2011). Thus infection by
R. felis may be misdiagnosed due to similar clinical signs
with other arthropod-borne rickettsios such as murine typhus
or some members of the SFG, as well as other infectious
diseases like dengue, malaria, brucellosis, leptospirosis, or
even other clinical conditions like Kawasaki disease
especially in countries with poor diagnostic facilities (Parola,
2011). Hence only advanced molecular techniques can
currently distinguish the infections which are very rare
methods of diagnosis in Africa in general and in particular in
east Africa (Brouqui et al., 2004).
Although C. felis is considered to be the only known
biological vector of R. felis, R. felis has recently been
detected in 12 species of fleas (C. canis, C. orientis,
Anomiopsyllus nudata, Archaeopsylla erinacei,
Ctenophthalmus sp., Xenopsylla cheopis X. rasilliensis,
Tunga penetrans, Ceratophyllus gallinae, Spilospsyllus
cuniculi and Echidnophaga gallinacean); 8 species of ticks
(Haemaphysalis flava, Rhipicephalus sanguineus, Ixodes
ovatus, and Carios capensis); and 3 species of mites (chigger
70
mites in South Korea and mesostigmata mites in Taiwan) in
different countries around the world (Eisen and Gage, 2012;
Hirunkanokpun et al., 2011; Parola, 2011; Reif and
Macaluso, 2009). The reported hosts for these vectors were
mainly cats, dogs and rodents, and more rarely opossums,
hedgehogs, horses, sheep, goats, gerbils and monkeys
(Dobler and Pfeffer, 2011; Eisen and Gage, 2012; Reif and
Macaluso, 2009). Even though, few confirmed human cases
have been described infection by R. felis is considered as
worldwide disease (Parola, 2011). It is more common in
tropical countries with warm climatic conditions.
In east African countries R. felis has recently been reported
in C. felis and P. irritans specimens collected from human
dwellings in the southwestern regions of the country in
Ethiopia, Rickettsia felis was detected in 21% of C. felis,
11% P. irritans and 6% C. canis collected from dogs and
cats in Ethiopia (Kumsa et al., 2013). In Kenya R. felis has
been reported in C. felis, C. canis, and P. irritans (Jiang et
al., 2013).
Human infection with R. felis in east African country has
been reported only in Kenyan in 3.4% of the afebrile patients
in western Kenya were found to be positive for R. felis using
molecular methods (Maina et al., 2012) and also Recently,
71
R. felis was detected in patients suffering from fever of
unknown cause with a prevalence of 3.7-7.2% in Kenya
(Richards et al., 2010). In eastern Africa countries R. felis is
considered as the second cause of fever in rural areas after
malaria (Mediannikov et al., 2013b). However, in many
eastern African countries studies on R. felis have not yet
been conducted and some authors argue that R. felis infection
should be suspected as one of the causes of fevers of
unknown etiologic agent in malaria endemic east African
countries (Mediannikov et al., 2013b; Richards et al., 2010).
72
infection is intimately associated with the introduction of
commensal rodents, such as R. rattus, R. norvegicus, and
Mus musculus, and their ectoparasites, particularly fleas
(Azad, 1990). R. typhi is transmitted to reservoir host rats
and humans by the contamination of the bite site or skin
abrasions with Rickettsia-containing flea feces during or
after blood feeding, as well as via the flea bite. R. typhi do
not have any harmful effects on the health and activity of
either the Xenopsylla cheopis vector or the reservoir host
rats. R. typhi infects endothelial cells in mammalian hosts
and midgut epithelial cells in the flea host (Azad, 1990;
Bitam, 2012; Bitam et al., 2010; Eisen and Gage, 2012).
R. typhi is transferred from a rodent reservoir by an
arthropod (often X. cheopis) to humans. After incubation
periods of six to 14 days it causes a relatively mild febrile
illness and clinical sings similar to caused by other infectious
diseases and thus cases may be overlooked without a
laboratory confirmed diagnosis (Azad, 1990; Eisen and
Gage, 2012). The most common clinical manifestations are
high fever, severe headache, chills, myalgia, weakness, and
nausea (Civen and Ngo, 2008). Murine typhus patients
usually present with undifferentiated febrile illness. The
pathognomonic rash is described as macular (49%),
73
maculopapular (29%), papular (14%), petechial (6%), and
morbiliform (3%), usually centrally distributed on the trunk,
but also found on the extremities (Azad, 1990; Eisen and
Gage, 2012). Physicians practicing in or near R. typhi-
endemic areas need to consider murine typhus in the
differential diagnosis of a febrile illness of unknown
aetiology (Azad, 1990; Eisen and Gage, 2012).
In east African countries R. typhi has been reported in 62%
of R. rattus collected from buildings in Addis Ababa in
Ethiopia (Azad and Beard, 1998) and is also endemic in
other east African countries like Kenya, Uganda, Tanzania,
Madagascar, Malawi, Zambia, Zimbabwe and Mozambique
(Azad, 1990; Azad and Beard, 1998; Azad et al., 1990).
In humans an overall seroprevalence of 8% by (Prabhu et al.,
2011) and 9.3% by (Dill et al., 2013) of typhus group
rickettsiae was reported in Tanzania and also 5.0%
seroprevalence of antibodies against R. typhi was reported in
Zamia (Okabayashi et al., 1999).
74
(McElroy et al., 2010; Mosbacher et al., 2010). B. henselae
belongs to the genus Bartonella which are hemotropic
facultative Gram-negative intracellular bacteria that infects
primarily red blood cells but can also be found associated
with host endothelial, dendritic, and CD34+ cells, which
include lymphopoietic stem and progenitor cells, small-
vessel endothelial cells, and embryonic fibroblasts (McElroy
et al., 2010). Ctenocephalides felis (cat flea) is reported as
the primary arthropod vector for B. henselae (Breitschwerdt
et al., 2010; Chomel and Kasten, 2010). Fleas are infected up
on feeding blood meal during the periods of bacteraemia
when the bacteria circulate in the blood of an infected animal
and then infected fleas shed infectious organisms in their
feces (Eisen and Gage, 2012; McElroy et al., 2010). The
domestic cat is the primary animal reservoir of B. henselae
and transmit the pathogen to humans via scratches or bites,
though naturally infected cats are asymptomatic (Dobler and
Pfeffer, 2011; Eisen and Gage, 2012). Previous studies show
that B. hensealae is more common in stray cats than pet cats
and also cats younger than one year old are important
sources of the bacterium (Eisen and Gage, 2012). In
addition, dogs and other animals have also been implicated
as sources of human infection (Chomel et al., 2006; Dobler
75
and Pfeffer, 2011). B. henselae is transmitted between
animals, and from animals to humans, by inoculation with
contaminated flea feces on an open wound, such as a scratch
or bite (Eisen and Gage, 2012).
In infected humans the main clinical symptoms of cat scratch
disease are benign lymphadenopathy, low-grade fever,
malaise and aching, headache, anorexia, and splenomegaly
(Breitschwerdt et al., 2010; Chomel et al., 2006; Chomel and
Kasten, 2010; Eisen and Gage, 2012). Batonella spp. are
more frequently associated with several symptoms
particularity in ocular infections and endocarditis
(Breitschwerdt et al., 2010; Chomel et al., 2006; Chomel and
Kasten, 2010). In immunocompromized humans including
HIV-positive persons B. henselae typically causes systemic
bartonellosis which manifests as bacillary angiomatosis
(tumor-like growth of blood vessels on the external skin
surface) and bacillary parenchymous peliosis (hepatitis).
These syndromes are uncommon in immunocompetent
individuals (Breitschwerdt et al., 2010; Chomel et al., 2006;
Eisen and Gage, 2012).
Among the very few reports available from eastern African
countries include an overall percentage of 6% (2/34)
B. henselae DNA in C. felis collected from cats in Oromia
76
(Kumsa et al., 2013), Ethiopia and 11% (5/46) prevalence of
antibodies against Bartonella spp. in cats from Addis Ababa
and Bartonella spp. related to B. elizabethae in small
mammals in northern Ethiopia (Meheretu et al., 2013) and
Bartonella spp. in bats in Kenya are the only reports so far
made revealing huge information gap on bartonellosis from
East Africa (Kosoy et al., 2010).
77
Humans are infected with Y. pestis by direct transmission via
infectious droplets or by contact with contaminated fluid or
tissue or indirectly through flea bites by inoculation of the
bacteria with contaminated flea feces on an open wound
(Eisen and Gage, 2012; McElroy et al., 2010). The human
flea (Pulex irritans) is reported to play an important role in
spreading plague by human-to-human transmission (Eisen
and Gage, 2012). Humans are extremely susceptible to
Y. pestis infection and can be infected by bites of infected
fleas, direct contact with infected animal tissues, secretions
or respiratory droplets, by consumption of raw or
insufficiently cooked meat products, inhalation of
aerosolized respiratory excreta of animals or patients with
the pneumonic form of infection (Eisen and Gage, 2012;
Perry and Fetherston, 1997). Cats are highly susceptible to
plague and can transmit infection directly to humans. In
addition, cats can disseminate the bacteria by transporting
infected fleas or from their hunting behaviour rodent
carcasses into the residential environment (McElroy et al.,
2010).
Plague is a serious illness in humans that is likely to be fatal
if left untreated and occurs in the three major forms in
humans (McElroy et al., 2010). The first type is the bubonic
78
plague is the most common type and it accounts for 80–90%
of all reported plague cases worldwide (Eisen and Gage,
2012; McElroy et al., 2010; Perry and Fetherston, 1997).
Bubonic plague is transmitted through the bite of an infected
flea, or direct contact between infectious host body fluids
and open skin lesions or abrasions on human skin. Bubonic
plague is characterized by a painful, severely inflamed
lymph node (regional lymphadenopathy) known as a bubo
accompanied by fever, headache, chills, malaise and extreme
exhaustion. Illness usually begins 2–10 days following
exposure (Eisen and Gage, 2012; McElroy et al., 2010; Perry
and Fetherston, 1997). The second type is septicemic plague
which can be primary septicemic plague, which is often
attributed to cutaneous exposure to Y. pestis, is less common
and is characterized by fever and sepsis without regional
lymphadenopathy. Septicemic plague also occurs
secondarily to bubonic plague (Eisen and Gage, 2012;
McElroy et al., 2010; Perry and Fetherston, 1997). The third
type is the pneumonic form (primary or secondary to
bacteraemia) of plague which can result from untreated
bubonic plague infection can spread through the bloodstream
to the lungs, resulting in secondary pneumonic plague is
acquired through inhalation of plague bacteria contained in
79
respiratory droplets or other materials, is the least common,
but the most severe and rapidly progressing form (Eisen and
Gage, 2012; McElroy et al., 2010; Perry and Fetherston,
1997). Pneumonic plague can spread from person-to-person
through respiratory droplets (Eisen and Gage, 2012; Perry
and Fetherston, 1997). Persons infected through respiratory
exposure develop primary pneumonic plague. Both primary
and secondary pneumonic plague is fulminant illnesses, with
rapid progression to hypoxia, hemoptysis, and shock.
Without prompt appropriate treatment, pneumonic plague is
usually fatal within 3–4 days (Eisen and Gage, 2012;
McElroy et al., 2010; Perry and Fetherston, 1997).
Plague has recently been recognized as a re-emerging
disease of humans in East African countries in remote areas
where the disease is endemic and where persons live in
unsanitary environmental conditions in close contact with rat
infested with fleas in Kenya, Uganda, Tanzania (Laudisoit et
al., 2007), Madagascar (Ratovonjato et al., 2014), Malawi,
Zambia, Zimbabwe and Mozambique east African countries
(Neerinckx et al., 2010). In northwest Uganda, which has
had recent plague outbreaks, cat fleas (Ctenocephalides felis)
have been reported as the most common fleas in the home
environment, which is suspected to be a major exposure site
80
for human plague in this country (Amatre et al., 2009;
Neerinckx et al., 2010). Very recently, on 4 November 2014,
WHO has reported an outbreak of plague in 16 districts in
seven regions in Madagascar in which a total of 119 cases of
plague have been confirmed, including 40 deaths. About 2%
of reported cases are of the pneumonic form of plague
(http://www.who.int/csr/don/21-november-2014-plague/en/).
81
African countries there is no any information about this
disease (Bitam, 2012; Parola et al., 2013).
82
reviewed by Billeter et al. (Billeter et al., 2008) and Tsai et
al. (Tsai et al., 2011). B. melophagi is one of the ruminant
associated Bartonella spp. recently isolated from sheep
blood in the USA (Bemis and Kania, 2007) and the same
bacteria were isolated from the blood of two female patients
with pericarditis and skin lesions in the USA (Maggi et al.,
2009). Furthermore, Kosoy et al., (2007) have deposited the
rpoB gene sequence of B. melophagi amplified from
M. ovinus in GenBank (accession number: EF605288) as
unpublished data that can‟t be accessible using PubMed
search engine. Even though much information about this
bacterium is not available, recently Kumsa et al. (Kumsa et
al., 2014a) reported an overall prevalence of 88.6%
B. melophagi DNA in Melophagus ovinus (sheep ked)
collected from sheep in Ethiopia. This is the only report from
Africa and thus further research is needed on this bacterium.
7.2. Q fever
There are several reports on the presence of Coxiella burnetii
in biting flies from many countries around the world (Nelder
et al., 2008). However, there is no information on the
occurrence of this bacterium in flies in African countries.
Studies are required to clarify the presence or absence of this
83
bacterium in flies in Africa including east Africa which
inhabits several species of haemtophagous biting flies.
84
absence of this bacterium in mosquitoes in east Africa
countries.
85
8. Vector-borne bacterial diseases in travellers returning
from East Africa
The diverse geography, ecosystems and climate of Africa
attracts huge population of international travellers for
various reasons including as a tourist destination, for
international aid or voluntary work, research or education,
family or friend visit and for business travel. The number of
international travellers to sub-Saharan Africa had increased
by 4% (1.3 million travellers) in the year 2011 (Mendelson
et al., 2014).These travellers from different countries of the
world during their stay in Africa are exposed to various types
of endemic infectious diseases in Africa including vector-
borne bacteria (Walker, 2003b). In international travellers
systemic febrile illness, a common syndrome for vector-
borne bacteria, was reported as the leading symptoms in sub-
Saharan Africa (Hunziker et al., 2012; Mendelson et al.,
2014). For instance systemic febrile illness was diagnosed in
27% (1,474) of travellers returning from east African
countries to their home country (Mendelson et al., 2014).
To mention some of the individual reports diagnosed in
home returning travellers from east African countries
R. africae has been molecularly confirmed in a 62 year old
French man returning after 2 months stay in western part of
86
Ethiopia (Stephany et al., 2009) and also R. africae was
diagnosed in Italian, USA and Canadian citizens human
patients returning from Zimbabuwae, Tanzania and Kenya
respectively (Raoult et al., 2001b). A case of relapsing fever
borreliosis with cutaneous eschar and radiculopathy in a 77
year old French woman traveller returning from Ethiopia
transmitted was reported to be transmitted by hard ticks
using molecular tools (Socolovschi et al., 2012a). In another
situation a fatal case of spotted fever group rickettsial
infection was reported in a missionary woman from the
United States who lived in a rural town in the central district
of Kenya approximately 70 km north of Nairob who had a
history of tick exposure and the patient initially
misdiagnosed and treated for malaria and then for the second
time was treated with sulfa-containing antibiotic that is
believed to exacerbated the clinical severity of rickettsial
infections in this patient (Rutherford et al., 2004). In USA
African tick bite fever was confirmed in 48-year-old woman
who returned with influenza like illness and inoculation
eschar after 2 weeks of returning from a 13-day trip where
she had spent time at a game farm in Zimbabwe (Owen et
al., 2006). Also recently Murine typhus was confirmed using
molecular tools in returned two patients that were infected in
87
Madagascar (Walter et al., 2012). Spotted fever group
rickettsial infections was confirmed at the Centers for
Disease Control and Prevention in USA in returned travellers
from east African countries in Zimbabwe, Rwanda,
Mozambique, Zambia and Tanzania (McQuiston et al.,
2004). Of the total 197 SFG and 1 TG rickettsiosis, 5 acute
Q fever and 1 bartonellosis cases confirmed in returned
European travellers from Sub-Saharan Africa from east
African countries SFG rickettsioses was diagnosed in
patients who travelled to Zimbabwe (n = 13) and Tanzania (n
= 7) and 2 patients infected with acute Q fever in Tanzania
(Jensenius et al., 2009). Furthermore, Ta et al. (Ta et al.,
2008) also reported Q fever in three febrile patient travellers
who returned home to Spain after short period stay in sub-
Saharan Africa and Mozambique.
In conclusion zoonotic vector-borne bacteria have emerged
as one of the important causes of diseases in home returning
international travellers from east African countries and there
is a possibility of misdiagnosis and mistreatment due to the
high prevalence of common symptoms like systemic febrile
illness associated with infections by many bacterial species
(Mendelson et al., 2014). Thus vector-borne zoonotic
88
bacteria need to be suspected in febrile patients returning
from east Africa (Mediannikov et al., 2013b).
89
dissected ticks, or the organs may be used for histological
testing. Immunodetection methods may also be used to
detect bacteria in haemolymph or organ smears. Slides are
air-dried and fixed in acetone before being treated with
polyclonal or monoclonal antibodies conjugated with
immunofluorescent labels (Brouqui et al., 2004; Parola et al.,
2013).
The gold standard for diagnosis of vector-borne bacterial
disease is by cultivation and identification of the etiologic
agent from an appropriate specimen (Brouqui et al., 2004).
Bacteria can be cultured from dead, frozen or live arthropods
after disinfecting with iodinated alcohol and rinsing in sterile
water, and drying on sterile paper in a laminar flow cabinet
the arthropod is macerated in 1 mL of cell culture medium
and inoculated into culturing medium for the target bacteria.
For instance a drop of hemolymph is used for cell culture
systems and shell vial technique used for culture of
Rickettsia species from live ticks (Brouqui et al., 2004;
Parola et al., 2013; Parola and Raoult, 2001; Raoult and
Roux, 1999). Appropriate antimicrobial agent that is not
effective against the target bacteria is added to the culture
media to prevent contamination of the cultures with gut
bacteria and fungi of the crushed arthropods. Bacteria like
90
Rickettsia, C. burnetii, and ehrlichiae can also be isolated by
use of the shell vial technique whereas borreliae are best
isolated and grown on the specialized growth medium BSK
II, which is a complex broth that contains long-chain fatty
acids, bovine serum albumin, and rabbit serum (Brouqui et
al., 2004; Parola et al., 2013; Parola and Raoult, 2001;
Raoult and Roux, 1999; Socolovschi et al., 2009).
Molecular methods using PCR based by amplification and
sequencing of 16S rRNA gene and other genes accepted for
the characterization of specific groups of bacteria is the gold
standard for the identification of vector-borne bacterial
agents in arthropods (Brouqui et al., 2004; Parola et al.,
2013; Parola and Raoult, 2001; Raoult and Roux, 1999;
Socolovschi et al., 2009). Identification strategies based on
PCR amplification followed by DNA sequencing of a
number of genes have been described for several bacteria
using genus-specific or species specific primers (Brouqui et
al., 2004; Parola et al., 2013; Raoult and Roux, 1999;
Socolovschi et al., 2009). Bacterial total DNA for PCR from
different arthropods such as lice, ticks, fleas, and flies can be
extracted by employing several methods like boiling
triturated arthropods in saline buffer or in PCR extraction
buffer, boiling hemolymph, phenol-chloroform extraction, or
91
recent extraction techniques by using a DNeasy Tissue kit
(QIAGEN, Hilden, Germany QIAamp Tissue Kit).
Molecular detection of bacteria in vectors has been
previously reviewed in detail (Brouqui et al., 2004; Parola et
al., 2013; Parola and Raoult, 2001; Raoult and Roux, 1999;
Socolovschi et al., 2009).
92
standard method for detection and to demonstrate the
presence of the spirochete (B. recurrentis and B. duttonii) in
the blood during the febrile period (Brouqui et al., 2004;
Cutler, 2010). Blood smears are stained according to Giemsa
or White. Mice or rats may also be inoculated
intraperitoneally, and they will demonstrate spirochetes in
their blood within a few days to 2 weeks (Brouqui et al.,
2004; Cutler, 2010). Immunological methods of diagnosis
relapsing fever are not satisfactory. Polymerase chain
reaction (PCR) is highly sensitive although not accessible to
many laboratories in endemic countries in east Africa
(Brouqui et al., 2004; Raoult and Roux, 1999).
Rickettsia spp. and Bartonella spp. diagnostics are based on
the detection of a specific antibody by indirect fluorescent
antibody testing (Brouqui et al., 2004; Parola et al., 2013;
Raoult and Roux, 1999). Cross-adsorption assay allowing
removal of the cross-reacting antibody is mandatory to avoid
cross reaction in closely related bacterial species (Brouqui et
al., 2004). Although sensitive and specific techniques such
as PCR and culture can be performed at reference
laboratories most cases worldwide are diagnosed by
serological analysis such as direct immunofuorescent assay
(DFA) (Raoult and Roux, 1999). The principal limitations
93
with serologic analysis include a usually negative result in
the acute phase when most patients first seek medical care,
poor sensitivity in cases treated early with doxycycline, and
an inability to distinguish between the various Rickettsia
species caused by cross-reactions (Brouqui et al., 2004;
Parola et al., 2013; Parola and Raoult, 2001; Raoult and
Roux, 1999).
94
of doxycycline can be effective (Badiaga and Brouqui, 2012;
Raoult and Roux, 1999).
B. recurrentis has been successfully treated with
chloramphenicol, penicillin, tetracycline, and erythromycin
the most efficient of which is the oral (a single dose of 0.5 g)
or iv administration of tetracycline and also erythromycin in a
single oral dose (0.5 g) is an equally effective therapy in
pregnant women and children of, 8 years of age (Badiaga
and Brouqui, 2012; Cutler, 2010; Raoult and Roux, 1999).
Treatment of B. quintana in humans is a challenge and the
recommended regimen for patients with chronic bacteraemia
is a combination of gentamicin (3 mg/kg intravenously once
daily for 14 days) and doxycycline (200 mg orally once
daily) for 28 days; for patients with endocarditis, this
regimen is extended to 42 days (Badiaga and Brouqui, 2012;
Raoult and Roux, 1999).
Plague can be treated with gentamicin, oxytetracycline,
tetracycline, chloramphenicol and doxycycline (Eisen and
Gage, 2012; McElroy et al., 2010; Perry and Fetherston,
1997). However, streptomycin still remains the drug of
choice due to its bacteriolytic activity it should be
administered with care to prevent the development of
endotoxic shock (Eisen and Gage, 2012; McElroy et al.,
95
2010). All patients suspected of having bubonic plague
should be placed in isolation until 2 days after starting
antibiotic treatment to prevent the potential spread of the
disease should the patient develop secondary plague
pneumonia (Eisen and Gage, 2012; McElroy et al., 2010;
Perry and Fetherston, 1997).
12. Conclusion
The present review demonstrated that vector-borne zoonotic
bacterial diseases occur in many east African countries and
are more common than are generally known and are rarely
diagnosed (Cutler, 2010; Eisen et al., 2012; Mediannikov et
al., 2013b). Even though, these diseases pose severe public
health problems in residents as well as in international
travellers and cause huge economic losses (Cutler, 2010;
Mediannikov et al., 2013b) they are under reported and most
disease are neglected. Thus, this review is intended to
contribute to the better understanding and alert clinicians,
microbiologists, scientific and medical community working
in east African countries and in other countries around the
world about the importance of vector-borne zoonotic
bacterial diseases transmitted by ticks, lice, fleas and flies.
Increasing the knowledge and awareness on vector-borne
96
bacterial diseases and vector control in east African countries
is very crucial for disease prevention. Public education about
vector-borne bacterial diseases and arthropod control need to
receive great attention and widely disseminated to the public
via various means including television and radio
advertisements and need promotion on the websites of health
authorities. Research on arthropod-borne zoonotic bacteria in
humans, domestic and wild animals, reservoir hosts and
arthropods need to be intensified and also need to consider
the detection of pathogen RNA so as so to detect live disease
causative agents in arthropods. It is also equally important
for the future studies to address the vector competence of
several arthropod species in which bacterial DNA have been
detected in different countries in east Africa by
demonstrating their ability to acquire the pathogen from an
infected host and transmit it to nave susceptible animals
during the subsequent feeding under experimental or
controlled conditions.
97
References
98
Azad, A.F., Beard, C.B., 1998. Rickettsial pathogens and their
arthropod vectors. Emerg. Infect. Dis. 4, 179-186.
Azad, A.F., Webb, L., Carl, M., Dasch, G.A., 1990. Detection of
rickettsiae in arthropod vectors by DNA amplification
using the polymerase chain reaction. Ann. N. Y. Acad.
Sci. 590, 557-563.
Badiaga, S., Brouqui, P., 2012. Human louse-transmitted
infectious diseases. Clin. Microbiol. Infect. 18, 332-337.
Baneth, G., 2014. Tick-borne infections of animals and humans: a
common ground. Int. J. Parasitol. 44, 591-596.
Barker, S.C., 1994. Phylogeny and classification, origins, and
evolution of host associations of lice. Int. J. Parasitol. 24,
1285-1291.
Barker, S.C., Murrell, A., 2004. Systematics and evolution of ticks
with a list of valid genus and species names. Parasitology
129 Suppl, S15-S36.
Beati, L., Keirans, J.E., 2001. Analysis of the systematic
relationships among ticks of the genera Rhipicephalus and
Boophilus (Acari: Ixodidae) based on mitochondrial 12S
ribosomal DNA gene sequences and morphological
characters. J. Parasitol. 87, 32-48.
Beati, L., Patel, J., Lucas-Williams, H., Adakal, H., Kanduma,
E.G., Tembo-Mwase, E., Krecek, R., Mertins, J.W.,
Alfred, J.T., Kelly, S., Kelly, P., 2012. Phylogeography
and demographic history of Amblyomma variegatum
(Fabricius) (Acari: Ixodidae), the tropical bont tick.
Vector. Borne. Zoonotic. Dis. 12, 514-525.
Beati, L., Raoult, D., 1993. Rickettsia massiliae sp. nov., a new
spotted fever group Rickettsia. Int. J. Syst. Bacteriol. 43,
839-840.
Bemis, D.A., Kania, S.A., 2007. Isolation of Bartonella sp. from
sheep blood. Emerg. Infect. Dis. 13, 1565-1567.
99
Billeter, S.A., Levy, M.G., Chomel, B.B., Breitschwerdt, E.B.,
2008. Vector transmission of Bartonella species with
emphasis on the potential for tick transmission. Med. Vet.
Entomol. 22, 1-15.
Bitam, I., 2012. Vectors of rickettsiae in Africa. Ticks. Tick.
Borne. Dis. 3, 382-386.
Bitam, I., Dittmar, K., Parola, P., Whiting, M.F., Raoult, D., 2010.
Fleas and flea-borne diseases. Int. J. Infect. Dis. 14, e667-
e676.
Borgnolo, G., Hailu, B., Ciancarelli, A., Almaviva, M.,
Woldemariam, T., 1993. Louse-borne relapsing fever. A
clinical and an epidemiological study of 389 patients in
Asella Hospital, Ethiopia. Trop. Geogr. Med. 45, 66-69.
Botros, B.A., Soliman, A.K., Salib, A.W., Olson, J., Hibbs, R.G.,
Williams, J.C., Darwish, M., el, T.A., Watts, D.M., 1995.
Coxiella burnetii antibody prevalences among human
populations in north-east Africa determined by enzyme
immunoassay. J. Trop. Med. Hyg. 98, 173-178.
Boutellis, A., Mediannikov, O., Bilcha, K.D., Ali, J., Campelo, D.,
Barker, S.C., Raoult, D., 2013. Borrelia recurrentis in
head lice, Ethiopia. Emerg. Infect. Dis. 19, 796-798.
Breitschwerdt, E.B., Kordick, D.L., 2000. Bartonella infection in
animals: carriership, reservoir potential, pathogenicity,
and zoonotic potential for human infection. Clin.
Microbiol. Rev. 13, 428-438.
Breitschwerdt, E.B., Maggi, R.G., Chomel, B.B., Lappin, M.R.,
2010. Bartonellosis: an emerging infectious disease of
zoonotic importance to animals and human beings. J. Vet.
Emerg. Crit Care (San. Antonio. ) 20, 8-30.
Brouqui, P., Bacellar, F., Baranton, G., Birtles, R.J., Bjoersdorff,
A., Blanco, J.R., Caruso, G., Cinco, M., Fournier, P.E.,
Francavilla, E., Jensenius, M., Kazar, J., Laferl, H., Lakos,
A., Lotric, F.S., Maurin, M., Oteo, J.A., Parola, P., Perez-
Eid, C., Peter, O., Postic, D., Raoult, D., Tellez, A.,
100
Tselentis, Y., Wilske, B., 2004. Guidelines for the
diagnosis of tick-borne bacterial diseases in Europe. Clin.
Microbiol. Infect. 10, 1108-1132.
Burger, T.D., Shao, R., Barker, S.C., 2014. Phylogenetic analysis
of mitochondrial genome sequences indicates that the
cattle tick, Rhipicephalus (Boophilus) microplus, contains
a cryptic species. Mol. Phylogenet. Evol. 76, 241-253.
Chang, C.C., Chomel, B.B., Kasten, R.W., Heller, R.M., Ueno, H.,
Yamamoto, K., Bleich, V.C., Pierce, B.M., Gonzales, B.J.,
Swift, P.K., Boyce, W.M., Jang, S.S., Boulouis, H.J.,
Piemont, Y., Rossolini, G.M., Riccio, M.L., Cornaglia, G.,
Pagani, L., Lagatolla, C., Selan, L., Fontana, R., 2000.
Bartonella spp. isolated from wild and domestic
ruminants in North America. Emerg. Infect. Dis. 6, 306-
311.
Cherry, N.A., Maggi, R.G., Cannedy, A.L., Breitschwerdt, E.B.,
2009. PCR detection of Bartonella bovis and Bartonella
henselae in the blood of beef cattle. Vet. Microbiol. 135,
308-312.
Chomel, B.B., Boulouis, H.J., Breitschwerdt, E.B., Kasten, R.W.,
Vayssier-Taussat, M., Birtles, R.J., Koehler, J.E., Dehio,
C., 2009. Ecological fitness and strategies of adaptation of
Bartonella species to their hosts and vectors. Vet. Res. 40,
29.
Chomel, B.B., Boulouis, H.J., Maruyama, S., Breitschwerdt, E.B.,
2006. Bartonella spp. in pets and effect on human health.
Emerg. Infect. Dis. 12, 389-394.
Chomel, B.B., Kasten, R.W., 2010. Bartonellosis, an increasingly
recognized zoonosis. J. Appl. Microbiol. 109, 743-750.
Civen, R., Ngo, V., 2008. Murine typhus: an unrecognized
suburban vectorborne disease. Clin. Infect. Dis. 46, 913-
918.
Cotte, V., Bonnet, S., Le, R.D., Le, N.E., Chauvin, A., Boulouis,
H.J., Lecuelle, B., Lilin, T., Vayssier-Taussat, M., 2008.
101
Transmission of Bartonella henselae by Ixodes ricinus.
Emerg. Infect. Dis. 14, 1074-1080.
Cutler, S., Abdissa, A., Adamu, H., Tolosa, T., Gashaw, A., 2012.
Borrelia in Ethiopian ticks. Ticks. Tick. Borne. Dis. 3, 14-
17.
Cutler, S.J., 2010. Relapsing fever--a forgotten disease revealed. J.
Appl. Microbiol. 108, 1115-1122.
Cutler, S.J., Bonilla, E.M., Singh, R.J., 2010a. Population structure
of East African relapsing fever Borrelia spp. Emerg.
Infect. Dis. 16, 1076-1080.
Cutler, S.J., Bonilla, E.M., Singh, R.J., 2010b. Population
structure of East African relapsing fever Borrelia spp.
Emerg. Infect. Dis. 16, 1076-1080.
Cutler, S.J., Bouzid, M., Cutler, R.R., 2007. Q fever. J. Infect. 54,
313-318.
Cutler, S.J., Browning, P., Scott, J.C., 2006. Ornithodoros
moubata, a soft tick vector for Rickettsia in east Africa?
Ann. N. Y. Acad. Sci. 1078, 373-377.
Dantas-Torres, F., Chomel, B.B., Otranto, D., 2012. Ticks and
tick-borne diseases: a One Health perspective. Trends
Parasitol. 28, 437-446.
De la Fuente, J., Estrada-Pena, A., 2012. Ticks and tick-borne
pathogens on the rise. Ticks. Tick. Borne. Dis. 3, 115-116.
de, J.J., Wilkinson, R.J., Schaeffers, P., Sondorp, H.E., Davidson,
R.N., 1995. Louse-borne relapsing fever in southern
Sudan. Trans. R. Soc. Trop. Med. Hyg. 89, 621.
Dehio, C., Sauder, U., Hiestand, R., 2004. Isolation of Bartonella
schoenbuchensis from Lipoptena cervi, a blood-sucking
arthropod causing deer ked dermatitis. J. Clin. Microbiol.
42, 5320-5323.
Dill, T., Dobler, G., Saathoff, E., Clowes, P., Kroidl, I., Ntinginya,
E., Machibya, H., Maboko, L., Loscher, T., Hoelscher,
102
M., Heinrich, N., 2013. High seroprevalence for typhus
group rickettsiae, southwestern Tanzania. Emerg. Infect.
Dis. 19, 317-320.
Dobler, G., Pfeffer, M., 2011. Fleas as parasites of the family
Canidae. Parasit. Vectors. 4, 139.
Eisen, R.J., Borchert, J.N., Mpanga, J.T., Atiku, L.A., MacMillan,
K., Boegler, K.A., Montenieri, J.A., Monaghan, A., Gage,
K.L., 2012. Flea diversity as an element for persistence of
plague bacteria in an East African plague focus. PLoS
One 7, e35598.
Eisen, R.J., Gage, K.L., 2012. Transmission of flea-borne zoonotic
agents. Annu. Rev. Entomol. 57, 61-82.
Estrada-Pena, A., Jongejan, F., 1999. Ticks feeding on humans: a
review of records on human-biting Ixodoidea with special
reference to pathogen transmission. Exp. Appl. Acarol.
23, 685-715.
Fischer, K., Walton, S., 2014. Parasitic mites of medical and
veterinary importance - is there a common research
agenda? Int. J. Parasitol.
Fournier, P.E., Ndihokubwayo, J.B., Guidran, J., Kelly, P.J.,
Raoult, D., 2002. Human pathogens in body and head lice.
Emerg. Infect. Dis. 8, 1515-1518.
Franck, S., Feldmeier, H., Heukelbach, J., 2003. Tungiasis: more
than an exotic nuisance. Travel. Med. Infect. Dis. 1, 159-
166.
Gillespie, J.J., Ammerman, N.C., Beier-Sexton, M., Sobral, B.S.,
Azad, A.F., 2009. Louse- and flea-borne rickettsioses:
biological and genomic analyses. Vet. Res. 40, 12.
Gumi, B., Firdessa, R., Yamuah, L., Sori, T., Tolosa, T., Aseffa,
A., Zinsstag, J., Schelling, E., 2013. Seroprevalence of
Brucellosis and Q-Fever in Southeast Ethiopian Pastoral
Livestock. J. Vet. Sci. Med. Diagn. 2.
Guptill, L., 2010. Bartonellosis. Vet. Microbiol. 140, 347-359.
103
Gutierrez, R., Cohen, L., Morick, D., Mumcuoglu, K.Y., Harrus,
S., Gottlieb, Y., 2014. Identification of Different
Bartonella Species in the Cattle Tail Louse (Haematopinus
quadripertusus) and in Cattle Blood. Appl. Environ.
Microbiol. 80, 5477-5483.
Halos, L., Jamal, T., Maillard, R., Girard, B., Guillot, J., Chomel,
B., Vayssier-Taussat, M., Boulouis, H.J., 2004. Role of
Hippoboscidae flies as potential vectors of Bartonella spp.
infecting wild and domestic ruminants. Appl. Environ.
Microbiol. 70, 6302-6305.
Hirunkanokpun, S., Thepparit, C., Foil, L.D., Macaluso, K.R.,
2011. Horizontal transmission of Rickettsia felis between
cat fleas, Ctenocephalides felis. Mol. Ecol. 20, 4577-4586.
Hoogstraal, H., 1956. African Ixodidae. 1. Ticks of the Sudan
(with special reference to equatorial province and with
preliminary reviews of the genera Boophilus, Margaropus
and Hyalomma). Department of the Navy, Bureau of the
Medicine and Surgery, United States Government Printing
Office, Washington D. C.
Hummel, P.H., 1976. Incidence in Tanzania of CF antibody to
Coxiella burneti in sera from man, cattle, sheep, goats and
game. Vet. Rec. 98, 501-505.
Hunziker, T., Berger, C., Staubli, G., Tschopp, A., Weber, R.,
Nadal, D., Hatz, C., Schlagenhauf, P., 2012. Profile of
travel-associated illness in children, Zurich, Switzerland.
J. Travel. Med. 19, 158-162.
Irwin, P.J., Jefferies, R., 2004. Arthropod-transmitted diseases of
companion animals in Southeast Asia. Trends Parasitol.
20, 27-34.
Jensenius, M., Davis, X., von, S.F., Schwartz, E., Keystone, J.S.,
Leder, K., Lopez-Velez, R., Caumes, E., Cramer, J.P.,
Chen, L., Parola, P., 2009. Multicenter GeoSentinel
analysis of rickettsial diseases in international travelers,
1996-2008. Emerg. Infect. Dis. 15, 1791-1798.
104
Jiang, J., Maina, A.N., Knobel, D.L., Cleaveland, S., Laudisoit,
A., Wamburu, K., Ogola, E., Parola, P., Breiman, R.F.,
Njenga, M.K., Richards, A.L., 2013. Molecular detection
of Rickettsia felis and candidatus rickettsia asemboensis in
fleas from human habitats, Asembo, Kenya. Vector.
Borne. Zoonotic. Dis. 13, 550-558.
Jongejan, F., Uilenberg, G., 2004. The global importance of ticks.
Parasitology 129 Suppl, S3-14.
Karger, A., Kampen, H., Bettin, B., Dautel, H., Ziller, M.,
Hoffmann, B., Suss, J., Klaus, C., 2012. Species
determination and characterization of developmental
stages of ticks by whole-animal matrix-assisted laser
desorption/ionization mass spectrometry. Ticks. Tick.
Borne. Dis. 3, 78-89.
Kelly, P.J., Matthewman, L.A., Mason, P.R., Raoult, D., 1993. Q
fever in Zimbabwe. A review of the disease and the results
of a serosurvey of humans, cattle, goats and dogs. S. Afr.
Med. J. 83, 21-25.
Kempf, M., Abdissa, A., Diatta, G., Trape, J.F., Angelakis, E.,
Mediannikov, O., La, S.B., Raoult, D., 2012. Detection of
Acinetobacter baumannii in human head and body lice
from Ethiopia and identification of new genotypes. Int. J.
Infect. Dis. 16, e680-e683.
Knobel, D.L., Maina, A.N., Cutler, S.J., Ogola, E., Feikin, D.R.,
Junghae, M., Halliday, J.E., Richards, A.L., Breiman,
R.F., Cleaveland, S., Njenga, M.K., 2013. Coxiella
burnetii in humans, domestic ruminants, and ticks in rural
western Kenya. Am. J. Trop. Med. Hyg. 88, 513-518.
Kosoy, M., Bai, Y., Lynch, T., Kuzmin, I.V., Niezgoda, M.,
Franka, R., Agwanda, B., Breiman, R.F., Rupprecht, C.E.,
2010. Bartonella spp. in bats, Kenya. Emerg. Infect. Dis.
16, 1875-1881.
105
Kumsa, B., Beyecha, K., Geloye, M., 2012a. Ectoparasites of
sheep in three agro-ecological zones in central Oromia,
Ethiopia. Onderstepoort J. Vet. Res. 79, E1-E7.
Kumsa, B., Parola, P., Raoult, D., Socolovschi, C., 2013.
Molecular Detection of Rickettsia felis and Bartonella
henselae in dog and cat fleas in Central Oromia, Ethiopia.
In.
Kumsa, B., Parola, P., Raoult, D., Socolovschi, C., 2014a.
Bartonella melophagi in Melophagus ovinus (sheep ked)
collected from sheep in northern Oromia, Ethiopia. Comp
Immunol. Microbiol. Infect. Dis. 37, 69-76.
Kumsa, B., Socolovschi, C., Parola, P., Rolain, J.M., Raoult, D.,
2012b. Molecular detection of Acinetobacter species in
lice and keds of domestic animals in Oromia Regional
State, Ethiopia. PLoS One 7, e52377.
Kumsa, B., Socolovschi, C., Raoult, D., Parola, P., 2014b. Spotted
fever group rickettsiae in ixodid ticks in Oromia, Ethiopia.
Ticks. Tick. Borne. Dis.
Kumsa, B.E., Mekonnen, S., 2011. Ixodid ticks, fleas and lice
infesting dogs and cats in Hawassa, southern Ethiopia.
Onderstepoort J. Vet. Res. 78, E1-E4.
Laudisoit, A., Leirs, H., Makundi, R.H., Van, D.S., Davis, S.,
Neerinckx, S., Deckers, J., Libois, R., 2007. Plague and
the human flea, Tanzania. Emerg. Infect. Dis. 13, 687-
693.
Liu, X.Y., Bonnet, S.I., 2014. Hard tick factors implicated in
pathogen transmission. PLoS Negl. Trop. Dis. 8, e2566.
Lorusso, V., Gruszka, K.A., Majekodunmi, A., Igweh, A.,
Welburn, S.C., Picozzi, K., 2013. Rickettsia africae in
Amblyomma variegatum ticks, Uganda and Nigeria.
Emerg. Infect. Dis. 19, 1705-1707.
Macaluso, K.R., Davis, J., Alam, U., Korman, A., Rutherford, J.S.,
Rosenberg, R., Azad, A.F., 2003. Spotted fever group
106
rickettsiae in ticks from the Masai Mara region of Kenya.
Am. J. Trop. Med. Hyg. 68, 551-553.
Maggi, R.G., Kosoy, M., Mintzer, M., Breitschwerdt, E.B., 2009.
Isolation of Candidatus Bartonella melophagi from human
blood. Emerg. Infect. Dis. 15, 66-68.
Maina, A.N., Knobel, D.L., Jiang, J., Halliday, J., Feikin, D.R.,
Cleaveland, S., Ng'ang'a, Z., Junghae, M., Breiman, R.F.,
Richards, A.L., Njenga, M.K., 2012. Rickettsia felis
infection in febrile patients, western Kenya, 2007-2010.
Emerg. Infect. Dis. 18, 328-331.
Maurin, M., Raoult, D., 1999. Q fever. Clin. Microbiol. Rev. 12,
518-553.
Mazigo, H.D., Bahemana, E., Konje, E.T., Dyegura, O., Mnyone,
L.L., Kweka, E.J., Kidenya, B.R., Heukelbach, J., 2012.
Jigger flea infestation (tungiasis) in rural western
Tanzania: high prevalence and severe morbidity. Trans. R.
Soc. Trop. Med. Hyg. 106, 259-263.
McElroy, K.M., Blagburn, B.L., Breitschwerdt, E.B., Mead, P.S.,
McQuiston, J.H., 2010. Flea-associated zoonotic diseases
of cats in the USA: bartonellosis, flea-borne rickettsioses,
and plague. Trends Parasitol. 26, 197-204.
McQuiston, J.H., Paddock, C.D., Singleton, J., Jr., Wheeling, J.T.,
Zaki, S.R., Childs, J.E., 2004. Imported spotted fever
rickettsioses in United States travelers returning from
Africa: a summary of cases confirmed by laboratory
testing at the Centers for Disease Control and Prevention,
1999-2002. Am. J. Trop. Med. Hyg. 70, 98-101.
Mediannikov, O., Abdissa, A., Socolovschi, C., Diatta, G., Trape,
J.F., Raoult, D., 2013a. Detection of a new Borrelia
species in ticks taken from cattle in Southwest Ethiopia.
Vector. Borne. Zoonotic. Dis. 13, 266-269.
Mediannikov, O., Fenollar, F., 2014. Looking in ticks for human
bacterial pathogens. Microb. Pathog.
107
Mediannikov, O., Socolovschi, C., Edouard, S., Fenollar, F.,
Mouffok, N., Bassene, H., Diatta, G., Tall, A., Niangaly,
H., Doumbo, O., Lekana-Douki, J.B., Znazen, A., Sarih,
M., Ratmanov, P., Richet, H., Ndiath, M.O., Sokhna, C.,
Parola, P., Raoult, D., 2013b. Common Epidemiology of
Rickettsia felis Infection and Malaria, Africa. Emerg.
Infect. Dis. 19, 1775-1783.
Medina-Sanchez, A., Bouyer, D.H., Alcantara-Rodriguez, V.,
Mafra, C., Zavala-Castro, J., Whitworth, T., Popov, V.L.,
Fernandez-Salas, I., Walker, D.H., 2005. Detection of a
typhus group Rickettsia in Amblyomma ticks in the state
of Nuevo Leon, Mexico. Ann. N. Y. Acad. Sci. 1063, 327-
332.
Meheretu, Y., Leirs, H., Welegerima, K., Breno, M., Tomas, Z.,
Kidane, D., Girmay, K., de Bellocq, J.G., 2013.
Bartonella prevalence and genetic diversity in small
mammals from ethiopia. Vector. Borne. Zoonotic. Dis. 13,
164-175.
Mendelson, M., Han, P.V., Vincent, P., von, S.F., Cramer, J.P.,
Loutan, L., Kain, K.C., Parola, P., Hagmann, S., Gkrania-
Klotsas, E., Sotir, M., Schlagenhauf, P., 2014. Regional
variation in travel-related illness acquired in Africa,
March 1997-May 2011. Emerg. Infect. Dis. 20, 532-541.
Merhej, V., Angelakis, E., Socolovschi, C., Raoult, D., 2014.
Genotyping, evolution and epidemiological findings of
Rickettsia species. Infect. Genet. Evol. 25, 122-137.
Merhej, V., Raoult, D., 2011. Rickettsial evolution in the light of
comparative genomics. Biol. Rev. Camb. Philos. Soc. 86,
379-405.
Morelli, G., Song, Y., Mazzoni, C.J., Eppinger, M., Roumagnac,
P., Wagner, D.M., Feldkamp, M., Kusecek, B., Vogler,
A.J., Li, Y., Cui, Y., Thomson, N.R., Jombart, T., Leblois,
R., Lichtner, P., Rahalison, L., Petersen, J.M., Balloux, F.,
Keim, P., Wirth, T., Ravel, J., Yang, R., Carniel, E.,
108
Achtman, M., 2010. Yersinia pestis genome sequencing
identifies patterns of global phylogenetic diversity. Nat.
Genet. 42, 1140-1143.
Morita, C., El Hussein, A.R., Matsuda, E., Abdel Gabbar, K.M.,
Muramatsu, Y., Abdel Rahman, M.B., Eleragi, A.M.,
Hassan, S.M., Chitambo, A.M., Ueno, H., 2004. Spotted
fever group rickettsiae from ticks captured in Sudan. Jpn.
J. Infect. Dis. 57, 107-109.
Mosbacher, M., Elliott, S.P., Shehab, Z., Pinnas, J.L., Klotz, J.H.,
Klotz, S.A., 2010. Cat scratch disease and arthropod
vectors: more to it than a scratch? J. Am. Board Fam.
Med. 23, 685-686.
Mura, A., Socolovschi, C., Ginesta, J., Lafrance, B., Magnan, S.,
Rolain, J.M., Davoust, B., Raoult, D., Parola, P., 2008.
Molecular detection of spotted fever group rickettsiae in
ticks from Ethiopia and Chad. Trans. R. Soc. Trop. Med.
Hyg. 102, 945-949.
Murrell, A., Dobson, S.J., Yang, X., Lacey, E., Barker, S.C., 2003.
A survey of bacterial diversity in ticks, lice and fleas from
Australia. Parasitol. Res. 89, 326-334.
Mutai, B.K., Wainaina, J.M., Magiri, C.G., Nganga, J.K.,
Ithondeka, P.M., Njagi, O.N., Jiang, J., Richards, A.L.,
Waitumbi, J.N., 2013. Zoonotic surveillance for
rickettsiae in domestic animals in Kenya. Vector. Borne.
Zoonotic. Dis. 13, 360-366.
Nakao, R., Qiu, Y., Igarashi, M., Magona, J.W., Zhou, L., Ito, K.,
Sugimoto, C., 2013. High prevalence of spotted fever
group rickettsiae in Amblyomma variegatum from Uganda
and their identification using sizes of intergenic spacers.
Ticks. Tick. Borne. Dis. 4, 506-512.
Neerinckx, S., Bertherat, E., Leirs, H., 2010. Human plague
occurrences in Africa: an overview from 1877 to 2008.
Trans. R. Soc. Trop. Med. Hyg. 104, 97-103.
109
Nelder, M.P., Lloyd, J.E., Loftis, A.D., Reeves, W.K., 2008.
Coxiella burnetii in wild-caught filth flies. Emerg. Infect.
Dis. 14, 1002-1004.
Nicholson, W.L., Allen, K.E., McQuiston, J.H., Breitschwerdt,
E.B., Little, S.E., 2010. The increasing recognition of
rickettsial pathogens in dogs and people. Trends Parasitol.
26, 205-212.
Okabayashi, T., Hasebe, F., Samui, K.L., Mweene, A.S., Pandey,
S.G., Yanase, T., Muramatsu, Y., Ueno, H., Morita, C.,
1999. Short report: prevalence of antibodies against
spotted fever, murine typhus, and Q fever rickettsiae in
humans living in Zambia. Am. J. Trop. Med. Hyg. 61, 70-
72.
Otranto, D., Dantas-Torres, F., Breitschwerdt, E.B., 2009.
Managing canine vector-borne diseases of zoonotic
concern: part one. Trends Parasitol. 25, 157-163.
Owen, C.E., Bahrami, S., Malone, J.C., Callen, J.P., Kulp-Shorten,
C.L., 2006. African tick bite fever: a not-so-uncommon
illness in international travelers. Arch. Dermatol. 142,
1312-1314.
Parola, P., 2011. Rickettsia felis: from a rare disease in the USA to
a common cause of fever in sub-Saharan Africa. Clin.
Microbiol. Infect. 17, 996-1000.
Parola, P., Davoust, B., Raoult, D., 2005a. Tick- and flea-borne
rickettsial emerging zoonoses. Vet. Res. 36, 469-492.
Parola, P., Inokuma, H., Camicas, J.L., Brouqui, P., Raoult, D.,
2001. Detection and identification of spotted fever group
Rickettsiae and Ehrlichiae in African ticks. Emerg. Infect.
Dis. 7, 1014-1017.
Parola, P., Paddock, C.D., Raoult, D., 2005b. Tick-borne
rickettsioses around the world: emerging diseases
challenging old concepts. Clin. Microbiol. Rev. 18, 719-
756.
110
Parola, P., Paddock, C.D., Socolovschi, C., Labruna, M.B.,
Mediannikov, O., Kernif, T., Abdad, M.Y., Stenos, J.,
Bitam, I., Fournier, P.E., Raoult, D., 2013. Update on tick-
borne rickettsioses around the world: a geographic
approach. Clin. Microbiol. Rev. 26, 657-702.
Parola, P., Raoult, D., 2001. Ticks and tickborne bacterial diseases
in humans: an emerging infectious threat. Clin. Infect.
Dis. 32, 897-928.
Perry, R.D., Fetherston, J.D., 1997. Yersinia pestis--etiologic
agent of plague. Clin. Microbiol. Rev. 10, 35-66.
Pfaffle, M., Littwin, N., Muders, S.V., Petney, T.N., 2013. The
ecology of tick-borne diseases. Int. J. Parasitol. 43, 1059-
1077.
Philip, C.B., Hoogstraal, H., Reiss-Gutfreund, R., Clifford, C.M.,
1966a. Evidence of rickettsial disease agents in ticks from
Ethiopian cattle. Bull. World Health Organ 35, 127-131.
Philip, C.B., Lackman, D.B., Bell, E.J., Hughes, L.E., 1966b.
Laboratory identification of typhus isolated by Reiss-
Gutfreund from Ethiopian livestock ticks. Am. J. Trop.
Med. Hyg. 15, 950-953.
Piesman, J., Eisen, L., 2008. Prevention of tick-borne diseases.
Annu. Rev. Entomol. 53, 323-343.
Porter, S.R., Czaplicki, G., Mainil, J., Guatteo, R., Saegerman, C.,
2011. Q Fever: current state of knowledge and
perspectives of research of a neglected zoonosis. Int. J.
Microbiol. 2011, 248418.
Prabhu, M., Nicholson, W.L., Roche, A.J., Kersh, G.J.,
Fitzpatrick, K.A., Oliver, L.D., Massung, R.F., Morrissey,
A.B., Bartlett, J.A., Onyango, J.J., Maro, V.P., Kinabo,
G.D., Saganda, W., Crump, J.A., 2011. Q fever, spotted
fever group, and typhus group rickettsioses among
hospitalized febrile patients in northern Tanzania. Clin.
Infect. Dis. 53, e8-15.
111
Qiu, Y., Nakao, R., Namangala, B., Sugimoto, C., 2013. First
genetic detection of Coxiella burnetii in Zambian
livestock. Am. J. Trop. Med. Hyg. 89, 518-519.
Raoult, D., Fournier, P.E., Fenollar, F., Jensenius, M., Prioe, T., de
Pina, J.J., Caruso, G., Jones, N., Laferl, H., Rosenblatt,
J.E., Marrie, T.J., 2001a. Rickettsia africae, a tick-borne
pathogen in travelers to sub-Saharan Africa. N. Engl. J.
Med. 344, 1504-1510.
Raoult, D., Fournier, P.E., Fenollar, F., Jensenius, M., Prioe, T., de
Pina, J.J., Caruso, G., Jones, N., Laferl, H., Rosenblatt,
J.E., Marrie, T.J., 2001b. Rickettsia africae, a tick-borne
pathogen in travelers to sub-Saharan Africa. N. Engl. J.
Med. 344, 1504-1510.
Raoult, D., La, S.B., Enea, M., Fournier, P.E., Roux, V., Fenollar,
F., Galvao, M.A., de, L., X, 2001c. A flea-associated
Rickettsia pathogenic for humans. Emerg. Infect. Dis. 7,
73-81.
Raoult, D., La, S.B., Kelly, P.J., Davoust, B., Gomez, J., 2005.
Bartonella bovis in cattle in Africa. Vet. Microbiol. 105,
155-156.
Raoult, D., Mouffok, N., Bitam, I., Piarroux, R., Drancourt, M.,
2013. Plague: history and contemporary analysis. J. Infect.
66, 18-26.
Raoult, D., Roux, V., 1997. Rickettsioses as paradigms of new or
emerging infectious diseases. Clin. Microbiol. Rev. 10,
694-719.
Raoult, D., Roux, V., 1999. The body louse as a vector of
reemerging human diseases. Clin. Infect. Dis. 29, 888-
911.
Raoult, D., Roux, V., Ndihokubwayo, J.B., Bise, G., Baudon, D.,
Marte, G., Birtles, R., 1997. Jail fever (epidemic typhus)
outbreak in Burundi. Emerg. Infect. Dis. 3, 357-360.
112
Ratovonjato, J., Rajerison, M., Rahelinirina, S., Boyer, S., 2014.
Yersinia pestis in Pulex irritans fleas during plague
outbreak, Madagascar. Emerg. Infect. Dis. 20, 1414-1415.
Reeves, W.K., Nelder, M.P., Cobb, K.D., Dasch, G.A., 2006.
Bartonella spp. in deer keds, Lipoptena mazamae
(Diptera: Hippoboscidae), from Georgia and South
Carolina, USA. J. Wildl. Dis. 42, 391-396.
Reif, K.E., Macaluso, K.R., 2009. Ecology of Rickettsia felis: a
review. J. Med. Entomol. 46, 723-736.
Reis, C., Cote, M., Le, R.D., Lecuelle, B., Levin, M.L., Vayssier-
Taussat, M., Bonnet, S.I., 2011. Vector competence of the
tick Ixodes ricinus for transmission of Bartonella birtlesii.
PLoS Negl. Trop. Dis. 5, e1186.
Richards, A.L., Jiang, J., Omulo, S., Dare, R., Abdirahman, K.,
Ali, A., Sharif, S.K., Feikin, D.R., Breiman, R.F., Njenga,
M.K., 2010. Human Infection with Rickettsia felis, Kenya.
Emerg. Infect. Dis. 16, 1081-1086.
Roux, V., Raoult, D., 1999. Body lice as tools for diagnosis and
surveillance of reemerging diseases. J. Clin. Microbiol.
37, 596-599.
Rutherford, J.S., Macaluso, K.R., Smith, N., Zaki, S.R., Paddock,
C.D., Davis, J., Peterson, N., Azad, A.F., Rosenberg, R.,
2004. Fatal spotted fever rickettsiosis, Kenya. Emerg.
Infect. Dis. 10, 910-913.
Sangare, A.K., Boutellis, A., Drali, R., Socolovschi, C., Barker,
S.C., Diatta, G., Rogier, C., Olive, M.M., Doumbo, O.K.,
Raoult, D., 2014. Detection of Bartonella quintana in
African body and head lice. Am. J. Trop. Med. Hyg. 91,
294-301.
Socolovschi, C., Honnorat, E., Consigny, P.H., Dougados, J.,
Passeron, A., Parola, P., Raoult, D., 2012a. Tick-borne
relapsing fever with cutaneous eschar and radiculopathy,
Ethiopia. J. Travel. Med. 19, 261-263.
113
Socolovschi, C., Matsumoto, K., Marie, J.L., Davoust, B., Raoult,
D., Parola, P., 2007. Identification of Rickettsiae, Uganda
and Djibouti. Emerg. Infect. Dis. 13, 1508-1510.
Socolovschi, C., Mediannikov, O., Raoult, D., Parola, P., 2009.
Update on tick-borne bacterial diseases in Europe.
Parasite 16, 259-273.
Socolovschi, C., Pages, F., Ndiath, M.O., Ratmanov, P., Raoult,
D., 2012b. Rickettsia species in african anopheles
mosquitoes. PLoS. One. 7, e48254.
Socolovschi, C., Pages, F., Raoult, D., 2012c. Rickettsia felis in
Aedes albopictus Mosquitoes, Libreville, Gabon. Emerg.
Infect. Dis. 18, 1687-1689.
Staley, G.P., Myburgh, J.G., Chaparro, F., 1989. Serological
evidence of Q fever in cattle in Malawi. Onderstepoort J.
Vet. Res. 56, 205-206.
Stephany, D., Buffet, P., Rolain, J.M., Raoult, D., Consigny, P.H.,
2009. Rickettsia africae infection in man after travel to
Ethiopia. Emerg. Infect. Dis. 15, 1867-1869.
Ta, T.H., Jimenez, B., Navarro, M., Meije, Y., Gonzalez, F.J.,
Lopez-Velez, R., 2008. Q Fever in returned febrile
travelers. J. Travel. Med. 15, 126-129.
Telford, S.R., III, Wormser, G.P., 2010. Bartonella spp.
transmission by ticks not established. Emerg. Infect. Dis.
16, 379-384.
Traversa, D., 2013. Fleas infesting pets in the era of emerging
extra-intestinal nematodes. Parasit. Vectors. 6, 59.
Tsai, Y.L., Lin, C.C., Chomel, B.B., Chuang, S.T., Tsai, K.H.,
Wu, W.J., Huang, C.G., Yu, J.C., Sung, M.H., Kass, P.H.,
Chang, C.C., 2011. Bartonella infection in shelter cats and
dogs and their ectoparasites. Vector. Borne. Zoonotic. Dis.
11, 1023-1030.
Vanderburg, S., Rubach, M.P., Halliday, J.E., Cleaveland, S.,
Reddy, E.A., Crump, J.A., 2014. Epidemiology of
114
Coxiella burnetii Infection in Africa: a OneHealth
systematic review. PLoS Negl. Trop. Dis. 8, e2787.
Vitale, G., Mansuelo, S., Rolain, J.M., Raoult, D., 2006. Rickettsia
massiliae human isolation. Emerg. Infect. Dis. 12, 174-
175.
Walker, A.R., Bouattour, A., Camicas, J.L., EstradaPena, A., Horak, I.C.,
Latif, A.A., Pegram, R.G., and Preston, P.M., 2003. Tick of
domestic Animals in Africa: a Guide to Identification of
species. Bioscience Reports, Edinburgh Scotland,U.K.
Walker, D.H., 2003. Rickettsial diseases in travelers. Travel. Med.
Infect. Dis. 1, 35-40.
Walter, G., Botelho-Nevers, E., Socolovschi, C., Raoult, D.,
Parola, P., 2012. Murine typhus in returned travelers: a
report of thirty-two cases. Am. J. Trop. Med. Hyg. 86,
1049-1053.
Yimer, M., Abera, B., Mulu, W., Bezabih, B., Mohammed, J.,
2014a. Prevalence and risk factors of louse- borne
relapsing fever in high risk populations in Bahir Dar city
Northwest, Ethiopia. BMC. Res. Notes 7, 615.
Yimer, M., Mulu, W., Ayalew, W., Abera, B., 2014b. Louse-borne
relapsing fever profile at Felegehiwot referral hospital,
Bahir Dar city, Ethiopia: a retrospective study. BMC. Res.
Notes 7, 250.
Yssouf, A., Flaudrops, C., Drali, R., Kernif, T., Socolovschi, C.,
Berenger, J.M., Raoult, D., Parola, P., 2013. Matrix-
assisted laser desorption ionization-time of flight mass
spectrometry for rapid identification of tick vectors. J.
Clin. Microbiol. 51, 522-528.
Yssouf, A., Socolovschi, C., Kernif, T., Temmam, S., Lagadec, E.,
Tortosa, P., Parola, P., 2014. First molecular detection of
Rickettsia africae in ticks from the Union of the Comoros.
Parasit. Vectors. 7, 444.
115
Table 1. Common tick species and their vectorized zoonotic bacteria in east Africa
Tick species Country Feeding hosts Human Bacteria detected Reference
biting
Ar. Persicus Ethiopia, South Sudan Domestic birds yes B. anserina (Cutler et al., 2012)
Am. cohaerens ETH, KEY, SOM, S.Sud, UGA, ERT Cattle and yes R. africae, C. (Kumsa et al., 2014b;
other ruminants burneti Mutai et al., 2013)
Am. gemma ETH, KEY SOM Animals Yes R. africae, C. (Kumsa et al., 2014b;
burneti Mutai et al., 2013)
Am. lepidum East Africa Animals - R. africae (Morita et al., 2004; Mura
et al., 2008; Socolovschi
et al., 2007)
Am. hebraeum Zimbabwe,Mozambique Cattle and Yes R. africae (Mutai et al., 2013)
other ruminants
Am. variegatum East Africa Animals Yes R . africae, C. (Kumsa et al., 2014b;
burneti Lorusso et al., 2013;
Mutai et al., 2013)
Ha. leachi East Africa Animals yes SF
Hy. dromedarii ETH, KEY, SOM, S.Sud, UGA, ERT Camels and Yes R. aeschlimannii (Morita et al., 2004)
other ruminants
Hy. impeltatum ETH, KEY, SOM, S.Sud, TAN, Large domestic No report No report (Walker et al., 2003)
ERT,Chad Animals
Hy. m. rufipes East Africa Animals Yes R. aeschlimannii (Kumsa et al., 2014b;
Mura et al., 2008)
Hy. truncatum East Africa Domestic Yes R. aeschlimannii (Kumsa et al., 2014b;
animals Morita et al., 2004; Mutai
et al., 2013)
Or. moubata East Africa Domestic pigs Yes Borrelia duttoni (Cutler, 2010; Cutler et
and poultry al., 2010b)
Or. savignyi ETH, KEY, SOM, S.Sud, UGA, ERT Camels and Yes No report (Walker et al., 2003)
cattle
Ot. megnini Kenya, Zimbabwe and Madagascar Domestic Yes No report (Walker et al., 2003)
animals
116
Rh. annulatus ETH, KEY and S.Sudan Ruminants No report R. africae, R. (Mutai et al., 2013)
aeschlimannii
Rh. decoloratus East Africa Animals Yes R. africae, C. (Kumsa et al., 2014b)
burneti
Rh. geigyi UGA and S.Sudan Ruminants No report No report (Walker et al., 2003)
Rh. microplus Zambia, Malawi, Tanzania and Cattle and Yes R . africae (Yssouf et al., 2014)
Zimbabwe, Mozambique, other animals
Madagascar and Kenya
Rh. In all East Africa south of Ethiopia Animals Yes R . africae (Mutai et al., 2013;
appendiculatus and Somalia Yssouf et al., 2014)
Rh. e. evertsi East Africa Domestic Yes R . africae, C. (Kumsa et al., 2014b)
animals burneti
Rh. lunulatus East Africa Domestic No report Not documented (Walker, 2003a)
animals
Rh. muhsamae ETH, KEY, S.Sud, UGA and TAN Animals No report Not documented (Walker, 2003a)
Rh. praetextatus ERT, ETH, SOM, KEY, S.Sud, UGA Animals yes R. massiliae, C. (Kumsa et al., 2014b)
and TAN burnetii
Rh. pravus ERT, ETH, SOM, KEY, S.Sud, UGA Animals No report Not documented (Walker, 2003a)
and TAN
Rh. pulchellus ERT, ETH, SOM, KEY and TAN Ruminants Yes R . africae, C. (Kumsa et al., 2014b)
burneti
Rh. sanguineus East Africa Dog and other Yes R. conorii (Walker et al., 2003)
animals
Rh. senegalensis S.Sudan and UGA Animals No report Not documented (Walker et al., 2003)
Rh. simus Zambia, Malawi, Mozambique, Cattle and No report Not documented (Walker et al., 2003)
Zimbabwe other ruminants
Rh. turanicus East Africa Animals No report Not documented (Walker et al., 2003)
Rh. zambeziensis Tanzania , Zambia, Mozambique, Cattle and No report Not documented (Walker et al., 2003)
Zimbabwe other ruminants
117
Table 2. Lice species of humans and domestic animals and their associated zoonotic bacteria reported from east
African countries
Lice species Order Host Bacteria detected Country Reference
Pediculus humanus Anoplura Humans Rickettsia prowazekii, East Africa (Badiaga and Brouqui,
Borrelia recurrentis, 2012; Barker, 1994;
Bartonella quintana, Raoult and Roux,
Acinetobacter baumannii 1999)
Pediculus capitis Anoplura Humans Borrelia recurrentis, East Africa (Badiaga and Brouqui,
Bartonella quintana 2012; Barker, 1994;
Acinetobacter baumannii Raoult and Roux,
1999)
Pthirus pubis Anoplura Humans No report East Africa (Badiaga and Brouqui,
2012; Barker, 1994)
Bovicola (Damalina) bovis Mallophaga Cattle No report East Africa (Barker, 1994)
Linognathus vituli Anoplura Cattle Acinetobacter soli East Africa (Kumsa et al., 2012b)
Solenopotes capillatus Anoplura Cattle No report East Africa (Barker, 1994)
Haematopinus eurysternus Anoplura Cattle No report (Barker, 1994)
Haematopinus quadripertusus Anoplura Cattle No report East Africa (Barker, 1994)
Haematopinus tuberculatus Anoplura Cattle No report (Barker, 1994)
Bovicola (Damalina) ovis Mallophaga Sheep No report East Africa (Barker, 1994)
Linognathus ovillus Anoplura Sheep No report East Africa (Barker, 1994)
Linognathus pedalis Anoplura Sheep No report East Africa (Barker, 1994)
Linognathus africanus Anoplura Sheep No report (Barker, 1994)
Bovicola (Damalinia) caprae Mallophaga Goat No report East Africa (Barker, 1994)
Linognathus stenopsis Anoplura Goat No report (Barker, 1994)
Linognathus africanus Anoplura Goat No report East Africa (Barker, 1994)
Bovicola (Werneckiella) equi, Mallophaga equines No report East Africa (Barker, 1994)
Haematopinus asini. Anoplura equines No report East Africa (Barker, 1994)
Haematopinus suis Anoplura Pig No report East Africa (Barker, 1994)
Trichodectes canis Mallophaga Dog No report East Africa (Kumsa and
Mekonnen, 2011)
Heterodoxus spiniger Mallophaga Dog Acinetobacter pitti East Africa (Kumsa et al., 2012b)
Linognathus steosus Anoplura Dog No report East Africa (Barker, 1994)
Felicola subrostrata Mallophaga Cat No report East Africa (Barker, 1994)
118
Fig. 1. Map of Africa showing countries that belong to the
119
Fig. 2. Map of Africa showing very few Coxiella burnetii
120
3. TICKS
121
122
Ticks are obligate haematophagous arthropods that act as
reservoirs and/or vectors for a wide range of emerging and
re-emerging human and animal diseases worldwide. Ticks
rank second only after mosquitoes as vectors of human
infectious diseases (27). Approximately 10% of the currently
recognized tick species carry human and animal pathogens
(11,13). Ticks are incriminated to transmit several bacteria
like spotted fever group (SFG) rickettsiae, Borrelia spp.,
Coxiella burnetii, Ehrlichia spp., and Bartonella spp
(26,27,35).
In Ethiopia several species of ticks are common and widely
distributed throughout all regions (39). The genera
Amblyomma (8 spp.), subgenus Boophilus (2 spp.), genus
Haemaphysalis (4 spp.), Hyalomma (9 spp.), genus
Rhipicephalus (8 spp.) and genus Ixodes (1 sp.) are known to
exist in Ethiopia (22,33). Ticks are considered to have more
veterinary significance than medical importance in the
country. Most of the studies on the role of ixodid ticks as
vectors of human pathogens in Ethiopia were conducted
before the discovery of well-advanced molecular tools
during the 1950s and 60s (28,29).
Even though there have been a number of studies on the
prevalence, impact on the commercial values of skin and
123
hides of animals, distribution and epidemiology of ixodid
ticks in Ethiopia (16,22,33), only few previous studies were
made on bacteria in ixodid ticks in Ethiopia (19). For
instance, only very few previous studies were made on
Rickettsia species in Ethiopian ixodid ticks (23). Likewise,
the only previous study on Borrelia sp. in ixodid ticks from
Ethiopia is the recent report of 7.3% new Borrelia sp. in
Amblyomma cohaerens collected from cattle in southwest
Ethiopia (21). And also the earlier finding of C. burnetii in
5.3% Amblyomma variegatum and in 10.8% Hyalomma
truncatum ticks before half of a centurary is the only
previous report from Ethiopia (28). Thus, in the current
study we intended to address the occurrence and molecular
identity of vector-borne bacteria including Rickettsia spp.,
Borrelia spp., Bartonella spp. and Coxiella burnetii in
several species of ixodid ticks actively feeding on domestic
animals in nine Districts in Oromia Regional State in
Ethiopia using molecular tools. For this purpose DNA
extracted by using standard techniques from ticks was first
screened individually for human pathogenic bacteria by
genus or species-specific genes and then when it was
necessary genes were amplified by standard PCR followed
by sequencing and then BLAST and phylogenetic analysis.
124
Results of our study revealed the presence of several species
of zoonotic bacteria such as Rickettsia africae the agent of
African Tick Bite Fever (ATBF), Rickettsia aeschlimannii
the agent of Spotted Fever (SF), Rickettsia massiliae the
agent of Spotted Fever (SF), Coxiella burnetii the agent of Q
fever, and new Borrelia spp., related and similar to the agent
of tick-borne relapsing fever (TBRF) in different ixodid tick
species collected from domestic animals in nine districts in
Ethiopia.
In Ethiopia morphological identification of ticks is not easy
due to the great diversity and high prevalence of
Rhipicephalus sanguineus group ticks which are closely
related and very similar species (13). Thus, to address this
issue we tried to identify ixodid ticks of Ethiopia by
employing morphological (12,42,43), protein profiling by
MALDI-TOF MS method (14,24,46) and molecular tools
using 12S RNA gene (2). Hence, after creating a reliable
reference database of spectra obtained from the legs of tick
specimens followed by blind test against the created
reference database we correctly identified 11 ixodid tick
species by the MALDI-TOF-MS method.
125
126
3.1. Article 2
127
136
Ticks and Tick-borne Diseases 6 (2015) 8–15
Original article
a r t i c l e i n f o a b s t r a c t
Article history: In Ethiopia, information on the transmission of human zoonotic pathogens through ixodid ticks remains
Received 16 May 2014 scarce. To address the occurrence and molecular identity of spotted fever group rickettsiae using molec-
Received in revised form 19 August 2014 ular tools, a total of 767 ixodid ticks belonging to thirteen different species were collected from domestic
Accepted 22 August 2014
animals from September 2011 to March 2014. Rickettsia africae DNA was detected in 30.2% (16/53) Ambly-
Available online 26 September 2014
ommma variegatum, 28.6% (12/42) Am. gemma, 0.8% (1/119) Am. cohaerens, 18.2% (4/22) Amblyomma
larvae, 6.7% (2/60) Amblyomma nymphs, 0.7% (1/139) Rhipicephalus (Boophilus) decoloratus and 25% (1/4)
Keywords:
nymphs of Rh. (Bo.) decoloratus. A markedly low prevalence of R. africae was recorded in both Am. cohaerens
Ixodid ticks
Domestic animals
and Rh. (Bo.) decoloratus (p < 0.0001) compared with that in Am. variegatum and Am. gemma. The preva-
Spotted fever rickettsiae lence of R. africae was markedly low in the western districts (Gachi and Abdela) (p < 0.0001); however, the
Oromia, Ethiopia prevalence of R. africae was relatively high in the central (Ada’a, Wolmara and Arsi) and eastern (Arero,
Moyale and Yabelo) districts, where Am. variegatum and Am. gemma were predominantly associated with
R. africae, respectively. R. aeschlimannii DNA was detected in 45.4% (5/11) Hyalomma marginatum rufipes
and 2.2% (1/46) Hy. truncatum. Moreover, the first report of R. massiliae DNA in 1.9% (1/52) Rhipicephalus
praetextatus ticks in Ethiopia is presented herein. Altogether, these results suggest that the transmission
of spotted fever group rickettsiae through ixodid ticks is a potential risk for human health in different
parts of Ethiopia. Clinicians in this country should consider these pathogens as a potential cause of febrile
illness in patients.
© 2014 Elsevier GmbH. All rights reserved.
Introduction the studies on the role of ixodid ticks as vectors of human pathogens
in Ethiopia were conducted before the discovery of well-advanced
Ticks are obligate hematophagous arthropods that act as reser- molecular tools during the 1950s and 60s (Burgdorfer et al., 1973;
voirs and vectors for a wide range of human and animal pathogens Philip et al., 1966). Indeed, the isolation of Rickettsia prowazekii, the
worldwide. Approximately 10% of the currently recognized tick causative agent of epidemic typhus, from Amblyomma ticks feeding
species carry human and animal pathogens. Currently, ticks and on Ethiopian cattle remains a mystery and had never been con-
mosquitoes are the major vectors for human and animal disease firmed in any other studies conducted in Ethiopia (Burgdorfer et al.,
agents (Jongejan and Uilenberg, 2004; Parola and Raoult, 2001). 1973).
Recent studies have indicated an increase in the spectrum of tick- Spotted fever group (SFG) rickettsiae are small, obligate intra-
borne pathogens affecting humans and animals (Dantas-Torres cellular, short rod, Gram-negative bacteria belonging to the genus
et al., 2012; de la Fuente and Estrada-Pena, 2012; Nicholson et al., Rickettsia, the family Rickettsiaceae and the order Rickettsiales
2010). (Parola et al., 2013). The genus Rickettsia comprises 3 main
Ticks are common and widely distributed throughout Ethiopia biogroups: the ‘spotted fever group’ (SFG), primarily transmitted
(Mekonnen et al., 2001). Several species of ixodid ticks have been through ixodid ticks, except R. felis and R. akari, which are vec-
identified in Ethiopia and are considered to have more veterinary tored through fleas and mites, respectively; the ‘typhus group’ (TG),
significance than medical importance (Kumsa et al., 2012). Most of transmitted through fleas and lice; and the ‘scrub typhus group’
(STG), primarily vectored through chiggers (Parola, 2011; Parola
et al., 2013). Currently, the genus Rickettsia comprises 31 species
that cause diseases in vertebrate hosts, including humans, domes-
∗ Corresponding author at: URMITE, UMR CNRS 7278, IRD 198, INSERM U1095,
tic animals, birds, and wildlife. Some ticks have also been implicated
Faculté de Médecine, 27 Bd Jean Moulin, 13385 Marseille cedex 5, France.
as SFG rickettsiae reservoirs, as these insects maintain rickettsiae
Tel.: +33 04 91 32 43 75; fax: +33 04 91 38 77 72.
E-mail address: philippe.parola@univ-amu.fr (P. Parola). both transstadially and transovarially.
http://dx.doi.org/10.1016/j.ttbdis.2014.08.001
1877-959X/© 2014 Elsevier GmbH. All rights reserved.
B. Kumsa et al. / Ticks and Tick-borne Diseases 6 (2015) 8–15 9
In humans, spotted fever rickettsia induces symptoms that Collection and identification of ticks
generally include fever, headache, myalgia, rash, local lym-
phadenopathy, and an eschar at the site of the tick bite, which Ticks were collected from September through November of
might be useful in diagnosis (Parola et al., 2013). Rickettsia africae, 2011. A thorough visual examination of the body surfaces of each
the causative agent of ATBF, typically causes symptoms, includ- study animal was conducted to establish the presence or absence
ing inoculation eschars, fever, regional lymphadenopathies and the of ticks. All observed ticks attached to the skin of each animal were
frequent lack of cutaneous rash or pale vesicular eruptions, and carefully removed using forceps or by hand to avoid any damage
the distribution of this pathogen is consistent with the geographi- to the body, and the specimen were individually placed into small,
cal distribution of Amblyomma ticks (Parola et al., 2013). Rickettsia pre-labeled plastic tubes containing 70% ethanol for subsequent
aeschlimannii is an emerging pathogen recently detected in human identification as previously described (Kumsa et al., 2012, 2014a).
patients in Morocco, South Africa, Algeria, Tunisia and Greece, All ticks from the same animal were placed into the same vial and
causing clinical symptoms resembling Mediterranean spotted fever transported to the Laboratory of the World Health Organization
caused by R. conorii (fever, generalized maculopapular rashes and Collaborative Center for Rickettsial Diseases and Arthropod-borne
an escar at the tick-bite site) (Parola et al., 2013). Rickettsia massil- Bacterial Diseases in Marseille, France.
iae is a pathogenic SFG rickettsia associated with clinical symptoms, The morphological identification of ticks and molecular studies
such as fever, eschar, night sweats, headache, maculopapular rash were performed from January 2012 through March 2014. All adult
and necrotic skin lesions, in human patients in America, Europe and ticks were identified at the species level, and the larvae and nymphs
Africa (Parola et al., 2013). were microscopically identified at the genus level using previously
Previous studies on rickettsiae in Ethiopian ixodid ticks have described morphological identification keys (Hoogstraal, 1956;
documented the presence of Rickettsia spp. in Amblyomma var- Walker et al., 2000, 2003). The sex and stage of each tick was deter-
iegatum, Amblyomma cohaerens and Rhipicephalus spp. in central mined, and photographs of the dorsal and ventral body parts of
Ethiopia (Philip et al., 1966) and Amblyomma spp. (Amblyomma each tick specimen were captured. The tick genera are abbreviated
gemma, Am. variegatum and Am. cohaerens) in the central and as previously described (Dantas-Torres, 2008).
eastern regions of Ethiopia (Burgdorfer et al., 1973). R. aeschli-
mannii, the agent of SFG rickettsiosis, had been detected in
DNA extraction from ticks
Hyalomma marginatum rufipes and R. africae, the agent of African
tick bite fever (ATBF), has been detected in Amblyomma lep-
Prior to DNA extraction, each tick specimen was rinsed twice in
idum and Am. variegatum ticks from eastern Ethiopia (Mura et al.,
sterile water for 15 min, dried on sterile filter paper, and longitu-
2008), and also R. africae had been detected in pools of Ambly-
dinally dissected into two equal halves. One half of each specimen
omma and Rhipicephalus ticks (Pader et al., 2012) (summarized in
was retained as reserve sample to avoid the risk of losing samples
Fig. 1).
for any reason during or after DNA extraction (Kumsa et al., 2014a,
In Ethiopia, where the environment is suitable for ticks, 84% of
2014b). Genomic DNA was individually extracted from a total of
the population is involved in agriculture and farmers and their fam-
767 tick specimens using the QIAamp DNA tissue extraction kit
ily members interact with ectoparasite-infested animals on daily
(Qiagen, Hilden, Germany) according to the manufacturer’s instruc-
life activities. Moreover, the confirmatory diagnosis of fever and
tions. In engorged tick DNA was extracted from small portion of
other diseases with unknown etiologies is not commonly practiced,
the anterior part of one half of the ticks so as to minimize the PCR
reflecting the lack of health centers in many regions of the coun-
inhibitory effect of large amount of blood in the abdomen. The DNA
try. Therefore, information concerning ectoparasite-borne bacteria
from each tick specimen was eluted in 100 �l of Tris–EDTA (TE)
is extremely important. To update the knowledge on tick-borne
buffer and stored at −20 ◦ C under sterile conditions to preclude
rickettsiae in Ethiopia, the aim of the present study was to address
contamination until the sample was used for PCR. To avoid cross-
the occurrence and molecular identity of Rickettsia species in ixo-
contamination among the samples during DNA extraction, the
did ticks collected from domestic animals in nine districts in Oromia
DNA extracting EZI Advanced XL Robot (Qiagen, Hilden, Germany)
Regional State in Ethiopia.
was thoroughly disinfected after each extraction according to the
manufacturer’s recommendations. The second half of each tick
Materials and methods specimen was stored at −80 ◦ C as a backup sample.
In this study, tick samples were collected from cattle, sheep, As a first step, all DNA samples were individually tested
dogs and cats in the following districts in Oromia Regional using a genus-specific qPCR system targeting the gltA gene
State in Ethiopia: Arsi (7◦ 56� 2.36�� N, 39◦ 39� 6.54�� E), Wol- (RKND03 system (Rolain et al., 2002): RKND03F-5� -GTG-AAT-
mara (9◦ 6� 17.87�� N, 38◦ 28� 8.50�� E), Kimbibit (9◦ 24� 39.95�� N, GAA-AGA-TTA-CAC-TAT-TTA-T-3� , RKND03R- 5� -GTA-TCT-TAG-
39◦ 21� 14.75�� E), Ada’a (9◦ 31� 60.00�� N, 38◦ 18� 0.00�� E), Bedele CAA-TCA-TTC-TAA-TAG-C-3� and RKND03P- 6-FAM-CTA-TTA-
(8◦ 27� 1.76�� N, 36◦ 21� 5.08�� E), Gachi (9◦ 14� 2.57�� N, 35◦ 54� 48.46�� E), TGC-TTG-CGG-CTG-TCG-GTT-C-TAMRA) in SFG rickettsiae and the
Arero (4◦ 43� 33.28�� N, 38◦ 45� 46.78�� E), Moyale (3◦ 31� 60.00�� N, Rpr274P gene in typhus group rickettsiae as previously described
39◦ 3� 0.00�� E) and Yabelo (4◦ 53� 41.91�� N, 38◦ 6� 0.59�� E). The (Mediannikov et al., 2010a; Socolovschi et al., 2012). Sterile water
districts are located in six zones in the central, southwestern was used as a negative control, and DNA from R. montanensis and
and southeast regions of the country, with various climates R. typhi were used as positive controls for SFG and typhus group
and agroecology. The livestock in the study areas are tradi- Rickettsia, respectively. All DNA samples positive for the gltA gene
tionally managed under extensive production systems (CSA, (RKND03 system) were further confirmed using species-specific
2008). genes for SFG Rickettsia mentioned below.
All the study animals were selected irrespective of sex and age. The Amblyomma spp. (n = 37) and Rhipicephalus (Boophilus)
The animals were categorized into two age groups, young (up to decoloratus (n = 2) DNA samples positive for SFG Rickettsia were fur-
one year) and adult (older than one year), according to a previous ther tested using a previously described R. africae species-specific
publication (Kumsa et al., 2012). qPCR targeting the ITS gene (Mediannikov et al., 2012a). Sterile
10 B. Kumsa et al. / Ticks and Tick-borne Diseases 6 (2015) 8–15
Fig. 1. Map of Ethiopia showing the areas in which SFG rickettsiae has previously been reported.
water was used as a negative control, and DNA from R. africae was the prevalence of Rickettsia DNA among different tick species in dif-
used as positive control. ferent districts. Statistical analysis was performed using EpiInfoTM 7
Hyalomma spp. DNA samples (n = 6) positive for SFG Rickettsia (CDC, 2008). The Mantel–Haenszel (MH) test in EpiInfoTM 7 was
were further tested using a previously described species-specific used to determine the relationships between the prevalence of
qPCR targeting the Sca1 gene (a cell-surface antigen) of R. aeschli- Rickettsia DNA in different tick species and genera and districts of
mannii with the same primers and probes used before (Socolovschi collection. A p-value of <0.05 was considered significant.
et al., 2012). Sterile water was used as a negative control, and DNA
from R. aeschlimannii was used as positive control.
Rhipicephalus sp. DNA sample (n = 1) positive for SFG Rickettsia Results
was further tested using a previously described species-specific
qPCR targeting hypothetical proteins of R. massiliae with the same Ticks were collected from a total of 244 animals (206 cattle, 29
primers and probes used before (Socolovschi et al., 2012). Sterile sheep, seven dogs and two cats) in nine districts: Abdela (37 cattle
water was used as a negative control, and DNA from R. massiliae and fifteen sheep), Gachi (29 cattle and three sheep), Ada’a (nine-
was used as positive control. teen cattle, seven dogs and two cats), Wolmara (seventeen cattle
and one sheep), Kimbibit (23 cattle and ten sheep), Arsi (31 cat-
tle), Arero (23 cattle), Moyale (eighteen cattle) and Yabelo (nine
Ethical statement cattle). Overall, 329 male ticks, 314 female (260 non-engorged
and 54 engorged) ticks, 98 nymphs and 26 larvae of ticks were
Ethical approval for the collection of ticks from domestic collected. Greater proportion of the ticks was collected from cat-
animals was obtained from the animal research ethics board tle 89.9% (689/767) while lower proportion was collected from
(Agreement # 14/160/550/2011) of the College of Veterinary sheep 6.4% (49/767), from dogs 3% (23/767) and from cats 0.8%
Medicine and Agriculture at Addis Ababa University. All neces- (6/767). From each infested animal from one up to ten ticks were
sary oral permits were obtained for the field studies, including collected. Further information on tick collection is presented on
permission from each animal owner and the administration and Table 1.
agricultural office of each district. The collection of ticks from ani- A total of 767 ixodid ticks comprising thirteen different species
mals was not harmful or contrary to the welfare of the animals. (Table 2) were screened for both spotted fever and typhus group
rickettsiae DNA using molecular methods. The overall prevalence
Data analysis of SFG rickettsiae was 6% (46/767) in all ticks collected from nine
Ethiopian districts. In the study 6.5% (45/689) of the ticks from cattle
Microsoft Excel was used for data management. Data were strat- and 16.7% (1/6) tick from one cat were positive for spotted fever
ified by districts for analysis and then descriptive statistics such as group rickettsiae, however; spotted fever group rickettsiae was not
percentages and means, were employed to estimate and compare detected in ticks taken from sheep (0/49) and dogs (0/23).
B. Kumsa et al. / Ticks and Tick-borne Diseases 6 (2015) 8–15 11
Table 1
Summary of species of ticks collected from different domestic animal in 9 districts in Oromia, Ethiopia.
Tick spp. District Animal spp. No. of ticks No. of each tick No. of
(number) collected spp. collected engorged ticks
(m = male, from one
f = female) animal
Amblyomma cohaerens (n = 119) Abdela Cattle (19) 39 (23m, 16f) 1–3 4 female
Sheep (8) 10 (6m, 4f) 1–2
Gachi Cattle (24) 49 (28m, 21f) 1–3 4 female
Sheep (1) 2 (1m, 1f) 2
Arsi Cattle (2) 2 (2m) 1
Arero Cattle (9) 12 (8m, 4f) 1–2 1 female
Moyale Cattle (2) 3 (2m, 1f) 1–2
Yabelo Cattle (1) 2 (1m, 1f) 2
Amblyomma gemma (n = 28) Ada’a Cattle (1) 1 (1m) 1
Arero Cattle (8) 13 (7m, 6f) 1–2 2 female
Moyale Cattle (11) 20 (10m, 10f) 1–2 2 female
Yabelo Cattle (5) 8 (4m, 4f) 1–2 1 female
Amblyomma lepidum (n = 2) Arero Cattle (1) 1 (1f) 1
Moyale Cattle (1) 1 (1f) 1
Amblyomma variegatum (n = 119) Gachi Cattle (3) 3 (3m) 1
Sheep (2) 2 (2m) 1
Ada’a Cattle (9) 13 (10m, 3f) 1–2
Wolmara Cattle (6) 7 (6m, 1f) 1–2 1 female
Arsi Cattle (19) 25 (20m, 5f) 1–2 2 female
Arero Cattle (2) 2 (2m) 1
Yabelo Cattle (1) 1 (1m) 1
Amblyomma spp. larva (n = 22) Abdela Cattle (8) 8 1
Sheep (2) 2 1
Gachi Cattle (5) 5 1
Ada’a Cattle (4) 4 1
Wolmara Cattle (2) 2 1
Yabelo Cattle (1) 1 1
Amblyomma spp. nymph (n = 60) Abdela Cattle (13) 13 1
Sheep (12) 13 1–2
Gachi Cattle (11) 11 1
Sheep (2) 2 1
Ada’a Cattle (6) 6 1
Cat (1) 1 1
Wolmara Cattle (1) 1 1
Arsi Cattle (10) 10 1
Arero Cattle (1) 1 1
Moyale Cattle (1) 1 1
Yabelo Cattle (1) 1 1
Rhpicephalus (Boophilus) decoloratus (n = 139) Abdela Cattle (14) 14 (14f) 1 3 female
Gachi Cattle (12) 14 (2m, 12f) 1–2 1 female
Sheep (1) 1 (1f) 1
Ada’a Cattle (8) 11 (3m, 8f) 1–2
Wolmara Cattle (15) 31 (16m, 15f) 1–3 2 female
Kimbibit Cattle (15) 20 (7m, 13f) 1–2 1 female
Sheep (1) 3 (3f) 3 2 female
Arsi Cattle (24) 31 (6m, 25f) 1–2 8 female
Arero Cattle (10) 10 (10f) 1
Moyale Cattle (3) 3 (3f) 1
Yabelo Cattle (1) 1 (1f) 1
Rhipicephalus (Boophilus)decoloratus larva (n = 4) Ada’a Cattle (2) 2 1
Kimbibit Cattle (2) 2 1
Rhipicephalus(Boophilus)decoloratus nymph (n = 31) Abdela Cattle (3) 3 1
Gachi Cattle (6) 6 1
Ada’a Cattle (4) 4 1
Wolmara Cattle (11) 11 1
Kimbibit Cattle (6) 6 1
Arero Cattle (1) 1 1
Rhipicephalus praetextatus (n = 52) Ada’a Cattle (7) 9 (4m, 5f) 1–2
Kimbibit Cattle (16) 23 (12m, 11f)) 1–3
Sheep (9) 12 (5m, 7f) 1–2
Arsi Cattle (6) 8 (5m, 3f) 1–2
Rhipicephalus eversti evetsi (n = 37) Abdela Cattle (1) 1(1m) 1
Gachi Cattle (3) 3 (2m, 1f) 1
Sheep (1) 1 (1f) 1
Ada’a Cattle (5) 8 (5m, 3f) 1–2
Wolmara Cattle (2) 4 (2m, 2f) 2
Sheep (1) 1 (1f) 1
Kimbibit Cattle (4) 7 (4m, 3f) 1–2
Arsi Cattle (6) 10 (6m, 4f) 1–2
Arero Cattle (1) 1 (1f) 1
Moyale Cattle (1) 1 (1f) 1 1 female
12 B. Kumsa et al. / Ticks and Tick-borne Diseases 6 (2015) 8–15
Table 1 (Continued)
Tick spp. District Animal spp. No. of ticks No. of each tick No. of
(number) collected spp. collected engorged ticks
(m = male, from one
f = female) animal
R. africae DNA was detected in 30.2% (16/53) Am. variegatum, recent scientific studies, and the results were confirmed through
28.6% (12/42) Am. gemma, 0.8% (1/119) Am. cohaerens, 18.2% (4/22) the amplification of two different target genes for each positive
Amblyomma larvae, 6.7% (4/60) Amblyomma nymphs, 0.7% (1/139) result using positive and negative controls as previously described
Rh. (Bo.) decoloratus and 25% (1/4) nymphs of Rh. (Bo.) decoloratus (Mediannikov et al., 2012a; Socolovschi et al., 2012).
(Table 2). The overall prevalence of R. africae was significantly The absence of spotted fever group rickettsiae in ticks collected
higher in both Am. variegatum and Am. gemma than in the other tick from sheep and dogs most probably reflect the very lower percent-
species (Am. cohaerens and Rh (Bo.) decoloratus) (Mantel–Haenszel age of sheep 11.9% (29/244) and dogs 2.9% (7/244) as compared
(MH), p < 0.0001). In the southwestern districts (Gachi and Abdela), to the very higher percentage of ticks from cattle as well as the
the overall prevalence of R. africae was markedly low compared greater number of cattle 84.4%(206/244) studied. Future studies
with the central and southeastern districts (MH, p < 0.0001). In the with representative number of domestic animals and their ticks are
central districts (Ada’a, Wolmara and Arsi), Am. variegatum was necessary to address the comparative importance of animal species
predominantly associated with R. africae, whereas Am. gemma was and their ticks in Ethiopia.
predominantly associated with this species in the southeastern dis- The detection of R. africae, the agent of African tick bite fever,
tricts (Arero, Moyale and Yabelo) (Fig. 2). This study is the first to in Am. variegatum confirms the results of a previous report associ-
report R. africae in Am. gemma, Rh (Bo). decoloratus, in larval and ating R. africae with this Ethiopian tick species (Mura et al., 2008).
nymphal ticks in Ethiopia. Statistically significant variation was not Similarly, this spotted fever group rickettsia, erroneously consid-
found (MH, p < 0.07) in the prevalence of R. africae between 3/54 ered as R. conorii, the agent of Mediterranean spotted fever, was
(5.5%) engorged and 4/260 (1.5%) non engorged female ticks. reported in Am. variegatum ticks from Ethiopia approximately 41
R. aeschlimannii DNA was detected in 45.4% (5/11) Hy. m. rufipes years ago (Burgdorfer et al., 1973; Philip et al., 1966). This erro-
and 2.2% (1/46) Hyalomma truncatum (Table 2). The prevalence of neous consideration of spotted fever group rickettsia as R. conorii,
R. aeschlimannii was significantly higher in Hy. m. rufipes than in vectored by Rhipicephalus spp., at that time in Ethiopia was due
Hy. truncatum ticks (MH, p = 0.001). to the fact that new techniques such as species sensitive and spe-
R. massiliae DNA was detected in 1.9% (1/52) Rhipicephalus prae- cific serology, shell vial assay and molecular methods that enabled
textatus ticks (Table 2). the isolation, identification and characterization of most of these
Spotted fever group rickettsia DNA was not detected in Haema- bacteria including R. africae, vectored principally by Amblyomma
physalis (Hae. leachi (n = 6) and Hae. spinulosa (n = 2) ticks collected spp., were developed only during the last 20 years after the 1990
from domestic animals. Similarly, SFG rickettsiae were not detected (Raoult et al., 2001). In consistent with our observation R. africae
in Am. lepidum, Rhipicephalus evertsi evertsi, Rhipicephalus pulchellus has been detected in Am. variegatum ticks collected from cattle in
and Rhipicephalus sanguineus ticks (Table 2). several African countries including in (7/26) in Kenya (Mutai et al.,
We did not detect any typhus group rickettsia DNA in the 767 2013), in (22/39) Uganda and 22/141 Nigeria (Lorusso et al., 2013),
ixodid ticks collected from domestic animals in Oromia. in 4–8% in Nigeria (Reye et al., 2012), in Senegal (Mediannikov et al.,
2010b), in 11/12 in Sudan (Morita et al., 2004) and in 6/6 Mali, 6/6
Discussion Niger and 1/13 Burundi (Parola et al., 2001). Similarly, in support of
our finding R. africae was detected in A. gemma ticks collected from
The overall prevalence of pathogenic SFG rickettsiae was cattle in Kenya (Mutai et al., 2013). Results of our study suggest that
detected in 6% (46/767) of ixodid ticks belonging to four genera A. gemma is potentially an important vector for R. africae in addition
(Amblyomma, Rhipicephalus and Hyalomma) collected from ani- to the already known Am. variegatum in Ethiopia. The identification
mals in nine districts, highlighting the importance of hard ticks for of R. africae in Am. gemma and Rh. (Bo.) decoloratus in the present
human health in Ethiopia. Our molecular strategy was based on study expands the current knowledge concerning tick species that
Table 2
Prevalence of SFG rickettsiae in ixodid ticks using species-specific gene qPCR in 9 districts in Oromia, Ethiopia.
Zones IluAba Bora East Showa West Showa North Showa Arsi Borana Total
District Abdela Gachi Ada’a Wolmara Kimbibit Arsi Arero Moyale Yabelo
Fig. 2. Map of Ethiopia showing the geographical distribution of SFG rickettsiae and their host ticks in the present study.
host R. africae in Ethiopia. The absence of significant variations in of 298 Amblyomma ticks comprising four different species in nine
the prevalence of R. africae between engorged and none engorged districts in Oromia; however, in the previous study (Mediannikov
female ticks most probably reflects the avoiding of lager portion of et al., 2013), only 109 Amblyomma ticks representing Am. cohaerens
abdomen of engorged ticks containing large blood. in a single district in southwest Ethiopia were studied.
We observed a strong geographic correlation between the high The low prevalence of R. africae in 1.1% Rh. (Bo.) decoloratus
prevalence of R. africae (in Am. variegatum in the central districts compared with the prevalence of this species in Amblyomma ticks
and Am. gemma in the southeastern districts) and the distribution (12.4%) is consistent with the argument that the rate of R. africae
of different Amblyomma species across Oromia, consistent with the infection in Rhipicephalus ticks is typically low, reflecting con-
previously established geographical distribution of Am. variegatum comitant co-feeding with Amlyomma ticks (Parola et al., 2013). In
as the predominant Amblyomma spp. in central Ethiopia (Kumsa addition, Rh. (Bo.) decoloratus is infected only when cohabiting with
et al., 2012; Mekonnen et al., 2001) and Am. gemma as the predom- Amblyomma ticks, regarded as the primary reservoir of R. africae
inant Amblyomma spp. in the southeastern and eastern regions of (Macaluso et al., 2003; Mediannikov et al., 2013). R. africae has also
Oromia, Ethiopia (Regassa, 2001). Am. cohaerens is the predomi- been reported in Rh (Bo.) decoloratus ticks in Liberia (Mediannikov
nant tick species in western Ethiopia, where the climate is humid et al., 2012b) and Nigeria (Ogo et al., 2012).
and wet for most of the year, whereas Am. variegatum is the pre- The prevalence of R. aeschlimannii in 45.4% of Hy. m. rufipes
dominant tick species in areas with high rainfall in central and ticks confirms a previous report in eastern Ethiopia (Mura et al.,
other regions of Ethiopia (Mekonnen et al., 2001), and Am. gemma 2008) and previous studies in other countries, reporting a preva-
is restricted to the semi-arid regions east of Rift Valley in Ethiopia lence of 44.8% in Senegal (Mediannikov et al., 2010a), 60% in Corsica
(Regassa, 2001). The high prevalence of R. africae in Am. gemma has (Matsumoto et al., 2004), 33.3% in Niger and 15% in Mali (Parola
also been reported in Kenya (Mutai et al., 2013). et al., 2001) and Camargue in southern France (Socolovschi et al.,
The markedly low prevalence of 1.9% (3/159) R. africae in west- 2012). R. aeschlimannii has been detected in Hy. truncatum in Sene-
ern Oromia (Gachi and Abdela districts), where Am. cohaerens is the gal (Mediannikov et al., 2010a), Sudan (Morita et al., 2004) and
predominant Amblyomma spp. (Kumsa et al., 2012), compared with Kenya (Mutai et al., 2013). The significantly lower prevalence of
that in the central (Ada’a, Wolmara and Arsi) (21.6%, 22/102) and R. aeschlimannii in Hy. truncatum suggests that Hy. m. rufipes is
southeastern (Arero, Moyale and Yabelo) (17.9%, 12/67) districts is the principal vector of this bacterium, as previously suggested
consistent with previous reports of low levels of SFG rickettsiae in (Demoncheaux et al., 2012; Parola et al., 2013).
Am. cohaerens compared with Am. variegatum and Am. gemma ticks Interestingly, this study is the first to report R. massiliae in 1.9%
in Ethiopia (Burgdorfer et al., 1973; Philip et al., 1966). In contrast, (1/52) Rh. praetextatus in Ethiopia. Consistent with this observation,
the absence of R. africae DNA in 109 Am. cohaerens has been R. massiliae has been previously detected in several Rhipicephalus
recently reported in southwestern Ethiopia (Mediannikov et al., spp. in other African countries, including 8.1% (2/37) in Rh. muh-
2013). This inconsistency likely reflects differences in the number samae in Mali (Parola et al., 2001), 8.2% (5/61) in Rh. senegalensis in
of Amblyomma ticks tested and the geographical locations exam- Guinea (Mediannikov et al., 2012b) and in 22.4% (11/49) in Rh. guil-
ined in the two studies. In the present study, we examined a total honi in Senegal (Mediannikov et al., 2010a). Rh. praetextatus is one of
B. Kumsa et al. / Ticks and Tick-borne Diseases 6 (2015) 8–15 15
the predominant tick species observed on ruminants in the central, Kumsa, B., Parola, P., Raoult, D., Socolovschi, C., 2014a. Bartonella melophagi in
north and eastern regions of Ethiopia (Mekonnen et al., 2001). Melophagus ovinus (sheep ked) collected from sheep in northern Oromia,
Ethiopia. Comp. Immunol. Microbiol. Infect. Dis. 37, 69–76.
Although published reports are not available on human hard Kumsa, B., Parola, P., Raoult, D., Socolovschi, C., 2014b. Molecular detection of Rick-
tick infestations in Ethiopia, Am. variegatum, Am. cohaerens and Rh. ettsia felis and Bartonella henselae in dog and cat fleas in Central Oromia, Ethiopia.
praetextatus have been reported to feed on humans (Dantas-Torres Am. J. Trop. Med. Hyg. 90, 457–462.
Lorusso, V., Gruszka, K.A., Majekodunmi, A., Igweh, A., Welburn, S.C., Picozzi, K., 2013.
et al., 2012; Jongejan and Uilenberg, 2004; Parola and Raoult, 2001). Rickettsia africae in Amblyomma variegatum ticks, Uganda and Nigeria. Emerg.
The significance of the present study is further strengthened Infect. Dis. 19, 1705–1707.
through recent reports of SFG rickettsiae in humans in Ethiopia, Macaluso, K.R., Davis, J., Alam, U., Korman, A., Rutherford, J.S., Rosenberg, R., Azad,
A.F., 2003. Spotted fever group rickettsiae in ticks from the Masai Mara region
including R. africae in a French man who stayed for 2 months in
of Kenya. Am. J. Trop. Med. Hyg. 68, 551–553.
western Ethiopia (Stephany et al., 2009) and 2.9% (3/102) SFG Rick- Matsumoto, K., Parola, P., Brouqui, P., Raoult, D., 2004. Rickettsia aeschlimannii in
ettsia DNA in dried thin blood smears prepared to test for malaria Hyalomma ticks from Corsica. Eur. J. Clin. Microbiol. Infect. Dis. 23, 732–734.
Mediannikov, O., Abdissa, A., Socolovschi, C., Diatta, G., Trape, J.F., Raoult, D., 2013.
in children with febrile illnesses at Soddo Christian Hospital in
Detection of a new Borrelia species in ticks taken from cattle in Southwest
Wolaitta Soddo in southern Ethiopia (Aarsland et al., 2012). Ethiopia. Vector Borne Zoonotic Dis. 13, 266–269.
The absence of any typhus group rickettsiae in ixodid ticks Mediannikov, O., Davoust, B., Socolovschi, C., Tshilolo, L., Raoult, D., Parola, P., 2012a.
in this study is consistent with previous studies that did not Spotted fever group rickettsiae in ticks and fleas from the Democratic Republic
of the Congo. Ticks Tick Borne Dis. 3, 371–373.
detect this bacterium in hard ticks in Ethiopia (Mediannikov et al., Mediannikov, O., Diatta, G., Fenollar, F., Sokhna, C., Trape, J.F., Raoult, D., 2010a. Tick-
2012a; Mura et al., 2008; Pader et al., 2012), suggesting that other borne rickettsioses, neglected emerging diseases in rural Senegal. PLoS Negl.
hematophagous arthropods might play a role in the transmission Trop. Dis., 4.
Mediannikov, O., Diatta, G., Zolia, Y., Balde, M.C., Kohar, H., Trape, J.F., Raoult, D., 2012.
of TG rickettsiae. Tick-borne rickettsiae in Guinea and Liberia. Ticks Tick Borne Dis. 3, 43–48.
In conclusion, the findings presented herein provide additional Mediannikov, O., Trape, J.F., Diatta, G., Parola, P., Fournier, P.E., Raoult, D., 2010b.
clear information on the geographic distribution and SFG rick- Rickettsia africae, Western Africa. Emerg. Infect. Dis. 16, 571–573.
Mekonnen, S., Hussein, I., Bedane, B., 2001. The distribution of ixodid ticks (Acari:
ettsia species detected in different ixodid ticks in various regions of Ixodidae) in central Ethiopia. Onderstepoort J. Vet. Res. 68, 243–251.
Ethiopia. These results suggest that physicians managing patients Morita, C., El Hussein, A.R., Matsuda, E., Abdel Gabbar, K.M., Muramatsu, Y., Abdel
with fever of unknown etiology in Ethiopia and those who care for Rahman, M.B., Eleragi, A.M., Hassan, S.M., Chitambo, A.M., Ueno, H., 2004. Spot-
ted fever group rickettsiae from ticks captured in Sudan. Jpn. J. Infect. Dis. 57,
travelers from Ethiopia should consider the presence of several SFG
107–109.
rickettsiae as potential causative agents. In addition, future studies Mura, A., Socolovschi, C., Ginesta, J., Lafrance, B., Magnan, S., Rolain, J.M., Davoust, B.,
are needed to address the isolation and culture of SFG rickettsiae Raoult, D., Parola, P., 2008. Molecular detection of spotted fever group rickettsiae
in ticks from Ethiopia and Chad. Trans. R. Soc. Trop. Med. Hyg. 102, 945–949.
from ticks collected from animals and vegetation (questing ticks),
Mutai, B.K., Wainaina, J.M., Magiri, C.G., Nganga, J.K., Ithondeka, P.M., Njagi, O.N.,
and in blood samples from human, domestic and other reservoir Jiang, J., Richards, A.L., Waitumbi, J.N., 2013. Zoonotic surveillance for rickettsiae
animals and determine the public health significance of this bac- in domestic animals in Kenya. Vector Borne Zoonotic Dis. 13, 360–366.
terium in Ethiopia. Nicholson, W.L., Allen, K.E., McQuiston, J.H., Breitschwerdt, E.B., Little, S.E., 2010.
The increasing recognition of rickettsial pathogens in dogs and people. Trends
Parasitol. 26, 205–212.
Ogo, N.I., de Mera, I.G., Galindo, R.C., Okubanjo, O.O., Inuwa, H.M., Agbede, R.I., Torina,
Conflict of interest
A., Alongi, A., Vicente, J., Gortazar, C., de la Fuente, J., 2012. Molecular identifica-
tion of tick-borne pathogens in Nigerian ticks. Vet. Parasitol. 187, 572–577.
The authors declare that no competing interests exist. Pader, V., Nikitorowicz, B.J., Abdissa, A., Adamu, H., Tolosa, T., Gashaw, A., Cutler, R.R.,
Cutler, S.J., 2012. Candidatus Rickettsia hoogstraalii in Ethiopian Argas persicus
ticks. Ticks Tick Borne Dis. 3, 338–345.
Acknowledgments Parola, P., 2011. Rickettsia felis: from a rare disease in the USA to a common cause of
fever in sub-Saharan Africa. Clin. Microbiol. Infect. 17, 996–1000.
Parola, P., Inokuma, H., Camicas, J.L., Brouqui, P., Raoult, D., 2001. Detection and
The authors would like to thank the domestic animal owners identification of spotted fever group Rickettsiae and Ehrlichiae in African ticks.
in the different districts of Oromia for allowing the collection of Emerg. Infect. Dis. 7, 1014–1017.
ticks from their animals. We would also like to thank the laboratory Parola, P., Paddock, C.D., Socolovschi, C., Labruna, M.B., Mediannikov, O., Kernif, T.,
Abdad, M.Y., Stenos, J., Bitam, I., Fournier, P.E., Raoult, D., 2013. Update on tick-
technicians from URMITE, Marseille, particularly Veronique Brice borne rickettsioses around the world: a geographic approach. Clin. Microbiol.
and Annick Bernard, for technical support. Rev. 26, 657–702.
Parola, P., Raoult, D., 2001. Ticks and tickborne bacterial diseases in humans: an
emerging infectious threat. Clin. Infect. Dis. 32, 897–928.
References Philip, C.B., Hoogstraal, H., Reiss-Gutfreund, R., Clifford, C.M., 1966. Evidence of rick-
ettsial disease agents in ticks from Ethiopian cattle. Bull. World Health Org. 35,
Aarsland, S.J., Castellanos-Gonzalez, A., Lockamy, K.P., Mulu-Droppers, R., Mulu, M., 127–131.
White, A.C., Cabada, M.M., 2012. Treatable bacterial infections are underrecog- Raoult, D., Fournier, P.E., Fenollar, F., Jensenius, M., Prioe, T., de Pina, J.J., Caruso,
nized causes of fever in Ethiopian children. Am. J. Trop. Med. Hyg. 87, 128–133. G., Jones, N., Laferl, H., Rosenblatt, J.E., Marrie, T.J., 2001. Rickettsia africae, a
Burgdorfer, W., Ormsbee, R.A., Schmidt, M.L., Hoogstraal, H., 1973. A search for the tick-borne pathogen in travelers to sub-Saharan Africa. N. Engl. J. Med. 344,
epidemic typhus agent in Ethiopian ticks. Bull. World Health Org. 48, 563–569. 1504–1510.
CDC, 2008. Centers for Disease Control and Prevention 1600 Clifton Rd. Atlanta, GA Regassa, A., 2001. Tick infestation of Borana cattle in the Borana Province of Ethiopia.
30333, USA 800-CDC-INFO (800-232-4636) TTY: (888)., pp. 232–6348. Onderstepoort J. Vet. Res. 68, 41–45.
CSA, 2008. Addis Ababa, Ethiopia, 2007/08. Reye, A.L., Arinola, O.G., Hubschen, J.M., Muller, C.P., 2012. Pathogen prevalence
Dantas-Torres, F., 2008. Towards the standardization of the abbreviations of genus in ticks collected from the vegetation and livestock in Nigeria. Appl. Environ.
names of ticks (Acari: Parasitiformes: Ixodida). Vet. Parasitol. 154, 94–97. Microbiol. 78, 2562–2568.
Dantas-Torres, F., Chomel, B.B., Otranto, D., 2012. Ticks and tick-borne diseases: a Rolain, J.M., Stuhl, L., Maurin, M., Raoult, D., 2002. Evaluation of antibiotic suscepti-
one health perspective. Trends Parasitol. 28, 437–446. bilities of three rickettsial species including Rickettsia felis by a quantitative PCR
de la Fuente, J., Estrada-Pena, A., 2012. Ticks and tick-borne pathogens on the rise. DNA assay. Antimicrob. Agents Chemother. 46, 2747–2751.
Ticks. Tick. Borne. Dis. 3, 115–116. Socolovschi, C., Reynaud, P., Kernif, T., Raoult, D., Parola, P., 2012. Rickettsiae of spot-
Demoncheaux, J.P., Socolovschi, C., Davoust, B., Haddad, S., Raoult, D., Parola, P., ted fever group, Borrelia valaisiana, and Coxiella burnetii in ticks on passerine
2012. First detection of Rickettsia aeschlimannii in Hyalomma dromedarii ticks birds and mammals from the Camargue in the south of France. Ticks Tick Borne
from Tunisia. Ticks Tick Borne Dis. 3, 398–402. Dis. 3, 355–360.
Hoogstraal, H., 1956. African Ixodidae. 1. Ticks of the Sudan (with special reference Stephany, D., Buffet, P., Rolain, J.M., Raoult, D., Consigny, P.H., 2009. Rickettsia africae
to equatorial province and with preliminary reviews of the genera Boophilus, infection in man after travel to Ethiopia. Emerg. Infect. Dis. 15, 1867–1869.
Margaropus and Hyalomma). Department of the Navy, Bureau of the Medicine Walker, A.R., Bouattour, A., Camicas, J.L., Estrada-Pena, A., Horak, I.C., Latif, A.A.,
and Surgery, United States Government Printing Office, Washington, DC. Pegram, R.G., Preston, P.M., 2003. Tick of Domestic Animals in Africa: A Guide to
Jongejan, F., Uilenberg, G., 2004. The global importance of ticks. Parasitology (Suppl. Identification of Species. Bioscience Reports, Edinburgh, Scotland, UK.
129), S3–S14. Walker, J.B., Keirans, J.E., Horak, I.G., 2000. The Genus Rhipicephalus (Acari, Ixodi-
Kumsa, B., Beyecha, K., Geloye, M., 2012. Ectoparasites of sheep in three agro- dae): A Guide to the Brown Ticks of the World. Cambridge University Press,
ecological zones in central Oromia, Ethiopia. Onderstepoort J. Vet. Res. 79, E1–E7. Cambridge.
3.2. Article 3
137
138
Manuscript
Click here to download Manuscript: C. burnetii in ixodid ticks in Oromia for PLOS Neglected Tropical Diseases__final.docx
3 Bersissa Kumsa, Cristina Socolovschi, Lionel Almeras, Didier Raoult, Philippe Parola *
6 Emergentes (URMITE), UM63, CNRS 7278, IRD 198 (Dakar, Sénégal), Inserm 1095, WHO
7 collaborative center for rickettsioses and other arthropod borne bacterial diseases, Faculté de
12 13385 Marseille cedex 5, France. Phone: + 33 (0) 4 91 32 43 75. Fax: + 33 (0)4 91 38 77 72.
14
15
16
17
18
19
21
22
23
24
25
1
26 Abstract
27 Background: Although ticks are widely distributed in all parts of Ethiopia, there is a paucity
28 of information on Coxiella burnetii in hard ticks from this country. The present study was
29 conducted from September 2011 to March 2014 to address the occurrence and genotypes of
30 C. burnetii using molecular methods in a total of 842 ticks collected from domestic animals.
32 tested for Coxiella burnetii by quantitative (q) real time PCR targeting two different genes
34 prevalence of 6.4% (54/842) of C. burnetii was recorded. Coxiella burnetii was detected in
39 C. burnetii DNA were observed in Am. gemma and Rh. pulchellus than in other tick species
41 higher (MH, p<0.0001) in ticks from southeastern districts (Arero, Moyale and Yabelo) than
42 from other districts. This study demonstrated the presence of C. burnetii genotype MST 18 in
43 ticks in southeastern districts, and genotype MST 20 was found in ticks in central districts.
45 of C. burnetii in Ethiopia. Clinicians in Ethiopia and travelers returning from Ethiopia with
46 fevers of unknown etiology should consider C. burnetii as one of the possible causes of
48
50
2
51 Author Summary
52 In Ethiopia, there is lack of information on Coxiella burnetii in hard ticks. Thus, the present
53 study was conducted to determine the occurrence and genotypes of Coxiella burnetii in ticks
54 collected from domestic animals in Ethiopia using molecular tools. Our study recorded an
55 overall frequency of 6.4% (54/842) C. burnetii in ticks. Coxiella burnetii was detected in
59 frequencies of C. burnetii DNA were observed in Am. gemma and Rh. pulchellus than in the
60 other tick species (Mantel-Haenszel (MH), p<0.0001). The overall frequency of C. burnetii
61 was significantly higher (MH, p<0.0001) in ticks from southeastern districts (Arero, Moyale
62 and Yabelo) than in other districts in Oromia. Our study showed the circulation of C. burnetii
63 genotype MST 18 in ticks in southeastern districts, and genotype MST 20 was found in ticks
64 in central districts. This study highlights the possible importance of ticks in the epidemiology
65 of C. burnetii in Ethiopia.
66
67
68
69
70
71
72
73
74
75
3
76 Introduction
77 Ticks are one of the major causes of health problems in both humans and animals [1]. Ticks
78 are vectors and reservoirs for several emerging and reemerging infectious pathogens of
79 medical and veterinary importance [2,3]. In Ethiopia, ticks are widely distributed in all agro-
80 ecological zones [4]. Most previously published data on ticks in Ethiopia focus on species
81 distribution, and thus only a very few molecular studies have been conducted on the
83 Coxiella burnetii, the causative agent of Q fever, is a small (3-5 µm) polymorphic obligate
86 mainly affects macrophages. It is able to build highly infective spore-like forms that are
87 resistant to environmental influences and therefore can stay infective for several months.
88 Thus, the bacterium is highly resistant to environmental stresses, such as high temperature,
89 osmotic pressure, and ultraviolet light. It can also survive standard disinfectants and is
90 resistant to many other environmental factors [5]. Due to its cosmopolitan distribution, high
92 symptoms and clinical signs and low infectious dose (1-10 bacteria), C. burnetii is considered
97 syndrome [7]. Syndromes, such as endocarditis, osteomyelitis, septic arthritis, infected aortic
98 aneurysms, chronic hepatitis or chronic pneumonia, have been described as the chronic type
99 of Q fever. In pregnant woman, infection may lead to premature delivery or abortion [5,8].
100 Farmers, veterinarians, abattoir and dairy workers and laboratory technicians are populations
4
101 that are at high risk of contracting infection as an occupational zoonosis [9]. Although
102 inhalation of C. burnetii is considered the main means of infection in humans, consumption
103 of raw milk and milk-products, transplacental infection, intradermal inoculation, blood
104 transfusion, and contact with urine, feces and semen of infected animals are also regarded as
106 major cause of environmental contamination and the source for human infection [5,7].
107 In animals, C. burnetii is known to cause infection in a wide range of species, including
108 ruminants, companion animals, birds and reptiles [5,7]. However, domestic ruminants are
109 implicated as the common source of infection for humans. As is the case of humans,
110 infections in animals are generally asymptomatic [9]. In small ruminants, C. burnetii causes
111 reproductive disorders, such as abortions, premature delivery, delivery of weak offspring,
112 stillbirth in the final stage of gestation, postpartum metritis and infertility [9,10]. In cattle, C.
113 burnetii generally causes less obvious clinical signs, but recent studies have shown that
114 abortion and irregular repeat breeding, metritis and infertility are important risk indicators. In
115 ruminants, C. burnetii is shed via birth fluids and the placenta during normal parturition or
116 abortion and in vaginal mucus, milk, feces, urine and semen. Inhalation or ingestion of C.
117 burnetii with contaminated feed and water from the environment is the major source of
119 Hard and soft ticks are one of the most important arthropods known to be naturally infected
120 with C. burnetii [5,6,8]. So far, C. burnetii has been reported in more than 40 different tick
121 species in several countries [5]. Ticks acquire C. burnetii during a blood feeding on infected
122 animals and can transmit the bacterium to other mammals during the next blood meal or by
123 aerogenic spread of dried tick fecal excretions, thus playing a role in the maintenance of C.
124 burnetii in the environment [11]. In infected ticks, C. burnetii multiplies in the middle gut
125 cells resulting in high titers of viable organisms that are expelled with feces. Ticks are
5
126 considered important reservoirs and potential vectors due to both transstadial and transovarial
127 transmission of C. burnetii to their offspring in some tick species [5,6]. Furthermore, some
128 previous authors have postulated that there is an increase in the virulence of C. burnetii after
129 passage through ticks. Ticks play an important role in the epidemiology of Q fever by
130 contaminating the environment by excreting C. burnetii via their feces, saliva and coxal
131 fluids. However, no evidence is available reporting the transmission of C. burnetii to humans
132 by blood-feeding ticks [6]. Recently, C. burnetii DNA has been detected in 8.3% of
133 Ornithodoros sonrai in Senegal [11], in fleas in Cyprus [12] and Egypt [13], and in biting flies
135 In sub-Saharan African countries, including Ethiopia, the importance of C. burnetii as a cause
136 of febrile illness in humans is poorly understood due to the lack of capacity of developed
137 laboratories [15]. Included in the few studies from African countries are reports of antibodies
138 against C. burnetii in the sera of 1%-24% of humans in 7 West African countries [16], 30.9%
139 of humans, 28.3% of cattle, 18.2% of sheep and 32% of goats in Kenya [17], 24.2% of goats
140 and 18.5% of sheep in Gambia [18] and C. burnetii DNA in the blood of 7.8% of cattle and
141 7.5% of goats in Tanzania [19] and 0.3% of febrile patients in Algeria and 0.4% in Senegal
143 The only previous study performed nearly half a century ago in Ethiopia [21] reported the
145 truncatum ticks. More recently, an epidemiological study highlighted the seroprevalence of
146 C. burnetii in 31.6% of cattle, 90% of camels and 54.2% of goats in eastern Ethiopia [22]. To
147 obtain better information about the involvement of tick species as reservoirs or vectors of Q
148 fever, in the present study, we aimed to determine the occurrence and genotype of C. burnetii
149 in different ticks species collected from domestic animals in different agro-ecological zones
150 and animal management systems in 9 districts in the Oromia Regional State of Ethiopia.
6
151 Materials and methods
153 Ixodid ticks were collected from cattle, sheep, dogs and cats in the Arsi, Wolmara, Kimbibit
154 Ada’a, Abdela, Gachi, Arero, Moyale and Yabelo districts in Oromia as described in a
155 previous publication [23]. These districts are located in 6 zones of the central, southwestern
156 and southeast parts of Ethiopia with various climates, agro-ecological and animal
157 management systems (Fig 1). Livestock comprising cattle, goats, sheep and camels in Arero,
158 Moyale and Yabelo districts (Borana zone) are kept under a pastoral type of animal
159 management. This system is characterized by very large animal populations, and livestock are
160 the main source of human food from milk and milk-byproducts and the main source of
161 income for farmers, whereas in all other study districts, animals, including cattle, sheep, goats
162 and equines, are kept under a crop-animal mixed type of management with extensive
163 production systems [22,24]. Animals were selected irrespective of their sex and age. Animals
164 were categorized into two age groups, young (up to one year) and adult (older than one year),
167 Ticks were collected from September through November of 2011. All the body surfaces of
168 each study animal were thoroughly examined visually to establish the presence or absence of
169 ticks. Ticks attached to the skin of each animal were carefully removed manually using
170 forceps or by hand to avoid any damage to the body and placed into separate, pre-labeled,
171 small, plastic tubes containing 70% ethanol for subsequent identification as previously
172 described. Ticks from the same animal were put in the same vial and transported to the
173 laboratory (URMITE, Marseille, France). Ticks were morphologically identified under a
174 microscope using identification keys described previously [25–27]. Species determination
175 was possible for adult specimens, whereas identification to the genus level was performed for
7
176 larvae and nymphs. For each tick specimen, the sex and developmental stage was determined,
177 and a photograph of the dorsal and ventral body parts was captured. Tick genera are
180 Prior to DNA extraction, each tick specimen was rinsed twice in sterile water for 15 minutes
181 and then dried on sterile filter paper. Each specimen was longitudinally cut into two equal
182 halves [29,30]. One half of each specimen was kept as reserve sample to avoid the risk of
183 losing samples for any reason during or after DNA extraction [23]. Genomic DNA was
184 individually extracted from a total of 842 tick specimens using the QIAamp DNA tissue
185 extraction kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The
186 DNA from each tick specimen was eluted in 100 µl of Tris EDTA (TE) buffer and stored at -
187 20°C under sterile conditions to preclude contamination until the sample was used for PCR.
188 To avoid cross-contamination among samples during DNA extraction, all parts of the DNA
189 extracting EZI Advanced XL Robot were disinfected after each batch of extraction as per
190 recommendations of the manufacturers. The second half of each tick specimen was stored at -
193 DNA samples were individually tested using C. burnetii-specific qPCR with primers and
194 probes designed for the amplification of the COX spacer sequence as previously described
195 [11]. DNA samples that were positive for the COX spacer sequence were further confirmed
196 by the more sensitive C. burnetii-species specific IS30A spacer as has been performed
197 previously using the same primers and probes [11,31]. Sterile water was used as a negative
198 control, whereas DNA from C. burnetii was used as a positive control. A sample was
199 considered positive when the PCRs were positive for the two different above-mentioned
8
201 Genotyping of C. burnetii detected in ticks
202 Multispacer sequence typing (MST) was used to determine the genotypes of C. burnetii in
203 randomly selected representative tick DNA samples that were positive for C. burnetii by
204 qPCR. Seven of the 10 spacer regions in the C. burnetii genome that exhibited higher
205 variation for differentiating the genotypes (Cox2, Cox5, Cox18, Cox37, Cox56, Cox57 and
206 Cox61 manufactured by Eurogentec, Seraing, Belgium) were selected for standard PCR using
207 similar PCR conditions and sequences of primers as described previously [11,31,32]. The
208 genotypes identified by MST were compared to genotypes included in the MST database
209 containing C. burnetii genotypes from countries in Europe and other parts of the world
211 cleaning of excess primers and nucleotides from DNA, sequencing, assembling and edition of
214 Ethical approval for the collection of ticks from domestic animals was obtained from the
215 animal research ethics board (Agreement # 14/160/550/2011) of the College of Veterinary
216 Medicine and Agriculture of Addis Ababa University. All necessary permits were obtained
217 from the administration and agricultural office of each district and from each animal owner.
219 Microsoft Excel was used for data management. Descriptive statistics, such as percentages
220 and means, were employed to summarize the proportions of ticks positive for C. burnetii
221 DNA. Statistical analyses were performed with EpiInfoTM7. Associations between the
222 prevalence of C. burnetii DNA among different tick species and genera and districts of
223 collection were determined using the Mantel-Haenszel (MH) test with the statistical software
224 EpiInfoTM7. All differences were considered significant at p-values < 0.05.
225
9
226 Results
227 Ticks in the present study were collected from a total of 245 animals (207 cattle, 29 sheep,
228 seven dogs and two cats) in nine districts: Abdela (37 cattle and fifteen sheep), Gachi (29
229 cattle and three sheep), Ada’a (20 cattle, seven dogs and two cats), Wolmara (seventeen cattle
230 and one sheep), Kimbibit (23 cattle and ten sheep), Arsi (31 cattle), Arero (23 cattle), Moyale
231 (eighteen cattle) and Yabelo (nine cattle). The 842 ticks collected consisted of 373 males, 343
232 females, 100 nymphs and 26 larvae. Overall, tick species collected comprised Am. cohaerens
233 (128: 76 m, 52 f), Am. gemma (49: 27 m, 22 f), Am. lepidum (2 f), Am. variegatum (62: 51 m,
234 11 f), Amblyomma larvae (22), Amblyomma nymphs (60), Rh. (Bo.) decoloratus (153: 37 m,
235 116 f), Rh. (Bo.) decoloratus larvae (4), Rh. (Bo.) decoloratus nymphs (31), Rh. praetextatus
236 (63: 34 m, 29 f), Rh. pulchellus (126: 69 m, 57 f), Hy. m. rufipes (14: 10 m, 4 f),
237 Rhipicephalus spp. nymphs (9), Hy. truncatum (53: 33 m, 20 f), Rh. sanguineus (14: 9 m, 5
238 f), Hae. leachci (6: 2 m,4 f) and Hae. spinulosa (2 f). The majority of the ticks were collected
239 from cattle 90.5% (762/842) and a small proportion was collected from sheep 6% (51/842),
240 from dogs 2.7% (23/842) and from cats 0.7% (6/842).
241 The present study recorded an overall frequency of 6.4% (54/842) of C. burnetii DNA in 7
242 species of ixodid ticks out of the total of 13 different species tested using molecular tools
244 C. burnetii DNA was detected in 28.6% (14/49) of Am. gemma, 25% (31/124) of Rh.
245 pulchellus, 7.1 % (1/14) of Hy. m. rufipes, 3.2 % (2/62) of Am. variegatum, 3.1% (4/128) of
246 Am. cohaerens, 1.6% (1/63) of Rh. praetextatus and 0.6% (1/153) of Rh. (Bo.) decoloratus
247 (Table 1). Photographs of ticks positive for C. burnetii is depicted on Fig 2. The overall
248 prevalence of C. burnetii DNA was significantly higher in both Am. gemma and Rh.
249 pulchellus than in the other positive tick species (Mantel-Haenszel (MH), p<0.0001). This
250 study is the first to report finding C. burnetii DNA in Am. gemma, Rh. pulchellus Rh (Bo).
10
251 decoloratus, Hy. m. rufipes and Rh. praetextatus ticks from Ethiopia. Overall, C. burnetii
252 DNA was most frequently found in the Rhipicephalus (7.4%; 33/444) and Amblyomma
253 (6.2%; 20/323) genera than in the Hyalomma (1.5%; 1/67) tick genus. However, C. burnetii
254 was not detected in Haemaphysalis genus (0/8) ticks (Table 1).
255 The overall proportion of positive C. burnetii DNA in ticks was 1.2% (1/80) in Kimbibit,
256 3.3% (4/121) in Arsi, 20% (21/105) in Arero, 27.8% (22/79) in Moyale and 15% (6/40) in
257 Yabelo districts (Table 1). Conversely, C. burnetii DNA was not detected in any ticks from
258 the Abdela (0/107), Gachi (0/107), Ada’a (0/128) and Wolmara (0/75) districts in western
259 and central Oromia (Table 1 and Fig 1). The overall frequency of positive C. burnetii DNA
260 was significantly higher in ticks collected from the Arero, Moyale and Yabelo (Borana zone)
261 southeastern districts than from the Kimbibit and Arsi central districts (Oromia regional state)
262 (MH, p<0.0001) (Fig 1). In this study, 7.1% (54/762) of the ticks from cattle were positive
263 for C. burnetii DNA; however, C. burnetii DNA was not detected in ticks taken from sheep
265 Multispacer sequence typing (MST) of C. burnetii from positive tick samples showed the
266 presence of two different genotypes: the first genotype, MST 18, was found in ticks from the
267 Borana zone (Arero, Moyale and Yabelo districts), whereas the second genotype, MST 20,
268 was detected in ticks from the central districts (Kimbibit and Arsi) (Fig 1). In southeastern
269 districts, genotype MST 18 was detected in Am. cohaerens in the Arero and Moyale districts,
270 in Hy. m. rufipes in the Moyale district, and in Rh. pulchellus and Am. gemma in the Arero,
271 Moyale and Yabelo districts. In the central districts, genotype MST 20 was detected in Rh.
272 praetextatus in the Kimbibit district and in Am. cohaerens, Am. variegatum and Rh. (Bo.)
274
275
11
276 Discussion
277 The overall frequency of C. burnetii (6.4%; 54/842) in 7 species of ixodid ticks collected
278 from cattle may highlight the possible importance of ticks in the epidemiology of C. burnetii
279 in Ethiopia. A reliable molecular strategy was used that has been validated in recent studies,
280 and our findings were confirmed by positive and negative controls and by amplification of
282 The detection of C. burnetii in Am. variegatum for the second time in Ethiopia confirms the
283 earlier report made some 48 years ago in central Ethiopia [21]. In line with our findings, C.
284 burnetii has been recently detected in 19.6% of Am. variegatum in Nigeria [33], in up to
285 37.6% of A. variegatum in two regions of Senegal [11], in 2.5% of pools of Am. variegatum
286 from cattle and in 20% of pools from dogs in Kenya [17]. These studies are the only reports
288 Our observation of C. burnetii in Am. gemma, Rh. (Bo.) decoloratus, Rh. pulchellus, Hy. m.
289 rufipes, Am. cohaerens and Rh. praetextatus for the first time in Ethiopia is additional
290 knowledge about the species of ticks that host C. burnetii in this country. Consistent with our
291 observation, C. burnetii has been reported in 30% of Rh. (Bo.) decoloratus and in 1.7 to
292 20.9% of Hy. m. rufipes in Senegal [11] and in 20% of pools of Rh. (Bo.) decoloratus in
293 Kenya [17]. Similarly, C. burnetii has recently been detected in different species of ixodid
294 ticks in France [31,34], Italy [35], Spain [36], Germany [37], Australia [38], Argentina [39],
295 Slovakia and Hungary [40], Japan [41] and in 8 species of ticks in China [42].
296 The finding of significantly higher C. burnetii in both Am. gemma and Rh. pulchellus
297 compared with all the other tick spp. most probably indicates the greater importance of these
298 two spp. in the epidemiology of this zoonosis than the other ticks. When considering the
299 established fact that these two tick species are the most predominant species in southeastern
300 Oromia in the Borana zone [4] where animals are kept under a pastoral type of management
12
301 and there is much closer contact between animals and pastoralists due to common
302 consumption of raw milk and meat, it is implied that there is a high likelihood of transmission
303 of C. burnetii to humans in the study districts of this zone. It is hypothesized that ticks
304 become infected during blood feeding on infected animals and then act as reservoirs of C.
305 burnetii and play an important role in maintaining the bacteria in the environment, which
306 may cause infection in wild vertebrates, domestic animals and humans [5,10].
307 However, contrary to our present finding, C. burnetii was not detected in ticks studied in
308 Hungary [43], in ticks in Germany [37] or in ticks in Sweden [44]. This variation is most
309 probably attributed to differences in animal management factors, for example that domestic
310 animals are housed within farms throughout their lifespan and are not allowed to graze on
311 pastures in some European countries and the systematic use of acaricides, such as
312 deltamethrin, that reduce tick populations and alter the ability of ticks to carry C. burnetii.
313 This is different from Ethiopia where animals graze in communal pastures throughout the
314 year that serve as an infectious source for blood feeding ticks, and there is no regular use of
315 acaricides as has been shown before [36]. In support of this fact, a recent serological study
316 reported antibodies against C. burnetii in 54.2% of goats, 31.6% of cattle and 90% of camels
318 In our study, a statistically significantly higher (MH, p<0.0001) prevalence of positive C.
319 burnetii DNA in ticks was observed in southeastern districts from the Borana zone (Arero,
320 Moyale and Yabelo) than in the central districts (Kimbibit and Arsi). This difference is most
321 probably attributable to factors such as the presence of larger population of cattle, goats,
322 sheep and camel that are usually managed by a pastoral type of management characterized by
323 unrestricted movement of animals in search of water and grazing pasture in districts in the
324 Borana zone. This is in contrast with the central districts that practice a crop-livestock mixed
325 type of farming with restricted movement of ruminants and equines and the absence of
13
326 camels. In support of the finding of the highest frequency of C. burnetii in camel-rearing
327 districts in the Borana zone of the present study, being a camel breeder was reported as a very
328 important risk factor for human Q-fever seropositivity in Chad [45]. Despite the greater
329 number of ticks tested in Abdela (0/107), Gachi (0/107), Ada’a (0/128) and Wolmara (0/75)
330 from the western and central districts of Oromia, C. burnetii was not detected (Table 1). In
331 agreement with our report, a significantly higher prevalence of C. burnetii was recorded in
332 ticks from areas where higher populations of ruminants are kept in Senegal [11] and Kenya
333 [17]. Also in support of our observations, the presence of ticks on animals was reported as the
334 most important risk factor for the occurrence of Q fever and abortion in northern Cyprus [46].
335 Furthermore, several previous studies linked higher populations of ruminants with outbreaks
337 For the first time in Ethiopia, this study provides new information on the presence of two
338 different genotypes of C. burnetii in ticks, MST 18 in ticks from animals in the Borana zone
339 and MST 20 in ticks from animals in the central districts. This finding is in line with previous
340 reports of several genotypes of C. burnetii in ticks [11]. Genotype MST18 was previously
341 identified from human and animal (sheep and goats) clinical samples in Hungary [49],
342 France, Italy, Romania, Greece, Slovakia, Germany and Spain [32,50]. Similarly, genotype
343 MST 20 has been identified in samples from cattle, goat and cow’s milk in Hungary [49],
344 from cow milk and milk products from several countries in Europe, including France, the
345 Netherlands, Portugal, Spain, Switzerland and the United Kingdom, in human clinical
346 samples from France, in a cow’s placenta from Germany and in rodents and in cow’s milk,
347 soil, and goat placental material from the United States [32] and other parts of the world
348 (Qatar and Saudi Arabia), suggesting the worldwide occurrence of this genotype, which
349 supports our finding. Previous studies have shown that genotype MST 20 is usually
350 associated with cattle and their products [49,50]. C. burnetii genotyping is important for
14
351 identifying the source of infection for humans and animals. It is important to note that
352 additional genotypes not detected by our sampling may be circulating in other parts of the
353 country. In the only previous report of C. burnetii genotypes from Africa, MST 6, 35, and 36
354 were detected in ticks in Senegal, and genotype MST 19 was recorded in human patients with
356 In conclusion, the findings of this study provide additional information on the geographic
357 distribution and genotypes of C. burnetii in different ticks in Ethiopia. The overall frequency
358 of C. burnetii in ticks from the Borana zone is higher than all other zones. Medical and
360 the country. Ethiopian physicians managing patients with fevers of unknown etiology and
361 travelers with unexplained fevers returning from Ethiopia should consider C. burnetii as one
362 of the potential causative agents. Future studies are necessary to address the isolation, culture,
363 and genotypes of C. burnetii in arthropods, human blood, domestic and other reservoir host
364 animals and the vector competence of these tick species for C. burnetii and its public health
366 Acknowledgments
367 The authors greatly acknowledge domestic animal owners in the different districts of Oromia
368 for their kindness in allowing us to collect ticks from their animals.
369
15
370 References
371
372 1. Jongejan F, Uilenberg G (2004) The global importance of ticks. Parasitology 129
373 Suppl: S3-14.
374 2. Dantas-Torres F, Chomel BB, Otranto D (2012) Ticks and tick-borne diseases: a One
375 Health perspective. Trends Parasitol 28: 437-446. S1471-4922(12)00121-3
376 [pii];10.1016/j.pt.2012.07.003 [doi].
377 3. De la Fuente J, Estrada-Pena A (2012) Ticks and tick-borne pathogens on the rise.
378 Ticks Tick Borne Dis 3: 115-116. S1877-959X(12)00030-1
379 [pii];10.1016/j.ttbdis.2012.03.001 [doi].
380 4. Regassa A (2001) Tick infestation of Borana cattle in the Borana Province of
381 Ethiopia. Onderstepoort J Vet Res 68: 41-45.
382 5. Angelakis E, Raoult D (2010) Q Fever. Vet Microbiol 140: 297-309. S0378-
383 1135(09)00338-1 [pii];10.1016/j.vetmic.2009.07.016 [doi].
384 6. Maurin M, Raoult D (1999) Q fever. Clin Microbiol Rev 12: 518-553.
389 8. Cutler SJ, Bouzid M, Cutler RR (2007) Q fever. J Infect 54: 313-318. S0163-
390 4453(06)00378-1 [pii];10.1016/j.jinf.2006.10.048 [doi].
391 9. Guatteo R, Seegers H, Taurel AF, Joly A, Beaudeau F (2011) Prevalence of Coxiella
392 burnetii infection in domestic ruminants: a critical review. Vet Microbiol 149:
393 1-16. S0378-1135(10)00480-3 [pii];10.1016/j.vetmic.2010.10.007 [doi].
394 10. Agerholm JS (2013) Coxiella burnetii associated reproductive disorders in domestic
395 animals--a critical review. Acta Vet Scand 55: 13. 1751-0147-55-13
396 [pii];10.1186/1751-0147-55-13 [doi].
397 11. Mediannikov O, Fenollar F, Socolovschi C, Diatta G, Bassene H, Molez JF, Sokhna
398 C, Trape JF, Raoult D (2010) Coxiella burnetii in humans and ticks in rural
399 Senegal. PLoS Negl Trop Dis 4: e654. 10.1371/journal.pntd.0000654 [doi].
403 13. Loftis AD, Reeves WK, Szumlas DE, Abbassy MM, Helmy IM, Moriarity JR, Dasch
404 GA (2006) Surveillance of Egyptian fleas for agents of public health
405 significance: Anaplasma, Bartonella, Coxiella, Ehrlichia, Rickettsia, and
406 Yersinia pestis. Am J Trop Med Hyg 75: 41-48. 75/1/41 [pii].
407 14. Nelder MP, Lloyd JE, Loftis AD, Reeves WK (2008) Coxiella burnetii in wild-caught
408 filth flies. Emerg Infect Dis 14: 1002-1004. 10.3201/eid1406.071691 [doi].
16
409 15. Vanderburg S, Rubach MP, Halliday JE, Cleaveland S, Reddy EA, Crump JA (2014)
410 Epidemiology of Coxiella burnetii Infection in Africa: a OneHealth systematic
411 review. PLoS Negl Trop Dis 8: e2787. 10.1371/journal.pntd.0002787
412 [doi];PNTD-D-13-01613 [pii].
413 16. Dupont HT, Brouqui P, Faugere B, Raoult D (1995) Prevalence of antibodies to
414 Coxiella burnetti, Rickettsia conorii, and Rickettsia typhi in seven African
415 countries. Clin Infect Dis 21: 1126-1133.
416 17. Knobel DL, Maina AN, Cutler SJ, Ogola E, Feikin DR, Junghae M, Halliday JE,
417 Richards AL, Breiman RF, Cleaveland S, Njenga MK (2013) Coxiella burnetii
418 in humans, domestic ruminants, and ticks in rural western Kenya. Am J Trop
419 Med Hyg 88: 513-518. ajtmh.12-0169 [pii];10.4269/ajtmh.12-0169 [doi].
420 18. Klaasen M, Roest HJ, van der Hoek W, Goossens B, Secka A, Stegeman A (2014)
421 Coxiella burnetii seroprevalence in small ruminants in The Gambia. PLoS
422 One 9: e85424. 10.1371/journal.pone.0085424 [doi];PONE-D-13-29780 [pii].
423 19. Qiu Y, Nakao R, Namangala B, Sugimoto C (2013) First genetic detection of Coxiella
424 burnetii in Zambian livestock. Am J Trop Med Hyg 89: 518-519. ajtmh.13-
425 0162 [pii];10.4269/ajtmh.13-0162 [doi].
433 22. Gumi B, Firdessa R, Yamuah L, Sori T, Tolosa T, Aseffa A, Zinsstag J, Schelling E
434 (2013) Seroprevalence of Brucellosis and Q-Fever in Southeast Ethiopian
435 Pastoral Livestock. J Vet Sci Med Diagn 2. 10.4172/2325-9590.1000109 [doi].
436 23. Kumsa B, Socolovschi C, Raoult D, Parola P (2014) Spotted fever group rickettsiae in
437 ixodid ticks in Oromia, Ethiopia. Ticks Tick Borne Dis . S1877-
438 959X(14)00170-8 [pii];10.1016/j.ttbdis.2014.08.001 [doi].
439 24. CSA (2008) Ethiopian agricultural sample survey report on livestock and livestock
440 characteristics vol II, 2007/08, Addis Ababa, Ethiopia..
441 25. Hoogstraal, H (1956) African Ixodidae. 1. Ticks of the Sudan (with special reference
442 to equatorial province and with preliminary reviews of the genera Boophilus,
443 Margaropus and Hyalomma). Washington D. C. Department of the Navy,
444 Bureau of the Medicine and Surgery, United States Government Printing
445 Office. 1101 p.
453 28. Dantas-Torres F (2008) Towards the standardization of the abbreviations of genus
454 names of ticks (Acari: Parasitiformes: Ixodida). Vet Parasitol 154: 94-97.
455 S0304-4017(08)00146-5 [pii];10.1016/j.vetpar.2008.03.006 [doi].
456 29. Kumsa B, Socolovschi C, Parola P, Rolain JM, Raoult D (2012) Molecular detection
457 of Acinetobacter species in lice and keds of domestic animals in Oromia
458 Regional State, Ethiopia. PLoS One 7: e52377. 10.1371/journal.pone.0052377
459 [doi];PONE-D-12-22871 [pii].
464 31. Socolovschi C, Reynaud P, Kernif T, Raoult D, Parola P (2012) Rickettsiae of spotted
465 fever group, Borrelia valaisiana, and Coxiella burnetii in ticks on passerine
466 birds and mammals from the Camargue in the south of France. Ticks Tick
467 Borne Dis 3: 355-360. S1877-959X(12)00104-5
468 [pii];10.1016/j.ttbdis.2012.10.019 [doi].
472 33. Reye AL, Arinola OG, Hubschen JM, Muller CP (2012) Pathogen prevalence in ticks
473 collected from the vegetation and livestock in Nigeria. Appl Environ
474 Microbiol 78: 2562-2568. AEM.06686-11 [pii];10.1128/AEM.06686-11 [doi].
480 35. Satta G, Chisu V, Cabras P, Fois F, Masala G (2011) Pathogens and symbionts in
481 ticks: a survey on tick species distribution and presence of tick-transmitted
482 micro-organisms in Sardinia, Italy. J Med Microbiol 60: 63-68.
483 jmm.0.021543-0 [pii];10.1099/jmm.0.021543-0 [doi].
484 36. Toledo A, Jado I, Olmeda AS, Casado-Nistal MA, Gil H, Escudero R, Anda P (2009)
485 Detection of Coxiella burnetii in ticks collected from Central Spain. Vector
486 Borne Zoonotic Dis 9: 465-468. 10.1089/vbz.2008.0070 [doi].
18
491 38. Cooper A, Stephens J, Ketheesan N, Govan B (2013) Detection of Coxiella burnetii
492 DNA in wildlife and ticks in northern Queensland, Australia. Vector Borne
493 Zoonotic Dis 13: 12-16. 10.1089/vbz.2011.0853 [doi].
494 39. Pacheco RC, Echaide IE, Alves RN, Beletti ME, Nava S, Labruna MB (2013)
495 Coxiella burnetii in ticks, Argentina. Emerg Infect Dis 19: 344-346.
496 10.3201/eid1902.120362 [doi].
497 40. Spitalska E, Kocianova E (2003) Detection of Coxiella burnetii in ticks collected in
498 Slovakia and Hungary. Eur J Epidemiol 18: 263-266.
499 10.1023/A:1023330222657 [doi].
511 44. Wallmenius K, Pettersson JH, Jaenson TG, Nilsson K (2012) Prevalence of Rickettsia
512 spp., Anaplasma phagocytophilum, and Coxiella burnetii in adult Ixodes
513 ricinus ticks from 29 study areas in central and southern Sweden. Ticks Tick
514 Borne Dis 3: 100-106. S1877-959X(11)00090-2
515 [pii];10.1016/j.ttbdis.2011.11.003 [doi].
522 47. De Lange MM, Schimmer B, Vellema P, Hautvast JL, Schneeberger PM, van
523 Duijnhoven YT (2014) Coxiella burnetii seroprevalence and risk factors in
524 sheep farmers and farm residents in The Netherlands. Epidemiol Infect 142:
525 1231-1244. S0950268813001726 [pii];10.1017/S0950268813001726 [doi].
526 48. Schimmer B, Schotten N, van EE, Hautvast JL, Schneeberger PM, van Duijnhoven
527 YT (2014) Coxiella burnetii seroprevalence and risk for humans on dairy
528 cattle farms, the Netherlands, 2010-2011. Emerg Infect Dis 20: 417-425.
529 10.3201/eid2003.131111 [doi].
530 49. Sulyok KM, Kreizinger Z, Hornstra HM, Pearson T, Szigeti A, Dan A, Balla E, Keim
531 PS, Gyuranecz M (2014) Genotyping of Coxiella burnetii from domestic
19
532 ruminants and human in Hungary: indication of various genotypes. BMC Vet
533 Res 10: 107. 1746-6148-10-107 [pii];10.1186/1746-6148-10-107 [doi].
534 50. Astobiza I, Tilburg JJ, Pinero A, Hurtado A, Garcia-Perez AL, Nabuurs-Franssen
535 MH, Klaassen CH (2012) Genotyping of Coxiella burnetii from domestic
536 ruminants in northern Spain. BMC Vet Res 8: 241. 1746-6148-8-241
537 [pii];10.1186/1746-6148-8-241 [doi].
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
20
581 Figure Legends
582
583 Fig 1. Map of Oromia showing the overall prevalence of C. burnetii in ticks and MST in each
585
586 Fig 2. Picture of ticks positive for C. burnetii in 9 districts in Oromia, Ethiopia.
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
21
626 Table 1. Prevalence of C. burnetii in ixodid ticks using species specific gene qPCR in 9
627 districts in Oromia, Ethiopia.
628
Zones IluAba Bora East Showa West North Arsi Borana
Showa
Showa
District Abdela Gachi Ada’a Wolmara Kimbibit Arsi Arero Moyale Yabelo Total
Am. cohaerens 0/52 0/57 - - - ½(50%) 1/12(8.3%) 2/3(66.7%) 0/2 4/128 (3.1%)
Cattle (n=58) 0/41 0/55 - - - ½(50%) 1/12(8.3%) 2/3(66.7%) 0/2 4/115 (3.5%)
Sheep(n=9) 0/11 0/2 - - - - - - - 0/13
Am. gemma - - 0/1 - - - 5/15(33.3% 7/23(30.4% 2/10(20%) 14/49(28.6%)
) )
Cattle (n=25) - - 0/1 - - - 5/15(33.3% 7/23(30.4% 2/10(20%) 14/49(28.6%)
) )
Am. lepidum - - - - - - 0/1 0/1 - 0/2
Cattle (n=2) - - - - - - 0/1 0/1 - 0/2
Am. variegatum - 0/5 0/14 0/9 - 2/31 (6.4%) 0/2 - 0/1 2/62 (3.2%)
Cattle (n=40) - 0/3 0/14 0/7 2/31 (6.4%) 0/2 - 0/1 2/60 (3.3%)
Sheep(n=2) 0/2 - - - - - - - 0/2
Amblyomma 0/10 0/5 0/4 0/2 - - - - 0/1 0/22
larva
Cattle (n=20) 0/8 0/5 0/4 0/2 - - - - 0/1 0/20
Amblyomma 0/26 0/13 0/7 0/1 - 0/10 0/1 0/1 0/1 0/60
nymph
Cattle (n=44) 0/13 0/11 0/6 0/1 - 0/10 0/1 0/1 0/1 0/44
Overall 0/88 0/80 0/26 0/12 - 3/43 (7%) 6/31 9/28 2/15(13.3%) 20/323(6.2%)
Amblyomma (19.3%) (32.1%)
spp.
Rh (Bo). 0/14 0/17 0/11 0/37 0/23 1/37 (2.7%) 0/10 0/3 0/1 1/153 (0.6%)
decoloratus
Cattle (n=104) 0/14 0/16 0/11 0/37 0/20 1/37 (2.7%) 0/10 0/3 0/1 1/149 (0.7%)
Rh. e. evertsi 0/2 0/4 0/11 0/6 0/7 0/12 0/1 0/1 - 0/44
Cattle (n=23) 0/2 0/3 0/11 0/4 0/7 0/12 0/1 0/1 - 0/41
22
Sheep(n=2) 0/1 0/2 - - - - - 0/3
Overall 0/19 0/27 0/63 0/54 1/76(1.3 1/62(1.6%) 15/72(20.8 12/46(26.1 4/25(16%) 33/444
Rhipicephalus %) %) %) (7.4%)
spp.
Hy. m. rufipes - - 0/2 0/3 0/2 - 0/2 1/5 (20%) - 1/14 (7.1%)
Cattle (n=5) - - 0/2 0/3 0/2 - 0/2 1/5 (20%) - 1/14 (7.1%)
Overall - - 0/31 0/9 0/4 0/16 0/2 1/5 (20%) - 1/67 (1.5%)
Hyalomma spp.
Hae. leachci - - 0/6 - - - - - - 0/6
Dog (1) - - 0/1 - - - - - - 0/1
Cat (2) - - 0/5 - - - - - - 0/5
Hae. spinulosa - - 0/2 - - - - - - 0/2
Dog (1) - - 0/2 - - - - - - 0/2
Overall 0/8 - - - - - - 0/8
Haemaphysalis
spp.
Overall 0/107 0/107 0/128 0/75 1/80(1.2 4/121(3.3 21/105(20 22/79(27.8 6/40(15%) 54/842(6.4%)
prevalence in %) %) %) %)
ticks
629
23
Figure
Click here to download high resolution image
Figure
Click here to download high resolution image
Figure
Click here to download high resolution image
3.3. Article 4
1
Aix Marseille Université, URMITE, UM63,
CNRS 7278, IRD 198, Inserm 1095, 13005
Marseille, France.
165
167
Manuscript
Amblyomma cohaerens male
Amblyomma cohaerens female
Amblyomma gemma male
Amblyomma gemma female
Amblyomma variegatum male Amblyomma variegatum female
Amblyomma spp. nymph
Amblyomma spp. larva
Rhipicephalus pulchellus male Rhipicephalus pulchellus female
Rhipicephalus (Bo.) decoloratus female
Rhipicephalus (Bo.) decoloratus male
Rhipicephalus (Bo.) decoloratus larva Rhipicephalus (Bo.) decoloratus nymph
Figure
100 Borrelia hermsii-AY597796-western North America
49 Borrelia hermsii-AY597795-western North America
Relapsing fever
49 Borrelia coriaceae-D82864-Japan related Borrelia spp.
Borrelia-anserina-DQ849626-Brazil
35
90 B.anserina-X75201-SW EDEN
Borrelia turicatae-AY934630-Florida-USA
42
99 Borrelia parkeri-AY934623-Florida-USA
Roe deer-Borrelia miyamotoi-FJ874925-Poland
Amblyomma americanum-Borrelia-lonestari-AY850064-USA
82
61 Rhipicephalus (Boophilus) microplus-Borrelia-sp.-EF141022-Brazil
100
54 81-Rh.pulchellus-Yabelo-Ethiopia
Borrelia spp. in
89 456Rh(Bo).decoloratus-W olmara-Ethiopia
Ethiopian
Homo sapiens-Borrelia persica-HM194741-Israel
Ornithodoros sonrai-Borrelia crocidurae-JX292921-Mali Rhipicephalus ticks
71
Borrelia duttonii-U28497-USA
100
Ornithodoros-moubata-GU357617-Borrelia-duttonii-Spain
45 Relapsing fever related
B.crocidurae-X75204-Sweden
40
21 Borrelia recurrentis-DQ346831-UK
Borrelia spp.
68 Borrelia recurrentis-D86618-Japan
99 Borrelia burgdorferi-X15661-USA
85 Borrelia burgdorferi-X15660-USA Lyme disease
100
Ixodid ticks-Borrelia andersonii-D83764-Japan
related Borrelia spp.
65 Borrelia valaisiana-AB022132-Japan
65
Borrelia tanukii-D85070-Japan
29 Borrelia japonica-D82853-Japan
Borrelia sinica-AB022138-Japan
20
Borrelia turdi-D85071-Japan
35
Borrelia garinii-AB035608-Japan
29
99 Borrelia garinii-D82846-Japan
96 Ixodes ricinus-Borrelia garinii-AB178326-Moscow-Russia
0.02
3.4. Article 5
193
194
Morphological, MALDI TOF Mass spectrometry
and molecular identification of ixodid tick species
in Oromia, Ethiopia
1
Aix Marseille Université, Unité de Recherche en Maladies
Infectieuses et Tropicales Emergentes (URMITE), UM63,
CNRS 7278, IRD 198 (Dakar, Sénégal), Inserm 1095, WHO
Collaborative Center for Rickettsioses and Other Arthropod-
Borne Bacterial Diseases, Faculté de Médecine, 27 bd Jean
Moulin, 13385 Marseille cedex 5, France.
2. Campus Universitaire IRD de Hann, Dakar, Senegal
195
196
Abstract
197
specimens were correctly identified. With the exception of
specimens from Rh. praetextatus species, relevant Log score
values (LSVs>1.8) of identification were obtained. The
morphological and MALDI-TOF MS identification were
confirmed by sequencing the 12S rRNA gene of 40 tick
specimens representing the 11 ixodid species. Taken
together, these results comfort that MALDI-TOF MS is a
reliable tool for quick tick species identification also
preserved in ethanol on condition that reference spectra
database was built with respective species specimens stored
in the same conditions.
198
1. Introduction
Ticks are obligate blood feeding arthropods belonging to the
phylum Arthropoda, class Arachnida and to the order Acari
(Jongejan and Uilenberg, 2004). They are grouped into three
families: Ixodidae (hard ticks) with more than 700 officially
recognized species, Argasidae (soft ticks) comprising
roughly 200 species and Nuttalliellidae with a single species
(Walker, 2003). Ticks can feed on every known class of
vertebrate (Estrada-Pena and De la Fuente, 2014). The blood
from mammals, birds, reptiles and amphibians is the only
critically important source of nutrition for growth,
development of organs for male, female, larvae and nymphs
of ticks (Parola and Raoult, 2001).
Ticks are capable of transmitting a broad range of human
and animal pathogens including viruses, bacteria, protozoa
and helminths. Some of these pathogens are of exceptional
importance because of high morbidity and mortality in both
humans and animals (Parola and Raoult, 2001). The
identification of tick species is a fundamentally important
task in epidemiological studies for examples to establish tick
species distribution maps, to characterize tick fauna and
seasonal pattern, to determine trophic preference of each tick
species and finally to assess tick species responsible of
199
pathogens transmission to humans and domestic animals
(Dantas-Torres et al., 2012).
Morphological identification of ticks at the genus and
species level using standard taxonomic keys is still a
reference method used by entomologists around the world
(Mediannikov and Fenollar, 2014). Acarologists have
continued to improve tick identification taxonomic keys for
different geographical regions of the worldwide (Jongejan
and Uilenberg, 2004). However, tick species identification at
immature developmental stages (e.g., eggs, larvae and
nymphs) is very difficult and sometime impossible due to the
lack of species specific morphological criteria or keys
(Mediannikov and Fenollar, 2014). The absence of tick
distinctive characters was also reported for sub-species from
the same group (Pegram et al., 1987). In addition, during tick
removing from host, important body parts for anatomical
identification of the specimen could be damaged notably
head parts like basis capituli, chelicerae and hypostome
(Karger et al., 2012; Yssouf et al., 2013a).
Thus, to resolve the aforementioned problems, alternative
methods like molecular tick identification were developed
(Mediannikov and Fenollar, 2014). Although mitochondrial
markers (12S and 16S ribosomal DNAs, cytochrome oxidase
200
subunit 1 (COX1) or cytochrome b) are frequently used for
tick species identification (Burger et al., 2014a; Burger et al.,
2014b), they are sometimes inefficient to distinguish closely
related tick species (Chitimia et al., 2009; Dvorak et al.,
2014; Murrell et al., 2000). Nuclear markers (18S, 28S,
internal transcribed spacers 1 and 2) were also described for
tick identification, but once more, they failed to discriminate
tick species that belong to the same group or complex
(Burger et al., 2013; Dobson and Barker, 1999). The
development of a universal gene sequence serving as a
“barcode” for identification of all tick species is not yet
available. Thus, this technically time-consuming and
expensive molecular approach is also confronted by the
absence of gene sequences for some tick species in the
GenBank database limiting its used (Mediannikov and
Fenollar, 2014).
During the last decade, matrix-assisted laser
desorption/ionization time-of-flight mass spectrometry
(MALDI-TOF MS) was successfully applied for the
identification and phylogenetic classification of
microorganisms (Neupert et al., 2005). This proteomic
approach was also used as taxonomic identification tool for
Drosophila species (Feltens et al., 2010). Recently, MALDI-
201
TOF MS was successfully applied for the identification of
different types of arthropods (Dvorak et al., 2014; Feltens et
al., 2010; Hoppenheit et al., 2014; Hoppenheit et al., 2013;
Kaufmann et al., 2012a; Kaufmann et al., 2012b; Kaufmann
et al., 2011; Steinmann et al., 2013; Yssouf et al., 2014a;
Yssouf et al., 2013b; Yssouf et al., 2014b). This proteomic
approach was also applied for tick identification using whole
body of fresh specimens (Karger et al., 2012). Yssouf et al
demonstrated that protein spectra from tick legs of fresh or
frozen specimens were sufficient for accurate tick species
identification by MALDI TOF protein profiling (Yssouf et
al., 2013a).
In Ethiopia ticks are very common and widely distributed
throughout all agroecological zones (Kumsa and Mekonnen,
2011). Ticks are routinely identified by morphological
method of identification by using taxonomic keys available
in the literature. Previous studies documented that the genus
Amblyomma (8 spp.), subgenus Boophilus, (2 spp.),
Haemaphysalis (4 spp.), Hyalomma (9 spp.), Rhipicephalus
(15 spp.), Ixodes (1 sp.), Argaus (1 sp.) and Ornithodorous
(2 spp.) exist in the country (Mekonnen et al., 2001). As
these ticks live in sympatry, misidentification has been
frequently reported due to difficulty in correct identification
202
leading to equivocal position (Mekonnen et al., 2001). Then,
MALDI-TOF MS could be a pertinent alternative to other
routine methods used for tick identification. To facilitate
sample storage and transportation, ticks collected in the field
are generally preserved in 70% ethanol in Ethiopia.
However, this mode of sample preservation was already
reported that it could alter MALDI-TOF MS profiles leading
to decrease of identification rate until the failing of paired
spectra matching with the reference MS database.
Thus, the aim of the present study was to determine the
discriminatory power of MALDI TOF MS for correct
identification of ixodid ticks preserved in 70% ethanol for
two years in Ethiopia.
203
2. Materials and Methods
2.1. Morphological identification of ticks
Ticks were collected from September through November of
2011 from domestic animals in Ethiopia by a veterinarian.
Ticks were carefully removed using forceps or by hand to
avoid any damage to their body, and the specimen were
placed into small pre-labelled plastic tubes containing 70%
ethanol for subsequent identification (Mekonnen et al.,
2001). Ticks from the same animal were placed into the
same vial and transported to the Laboratory of the World
Health Organization Collaborative Center for Rickettsial
Diseases and Arthropod-borne Bacterial Diseases in
Marseille, France. Ticks were morphologically identified at
the species level under microscope using previously
established taxonomic identification keys (Hoogstraal, 1956;
Walker, 2003; Walker, 2000). For each tick, photographs of
the dorsal and ventral sides were captured and the gender
was determined. The tick genera and species are abbreviated
as previously described (Dantas-Torres, 2008). Then adult
tick specimens were selected for MALDI-TOF protein
profiling study.
204
2.2. MALDI-TOF MS data and sample preparation
2.2.1. Sample preparation
All the ixodid ticks studied by MALDI TOF MS in the
present study were preserved in 70% ethanol for about 2
years. Each specimen was kept for 10 min in decreasing
ethanol concentration solutions from 70% to 10% (v/v),
followed by a wash in distilled water prior to a sample
drying on sterile filter paper as previously described (Yssouf
et al., 2013a). Then, 4 legs per tick specimen were cut with
sterile scalpels, and the remaining body parts were kept for
subsequent DNA extraction. Legs from individual tick were
manually crushed and homogenized in 40 μl of 10% formic
acid and 40 µl of 50% acetonitrile in 1.5-mL micro-
centrifuge tubes using pellet pestles (Fischer Scientific). A
quick spin was performed to discard debris and 1 µl of the
supernatant was deposited onto a spot on an MSP 96 target
polished steel micro Scout target plate (Bruker Daltonics,
Wissembourg, France), in quadruplicate. After air-drying, 1
µL of matrix solution composed of saturated α-cyano-4-
hydroxycynnamic acid (Sigma, Lyon. France), 50%
acetonitrile (v/v), 2.5% trifluoroacetic acid (v/v) (Aldrich,
Dorset, UK) and HPLC-grade water was directly overlaid on
each spot sample on the target plate, dried for several
205
minutes at room temperature and introduced into the
MALDI-TOF-MS machine. To control loading on mass
spectra steel, matrix quality and MALDI-TOF apparatus
performance, matrix solution was loaded in duplicate onto
each MALDI-TOF plate with or without a bacterial test
standard (Bruker protein Calibration Standard I).
206
different tick species were determined by analyzing and
comparing the average spectral profiles obtained from the
four spots for each individual tick specimen using the
ClinProTools 2.2 software (Bruker Daltonics). After the
validation of reproducibility and specificity of spectra, they
were loaded into the MALDI Biotyper 3.0 to create a
spectral database. The database was composed of a total of
11 species (Am. cohaerens n=4, Am. gemma n=5,
Am. variegatum n=4, Hy. truncatum n=4, Hy. m. rufipes n=3,
Rh (Bo).decoloratus n=3, Rh. e. evertsi n=4, Rh. pulchellus
n=4, Rh. sanguienus n=3, Rh. bergeoni n=4 and
Rh. praetextatus n=3) of ixodid ticks preserved in 70%
ethanol with 3 to 5 specimens for each species including
both gender. These specimens were used to create the
database 2 (Table 1).
207
R. africae, Rh. sanguineus, Hyalomma marginatum rufipes,
Ixodes ricinus, D. marginatus and D. reticulatus ),
30 mosquito species (Anopheles gambiae molecular form
M and An. gambiae molecular form S, An. funestus,
An. ziemanni, An. arabiensis, An. wellcomei, An. rufipes,
An. pharoensis, An. coustani, An. claviger, An. hyrcanus,
An. maculipennis, Culex quinquefasciatus, Cx. pipiens,
Cx. modestus,Cx. insignis, Cx. neavei, Ae. albopictus, Aedes
excrucians, Ae vexans, Ae. rusticus, Ae. dufouri,
Ae. cinereus, Ae. fowleri, Ae. aegypti, Ae. caspius, Mansonia
uniformis, Orthopodomyia reunionensis, Coquillettidia
richiardii and Lutzia tigripes,), and other arthropods
including louse (Pediculus humanus corporis ), triatomine
(Triatoma infestans) and bedbugs (Cimex lectularius), as
well as the spectra obtained from the bodies (without the
abdomens) of 5 flea species (Ctenocephalides felis, Ct.
Ct. canis, Archaeopsylla erinacei, Xenopsylla cheopis and
Stenoponia tripectinata) (Yssouf et al., 2013a; Yssouf et al.,
2014a; Yssouf et al., 2013b; Yssouf et al., 2014b). Database
2 includes Database 1 which was created from the leg
protein spectra of 11 tick species stored in ethanol (Table 1).
The reliability of the identification was estimated based on
the Log Score values (LSVs) exhibited by the MALDI-
208
Biotyper software, ranged from 0 to 3. These LSVs
correspond to the degree of homology between the query
mass spectra and the reference spectra. An LSV was
obtained for each spectrum of the samples tested.
209
under sterile conditions to preclude contamination until the
sample was used for PCR as has been used previously
(Kumsa et al., 2014a; Kumsa et al., 2014b). For molecular
identification of tick species, DNA samples of 1-7 individual
ticks per each tick species that were used to create the
database were subjected to standard PCR in an automated
DNA thermal cycler to amplify a fragment of the 360-base
pair from the mitochondrial 12S RNA gene as described
before (Beati and Keirans, 2001). The detection of amplified
products, the removal of excess primers and nucleotides
from the DNA, sequencing, the assembly and edition of
sequences and BLAST analysis were all performed as
previously described (Kumsa et al., 2014b; Kumsa et al.,
2012b). Phylogenetic tree was constructed using Mega 5
software (Molecular Evolution Genetic Analysis; The
Biodesign Institute, Tempe, AZ).
Ethical statement
Ethical approval for the collection of ticks from domestic
animals was obtained from the animal research ethics board
(Agreement # 14/160/550/2011) of the College of Veterinary
Medicine and Agriculture of Addis Ababa University. All
necessary permits were obtained from the administration and
210
agricultural office of each district and from each animal
owner. The collection of ticks in the field did not involve
privately owned, wildlife, national park or other protected
areas and endangered or protected species.
211
3. Results
3.1. Morphological identification of ticks
Ticks for the present study were collected from domestic
animals in Gachi, Abdalla, Wolmara, Ada‟a, Kimbibit, Arsi,
Arero, Moyale and Yabelo districts in Oromia regional state
in Ethiopia. A total of 2334 ixodid ticks collected from
domestic animals were morphologically identified using
taxonomic keys in to 12 species including Am. cohaerens
(n=496:364 male(m), 132 female(f)), Am. gemma
(n=122:70m, 52f), Am. variegatum (n=157: 146m, 11f), Hy.
truncatum (n=79: 62m, 17f), Hy. m. rufipes (n=29: 18m,
11f), Rh (Bo). decoloratus (n=495: 70m, 425f), Rh. e. evertsi
(n=65: 36m, 29f), Rh. pulchellus (n=751: 300m, 451f),
Rh. sanguienus (n=14: 9m, 5f), Rh. bergeoni (n=76: 44m,
32f), Rh. praetextatus (n=47: 33m, 14f) and Hae. leachi
(n=3: 0m, 3f). The photographs of each tick species
identified in the present study are depicted on figure 1. From
these morphologically identified ticks, a total of 85 ticks
from 3 to 11 specimens from each of the 12 tick species were
studied by MALDI TOF MS and molecular methods.
212
3.2. MALDI-TOF MS data
The legs from 85 tick specimens identified morphologically
in 12 species were subjected to MALDI-TOF MS to assess
intra-species reproducibility and inter-species specificity
(Table 1). MALDI-TOF MS analysis yielded unique
reproducible protein profile spectra for each of the 12 tick
species tested with the mass ranging from 2-20 kDa (Figure
1). Likewise, male and female ticks belonging to the same
tick species revealed visually similar protein profiles. To
confirm the intra-species reproducibility and inter-species
specificity of MS spectra and to visualize the distances
between the different tick species, MS protein profiles of 3
to 5 specimens per tick species from both genders were used
to generate a dendrogram (Figure 2). Clustering analysis
revealed that all specimens of the same species were grouped
on distinct branches. In addition, all specimens from the
same genus gathered in the same part of the dendrogram,
excepted ticks from the Rh. (Bo). decoloratus which were
not grouped with the species from Rhipicephalus geneses,
but were found in the same part as specimens from the
Amblyoma geneses (Figure 2).
213
3.3. Blind tests
MS spectra from the 11 tick species were queried
successively against the MS reference Database 1 and
Database 2 (i.e., Database 2 = Database 1 plus the leg spectra
from these 11 tick species stored in ethanol, for details see
M&M). The blind test of the 41 MSP from the legs of the
ticks preserved in ethanol against Database 1 produced the
LSVs ranging from 1.144 to 1.705. Interestingly, the tick
specimens which tick species counterparts were in the
Database 1 had higher LSVs. Nevertheless, these LSVs were
to low to be considered as reliable (LSVs < 1.8, threshold
value of identification).
Then, a total of 44 leg MS spectra, from 11 tick species with
3 to 5 specimens per species preserved in ethanol, were used
to implement MS reference Database 1, which was then
renamed Database 2 (Table 1). As solely 3 Ha. leachi
specimens were selected, they were used to control
misidentification rather than added them in the database. The
41 remaining MS spectra were then tested against Database
2, yielding to 100% of the MSP tested correctly identified at
the species level (Table 1). All queried specimens matched
with their counterpart species specimens preserved in ethanol
as first top-ranking hits. LSVs ranged from 1.698 to 2.626
214
(Table 1). With the exception of the specimens from the Rh.
praetextatus species, all the other specimens queried against
the Database 2 were identified with a LSVs of greater than
1.8 (this score was previously defined as the threshold values
of confident identification in ticks, Yssouf 2013).
215
respective homologies available in the GenBank (Table 2).
However, the remaining other four tick species (Am.
cohaerens, Am. gemma, Rh. bergeoni and Rh. praetextatus)
their corresponding 12S RNA gene sequence homology is
not available in the GenBank. BLAST analysis of the 12S
RNA gene indicated that these 4 tick species have high
intraspecies similarity among the specimens ranging from
98.9-100% (Table 2). However, these 4 tick species showed
only 92.9 to 94.7% sequence similarity to their closet tick
species available in the GenBank (Table 2). Thus, these 4
12S RNA gene sequences of Ethiopian ixodid ticks were
submitted to the GenBank for the first time as new sequences
under accession numbers, respectively.
A phylogenetic tree was constructed from the 12S rRNA
gene sequences of ticks of the present study and the
sequences of their homology in the GenBank using the
Neighbor-joining statistics of Mega 5. Result of this
phyogenetic analysis revealed that all the individual ticks
that belong to the same species cluster together with their
respective homology sequences in the GenBank for all the
11 ixodid species of Ethiopian ticks (Figure 3).
216
4. Discussion
Result of the morphological identification of ixodid ticks
revealed the presence of Am. cohaerens, Am. variegatum,
Am. gemma, Hy. truncatum, Hy. m. rufipes, Rh. (Bo.)
decoloratus, Rh. e. evertsi, Rh. pulchellus, Rh. sanguienus,
Rh. bergeoni and Rh. praetextatus (Fig 3) in Ethiopia. This
morphological finding was supported by MALDI TOF MS
technique that yielded reproducible protein profiles spectra
specific for each of the 11 tick species and was further
confirmed by molecular identification by amplification and
sequencing of the 12S rRNA gene of each tick species and
phylogenetic analysis. In support of our observation, several
previous morphological studies have reported the presence
of many species of ixodid ticks collected from domestic
animals in Ethiopia (Bekele, 2002; Mekonnen et al., 2001;
Regassa, 2001).
Recently, in our laboratory MALDI-TOF MS technique was
used to identify fresh tick (Yssouf et al., 2013a) and fresh
mosquitoes (Yssouf et al., 2014a; Yssouf et al., 2013b)
species from the spectra of protein extracted from their legs.
This prompted us to use MALDI-TOF MS technique to
identify Ethiopian ixodid ticks preserved in 70% ethanol.
Result of the first blind test against database 1 suggest that it
217
is not applicable to identify tick species or other arthropods
by MALDI-TOF MS technique when the reference database
is created from fresh samples is tested against 70% ethanol
preserved ticks. This observation was further strengthened
by the result of blind test 1 of less than 1.8 LSV recorded for
all the of the 70% ethanol preserved 3 species of Ethiopian
ticks (Amblyomma variegatum, Hyalomma m. rufipes,
Rhipicephalus sanguineus) that were tested against database
1 which was previously created in our laboratory from
spectra of legs of fresh arthropods including these 3
aforementioned species of ticks. Even though it was lower
than the set cutoff value for species differentiation by
MAMDI TOF, the LSVs for these 3 tick spp. is better than
all the other LSVs recorded in the blind test 1 suggesting the
ticks belong to the same species. This finding is consistent
with previous report of absence of similarities in the spectral
profile level between the fresh specimens and those stored in
70% ethanol of the same species of fleas (Yssouf et al.,
2014b) and mosquitoes (Kaufmann et al., 2011).
Results of the blind test 2 against database 2 (which contains
all the 12 Ethiopian tick species preserved in 70% ethanol)
demonstrated that all the 11 ixodid tick species produced
distinct, consistent and reproducible species-specific protein
218
spectra obtained from the legs of ticks preserved in 70%
ethanol. The tick reference database of our laboratory is now
upgraded by adding 11 different Ethiopian ixodid tick
species preserved in 70% ethanol. Overall, 10 tick species
specimens were identified by MALDI-TOF MS with spectra
scores >1.8 which is considered as reliable scores value for
the identification of bacterial genera by MALDI-TOF-MS as
has been used previously (Yssouf et al., 2013a). The finding
of species-specific protein spectra for both male and female
ticks that belong to the same species is also consistent with
the previous report from our laboratory (Yssouf et al.,
2013a). In support of our findings MALDI-TOF MS data
analysis has recently been used to identify various types of
arthropods including ticks (Hoppenheit et al., 2014;
Hoppenheit et al., 2013; Kaufmann et al., 2012b; Kaufmann
et al., 2011; Muller et al., 2013). The slightly lower LSV of
1.698-1.740 than the cutoff value of 1.8 recorded for the Rh.
praetextatus in our study could be attributable to low
spectral quality or strain variation among ticks of different
districts in Ethiopia. However, this LSVs still seems good
when compared to the lower LSVs of 1.3- 1.7 recorded for
up to 76% of ticks removed from patients previously (Yssouf
et al., 2013a).
219
This study is the first to report the use of MALDI-TOF-MS
and molecular tools for identification of ixodid ticks
preserved in 70% ethanol in Africa.
The observation of the hierarchical grouping of the mass
spectra of each individual tick specimen clustered according
to their respective genera and species on the dendrogram
(Fig. 2) using the MSP dendrogram function of the MALDI-
Biotyper 3.0 is in line with previous findings in several
arthropods (Karger et al., 2012; Kaufmann et al., 2011;
Yssouf et al., 2013a; Yssouf et al., 2014a; Yssouf et al.,
2013b; Yssouf et al., 2014b). In this study the clustering of
the specimens of the Rh. (Bo). decoloratus together with the
specimens from the Amblyomma genus in steady of the
grouping together with Rhipicephalus genus most probably
suggests that MSP dendrogram function of the
MALDIBiotyper 3.0 may not be good method for tick
phylogenetic study as has been argued previously (Karger et
al., 2012; Yssouf et al., 2013a).
The demonstration of satisfactory and reproducible data for
species identification of ixodid ticks from samples preserved
in 70% ethanol by MALDI-TOF MS is important for the fact
that 70% ethanol is the most widely utilized convenient
medium for storing samples for entomological surveys and
220
for processing by molecular techniques under field condition
around the world and in African countries including
Ethiopia. This finding is in line with the recent report of
identification of mosquitoes species preserved in alcohol
(Kaufmann et al., 2012a; Kaufmann et al., 2011).
The use of legs of adult ticks to identify tick species by
MALDI-TOF MS further confirms the previous report in our
laboratory where legs of fresh ticks and mosquitoes were
used for species identification (Yssouf et al., 2013a; Yssouf
et al., 2014a; Yssouf et al., 2013b). The use of protein profile
spectra extracted from legs of ticks for species identification
by MALDI-TOF MS has the advantage over the use of the
whole body or the abdomen of ticks due to the already
established fact that in hematophagous arthropods the
presence of blood has a considerable negative impact on
MALDI-TOF patterns and reduce the intensity of the
biomarker masses (Kaufmann et al., 2012a; Kaufmann et al.,
2011). However, our observations during the study period
(data not shown) show that the use of legs is more feasible
for large and medium body sized genera of ticks. In the case
of larvae and nymphs of ticks and small size genera of ticks
like Haemaphysalis species ticks the other body parts
excluding the abdomen seem to be suitable.
221
Among the ixodid ticks we have identified by MALDI-TOF
MS and 12S rRNA gene the species Am. cohaerens, Am.
variegatum, Rh (Bo). decoloratus, Rh. pulchellus, Rh. e.
eversti, Rh. praetextatus, and Hy. m. rufipes are widely
distributed ticks with great economic and public health
importance in different parts of Ethiopia (Mekonnen et al.,
2001; Stephany et al., 2009). For instance, recently the
detection of emerging and remerging human pathogenic
bacteria including spotted fever group rickettsiae, Coxiella
burnetii and new Borrelia spp. were reported from these
ticks (Kumsa et al., 2014c).
The 12S rRNA gene has been previously used as a reliable
tool for identification of different species of ixodid ticks
including Rhipicephalus ticks (Beati and Keirans, 2001). The
phylogenetic tree constructed from the 12S rRNA gene
sequences of tick and homology sequences of ticks in the
GenBank presented herein is consistent with previous studies
(Beati and Keirans, 2001). The phylogenetic tree constructed
from the 12S rRNA gene sequences of ticks of this study
presented herein is in agreement and confirmed our
morphological identification of ticks and MALDI-TOF MS
tick identification. Interestingly, all the tick spp. for which
nucleotide homology is available in the GenBank showed
222
high nucleotide identities of 99-100% with their respective
reference species in the GenBank and had high bootstrap
values in the 12S gene phylogenetic tree (Fig 3) (Beati and
Keirans, 2001). However, on the contrary in all the 4 tick
species (Am. cohaerens, Am. gemma, Rh. bergeoni and
Rh. praetextatus) for which identical 12S gene is missing in
the GenBank demonstrated low nucleotide homology with
reference 12S gene sequence in the GenBank to the closet
tick spp. and low bootstrap value in the 12S gene
phylogenetic tree and cluster near to the related tick spp.
(Fig. 3).
In conclusion the present study shows that protein profiling
by MALDI-TOF MS is suitable for identification of ixodid
ticks preserved in 70%. Further studies are needed to extend
the MALDI-TOF MS method of species identification to
other arthropods of public and economic importance.
Acknowledgements
The authors greatly acknowledge domestic animal owners in
the different districts of Oromia for their kindness in
allowing us to collect ticks from their animals.
223
References
Beati, L., Keirans, J.E., 2001. Analysis of the systematic
relationships among ticks of the genera
Rhipicephalus and Boophilus (Acari: Ixodidae) based
on mitochondrial 12S ribosomal DNA gene
sequences and morphological characters. J. Parasitol.
87, 32-48.
224
Chitimia, L., Lin, R.Q., Cosoroaba, I., Braila, P., Song, H.Q.,
Zhu, X.Q., 2009. Molecular characterization of hard
ticks from Romania by sequences of the internal
transcribed spacers of ribosomal DNA. Parasitol.
Res. 105, 1479-1482.
225
Fotso, F.A., Mediannikov, O., Diatta, G., Almeras, L.,
Flaudrops, C., Parola, P., Drancourt, M., 2014.
MALDI-TOF mass spectrometry detection of
pathogens in vectors: the Borrelia
crocidurae/Ornithodoros sonrai paradigm. PLoS
Negl. Trop. Dis. 8, e2984.
Karger, A., Kampen, H., Bettin, B., Dautel, H., Ziller, M.,
Hoffmann, B., Suss, J., Klaus, C., 2012. Species
determination and characterization of developmental
stages of ticks by whole-animal matrix-assisted laser
226
desorption/ionization mass spectrometry. Ticks. Tick.
Borne. Dis. 3, 78-89.
227
Kumsa, B., Socolovschi, C., Parola, P., Rolain, J.M., Raoult,
D., 2012b. Molecular detection of Acinetobacter
species in lice and keds of domestic animals in
Oromia Regional State, Ethiopia. PLoS One 7,
e52377.
228
Parola, P., Raoult, D., 2001. Ticks and tickborne bacterial
diseases in humans: an emerging infectious threat.
Clin. Infect. Dis. 32, 897-928.
229
Yssouf, A., Flaudrops, C., Drali, R., Kernif, T., Socolovschi,
C., Berenger, J.M., Raoult, D., Parola, P., 2013a.
Matrix-assisted laser desorption ionization-time of
flight mass spectrometry for rapid identification of
tick vectors. J. Clin. Microbiol. 51, 522-528.
230
Figure Legends
231
Table 1. Ixodid tick species used to create the reference database 2 of MALDI-TOF spectra and
arthropods used in the blind test.
No. of
No. of
specimens
specimens ID log score-
used for No. No.
Species Origin used to values*
the blind female male
create the [low-High]
test
database
procedure
Am. cohaerens Oromia 4 4 3 5 [1.823 - 1.963]
Am. gemma Oromia 5 2 2 5 [2.222 - 2.418]
Am. variegatum Oromia 4 4 2 6 [1.833 - 2.023]
Hy. truncatum Oromia 4 3 2 5 [1.965 - 2.003]
Hy. m. rufipes Oromia 3 4 1 6 [1.821 - 2.244]
Rh (Bo).
Oromia 3 5 7 1 [1.808 - 2.368]
decoloratus
Rh. e. evertsi Oromia 4 3 2 5 [2.013 - 2.077]
Rh. pulchellus Oromia 4 4 4 4 [1.843 - 2.173]
Rh. sanguienus Oromia 3 3 1 5 [2.051 - 2.134]
Rh. bergeoni Oromia 4 7 4 7 [1.801 - 2.626]
Rh. praetextatus Oromia 3 5 2 6 [1.698 - 1.740]
Ha. leachi# / 3 3 0 [0.538 -1.064]
Total 41 44 30 55
*Range of log score values. #Species not included in the database.
207
Table 2. Result of BLAT analysis of the 12S RNA gene of ticks
showing the sequence similarities among each specimen of the same
species and sequence similarity to their homology in the GenBank.
Tick species Number Intraspecies % Nearest homology sequence in % similarity to
speciemens GenBank
208
209
210
211
4. FLEAS
237
238
Fleas (order Siphonaptera) are holometabolous obligate
239
distribution and great economic significance of fleas in
the targeted genes and then the identity of the bacteria was
240
4.1. Article 6
241
248
Am. J. Trop. Med. Hyg., 90(3), 2014, pp. 457–462
doi:10.4269/ajtmh.13-0010
Copyright © 2014 by The American Society of Tropical Medicine and Hygiene
Molecular Detection of Rickettsia felis and Bartonella henselae in Dog and Cat Fleas
in Central Oromia, Ethiopia
Bersissa Kumsa, Philippe Parola, Didier Raoult, and Cristina Socolovschi*
Aix Marseille Université, Marseille, France; Department of Parasitology, College of Veterinary Medicine and Agriculture,
Addis Ababa University, Bishoftu, Ethiopia
Abstract. Fleas are important vectors of several Rickettsia and Bartonella spp. that cause emerging zoonotic diseases
worldwide. In this study, 303 fleas collected from domestic dogs and cats in Ethiopia and identified morphologically as
Ctenocephalides felis felis, C. canis, Pulex irritans, and Echidnophaga gallinacea were tested for Rickettsia and Bartonella
DNA by using molecular methods. Rickettsia felis was detected in 21% of fleas, primarily C. felis, with a similar prevalence
in fleas from dogs and cats. A larger proportion of flea-infested dogs (69%) than cats (37%) harbored at least one C. felis
infected with R. felis. Rickettsia typhi was not detected. Bartonella henselae DNA was detected in 6% (2 of 34) of C. felis
collected from cats. Our study highlights the likelihood of human exposure to R. felis, an emerging agent of spotted fever,
and B. henselae, the agent of cat-scratch disease, in urban areas in Ethiopia.
457
458 KUMSA AND OTHERS
receives an average annual rainfall of 8.00 mm3, has average using genus-specific quantitative PCR (qPCR) for spotted
maximum and minimum temperatures of 30.7 °C and 8.5 °C, fever group Rickettsia by targeting the gltA gene (RKND03
respectively, and a mean relative humidity of 61.3%.29 system)19,20 in accordance with the manufacturer’s recommen-
Fleas were collected during September–November 2011. dations and protocols (Applied Biosystems, Foster City, CA).
In the study area, the main rainy season is during mid-June– In addition, each DNA sample was tested for the bioB gene by
September, and the dry season is during November–May. All using a second qPCR with primers and a probe specific for R.
study animals were selected irrespective of sex, age, and breed felis.20,25,31 All DNA samples positive for either the gltA or
by using a simple random sampling method. The sex (female, bioB genes were further confirmed by a third qPCR specific
male) and age (young, adult) were recorded for each study for the orfB gene, which is highly sensitive and specific for R.
animal. The animals were categorized into two age groups, felis, by using described primers and probes.20,31 A sample was
young (£ 1 year) and adult (> 1 year) according to Kumsa and considered positive for R. felis when it was showed a positive
Mekonnen.1 In the town of Bishoftu, dogs generally stay in result for at least two of the genes described above.
gardens and their owners provide food. Cats also usually stay In addition, 21 randomly selected R. felis DNA-positive
in gardens and are provided with food, water, and sleeping samples were subjected to a standard PCR analysis specific
areas by the owners. In the present study, stray dogs and cats for the ompA gene with primers 190.70, 190.180, and 190.701
were not sampled. (Eurogentec, Seraing, Belgium), which amplify a 629–632-
Study methods. A cross-sectional study type was used to basepair fragment.20,31 Sterile water was used as a negative
determine the species of fleas infesting dogs and cats in Bishoftu control and DNA from R. montanensis was used as a positive
by using a house-to-house survey. All collected fleas were tested control. Sequencing was performed as described.32 Sequences
by using molecular methods for Bartonella spp. and spotted were aligned by using ChromasPro version 2.31 (http://en.bio-
fever and typhus group Rickettsia spp. soft.net/dna/chromas.html) and then analyzed and compared
Flea collection and identification. The skin of each study with the Rickettsia sequences available in GenBank by using
dog and cat was first rubbed with a piece of cotton soaked in the BLAST search tool.
ether and then combed for 10–15 minutes by using a one- All flea samples were also individually tested by genus-
toothed standard flea comb (0.08 cm distance between teeth), specific qPCR for typhus group Rickettsia spp. by using a
which other previous investigators recommended for flea col- PCR specific for the Rpr274P20,32 gene according to the man-
lection from dogs and cats.1,30 After combing, the flea comb ufacturer’s protocol using a 7900HT Fast Real-Time PCR
was held over a white plastic tray, and the fleas were collected System (Applied Biosystems, Foster City, CA). Sterile water
from the tray by using forceps. Echidnophaga gallinacea fleas, was used as a negative control and DNA from an R. typhi cell
usually found firmly attached to the skin around the eyelids, culture was used as a positive control.
snout, ears, lips and heads of the infested animals, were manu- All samples were individually tested for the presence of the
ally collected y using forceps and gloved hands. The collected 16S–23S ribosomal RNA intergenic spacer gene by qPCR
fleas (minimum = 1 and maximum = 23) from each infested with a 21-basepair probe specific for the genus Bartonella as
dog and cat were placed into pre-labeled small plastic tubes. described.8,25 Sterile water was used as a negative control and
All collected fleas from each animal were placed in separate DNA from B. elizabethae was used as a positive control. All
vials containing 70% ethanol and transported to the Labora- DNA samples positive for the intergenic spacer gene were
tory of the World Health Organization Collaborative Center further tested for B. henselae by qPCR specific for the pap31
for Rickettsial Diseases and Arthropod-borne Bacterial Dis- gene, which is specific for this species, as described.8,25
eases in Marseille, France. Ethics. Ethical approval for the collection of fleas from
Morphologic identification of fleas and molecular studies domestic animals was obtained from the animal research ethics
were performed from the end of 2011 through 2012. The flea board (Agreement # 14/160/550/2011) of the College of Veter-
species and sex were determined by using a microscope inary Medicine and Agriculture of Addis Ababa University.
according to standard morphologic identification keys as All necessary oral permits were obtained for the described field
described.1 The head shape, length of the first and second studies, including permission from the administration and agri-
spines of the genal comb, number of setae-bearing notches of cultural office of the district and from each animal owner. Flea
the dorsal aspect of the hind tibiae, and the number of setae on collection from the animals was not harmful to the animals.
the first segment of the neck were used to differentiate C. felis Data analysis. Microsoft (Redmond, WA) Excel was used
from C. canis as described.1 Photographs of body parts of each for data management. Descriptive statistics, such as percentages
flea were made. and means, were used to summarize the proportions of flea
DNA extraction from fleas. Before DNA extraction, each infestations among the animals and infection of fleas with
flea specimen was rinsed twice in sterile water for 15 minutes R. felis. Statistical analysis was performed by using EpiInfo 7
and dried on sterile filter paper. Each specimen was dissected (Centers for Disease Control and Prevention, Atlanta, GA),
laterally into two halves. Genomic DNA was extracted from Associations between the sex and age groups of the dogs and
each specimen by using the QIAamp DNA Tissue Extraction cats, species and sex of fleas, and prevalence of flea infestation
Kit (QIAGEN, Hilden, Germany) according to the manufac- and R. felis in fleas were determined by using the Mantel-
turer’s instructions. The DNA from each flea was eluted in Haenszel test and EpiInfo 7. A P value £0.05 was con-
100 mL of Tris-EDTA buffer and stored at −20 °C under ster- sidered significant.
ile conditions to preclude contamination until the sample was
used for polymerase chain reaction. The second half of each RESULTS
flea was stored at −80 °C as a backup sample.
Molecular detection of Rickettsia and Bartonella species. Identification of fleas. In this study, 64% (30 of 47) of dogs
The DNA extracted from the fleas was individually tested by and 50% (8 of 16) of cats were infested with at least with one
BARTONELLA HENSELAE AND RICKETTSIA FELIS IN FLEAS FROM ETHIOPIA 459
Table 1
Species composition and overall percentage of flea infestations by sex and age of dogs and cats in Oromia, Ethiopia*
Host factors Host sex Host age
Dog Ctenophalides felis 11/19 (58) (90:77) 18/28 (64) (114:38) 12/17 (71) (79:24) 17/30 (57) (112:27) 29/47 (62) (191:51)
C. canis 2/19 (10) (2:4) 3/28 (11) (4:6) 3/17 (18) (4:9) 2/30 (7) (2:1) 5/47 (11) (6:10)
Pulex irritans 2/19 (10) (1:1) 6/28 (21) (2:5) 4/17 (23) (2:3) 4/30 (13) (1:3) 8/47 (17) (3:6)
Echidnophaga gallinacea 0/19 (0) 2/28 (7) (2:0) 1/17 (6) (1:0) 1/30 (3) (1:0) 2/47 (4) (2:0)
Overall in dogs 11/19 (58) (80:18) 19/28 (68) (122:49) 13/17 (76) (86:36) 17/30 (57) (116:31) 30/47 (64) (202:67)
Cat C. felis 6/10 (60) (19:12) 2/6 (33) (2:1) 6/7 (86) (17:9) 2/9 (22) (4:4) 8/16 (50) (21:13)
*Values are no. infested/no. tested (%) (no. females:no. males).
species of flea (Table 1). Of the 269 fleas (202 females and C. felis and E. gallinacea (3%, 1 of 30) mixed species infesta-
67 males collected from dogs, 242 C. felis (191 males and tions recorded in dogs was lower than all other types of mixed
51 females), 16 C. canis (6 females and 10 males), 9 P. irritans species infestations (Table 2).
(three females and six males) and 2 E. gallinacea (two Molecular detection of Rickettsia spp. in fleas. Molecular
females) specimens were distinguished by morphologic iden- analysis of 303 fleas collected from dogs and cats identified
tification. In female dogs, an overall prevalence of 58% (11 of R. felis DNA in 63 fleas (21%) (Table 3). In our study, R. felis
19) and 98 fleas (80 females and 18 males) was recorded. The DNA was detected in 21% (56 of 269) of fleas from dogs and
corresponding values for male dogs were 68% (19 of 28) and in 21% (7 of 34) of fleas from cats. The overall prevalence of
171 fleas (122 females and 49 males), respectively (Table 1). R. felis was not significantly different between fleas obtained
An overall prevalence of 76% (13 of 17) and 57% (17 of 30) from dogs and cats. In dogs, 11% (1 of 9) of P. irritans and 6%
was observed in young and adult dogs, respectively (Table 1). (1/16) of C. canis specimens were positive for R. felis DNA.
A total of 122 fleas (86 females and 36 males) and 147 fleas However, R. felis DNA was not detected in E. gallinacea from
(116 females and 31 males) were collected from young and dogs (Table 3). The overall percentage of R. felis was higher
adult dogs, respectively (Table 1). There was no statistically in C. felis than in C. canis (29 of 47 versus 5 of 47; P < 0.0001).
significant difference in the overall prevalence of flea infes- However, no statistically significant differences were observed
tations between dogs of different ages (13 of 17 versus 17 of in the overall prevalence of R. felis between female and male
30; P = 0.2, by Mantel-Haenszel test) and sex (11 of 19 versus C. felis from dogs (46 of 191 versus 8 of 51; P = 0.2, by Mantel-
19 of 28; P = 0.5, by Mantel-Haenszel test) (Table 1). Haenszel test) and cats (4 of 21 versus 3 of 13; P = 0.8, Mantel-
A total of 34 (21 females and 13 males) C. felis were identi- Haenszel test) (Table 3).
fied on cats (Table 1). In this study, C. felis was the most The overall prevalence of R. felis was 21% (21 of 98) and
prevalent flea species on both hosts; overall prevalence of 20% (35 of 171) among fleas collected from female and male
62% (29 of 47) on dogs (29 of 47 versus 8 of 47; P = 0.00002, dogs, respectively. The corresponding values for fleas col-
by Mantel-Haenszel test) and 50% (8 of 16) on cats. Pulex lected from young and adult dogs were 19% (23 of 122) and
irritans, with an overall prevalence of 17% (8 of 47), was the 22% (33 of 147), respectively (Table 3). No statistically signif-
second most prevalent species, followed by C. canis; the least icant differences were observed in the overall prevalence of
prevalent species on dogs was E. gallinacea (Table 1). R. felis between the fleas collected from female versus male
For dogs (191 of 242 versus 51 of 242; P < 0.0001, by Mantel- dogs and young versus adult dogs.
Haenszel test) and cats (21 of 34 versus 13 of 34; P = 0.05, by Analysis of flea species positive for R. felis DNA on the
Mantel-Haenszel test), the number of female C. felis was sig- infested animals showed that 69% (20 of 29) of dogs infested
nificantly higher than the number of males (Table 1). Further- with C. felis harbored at least one flea that contained R. felis
more, the prevalence of C. felis monospecies in dogs (60%) (18 DNA, whereas 7% (2 of 29) of dogs infested with C. felis
of 30 versus 5 of 30; P = 0.0006, by Mantel-Haenszel test) and harbored fleas that contained R. felis DNA (Table 4). How-
cats (100%, 8 of 8) was significantly higher than all other types ever, in the case of C. canis, P. irritans, and E. gallinacea, only
of mixed species infestations (Table 2). The prevalence of a small percentage of infested dogs harbored fleas that
contained R. felis DNA (Table 4). In contrast to dogs, only
37% (3 of 8) of cats infested with C. felis harbored at least one
flea positive for R. felis DNA (Table 4).
Table 2
The mean ± SD qPCR cycle threshold (Ct) values for posi-
Percentage of mono- and multiple flea species infestations in positive
dogs and cats in Oromia, Ethiopia*
tive samples were 21.1 ± 6.6 for the orfB gene, 28.4 ± 3.1 for
Flea spp. Dog Cat Overall
the bioB, and 31.3 ± 2.3 for the gltA gene.
We amplified a 629–632-baspair fragment of the ompA
Ctenocephalides felis 18/30 (60)† 8/8 (100) 26/38 (68)
gene from 18 samples that were representative of all of the
C. felis and C. canis 2/30 (7) – 2/30 (7)
C. felis and Pulex irritans 5/30 (17)† – 5/30 (17) species of fleas collected. A BLAST analysis of the ompA
C. felis, C. canis, and P. irritans 3/30 (10) – 3/30 (10) gene sequence from all flea species showed 99.6–100% nucle-
C. felis and Echidnophaga 1/30 (3) – 1/30 (3) otide homology with the GenBank reference sequence of
gallinacea R. felis obtained from C. felis in Yucatan, Mexico (GenBank
E. gallinacean 1/30 (3) – 1/30 (3)
Overall frequency 30/30 (100) 8/8 (100) 38/38 (100) accession no. AJ563398), and one sample was 100% simi-
lar to R. felis originating from Guatemala and Costa Rica
*Values are no. positive/no. tested %).
†P = 0.0006, by Mantel-Haenszel test. (GenBank accession no. JN990593).
460 KUMSA AND OTHERS
Table 3
Prevalence of Rickettsia felis in different flea species collected from dogs and cats in Oromia, Ethiopia*
Host factor Host sex Host age
Host Flea spp. Flea sex Female host Male host Young host Adult host Total hosts
Dog Ctenocephalides felis F 20/77 (26) 26/114 (23) 20/79 (25) 26/112 (23) 46/191 (24)
M 1/13 (8) 7/38 (18) 2/24 (8) 6/27 (22) 8/51 (16)
Total 21/90 (23) 33/152 (22) 22/103 (21) 32/139 (23) 54/242 (22)
C. canis F 0/2 (0) 0/4 (0) 0/4 (0) 0/2 (0) 0/6 (0)
M 0/4 (0) 1/6 (17) 0/9 (0) 1/1 (100) 1/10 (10)
Total 0/6 (0) 1/10 (10) 0/13 (0) 1/3 (33) 1/16 (6)
Pulex irritans F 0/1 (0) 1/2 (50) 1/2 (50) 0/1 (0) 1/3 (33)
M 0/1 (0) 0/5 (0) 0/3 (0) 0/3 (0) 0/6 (0)
Total 0/2 (0) 1/7 (14) 1/5 (20) 0/4 (0) 1/9 (11)
Echidnophaga gallinacea F – 0/2 (0) 0/1 (0) 0/1 (0) 0/2 (0)
M – – – – –
Total – 0/2 (0) 0/1 (0) 0/1 (0) 0/2 (0)
Overall: dogs 21/98 (21) 35/171 (20) 23/122 (19) 33/147 (22) 56/269 (21)
Cat C. felis F 4/19 (21) 0/2 (0) 2/17 (12) 2/4 (50) 4/21 (19)
M 3/12 (25) 0/1(0) 2/9 (22) 1/4 (25) 3/13 (23)
Total 7/31 (23) 0/3 (0) 4/26 (15) 3/8 (37) 7/34 (21)
Overall: dogs and cats 28/129 (22) 35/174 (20) 27/148 (18) 36/155 (23) 63/303 (21)
*Values are no. positive/no. tested (%).
No typhus group Rickettsia species were detected in any of been reported, a variation that is most likely attributable to
the 303 fleas collected from dogs and cats in Bishoftu in cen- differences in geographic and animal factors. The observation
tral Oromia. of a higher overall percentage of R. felis in C. felis than in
Molecular detection of Bartonella spp. in fleas. Two of 303 other flea species and detection of R. felis in C. canis and
fleas tested were positive for Bartonella spp. by Bartonella P. irritans collected from dogs in our study is consistent with
qPCR and B. henselae-specific qPCR, and had mean ± SD those of other reports.18,19 The absence of a significant differ-
Ct values of 31.0 ± 3.2 and 21.2 ± 1.1, respectively. Bartonella ence in the overall prevalence of R. felis between fleas from
henselae DNA was detected in 6% (2 of 34) C. felis collected female and male dogs and fleas from young and adult dogs in
from cats. One C. felis was coinfected with R. felis. this study suggests that the sex and age of the host does not
affect the prevalence of R. felis in fleas.
DISCUSSION Average Ct values for the orfB, bioB, and gltA genes in
detection of R. felis DNA in fleas in the present study are
We detected R. felis in C. canis collected from dogs. The consistent with reported values.31 This variation in Ct values
results of our study also provide additional molecular and most likely reflects differences in sensitivity and specificity
geographic evidence for R. felis in C. felis and P. irritans among different assays and specific gene targets for the DNA
specimens collected from dogs and cats in central Oromia. of this bacterium. In addition, fifty randomly selected second
Our molecular findings were validated by the facts that all half-cut female flea backup samples were negative for R. felis
negative control samples had negative results and all positive DNA, confirming our initial negative observations for the
control samples had positive findings in all quantitative and corresponding first half-cut female flea samples.
conventional PCRs. The finding of a higher proportion of flea-infested dogs
The detection of an overall percentage of 21% of R. felis (69%) than infested cats (37%) carrying C. felis that contain
DNA in fleas collected from dogs and cats is comparable to R. felis DNA is consistent with previous findings.22 The sig-
the prevalence of R. felis reported by investigators in several nificantly higher count, prevalence, and number of species of
countries.14,18,21,33 However, overall prevalence values that fleas on dogs than on cats in this and previous studies1 under-
are lower17,25 and higher12,22 than our results for R. felis have scores the need for additional studies to clearly determine the
relative importance of dogs and cats in the transmission of
Table 4 R. felis. In addition, we detected R. felis in one of two C. felis
Proportion of infested dogs and cats harboring fleas infected with collected from a goat in Ethiopia, but R. felis DNA was not
Rickettsia felis and Bartonella henselae in Oromia, Ethiopia* detected in one C. felis flea taken from a cow.
R. felis B. henselae Furthermore, we report detection of B. henselae in C. felis
Flea spp. Infection status Dog Cat Cat collected from cats in Ethiopia. Bartonella henselae DNA was
Ctenocephalides At least 1 infected 20/29 (69) 3/8 (37) 1/8 (12) detected in only in 6% (2 of 34) of C. felis fleas collected from
felis All infected 2/29 (7) 0/8 (0) 1/8 (12) cats in our study. This finding suggests that cats are important
None infected 7/29 (24) 5/8 (62) 6/8 (75) sources of this bacterium and is consistent with reports of
C. canis At least 1 infected 0/5 (0) – – B. henselae in fleas from several countries worldwide.18,25,26,33
All infected 1/5 (20)
Bartonella henselae is the major cause of cat-scratch disease,
– –
None infected 4/5 (80) – –
Pulex irritans At least 1 infected 0/8 (0) – – for which cats are implicated as the main animal reser-
All infected 1/8 (12) – – voir.6,7,24,25,34 The finding of a single C. felis coinfected with
None infected 7/8 (87) – – B. henselae and R. felis in this study implies the possibility of
Echidnophaga None infected 2/2 (100) – – dual infections in exposed humans. Benign lymphadenopathy,
gallinacea
low-grade fever, malaise and aching, headache, anorexia, and
BARTONELLA HENSELAE AND RICKETTSIA FELIS IN FLEAS FROM ETHIOPIA 461
splenomegaly are the major clinical signs in patients affected In conclusion, our study shows that C. felis is the most
with cat-scratch disease.6,7,24 common ectoparasite of dogs and cats in urban central
The high overall prevalence of flea infestation in dogs Oromia, Ethiopia. This study provides molecular evidence of
(64%) and cats (50%) in the present study supports the obser- R. felis and B. henselae in the fleas collected from dogs and
vation of a prevalence of 99.5% in dogs and 91.5% in cats in cats in this region. To our knowledge, this is the first study to
Hawassa in southern Ethiopia.1 These observations suggest report the presence of B. henselae in fleas collected from cats
the presence of favorable climatic conditions important for in Ethiopia. The high prevalence of flea infestations in dogs
flea survival, reproduction, and development on dogs and cats and cats and the finding that a high percentage of fleas are
in the study area. These findings also indicate that in Ethiopia, infected with these important zoonotic bacteria in this study
dogs and cats are not generally treated for any ectoparasites, suggest that R. felis and B. henselae may be more important in
including fleas. Ethiopia than previously believed. Our study demonstrates
The absence of statistically significant variations in the that dogs, cats and their fleas may play important roles in
overall prevalence of flea infestation among dogs of different increasing the risk of human exposure to these pathogenic
age and sex groups in our study is consistent with those of bacteria in this country. Therefore, R. felis and B. henselae
earlier observations.5,35,36 Previous investigators argued that should be considered in the differential diagnosis of any fever
differences in the prevalence of flea infestation is associated of unknown cause among patients in this country. Future
with host habitat and environmental factors rather than host- studies are required to address the isolation and culture of
dependent factors.36 Female C. felis were more abundant than the bacteria, serologic and molecular prevalence, clinical
males in dogs and cats, a result consistent with previous find- signs, treatment, and potential risk factors in humans in
ings.1,3,5,35,37–39 Factors such as the longer time required by Ethiopia. Additional epidemiologic studies in different animal
female fleas to imbibe large quantities of blood for reproduc- and arthropod species and the public health significance of
tion, higher female survival rates, and the greater ability of these zoonotic bacteria in different agroecologic zones are
females to evade capture during host grooming have been urgently needed in this country.
suggested as possible explanations.1,3,5,35–39
The higher overall percentage, overall flea count. and four Received January 8, 2013. Accepted for publication June 22, 2013.
flea species on dogs compared with cats (C. felis only) in the Published online January 20, 2014.
present study is consistent with findings of a previous report1
Acknowledgments: We thank dog and cat owners in Bishoftu for
and the following reports: three species on dogs and one spe- allowing us to collect fleas from their animals, and laboratory techni-
cies on cats in Libya,3 three species on dogs and one species cians at URMITE (Marseille, France), particularly Veronique Brice,
on cats in Albania,35 and three species on dogs and two spe- Marielle Bedotto, and Annick Bernard, for technical support.
cies on cats in Hungary.5 These finding suggest that dogs are Authors’ addresses: Bersissa Kumsa, Didier Raoult, and Cristina
the preferred hosts of fleas, which is most likely caused by Socolovschi, URMITE, UMR, CNRS 6236, IRD 198, Faculté de
stronger grooming behavior of cats in removing fleas from Médecine, Aix Marseille Université, Marseille, France, E-mails:
their bodies.36,37 In general, dogs have thicker, longer, and bersissak@yahoo.com, didier.raoult@gmail.com, and cristina
.socolovschi@univ-amu.fr. Philippe Parola, Service des Maladies
denser fur than cats, which favors infestations. Infectieuses et Tropicales, Hôpital Nord, Marseille, France, E-mail:
The predominance of C. felis over all other flea species in philippe.parola@gmail.com.
both host species is in consistent with results of an earlier
study in Ethiopia1 and with other studies worldwide.5,37–39
Ctenocephalides felis is generally considered the predominant REFERENCES
species found on dogs and cats, and has replaces C. canis as
the most common flea species on domestic dogs in many 1. Kumsa B, Mekonnen S, 2011. Ixodid ticks, fleas and lice infest-
ing dogs and cats in Hawassa, southern Ethiopia. Onderstepoort
countries.37 This flea species is better adapted than C. canis J Vet Res 78: 1–4.
to a wider range of environmental factors.38,39 Furthermore, 2. Dobler G, Pfeffer M, 2011. Fleas as parasites of the family
previous studies have shown that C. felis is the most common Canidae. Parasit Vectors 4: 139.
flea in urban areas, whereas C. canis is the most common in 3. Kaal JF, Baker K, Torgerson PR, 2006. Epidemiology of flea
rural areas.30 The finding that C. canis was the second most infestation of ruminants in Libya. Vet Parasitol 141: 313–318.
4. Bitam I, Dittmar K, Parola P, Whiting MF, Raoult D, 2010. Fleas
abundant species, followed by P. irritans and E. gallinacea, on and flea-borne diseases. Int J Infect Dis 14: e667–e676.
dogs in our study is consistent with results of reports from 5. Farkas R, Gyurkovszky M, Solymosi N, Beugnet F, 2009. Prev-
Hungary,5 Albania,35 Spain,36 and Iran.39 Consistent with pre- alence of flea infestation in dogs and cats in Hungary com-
vious reports,1,3,36 E. gallinacea is a species frequently found bined with a survey of owner awareness. Med Vet Entomol
23: 187–194.
on birds and only occasionally reported on dogs and cats in 6. Chomel BB, Boulouis HJ, Maruyama S, Breitschwerdt EB, 2006.
various regions because of transient infestations from contact Bartonella spp. in pets and effect on human health. Emerg
with birds. In our study, the absence of X. cheopis, the tropi- Infect Dis 12: 389–394.
cal rat flea, is explained by the fact that this flea species is 7. Chomel BB, Kasten RW, 2010. Bartonellosis, an increasingly
rarely found on carnivores except during seasons of higher rat recognized zoonosis. J Appl Microbiol 109: 743–750.
8. Rolain JM, Bourry O, Davoust B, Raoult D, 2005. Bartonella
density and occurrence of disease epizootics. In addition, quintana and Rickettsia felis in Gabon. Emerg Infect Dis 11:
highland agroecology is more favorable for X. cheopis than 1742–1744.
areas such as Bishoftu, which is located in midland altitudes. 9. Parola P, 2011. Rickettsia felis: from a rare disease in the USA to
Possible explanations for the absence of C. felis strongylus in a common cause of fever in sub-Saharan Africa. Clin Microbiol
Infect 17: 996–1000.
our study are that this flea species is adapted to domestic 10. Hirunkanokpun S, Thepparit C, Foil LD, Macaluso KR, 2011.
ruminants, as observed in Libya3 or is restricted to rural areas Horizontal transmission of Rickettsia felis between cat fleas,
and stray dogs and cats in Ethiopia. Ctenocephalides felis. Mol Ecol 20: 4577–4586.
462 KUMSA AND OTHERS
11. Socolovschi C, Mediannikov O, Sokhna C, Tall A, Diatta G, 26. Tsai YL, Lin CC, Chomel BB, Chuang ST, Tsai KH, Wu WJ,
Bassene H, Trape JF, Raoult D, 2010. Rickettsia felis-associated Huang CG, Yu JC, Sung MH, Kass PH, Chang CC, 2011.
uneruptive fever, Senegal. Emerg Infect Dis 16: 1140–1142. Bartonella infection in shelter cats and dogs and their ectopar-
12. Richards AL, Jiang J, Omulo S, Dare R, Abdirahman K, Ali A, asites. Vector Borne Zoonotic Dis 11: 1023–1030.
Sharif SK, Feikin DR, Breiman RF, Njenga MK, 2010. Human 27. Tiao N, Darrington C, Molla B, Saville WJ, Tilahun G, Kwok
infection with Rickettsia felis, Kenya. Emerg Infect Dis 16: OC, Gebreyes WA, Lappin MR, Jones JL, Dubey JP, 2013.
1081–1086. An investigation into the seroprevalence of Toxoplasma
13. Maina AN, Knobel DL, Jiang J, Halliday JO, Feikin DR, gondii, Bartonella spp., feline immunodeficiency virus (FIV),
Cleaveland S, Ng’ang’a Z, Junghae M, Breiman RF, Richards and feline leukaemia virus (FeLV) in cats in Addis Ababa,
AL, Njenga MK, 2012. Rickettsia felis infection in febrile Ethiopia. Epidemiol Infect 141: 1029–1033.
patients, western Kenya, 2007–2010. Emerg Infect Dis 18: 28. Meheretu Y, Leirs H, Welegerima K, Breno M, Tomas Z, Kidane
328 –331. D, Girmay K, de Bellocq JG, 2013. Bartonella prevalence and
13. Hirunkanokpun S, Thepparit C, Foil LD, Macaluso KR, 2011. genetic diversity in small mammals from Ethiopia. Vector
Horizontal transmission of Rickettsia felis between cat fleas, Borne Zoonotic Dis 13: 164–175.
Ctenocephalides felis. Mol Ecol 20: 4577–4586. 29. Central Statistical Agency, 2008. Ethiopian Agricultural Sample
14. Gilles J, Just FT, Silaghi C, Pradel I, Lengauer H, Hellmann K, Survey Report on Livestock and Livestock Characteristics.
Pfister K, 2008. Rickettsia felis in fleas, France. Emerg Infect Volume II, 2007/08. Addis Ababa, Ethiopia: Central Statis-
Dis 14: 684–686. tical Agency.
15. Maioli G, Horta MC, Ogrzewalska M, Capelli G, Souza SO, 30. Marchiondo AA, Holdsworth PA, Green P, Blagburn BL, Jacobs
Richtzenhain LJ, Labruna MB, 2009. First detection of Rickettsia DE, 2007. World Association for the Advancement of Veteri-
felis in Ctenocephalides felis fleas from Italy. Clin Microbiol nary Parasitology (W.A.A.V.P.) guidelines for evaluating the
Infect 15 (Suppl 2): 222–223. efficacy of parasiticides for the treatment, prevention and con-
16. Nogueras MM, Pons I, Ortuno A, Lario S, Segura F, 2011. Rickettsia trol of flea and tick infestation on dogs and cats. Vet Parasitol
felis in fleas from Catalonia (northeast Spain). Vector Borne 145: 332–344.
Zoonotic Dis 11: 479–483. 31. Socolovschi C, Pages F, Ndiath MO, Ratmanov P, Raoult D, 2012.
17. Mokhtar AS, Tay ST, 2011. Molecular detection of Rickettsia felis, Rickettsia species in African Anopheles mosquitoes. PLoS ONE
Bartonella henselae, and B. clarridgeiae in fleas from domestic 7: e48254.
dogs and cats in Malaysia. Am J Trop Med Hyg 85: 931–933. 32. Kumsa B, Socolovschi C, Parola P, Rolain JM, Raoult D, 2012.
18. Boudebouch N, Sarih M, Beaucournu JC, Amarouch H, Hassar Molecular detection of Acinetobacter species in lice and keds
M, Raoult D, Parola P, 2011. Bartonella clarridgeiae, B. of domestic animals in Oromia Regional State, Ethiopia. PLoS
henselae and Rickettsia felis in fleas from Morocco. Ann Trop ONE 7: e52377.
Med Parasitol 105: 493–498. 33. Tsai KH, Huang CG, Fang CT, Shu PY, Huang JH, Wu WJ, 2011.
19. Mediannikov O, Abdissa A, Diatta G, Trape JF, Raoult D, 2012. Prevalence of Rickettsia felis and the first identification of
Rickettsia felis in fleas, southern Ethiopia, 2010. Emerg Infect Bartonella henselae Fizz/CAL-1 in cat fleas (Siphonaptera:
Dis 18: 1385–1386. Pulicidae) from Taiwan. J Med Entomol 48: 445– 452.
20. Socolovschi C, Pages F, Raoult D, 2012. Rickettsia felis in Aedes 34. Saisongkorh W, Rolain JM, Suputtamongkol Y, Raoult D, 2009.
albopictus mosquitoes, Libreville, Gabon. Emerg Infect Dis 18: Emerging Bartonella in humans and animals in Asia and
1687–1689. Australia. J Med Assoc Thai 92: 707–731.
21. Abramowicz KF, Wekesa JW, Nwadike CN, Zambrano ML, 35. Xhaxhiu D, Kusi I, Rapti D, Visser M, Knaus M, Lindner T,
Karpathy SE, Cecil D, Burns J, Hu R, Eremeeva ME, 2012. Rehbein S, 2009. Ectoparasites of dogs and cats in Albania.
Rickettsia felis in cat fleas, Ctenocephalides felis parasitiz- Parasitol Res 105: 1577–1587.
ing opossums, San Bernardino County, California. Med Vet 36. Gracia MJ, Calvete C, Estrada R, Castillo JA, Peribanez MA,
Entomol 26: 458–462. Lucientes J, 2008. Fleas parasitizing domestic dogs in Spain.
22. Troyo A, Álvarez D, Taylor L, Abdalla G, Calderón-Arguedas Vet Parasitol 151: 312–319.
Ó, Zambrano ML, Dasch GA, Lindblade K, Hun L, Eremeeva 37. Fuehrer HP, Igel P, Treiber M, Baumann TA, Riedl J, Swoboda
ME, Estévez A, 2012. Rickettsia felis in Ctenocephalides felis P, Joachim A, Noedl H, 2012. Ectoparasites of livestock, dogs,
from Guatemala and Costa Rica. Am J Trop Med Hyg 86: and wild rodents in the Chittagong Hill Tracts in southeastern
1054–1056. Bangladesh. Parasitol Res 111: 1867–1870.
23. Raoult D, La Scola B, Enea M, Fournier PE, Roux V, Fenollar F, 38. Slapeta J, King J, McDonell D, Malik R, Homer D, Hannan P,
Galvao MA, de Lamballerie X, 2001. A flea-associated Rickettsia Emery D, 2011. The cat flea (Ctenocephalides f. felis) is the
pathogenic for humans. Emerg Infect Dis 7: 73–81. dominant flea on domestic dogs and cats in Australian veteri-
24. Guptill L, 2010. Bartonellosis. Vet Microbiol 140: 347–359. nary practices. Vet Parasitol 180: 383–388.
25. Rolain JM, Franc M, Davoust B, Raoult D, 2003. Molecu- 39 Tavassoli M, Ahmadi A, Imani A, Ahmadiara E, Javadi S,
lar detection of Bartonella quintana, B. koehlerae, B. henselae, Hadian M, 2010. Survey of flea infestation in dogs in differ-
B. clarridgeiae, Rickettsia felis, and Wolbachia pipientis in cat ent geographical regions of Iran. Korean J Parasitol 48:
fleas, France. Emerg Infect Dis 9: 338–342. 145–149.
5. LICE AND SHEEP
KED
249
Lice and sheep keds (Melophagus ovinus) are two of the most
(18,19).
251
with small ruminants during the night to protect them from
252
5.1. Article 7
doi:10.1371/journal.pone.0052377.
253
264
Molecular Detection of Acinetobacter Species in Lice and
Keds of Domestic Animals in Oromia Regional State,
Ethiopia
Bersissa Kumsa1,2, Cristina Socolovschi2, Philippe Parola2, Jean-Marc Rolain2, Didier Raoult2*
1 Department of Parasitology, College of Veterinary Medicine and Agriculture, Addis Ababa University, Bishoftu, Ethiopia, 2 Aix Marseille Université, URMITE, UM63, CNRS
7278, IRD 198, Inserm 1095, Marseille, France
Abstract
This study was conducted to determine the presence of Acinetobacter and Rickettsia species DNA in lice and Melophagus
ovinus (sheep ked) of animals from Oromia Regional State in Ethiopia. From September through November 2011, a total of
207 cattle, 85 sheep, 47 dogs and 16 cats were examined for ectoparasites. Results of morphological identification revealed
several species of ectoparasites: Linognathus vituli (L. vituli), Bovicola bovis (B. bovis) and Solenopotes capillatus (S. capillatus)
on cattle; B. ovis and Melophagus ovinus (M. ovinus) on sheep; and Heterodoxus spiniger (H. spiniger) on dogs. There was a
significantly (p#0.0001) higher prevalence of L. vituli observed in cattle than both S. capillatus and B. bovis. Molecular
identification of lice using an 18S rRNA gene analysis confirms the identified lice species by morphological methods. We
detected different Acinetobacter species among lice (11.1%) and keds (86.4%) including A. soli in L. vituli of cattle, A. lowffii in
M. ovinus of sheep, A. pittii in H. spiniger of dogs, 1 new Acinetobacter spp. in M. ovinus and 2 new Acinetobacter spp. in H.
spiniger of dogs using partial rpoB gene sequence analysis. There was a significantly higher prevalence of Acinetobacter spp.
in keds than in lice (p#0.00001). Higher percentage of Acinetobacter spp. DNA was detected in H. spiniger than in both B.
ovis and L. vituli (p#0.00001). Carbapenemase resistance encoding genes for blaOXA-23, blaOXA-24, blaOXA-58, blaNDM-1
and blaOXA-51 were not found in any lice and keds. These findings suggest that synanthropic animals and their
ectoparasites might increase the risk of human exposure to zoonotic pathogens and could be a source for Acinetobacter
spp. infections in humans. However, additional epidemiological data are required to determine whether ectoparasites of
animals can act as environmental reservoirs and play a role in spreading these bacteria to both animal and human hosts.
Citation: Kumsa B, Socolovschi C, Parola P, Rolain J-M, Raoult D (2012) Molecular Detection of Acinetobacter Species in Lice and Keds of Domestic Animals in
Oromia Regional State, Ethiopia. PLoS ONE 7(12): e52377. doi:10.1371/journal.pone.0052377
Editor: Patrick C. Y. Woo, The University of Hong Kong, China
Received August 1, 2012; Accepted November 12, 2012; Published December 17, 2012
Copyright: � 2012 Kumsa et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The authors have no support or funding to report.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: didier.raoult@gmail.com
renamed as A. pittii, Acinetobacter genomic sp. 13TU as A. nosocomialis districts. In addition, October is a post rainy month with only few
and Acinetobacter genomic sp. G13 as A.lwoffii [16]. days of raining while November is a month during which the dry
In humans, members of the genus Acinetobacter have emerged as season commences hence is without raining or rarely for some
opportunistic pathogens, and are frequently implicated in various days. The farming system in all the districts is characterized by a
types of infections, especially in immunocompromised individuals mixed crop-livestock production system. The livestock in the study
and intensive health care units [14] throughout the world. In areas are traditionally managed under extensive production
tropical countries, they are associated with severe community- systems [32].
acquired infections [13]. They have the capability to survive for
prolonged periods under a wide range of environmental conditions Collection and morphological identification of
[13]. A. baumannii is described as the most common species in this ectoparasites
genus frequently associated with outbreaks and has been Lice and flies were carefully removed manually, using forceps or
repeatedly reported to develop a high level of resistance against by hand, to avoid any damage to the body and were then placed in
all available classes of antimicrobial drugs in many parts of the vials containing 70% ethanol for subsequent identification. In the
world [12]. A. pittii is reported as the second most commonly case of lice collected from dogs, the dog’s body was brushed for ten
isolated Acinetobacter species after A. baumannii in human patients minutes with a flea comb as previously described [8]. All lice and
[16,17]. More recently other less known species such as A. lwoffii flies from the same animal were put in the same vial and
and A. soli have been associated with serious infections and transported to the Laboratory of the World Health Organization
considered as emerging pathogens [14,19,20]. For instance A. Collaborative Center for Rickettsial Diseases and Arthropod-
lwoffii has been associated with acute gastroenteritis in USA [21],
borne Bacterial Diseases located in Marseille, France. Morpho-
multidrug resistant A. lwoffii in southern Thailand [22], A. soli
logical identification of ectoparasites and molecular studies were
outbreak as a cause of infection in neonatal intensive health care
performed from the end of 2011 through 2012. All of the
unit in Korea [23] and nosocomial bloodstream infections due to
ectoparasites were identified to the species level using a microscope
A. pittii in United States were reported.
and the morphological identification keys described [1,3].
In animals infection due to Acinetobacter species is considered as
Photographs of the dorsal and ventral body parts of each
an emerging problem due to escalating in the number of reports
ectoparasite were captured (Fig. 1), and the number and sex of
from many countries of the world. Nosocomial infection by A.
each louse and fly was determined.
baumannii in dogs and cats in intensive care unit in Switzerland
[24], infection due to A. baumannii in pets and horses in Switzerland
[25], multidrug resistant Acinetobacter spp. in veterinary clinics in Molecular identification of ectoparasites
Germany [12], detection of A. baumannii in samples from cattle and Prior to DNA extraction, each specimen was rinsed twice in
pigs slaughtered for human consumption in major Scottish sterile water for 15 minutes and then dried on sterile filter paper.
abattoirs [26], carbapenemase producing Acinetobacter spp. in cattle Each specimen was longitudinally cut into two equal halves.
from France [27] and OXA-23 producing Acinetobacter spp. from Genomic DNA was extracted from each specimen using the
faeces of horses in Belgium [28] are some of the current reports QIAamp DNA tissue extraction kit (Qiagen, Hilden, Germany) as
that notify the growing importance of Acinetobacter spp. in per the instructions of the manufacturer. DNA from each
veterinary medicine mirroring the situation happening in human ectoparasite was eluted in 200 ml of TE buffer and stored at
medicine. 220uC under sterile conditions to preclude any contamination
In arthropods Acinetobacter spp. have been detected in different until the sample was used for PCR. The second half of each louse
species including tsetse flies, sand flies, mosquitoes, fleas and ticks and fly was kept at 280uC as a backup sample.
in many countries of the world. Details on Acinetobacter spp. For molecular species identification, DNA samples of 3 to 8
detected in arthropods, species of vectors and methods of detection randomly selected individual lice per species were subjected to
is presented (Table 1). Also, recently several investigators detected standard PCR in an automated DNA thermal cycler to amplify a
A. baumannii in both body and head lice of humans from some fragment of the 18S rRNA gene as described [33]. The PCR was
countries [15,29,30,31]. Despite the worldwide distribution and carried out in a Peltier PTC-200 model thermal cycler (MJ
great economic significance of these ectoparasites, information is Research Inc, Watertown, Mass.). The amplified products were
not available on the occurrence of Acinetobacter species in the lice detected by electrophoresis on 2% agarose gels in TBE 0.56
and flies found on domestic animals. In addition, there is little buffer, stained with ethidium bromide and visualized using
known about Rickettsia species in the lice and flies of domestic ultraviolet (UV) transillumination. A DNA molecular weight
animals in Ethiopia. Therefore, the current study investigated the marker (Boehringer-Mannheim VI, Germany) was used to
presence of these bacteria in arthropods of domestic animals in six estimate the size of the products. The positive controls consisted
districts in Oromia Regional State, Ethiopia. of one sample from Pediculus humanus capitis collected in Mali and
one sample of P. humanus humanus collected in Algeria. These
Materials and Methods samples were included in the PCR assay, and sterile water was
used as negative control.
Study areas and animals The PCR products were cleaned of excess primers and
Lice and Melophagus ovinus were collected from indigenous cattle, nucleotides using a QIAquick Spin PCR Purification Kit (Qiagen)
sheep and dogs in six different districts in Oromia Regional State, as per instructions of the manufacturer. Purified DNA was
Ethiopia: Asalla, Walmara, Shano, Ada’a, Bedele and Gachi. The sequenced using the BigDye Terminator Cycle Sequencing Ready
districts are located in 5 zones in the central, southeast and Reaction Kit (ABI PRISM, PE Applied Biosystems, Foster City,
southwestern parts of the country, with various climates and CA). All obtained sequences were assembled and edited using
agroecology (Table 2). Ectoparasite collections were performed Chromas Pro1.34 (Technelyium Pty. Ltd., Tewantin). The
from September through November of 2011. A long rainy season sequences of the 18S rRNA genes were then subjected to BLAST
from July to September, a short rainy season from March to May analysis to determine similarities to those available in GenBank
and a dry season from November to April prevail in all the study and to construct a phylogenetic tree using Mega 5 software
Table 1. Summary of Acinetobacter spp. detected in various species of arthropods from different countries of the world.
doi:10.1371/journal.pone.0052377.t001
(Molecular Evolution Genetic Analysis; The Biodesign Institute, asite DNA was tested for spotted fever group Rickettsia with primers
Tempe, AZ). targeting the gltA gene specific for this group [34] and for typhus
group Rickettsia with primers targeting the Rpr274P gene, which
Molecular detection of Acinetobacter and Rickettsia encodes a hypothetical protein [35]. Sterile water was used as
species negative control; A. baumannii, R. montanensis and R. typhi DNA were
All the specimens were individually tested for the presence of used as positive controls. The samples were considered positive
Acinetobacter species DNA targeting the rpoB gene [30] by real-time when cycle thresholds (Ct) were ,35.
quantitative (q) PCR as per the instructions of the manufacturer
(Applied Biosystems, Foster City, CA). In addition, the ectopar-
Table 2. Description of study districts and number of study animals in Oromia Regional State.
Annual
Km from rainfall
Addis Altitude in Av. Annual No. Animals
District Zone Ababa Agroecology in m.a.s.l Coordinates mm Temp in 6C examined
Asalla Arsi 175 southeast Highland 2500 7u569589.5799N 2000–4000 20–30 Sheep = 39
39u8923.4299 E
Shano North Showa 70 Northeast Highland 2861 9u20.00.0099N 945.4 6.1–18.5 Cattle = 38
39u18900.0099E Sheep = 19
Walmara West Showa 45 west Highland 2500 9u7950.6499N 1060 4.6–23.3 Cattle = 42
38u28938.4699E
Ada’a East Showa 47 east Midland 1911 8u44937.6999N 1911 13–26.5 Cattle = 53
38u59919.2899E Dogs = 47
Cats = 16
Bedele Illubabora 483 southwest Midland 1974 8u,2791.7699N 1400 12.5–27.5 Cattle = 32
36u2195.0899E Sheep = 21
Gachi Illubabora 460 southwest Midland 1751 9u1492.5799N 1300 13–28 Cattle = 42
35u 4948.4699E Sheep = 6
Figure 1. Photographs of morphologically identified 5 species of lice and Melophagaus ovinus collected from domestic animals. A,
Bovicola ovis of sheep; B, Bovicola bovis of cattle; C, Heterodoxus spiniger of dog; D, Linognathus vituli of cattle; E, Solenopotes capillatus of cattle; F,
Melophagus ovinus (sheep ked) of sheep.
doi:10.1371/journal.pone.0052377.g001
Molecular detection of carbapenemase encoding genes Ectoparasite collections are not harmful and are not against the
All the DNA of lice (n = 82) and flies (n = 19) positive for welfare of animals. No collection had been done from privately-
Acinetobacter species were tested for the presence of carbapenemase owned, wildlife, national park or other protected areas and
encoding genes by qPCR targeting blaOXA-23, blaOXA-24, endangered or protected species.
blaOXA-58 and blaNDM-1 using primers, probes and all
conditions as has been described before [15,36]. In addition, a Data analysis
total of 32 lice and 10 flies with sufficient amount of DNA were Microsoft Excel was used for data management. Descriptive
evaluated for carbapenemase encoding genes by standard PCR statistics such as percentages and means were employed to
targeting blaOXA-51 and blaOXA-23 with primers and all summarize the proportions of infestations with lice and keds.
conditions as described before [15]. Statistical analysis was performed with EpiInfoTM7 and a P-value
of ,0.05 was considered significant.
Molecular identification of Acinetobacter spp. by partial
rpoB gene Results
A total of 32 lice and 10 keds DNA of sufficient amount and
positive for Acinetobacter spp. by qPCR were further subjected to
Morphological identification of ectoparasites
standard PCR targeting partial rpoB gene (zone 1) to identify A total of 207 cattle, 85 sheep, 47 dogs and 16 cats in six
Acinetobacter spp. using the primers and all conditions as described districts were examined for the presence of lice and sheep keds
before [17]. Sterile water and A. genomosp DNA were used as (Table 2). The results of the morphological identification of the
negative and positive controls, respectively. Detection of amplified collected lice and flies from infested animals revealed a total of 408
products, cleaning of excess primers and nucleotides from DNA, Linognathus vituli, 4 Bovicola bovis and 3 Solenopotes capillatus from
sequencing, assembling and edition of sequences, BLAST analysis cattle; 22 Heterodoxus spiniger from dogs; 298 Bovicola ovis and 22
and rpoB gene phylogenetic tree construction with Maximum Melophagus ovinus from sheep (Fig. 1). Furthermore, the study
likelihood statistics of Mega 5 were all performed using similar showed that out of all of the examined cattle, 19.3% (40/207) were
methods as described for 18S rRNA gene for lice above. infested with L. vituli, 0.5% (1/207) with S. capillatus and 0.5% (1/
207) with B. bovis. In cattle, there was a significantly higher
prevalence of L. vituli than both S. capillatus and B. bovis
Ethical statement (p#0.0001). Of the total examined sheep, 48.2% (41/85) were
Ethical approval for the collection of lice and flies from domestic infested with B. ovis and 21.05% (4/19) were positive for M. ovinus.
animals was obtained from the animal research ethics board In addition, 19.1% (9/47) of the examined dogs were infested with
(Agreement # 14/160/550/2011) of the College of Veterinary H. spiniger. Alternatively, lice were not detected on any of the cats
Medicine and Agriculture of Addis Ababa University. All studied.
necessary oral permits were obtained for the described field
studies, including permission of administration and agricultural
office of each Ethiopian district and from each animal owner.
Molecular identification of ectoparasites higher prevalence of Acinetobacter spp. in flies than in lice (82/735
Molecular identification of the lice based on the 18S rRNA gene vs. 19/22; p#0.00001). The study showed that a higher
analysis was used to confirm the species of lice identified by percentage of Acinetobacter spp. DNA was detected in H. spiniger
morphological methods. A BLAST analysis of 18S rRNA gene of dogs than in B. ovis and L. vituli (15/22 vs. 19/298; 47/408;
sequences of lice from dogs morphologically identified as H. spiniger p#0.00001). The prevalence of Acinetobacter spp. in L. vituli
(n = 5) showed 100% (509/509) similarity to the GenBank collected from cattle was significantly higher than in B. ovis from
reference of H. spiniger collected from Japan (GU569166) (Fig. 2). sheep (47/408 vs 19/298; p = 0.02). Acinetobacter spp. was not
Likewise, a BLAST analysis of 18S rRNA gene sequences of lice detected in any of the 4 Bovicola bovis collected from cattle (Table 3).
that were collected from cattle and were morphologically Results of our study revealed that out of the total lice infested
identified as L. vituli (n = 7) showed 99.3% (553/557) similarity animals, 88.9% (8/9) of dogs were infested with H. spiniger, 45%
to Linognathus vituli collected in Australia (GenBank Access. No. (18/40) of cattle were infested with L. vituli and 34.4% (14/41) of
AY077774). Four mutations were detected between our sequence sheep were infested with B. ovis and harbored at least 1 louse
and the reference (AY077774) at 283 bp (T-G), 327 bp (T-A), positive for Acinetobacter spp. (Table 4). Similarly, Acinetobacter spp.
352 bp (T-C), and 420 bp (T-C). This sequence was submitted to was detected at least in one M. ovinus in all the 4/4 (100%) infested
GenBank under accession number JX401573. sheep. Highest proportion of Acinetobacter spp. was observed in
Interestingly, a sequence analysis of the 18S rRNA gene of 3 lice cattle infested with L. vituli in Walmara and Gachi districts whereas
from cattle morphologically identified as Solenopotes capillatus highest percentage of Acinetobacter spp. was noted in sheep infested
showed 92.2% (536/581) homology with the GenBank sequence with B. ovis in Shano and Asalla districts (Table 4).
of Neohaematopinus sciuri (AF423798) and 92% (520/565) homology A molecular investigation of the 735 lice collected from
with Hoplopleura hirsute collected in Japan from the cotton rat domestic animals and 22 Melophagus ovinus collected from sheep
Sigmodon hispidus (GU569181). using qPCR did not produced any positive results for either
A BLAST analysis of the 18S rRNA gene of 4 Bovicola bovis lice spotted fever or typhus group Rickettsia species.
from cattle showed 99.8% (494/495) similarity to B. ovis (GenBank
Access No. GU569184) and 99.4% (510/513) similarity to Molecular identification of Acinetobacter spp
Geomydoecus craigi (GenBank Access No. AF385040) (Fig. 2). There We succeeded in the amplification of 350 bp fragment of partial
are no 18S rRNA gene sequences available in GenBank for S. rpoB gene from a total of 10 samples including one from L. vituli of
capillatus and Bovicola bovis. Thus, these two sequences were cattle, 3 from M. ovinus of sheep and 6 from H. spiniger of dogs
submitted to GenBank under accession numbers JX184910 and (Table 5). BLAST analysis of partial rpoB gene sequence and rpoB
JX184911, respectively. An analysis of the 18S rRNA gene of phylogenetic tree showed the presence of A. soli in L. vituli of cattle,
Bovicola ovis (n = 8) from sheep revealed 100% (495/495) similarity A. lwoffii in M. ovinus of sheep, 1 new Acinetobacter sp. in M. ovinus of
with Bovicola ovis detected in Japan and Australia with the sheep, A. pittii in H. spiniger of dogs and 2 new Acinetobacter sp. in H.
sequences GU569184 and AY077769, respectively. spiniger of dogs (Table 5 and Fig. 3). Five of these 10 sequences
An analysis of the 18S rRNA gene from Pediculus humanus were submitted to GenBank under accession numbers KC130085-
humanus (n = 1) from Mali revealed 99.4% (496/499) similarity 89, respectively.
with Pediculus humanus corporis detected in France with Accessions
Nos. AF139479 and AF139478, respectively. An analysis of the Detection of carbapenemase encoding genes in
18S rRNA gene from Pediculus humanus capitis (n = 1) from Algeria
Acinetobacter species
revealed 99.8% (518/520) similarity with the Pediculus humanus
A molecular investigation of 82 lice and 19 keds positive for
capitis detected in Thailand, Portugal and China with Accessions
Acinetobacter spp. by qPCR did not produced any positive results for
Nos. AY236418, AY236414 and AY236417, respectively.
blaOXA-23, blaOXA-24, blaOXA-58 and blaNDM-1 genes
A phylogenetic tree was constructed from the 18S rRNA gene
encoding for carbapenemase resistance. Likewise, investigation
sequences of lice collected in this study and the most similar
of 32 lice and 10 keds DNA by standard PCR never produced any
sequences in the GenBank using the Neighbor-joining statistics of
positive results for both blaOXA-23 and blaOXA-51 genes
Mega 5. H. spiniger reference lice and our lice sample from dogs
encoding for carbapenemase resistance.
that belong to the family Boophidae in the suborder Amblycera
formed one group on the phylogenetic tree (Fig. 2). Reference
Linognathus vituli and our lice sample from cattle that belong to the Discussion
family Linognathidae under the suborder Anoplura clustered In this study, we detected, for the first time, Acinetobacter spp. in
together. The 3 individual S. capillatus lice from cattle, which the lice and flies of animals from Ethiopia. In addition, we
shared the same family and suborder with L. vituli, clustered near obtained two new sequences of 18S rRNA genes from Bovicola bovis
the cattle louse. Both the control and the reference P. humanus and Solenopotes capillatus collected from cattle. We believe that the
capitis and P. humanus humanus isolated from human head and body results of this study are valid, as all of the negative control samples
lice, belonging to the family Pediculidae under the suborder produced negative results and all the positive control samples
Anoplura, formed a separate group on the phylogenetic tree that tested positive for both Acinetobacter spp. and Rickettsia spp.
was close to L. vituli from cattle. The Bovicola ovis reference lice and detection in the lice and flies of these animals. These findings
our lice sample from sheep that belong to the family Trichodec- confirmed that our molecular study conditions precluded any
tidae in the suborder Ischnocera clustered in a separate group accidental cross-contamination of samples during the study period.
close to the four B. bovis cattle lice, which belong to the same Similarly, in our 18S rRNA gene study of lice, we obtained
genus, family and suborder (Fig. 2). sequences from both positive controls, Pediculus humanus capitis and
Pediculus humanus humanus, confirming the appropriateness of our
Detection of Acinetobacter and Rickettsia species in lice working conditions.
and M. ovinus In the study, the species of lice morphologically identified as L.
We detected Acinetobacter spp. in a total of 82 lice (11.1%) and 19 vituli, B. bovis and S. capillatus from cattle; H. spiniger from dogs; and
flies (86.4%) (Table 3). In our study, there was a significantly B.ovis from sheep (Fig. 1) were confirmed molecularly using 18S
Figure 2. Phylogenetic tree based on 18S gene sequences of lice species collected from domestic animals. Accession numbers in red
color are 18S gene sequence of lice of animals from Ethiopia recently deposited in the GenBank. Minimum evolution method was used to build the
phylogentic tree. Bootstrap values are indicated at the nodes.
doi:10.1371/journal.pone.0052377.g002
rRNA gene sequences. The 18S rRNA gene has been used such as a rounded head, reddish-brown color and dark transverse
previously as an important tool to investigate human lice bands on the abdomen [1] (Fig. 1). S. capillatus from cattle are the
phylogeny [33], to study the evolution of sucking lice [37] and smallest sucking lice, with morphological characteristics that
to study the phylogeny of lice [38]. The phylogenetic tree include a hexagonal sterna plate on the thorax, a weak front pair
constructed from the 18S rRNA gene sequences of the collected of legs and prominent abdominal tubercles [3]. In Ethiopia, both
lice and reference lice in the GenBank presented here supports the species are typically reported with low prevalence [4]. The finding
information from previous studies [37,38]. in this study that there was a significantly (p#0.0001) higher
In this study, we sequenced the 18S rRNA genes of B. bovis and prevalence of L. vituli than both S. capillatus and B. bovis in cattle is
S. capillatus from cattle for the first time. B. bovis is the only biting in agreement with previous work [4]. Our findings in sheep, dogs
lice species in cattle, and it has important morphological features and cats of this study are also in line with previous reports [5,6,8].
Table 3. Percentage of Acinetobacter spp. detected by qPCR in lice and flies collected from domestic animals in six districts in
Oromia.
District (Number of positive samples for Acinetobacter spp./Number of tested samples) (%)
Lice spp. Asalla (%) Shano (%) Walmara (%) Ada’a (%) Bedele (%) Gachi (%) Total (%)
doi:10.1371/journal.pone.0052377.t003
The observation that 11.1% of the lice in the current study were been detected with a prevalence of 5.1% in the feces of domestic
found to contain Acinetobacter spp. is lower than the earlier report of animals in Senegal [15].
47% in human head lice and 71% human body lice from Ethiopia Most earlier investigators suggested that it is still yet not exactly
[31], 21% in human body lice [29] and 33% in human head lice determined how both body and head human lice acquired A.
from Paris [30]. Recently, a low prevalence (4%) of A. baumannii in baumannii infection [30,31]. However, some authors argued
head lice was reported from Senegal [15]. However, the overall undiagnosed transient A. baumannii bacteremia in infested patients
percentage of 86.4% (19/22) of Acinetobacter spp. detected in M. as a source of infection for body lice but they stated it is not
ovinus of sheep in our study is greater than that found in the possible to rule out the possibility of acquiring A. baumannii
aforementioned reports. The variation among study districts in the infection in human body lice from external environmental
percentage of Acinetobacter spp. in lice and keds in infested animals contamination [29]. Other investigators pointed out that bacterial
is most probably attributed to differences in agroecology, animal spp. such as Acinetobacter abundant in the environment reside in the
management and factors like age, sex, physiological status or gut of several species of arthropods as transient or natural flora
presence of other concurrent diseases that may favor infection of [39] acquired commonly by vertical transmission [40] and also are
lice or the animal by the bacteria. This observation coincides with maintained and spread by several mechanisms of horizontal
the findings of differences in the prevalence of infection in lice of transmission including mating, cofeeding, or contact with
humans by A. baumannii among different study areas with different contaminated faeces [41,42,44]. Having all these facts in mind
altitudes in south western Ethiopia [31]. and due to the ubiquitous nature of Acinetobacter spp. in the
In line with our findings, Acinetobacter spp. were detected in many environment including on the skin and in the faces of animals [15]
species of arthropods in different parts of the world (Table 1). For lice and keds possibly had acquired Acinetobacter spp. infection from
instance, Acinetobacter spp. have been detected in 13% of Lutzomyia the skin, faces or transient Acinetobacter spp. bacteremia of their host
longipalpis (sand flies) in Brazil [39], 7.7% of Glossina palpalis palpalis animals. Moreover, it is not possible to rule out infection of lice
(tsetse fly) in Angola [40], 8.7% in the gut of the Prionoplus reticularis and keds by vertical route and also probably animals may play a
larvae (wood feeding beetle) in New Zealand [41], up to 75% in role as reservoirs for these bacteria.
Bacterria cockerelli (potato psyllid) in the USA [42], 18.9% in We did not detect Rickettsia DNA in the lice and flies in our
chewing lice of pocket gophers in the USA [43] and 1% in Bemisia study. This finding contrasts with previous work that reported
tabaci (tobacco whitefly) in India [44]. Moreover, A. baumannii have detecting R. helvetica in M. ovinus from sheep, in Linognathus stenopsis
Table 4. Proportion of infested animals positive for Acinetobacter spp. in lice and keds by qPCR in six districts in Oromia.
District No. animals positive for Acinetobacter spp./No. animals infested with lice or fly
B. ovis (Sheep) L. vituli (Cattle) S. capillatus (Cattle) B. ovis (Cattle) H. spiniger (Dog) M. vinus (Sheep)
doi:10.1371/journal.pone.0052377.t004
Figure 3. Phylogenetic tree based on partial rpoB gene sequences of Acinetobacter species. Maximum Likelihood method was used to
build the phylogentic tree. Bootstrap values are indicated at the nodes. Bold indicates the taxonomic position of Acinetobacter species identified in
this study.
doi:10.1371/journal.pone.0052377.g003
from goats, and Rickettsia spp. in Haematopinus eurysternus from cattle study is in line with the previous finding of absence of resistance to
in Hungary [45,46]. several antimicrobials in A. soli in intensive health care units of
The absence of any positive results for blaOXA-23, blaOXA- neonates in Brazil [23], full susceptibility to several antibiotics of A.
24, blaOXA-58, blaNDM-1 and blaOXA-51 genes encoding for lwoffii from acute gastroenteritis in USA [21] and absence of
carbapenemase resistance by both qPCR and standard PCR in blaOXA-like genes encoding for carbapenemase resistance in A.
Acinetobacter spp. in the lice and keds of domestic animals in our baumannii from faeces of domestic animals in Senegal [15]. On the
Table 5. Summary of BLAST analysis of partial rpoB gene sequences obtained from lice and keds of domestic animals in six
districts in Oromia, Ethiopia.
Lice/fly spp Host spp Length (bp) Nearest match in GenBank % similarity
doi:10.1371/journal.pone.0052377.t005
other hand our findings contrast the previous multidrug resistance in the world are needed as has been done during the last decade
in A. baumannii reported from several countries of the world for Bartonella species, ectoparasites and their animal hosts [52].
[12,36,47]. This finding also contradicts the observations of Further studies are also required to isolate and determine the
multidrug-resistant isolates of Acinetobacter spp. from a range of human health significance of the new Acinetobacter spp. detected in
environmental sources in South Korea [20] and the presence of the lice and keds from animals.
resistance to antibiotics in A. baumannii and other Acinetobacter spp. To our knowledge, this study is the first to report the presence of
from blood cultures in Norway [48]. This variation is most DNAs from different Acinetobacter spp. in various species of lice
probably attributed to differences in strains of the bacteria among collected from domestic animals and in flies collected from sheep.
various studies and other many factors contributing for emergence Our study demonstrates that Acinetobacter spp. are not only
of resistance. common as hospital pathogens and in human lice but they can
Acinetobacter spp. identification study from DNA of lice and keds also be detected in the ectoparasites of animals. Our findings
of animals in Oromia uncovered the occurrence of 3 previously suggest that synanthropic animals and their ectoparasites might
described species and 3 new Acinetobacter spp. (Table 5 and Fig. 3). play a role to increase the risk of human exposure to zoonotic
All the 3 previously described species, we detected: A. soli from L. pathogens and could be a source for Acinetobacter spp. infections in
vituli of cattle, A. lowffii from keds of sheep and A. pittii from H. humans. However, additional epidemiological data are required to
spiniger of dogs with high nucleotide sequence identities of 98– justify the significance of this finding and to determine whether
100% with their respective reference species in the GenBank ectoparasites from animals can act as environmental reservoirs and
(Table 5). This finding supports the criteria established for play a role in spreading these bacteria to both animal and human
Acinetobacter spp. identification using partial rpoB gene sequence hosts.
analysis [17,18]. In line with our observation higher predominance
of other Acinetoabacter spp. (24.8%) than A. baumannii (only 8.8%)
from human blood culture isolates was recently reported from
Acknowledgments
Norway [48]. Furthermore, these species are nowadays reported to The authors greatly acknowledge all animal owners in Oromia and Yonas
cause various types of human infections worldwide Abiyi, Teferi Banti, Fikadu and Gurara Megerssa Veterinary professionals
[21,23,49,50,51] and are implicated as emerging Acinetobacter spp. of Oromia Regional laboratories for their help during sample collection.
We also identified two new Acinetobacter species from H. spiniger of Also Amina Boutellis and Seydina M. Diene PhD students at the Faculty of
dogs and one from ked of sheep (Table 5 and Fig. 3). All these 3 Medicine of Aix Marseille University, are gratefully acknowledged for their
professional help during laboratory work of the study.
new Acinetobacter spp. demonstrated low nucleotide homology with
reference rpoB sequence in the GenBank and low bootstrap value
in the rpoB phylogenetic tree. Results of our study suggest the Author Contributions
presence of specific Acinetobacter species in lice and keds of domestic Conceived and designed the experiments: DR PP. Performed the
animals unlike A. baumannii in human lice. We believe that experiments: BK CS. Analyzed the data: CS BK DR JMR. Contributed
additional in depth epidemiological studies involving other species reagents/materials/analysis tools: DR. Wrote the paper: BK CS DR PP
of lice and ectoparasites of different animal species from vast areas JMR.
References
1. Wall R, Shearer D (1997) Veterinary Entomology; 1, editor. London: Chapman 4. Kumsa B, Bekele S (2008) Lice Infestation On Cattle In Endegagn District,
and Hall. 438 p. Southern Ethiopia: Species Composition, Prevalence And Seasonal Pattern. Bull
2. Small RW (2005) A review of Melophagus ovinus (L.), the sheep ked. Vet Parasitol Anim Hlth Prod Afr 56: 213–222.
130: 141–155. 5. Kumsa B, Beyecha K, Geloye M (2012) Ectoparasites of sheep in three agro-
3. Taylor MA, Coop RL, Wall R (2007) Veterinary Parasitology: Wiley-Blackwell. ecological zones in central Oromia, Ethiopia Onderstepoort J Vet Res 79(1): 7
847 p. pages.
6. Sertse T, Wossene A (2007) A study on ectoparasites of sheep and goats in
eastern part of Amhara region, Northeast Ethiopia. Small Rumin Res 69: 55–61.
7. Beyecha K, Beyene D, Kumsa B (2012) Ectoparasites of goats in three 35. Walter G, Botelho-Nevers E, Socolovschi C, Raoult D, Parola P (2012) Murine
agroecologies in central Oromia, Ethiopia. Comp Clin Pathol DOI 10.1007/ typhus in returned travelers: a report of thirty-two cases. Am J Trop Med Hyg
s00580-012-1563-x. 86: 1049–1053.
8. Kumsa B, Mekonnen S (2011) Ixodid ticks, fleas and lice infesting dogs and cats 36. Kusradze I, Diene SM, Goderdzishvili M, Rolain JM (2011) Molecular detection
in Hawassa, southern Ethiopia. Onderstepoort J Vet Res 78: 4 pages. of OXA carbapenemase genes in multidrug-resistant Acinetobacter baumannii
9. Yacob HT, ATAKLTY H, Kumsa B (2008) Major ectoparasites of cattle in and isolates from Iraq and Georgia. Int J Antimicrob Agents 38: 164–168.
around Mekelle, northern Ethiopia. Entom Res 28: 26–30. 37. Light JE, Smith VS, Allen JM, Durden LA, Reed DL (2010) Evolutionary
10. Chanie M, Negash T, Sirak A (2010) Ectoparasites are the major causes of history of mammalian sucking lice (Phthiraptera: Anoplura). BMC Evol Biol 10:
various types of skin lesions in small ruminants in Ethiopia. Trop Anim Health 292.
Prod 42: 1103–1109. 38. Barker SC, Whiting M, Johnson KP, Murrell A (2003) Phylogeny of the lice
11. Abebayehu T, Kibrom M (2010) Study on ectoparasitic defects of processed (Insecta, Phthiraptera) inferred from small subunit rRNA. Zoologica Scripta 32:
skins at Sheba Tannery, Tigray, Northern Ethiopia. Trop Anim Health Prod 42: 407–414.
1719–1722. 39. Gouveia C, Asensi MD, Zahner V, Rangel EF, Oliveira SM (2008) Study on the
12. Zordan S, Prenger-Berninghoff E, Weiss R, van der Reijden T, van den Broek P, bacterial midgut microbiota associated to different Brazilian populations of
et al. (2011) Multidrug-resistant Acinetobacter baumannii in veterinary clinics, Lutzomyia longipalpis (Lutz & Neiva) (Diptera: Psychodidae). Neotrop Entomol 37:
Germany. Emerg Infect Dis 17: 1751–1754. 597–601.
13. Peleg AY, Seifert H, Paterson DL (2008) Acinetobacter baumannii: emergence of a 40. Geiger A, Fardeau ML, Grebaut P, Vatunga G, Josenando T, et al. (2009) First
successful pathogen. Clin Microbiol Rev 21: 538–582. isolation of Enterobacter, Enterococcus, and Acinetobacter spp. as inhabitants of the
14. Turton JF, Shah J, Ozongwu C, Pike R (2010) Incidence of Acinetobacter species tsetse fly (Glossina palpalis palpalis) midgut. Infect Genet Evol 9: 1364–1370.
other than A. baumannii among clinical isolates of Acinetobacter: evidence for 41. Reid NM, Addison SL, Macdonald LJ, Lloyd-Jones G (2011) Biodiversity of
emerging species. J Clin Microbiol 48: 1445–1449. active and inactive bacteria in the gut flora of wood-feeding huhu beetle larvae
15. Kempf M, Rolain JM, Diatta G, Azza S, Samb B, et al. (2012) Carbapenem (Prionoplus reticularis). Appl Environ Microbiol 77: 7000–7006.
Resistance and Acinetobacter baumannii in Senegal: The Paradigm of a Common 42. Nachappa P, Levy J, Pierson E, Tamborindeguy C (2011) Diversity of
Phenomenon in Natural Reservoirs. PLoS One 7: e39495. endosymbionts in the potato psyllid, Bactericera cockerelli (Triozidae), vector of
16. Nemec A, Krizova L, Maixnerova M, van der Reijden TJ, Deschaght P, et al. zebra chip disease of potato. Curr Microbiol 62: 1510–1520.
(2011) Genotypic and phenotypic characterization of the Acinetobacter calcoaceticus- 43. Reed DL, Hafner MS (2002) Phylogenetic analysis of bacterial communities
Acinetobacter baumannii complex with the proposal of Acinetobacter pittii sp. nov. associated with ectoparasitic chewing lice of pocket gophers: a culture-
(formerly Acinetobacter genomic species 3) and Acinetobacter nosocomialis sp. nov. independent approach. Microb Ecol 44: 78–93.
(formerly Acinetobacter genomic species 13TU). Res Microbiol 162: 393–404. 44. Singh ST, Priya NG, Kumar J, Rana VS, Ellango R, et al. (2012) Diversity and
17. Gundi VA, Dijkshoorn L, Burignat S, Raoult D, La Scola B (2009) Validation of phylogenetic analysis of endosymbiotic bacteria from field caught Bemisia tabaci
partial rpoB gene sequence analysis for the identification of clinically important from different locations of North India based on 16S rDNA library screening.
and emerging Acinetobacter species. Microbiology 155: 2333–2341. Infect Genet Evol 12: 411–419.
18. La Scola B, Gundi VA, Khamis A, Raoult D (2006) Sequencing of the rpoB gene
45. Hornok S, de la Fuente J, Biro N, Fernandez de Mera IG, Meli ML, et al. (2011)
and flanking spacers for molecular identification of Acinetobacter species. J Clin
First molecular evidence of Anaplasma ovis and Rickettsia spp. in keds (Diptera:
Microbiol 44: 827–832.
Hippoboscidae) of sheep and wild ruminants. Vector Borne Zoonotic Dis 11:
19. Narciso-da-Rocha C, Vaz-Moreira I, Svensson-Stadler L, Moore ER, Manaia
1319–1321.
CM (2012) Diversity and antibiotic resistance of Acinetobacter spp. in water from
46. Hornok S, Hofmann-Lehmann R, de Mera IG, Meli ML, Elek V, et al. (2010)
the source to the tap. Appl Microbiol Biotechnol.
Survey on blood-sucking lice (Phthiraptera: Anoplura) of ruminants and pigs
20. Choi JY, Kim Y, Ko EA, Park YK, Jheong WH, et al. (2012) Acinetobacter species
with molecular detection of Anaplasma and Rickettsia spp. Vet Parasitol 174: 355–
isolates from a range of environments: species survey and observations of
358.
antimicrobial resistance. Diagn Microbiol Infect Dis 74: 177–180.
47. Higgins PG, Dammhayn C, Hackel M, Seifert H (2010) Global spread of
21. Regalado NG, Martin G, Antony SJ (2009) Acinetobacter lwoffii: bacteremia
carbapenem-resistant Acinetobacter baumannii. J Antimicrob Chemother 65: 233–
associated with acute gastroenteritis. Travel Med Infect Dis 7: 316–317.
238.
22. Nakwan N, Wannaro J, Nakwan N (2011) Multidrug-resistant Acinetobacter lwoffii
infection in neonatal intensive care units. Res and Reports in Neonatatology. 48. Karah N, Haldorsen B, Hegstad K, Simonsen GS, Sundsfjord A, et al. (2011)
23. Pellegrino FL, Vieira VV, Baio PV, dos Santos RM, dos Santos AL, et al. (2011) Species identification and molecular characterization of Acinetobacter spp. blood
Acinetobacter soli as a cause of bloodstream infection in a neonatal intensive care culture isolates from Norway. J Antimicrob Chemother 66: 738–744.
unit. J Clin Microbiol 49: 2283–2285. 49. Wisplinghoff H, Paulus T, Lugenheim M, Stefanik D, Higgins PG, et al. (2012)
24. Francey T, Gaschen F, Nicolet J, Burnens AP (2000) The role of Acinetobacter Nosocomial bloodstream infections due to Acinetobacter baumannii, Acinetobacter pittii
baumannii as a nosocomial pathogen for dogs and cats in an intensive care unit. and Acinetobacter nosocomialis in the United States. J Infect 64: 282–290.
J Vet Intern Med 14: 177–183. 50. Vaz-Moreira I, Novo A, Hantsis-Zacharov E, Lopes AR, Gomila M, et al. (2011)
25. Endimiani A, Hujer KM, Hujer AM, Bertschy I, Rossano A, et al. (2011) Acinetobacter rudis sp. nov., isolated from raw milk and raw wastewater. Int J Syst
Acinetobacter baumannii isolates from pets and horses in Switzerland: molecular Evol Microbiol 61: 2837–2843.
characterization and clinical data. J Antimicrob Chemother 66: 2248–2254. 51. Debarry J, Garn H, Hanuszkiewicz A, Dickgreber N, Blumer N, et al. (2007)
26. Hamouda A, Findlay J, Al Hassan L, Amyes SG (2011) Epidemiology of Acinetobacter lwoffii and Lactococcus lactis strains isolated from farm cowsheds possess
Acinetobacter baumannii of animal origin. Int J Antimicrob Agents 38: 314–318. strong allergy-protective properties. J Allergy Clin Immunol 119: 1514–1521.
27. Poirel L, Bercot B, Millemann Y, Bonnin RA, Pannaux G, et al. (2012) 52. Saisongkorh W, Rolain JM, Suputtamongkol Y, Raoult D (2009) Emerging
Carbapenemase-producing Acinetobacter spp. in Cattle, France. Emerg Infect Dis Bartonella in humans and animals in Asia and Australia. J Med Assoc Thai 92:
18: 523–525. 707–731.
28. Smet A, Boyen F, Pasmans F, Butaye P, Martens A, et al. (2012) OXA-23- 53. Murrell A, Dobson SJ, Yang X, Lacey E, Barker SC (2003) A survey of bacterial
producing Acinetobacter species from horses: a public health hazard? J Antimicrob diversity in ticks, lice and fleas from Australia. Parasitol Res 89: 326–334.
Chemother. 54. Pidiyar VJ, Jangid K, Patole MS, Shouche YS (2004) Studies on cultured and
29. La Scola B, Raoult D (2004) Acinetobacter baumannii in human body louse. Emerg uncultured microbiota of wild culex quinquefasciatus mosquito midgut based on
Infect Dis 10: 1671–1673. 16s ribosomal RNA gene analysis. Am J Trop Med Hyg 70: 597–603.
30. Bouvresse S, Socolovshi C, Berdjane Z, Durand R, Izri A, et al. (2011) No 55. Dinparast Djadid N, Jazayeri H, Raz A, Favia G, Ricci I, et al. (2011)
evidence of Bartonella quintana but detection of Acinetobacter baumannii in head lice Identification of the midgut microbiota of An. stephensi and An. maculipennis for
from elementary schoolchildren in Paris. Comp Immunol Microbiol Infect Dis their application as a paratransgenic tool against malaria. PLoS One 6: e28484.
34: 475–477. 56. Sant’Anna MR, Darby AC, Brazil RP, Montoya-Lerma J, Dillon VM, et al.
31. Kempf M, Abdissa A, Diatta G, Trape JF, Angelakis E, et al. (2012) Detection of (2012) Investigation of the bacterial communities associated with females of
Acinetobacter baumannii in human head and body lice from Ethiopia and Lutzomyia sand fly species from South America. PLoS One 7: e42531.
identification of new genotypes. Int J Infect Dis. 57. Geiger A, Fardeau ML, Njiokou F, Joseph M, Asonganyi T, et al. (2011)
32. Central Statistics Agency (2008) Ethiopian agricultural sample survey report on Bacterial diversity associated with populations of Glossina spp. from Cameroon
livestock and livestock characteristics, Vol II, 2007/08, Addis Ababa, Ethiopia. and distribution within the Campo sleeping sickness focus. Microb Ecol 62: 632–
33. Yong Z, Fournier PE, Rydkina E, Raoult D (2003) The geographical segregation 643.
of human lice preceded that of Pediculus humanus capitis and Pediculus humanus 58. Zouache K, Raharimalala FN, Raquin V, Tran-Van V, Raveloson LH, et al.
humanus. C R Biol 326: 565–574. (2011) Bacterial diversity of field-caught mosquitoes, Aedes albopictus and Aedes
34. Socolovschi C, Barbarot S, Lefebvre M, Parola P, Raoult D (2010) Rickettsia aegypti, from different geographic regions of Madagascar. FEMS Microbiol Ecol
sibirica mongolitimonae in traveler from Egypt. Emerg Infect Dis 16: 1495–1496. 75: 377–389.
Oromia, Ethiopia.
265
274
Author's personal copy
a r t i c l e i n f o a b s t r a c t
Article history: Melophagus ovinus (sheep ked) is one of the most common ectoparasites that contributes
Received 11 September 2013 to enormous economic losses in the productivity of sheep in many countries. The present
Received in revised form 28 October 2013
study was conducted from January 2012 to July 2013 on M. ovinus collected from sheep at
Accepted 1 November 2013
three sites in Ethiopia. Of the sheep studied, 65.7% (88/134) were infested with M. ovinus.
The prevalence of M. ovinus was 76% (76/100), 47% (8/17) and 23.5% (4/17) at the Kimbibit,
Keywords:
Chacha and Shano sites, respectively. An overall number of 229 M. ovinus specimens (138
Sheep
Bartonella melophagi females, 86 males and five pupae) and 554 M. ovinus specimens (272 females, 282 males)
Melophagus ovinus were collected from young and adult sheep, respectively. Bartonella DNA was detected in
Sheep ked 89% (694/783) of M. ovinus using a quantitative Bartonella genus-specific PCR assay tar-
Molecular detection geting the 16S/23S rRNA intergenic spacer region. The sequencing of the PCR products of
Oromia, Ethiopia fragments of the gltA and rpoB genes showed 99.6–100% and 100% homology, respectively,
with B. melophagi. Statistically significant variation was not noted in the overall prevalence
of Bartonella DNA between female and male M. ovinus. All of the sheep infested with M. ovi-
nus 100% (88/88) harbored at least one M. ovinus specimen that contained Bartonella DNA.
This study highlights that B. melophagi in M. ovinus from sheep in highlands in Ethiopia
possibly has certain zoonotic importance.
© 2013 Elsevier Ltd. All rights reserved.
0147-9571/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.cimid.2013.11.001
Author's personal copy
70 B. Kumsa et al. / Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 69–76
USA [7], Anaplasma ovis in Hungary [8] and Borrelia burgdor- animals and humans is normally very high; and humans
feri sensu lato in China [9] have been detected in M. ovinus. usually consume raw or insufficiently cooked meat, blood
Bartonella spp. are small hemotropic, facultative, Gram- and milk of animals, information on vector-borne bacteria
negative and arthropod-borne intracellular alphapro- is extremely important. Thus, the current study aimed to
teobacteria that belong to the family Bartonellaceae. determine the presence of and to identify Bartonella spp.
Bartonella are aerobic bacilli, oxidative negative, fastidious, in M. ovinus feeding on sheep at three different sites in
slow growing and highly adapted to their vertebrate hosts. Ethiopia.
The bacteria infect the erythrocytes and endothelial cells of
their mammalian hosts. To date, the genus Bartonella com- 2. Materials and methods
prises more than 30 different species and subspecies that
infect a wide variety of mammalian hosts [10,11]. 2.1. Study areas and animals
In humans Bartonella spp. are implicated in causing
several emerging and reemerging diseases manifested by In this study, sheep from three sites in Kimbibit, Shano
a wide variety of clinical signs [12,13]. At least 13 Bar- and Chacha were studied for M. ovinus infestation. Shano
tonella spp. were reported to cause diseases in humans, (9 19� 60.00�� N 39◦ 18� 0.00�� E), Kimbibit (9◦ 24� 39.95�� N
◦
of which B. bacilliformis (Carrion’s disease), B. quintana 39◦ 21� 14.75�� E), and Chacha (9◦ 31� 20.23�� N 39◦ 27� 5.40�� E)
(trench fever) and B. henselae (cat scratch disease) were are located 70, 110 and 130 km, respectively, to the north of
described as the most common species of the genus [13,14]. Addis Ababa. These areas have a highland type of agroecol-
Other species, such as B. elizabethae, B. vinsonii subsp. ogy, with altitudes of 1880, 3000 and 2700 meter above
berkhoffii, B. vinsonii subsp. arupensis, B. koehlerae, B. alsat- sea level (m.a.s.l.), respectively. Kimbibit and Shano are
ica, B. washoensis, B. rochalimae, and B. tamiae, have been located in Oromia regional state, whereas Chacha is located
associated with chronic bacteremia and/or endocarditis, in Amhara regional state. Sheep is the major animal species
bacillary angiomatosis, peliosis hepatitis, retinitis, uveitis kept in Ethiopian highlands with a cool climate [32]. At
and myocarditis [10,15,13,14,16]. all study sites, the main rainy season is from mid-June to
Many species of domestic (cat, dog, cattle and sheep) September, and the dry season extends from November to
and wild (rodents, deer and elk) animals play key role May. The farming system at all sites is characterized by a
as reservoir hosts of Bartonella spp. In the mammalian mixed crop–livestock production system. The livestock in
reservoir hosts, Bartonella spp. usually cause persis- the study areas are traditionally managed under extensive
tent asymptomatic bacteremia [17]. However, recently, production systems [32].
ruminant-associated Bartonella spp., including B. chomeli, B. M. ovinus was collected from sheep in Shano, Kimbibit
bovis, B. schoenbuchensis and B. capreoli, have been reported and Chacha during January and February 2013. All study
to infect cattle in France [18,19], the USA [20], Caledonia sheep were selected irrespective of their sex and age. Sex
[21], China [22] and Poland [23]. These cited investigations (female or male) and age (young or adult) were recorded
in domestic ruminants are part of an escalating number of for each study animal. The animals were categorized into
reports of Bartonella spp. in veterinary medicine, mirroring two age groups, young (up to one year) and adult (older
the situation in human medicine. than one year), according to a previous publication [30,31].
Many different species of hematophagous arthropods,
including lice (Pediculus humunus humunus), fleas (Cteno- 2.2. Collection and identification of M. ovinus
cephalides felis), sand flies (Lutzomyia verrucarum), biting
flies (Haematobia spp.) [24], deer keds (Lipoptena cervi) M. ovinus was carefully removed manually, using for-
and ticks (Ixodes ricinus), are implicated in mediating the ceps or by hand to avoid any damage to the body, and were
transmission of Bartonella spp. between reservoir hosts then placed in small plastic vials containing 70% ethanol
and humans (and other susceptible hosts) [25,26,10,17]. for subsequent identification. All collected M. ovinus spec-
Although there are several reports on the presence of Bar- imens from the same infested animal were placed into
tonella spp. in lice, fleas, biting flies and ticks from domestic pre-labeled, separate small plastic vials and transported to
animals from many countries around the world, there is the Laboratory of the World Health Organization Collabo-
paucity of information on the occurrence of Bartonella spp. rative Center for Rickettsial Diseases and Arthropod-borne
of medical and veterinary importance in M. ovinus world- Bacterial Diseases in Marseille, France. The sex and stage of
wide. The only available reports on Bartonella spp. from M. ovinus was determined using a microscope according
Ethiopia are 6% (2/34) B. henselae in C. felis collected from to standard morphological identification keys, as previ-
cats [27], an 11% (5/46) prevalence of antibodies against ously described [30]. Photographs of the dorsal and ventral
Bartonella spp. in cats from Addis Ababa [28] and Bartonella body parts of each M. ovinus specimen were captured
spp. related to B. elizabethae in small mammals in northern (Figs. 1 and 2).
Ethiopia [29].
In Ethiopia, all previous studies on M. ovinus focused 2.3. DNA extraction from M. ovinus
on the organism’s prevalence and impact on the commer-
cial value of the skin of sheep [30,31]. However, studies Before DNA extraction, each M. ovinus specimen was
on the role of M. ovinus as vectors of pathogens of veteri- rinsed twice in sterile water for 15 min and then dried
nary and medical importance are remarkably scarce in the on sterile filter paper. Each specimen was longitudinally
country. In a country such as Ethiopia, where ectoparasites cut into two equal halves. One-half of each specimen was
of domestic animals are very common; contact between kept at −80 ◦ C as reserve sample to avoid risks of losing
Author's personal copy
B. Kumsa et al. / Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 69–76 71
Fig. 1. Melophagus ovinus female and male collected from sheep, Oromia, Ethiopia.
samples for any reason during or after DNA extraction. Rh. sanguineus kept at the WHO Collaborative Center for
Genomic DNA was individually extracted from a total of Rickettsial Diseases (Marseille, France) was included in
783 M. ovinus one-half specimens using the QIAamp DNA each batch of extraction to rule out cross contamination
tissue extraction kit (Qiagen, Hilden, Germany) according among samples. The second half of each M. ovinus specimen
to the manufacturer’s instructions. The DNA from each M. was stored at −80 ◦ C as a backup sample.
ovinus specimen was eluted in 100 �l of TE buffer and
stored at −20 ◦ C under sterile conditions to preclude con- 2.4. Molecular detection of Bartonella spp. using ITS gene
tamination until the sample was used for PCR [30]. To in M. ovinus
avoid cross-contamination among samples during DNA
extraction, all parts of the DNA extracting EZI Advanced As a first step, all DNA samples were individually
XL Robot were disinfected after each batch of extraction tested for the presence of the 16S/23S rRNA intergenic
as per recommendations of the manufacturers. Further- spacer-region ITS gene specific for the genus Bartonella by
more, one non-infected tick from laboratory colonies of real-time quantitative PCR (qPCR) with a 21-bp probe, as
previously described [33,34], following the instructions of
the manufacturer (Applied Biosystems, Foster City, CA).
Sterile water and DNA extracted from non-infected Rh.
sanguineus were used as negative controls, and B. eliza-
bethae DNA was used as a positive control. The analytical
sensitivity of qPCR assay targeting ITS gene that could
detect the number of copies of B. melophagi per ked was
determined in triplicate reactions with ten-fold serial dilu-
tions of a known titration of Bartonella quintana suspension.
The samples were considered positive when the cycle
threshold (Ct) was <35, due to the data obtained after the
analysis of the analytical sensitivity of qPCR targeting ITS
gene. To verify the quality of our DNA extraction randomly
selected samples negative 50.6% (40/79) for Bartonella spp.
were further subject to standard PCR to amplify 18S rRNA
gene of M. ovnus as described before [30].
72 B. Kumsa et al. / Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 69–76
based on qPCR analysis of the ITS gene were further sub- overall prevalence of M. ovinus infestations between sheep
jected to a standard PCR targeting the citrate synthase (gltA) of different sex (MH, p = 0.75) and age (MH, p = 0.96) groups.
gene and the rpoB gene using primers and all conditions In the current study, at all three sites, there was
described before [35,21,36]. Sterile water and B. elizabethae no statistically significant difference between the overall
DNA were used as negative and positive controls, respec- numbers of female and male M. ovinus specimens collected
tively. The detection of amplified products, the removal of from female compared with male sheep (MH, p = 0.67),
excess primers and nucleotides from the DNA, sequencing, young compared with adult sheep (MH, p = 0.1) and in the
the assembly and edition of sequences and BLAST analysis overall collection (MH, p = 0.19). The female-to-male ratio
were all performed using similar methods described before of M. ovinus was 1:02, 1:9, 1:4 and 1:1 for the Kimbibit,
[30]. Chacha and Shano sites and the overall collection, respec-
tively.
2.6. Data analysis
3.2. Molecular detection of Bartonella spp. using ITS gene
Microsoft Excel was used for data management. in M. ovinus
Descriptive statistics, such as percentages and means, were
An overall prevalence of 88.6% (694/783) for Bartonella
employed to summarize the proportions of infestations
spp. DNA was recorded in M. ovinus collected from sheep
with M. ovinus and infections of M. ovinus with Bartonella
in northern Oromia using molecular tools (Table 2). The
spp. Statistical analysis was performed with EpiInfoTM 7.
prevalence of Bartonella spp. DNA was 88.4% (586/663),
Associations between the sex and age groups of the sheep,
94.9% (93/98) and 68.2% (15/22) at the Kimbibit, Chacha
the sites of collection and the sex of M. ovinus and the preva-
and Shano sites, respectively (Fig. 3) (Table 2). A statisti-
lence of M. ovinus infestation in sheep and Bartonella spp. in
cally significant difference in the prevalence of Bartonella
M. ovinus were determined by the Mantel–Haenszel (MH)
spp. at the three sites (MH, p ≥ 0.05) was not observed. Sim-
test using EpiInfoTM 7. A p-value of <0.05 was considered
ilarly, statistically significant variation was not noted in the
significant.
overall prevalence of Bartonella spp. between female and
male M. ovinus collected from sheep at the three sites or
2.7. Ethical statement
between different sex and age groups (Table 2).
The overall prevalence of Bartonella spp. was 74.2%
Ethical approval for the collection of M. ovinus from
(187/252) and 95.5% (507/531) among M. ovinus speci-
sheep was obtained from the animal research ethics board
mens collected from female and male sheep, respectively.
(Agreement # 14/160/550/2011) of the College of Veteri-
The corresponding figures among M. ovinus specimens col-
nary Medicine and Agriculture of Addis Ababa University.
lected from young and adult sheep were 96.5% (222/230)
All necessary oral permits were obtained for the field
and 85.3% (472/553), respectively (Table 2). Statistically
studies, including permission from the administration and
significant differences were not found in the overall preva-
agricultural office of each site and from each animal owner.
lence of Bartonella spp. between M. ovinus specimens
M. ovinus collection from sheep was not harmful and not
collected from female compared with male sheep and
contrary to the welfare of the animals.
young compared with adult sheep (MH, p ≥ 0.05).
An analysis of M. ovinus specimens that were positive
3. Results
for Bartonella spp. DNA on the infested sheep revealed that
100% (88/88) of the sheep infested with M. ovinus harbored
3.1. Prevalence of M. ovinus in sheep
at least one M. ovinus specimen containing Bartonella spp.
DNA. The mean qPCR Ct values and standard deviations
In the present study, 88/134 (65.7%) sheep were infested
(SDs) for Bartonella spp. DNA-positive samples using the
with M. ovinus (Table 1). An overall number of 783 M. ovinus
ITS gene were 22.2 ± 4.04 (ITS gene: mean copies/half of ked
specimens (410 females, 368 males and five pupae) were
1.7 × 105 [minimum 9 × 103 , maximum 3 × 106 ]), 23 ± 4.1
collected from infested sheep. The prevalence of M. ovi-
(ITS gene: mean copies/half of ked 9.6 × 104 [minimum
nus was 76% (76/100), 47% (8/17) and 23.5% (4/17) at the
5 × 103 , maximum 1.8 × 106 ]), 30 ± 4.2 (ITS gene: mean
Kimbibit, Chacha and Shano sites, respectively (Table 1).
copies/half of ked 0.6 × 103 [minimum 0.3 × 102 , maximum
The prevalence of M. ovinus was significantly higher in
1.3 × 104 ]), and 22.6 ± 4.1 (ITS gene: mean copies/half of
Kimbibit than at both the Chacha (MH test, p = 0.01) and
ked 1.3 × 105 [minimum 6.7 × 103 , maximum 2.4 × 106 ])
the Shano (MH test, p = 0.00001) sites. In female sheep, an
for females, males, pupae and overall samples, respectively.
overall prevalence of 63.1% (36/57) and 252 M. ovinus spec-
Visualization of strong DNA bands of amplified PCR
imens (135 f, 116 m and one pupa) were recorded. The
products using electrophoresis in all the 40 DNA samples
corresponding figures for male sheep were 65.8% (52/79)
tested for 18S rRNA gene suggested that our DNA extraction
prevalence and 531 M. ovinus specimens (275 f, 252 m and
conditions were appropriate and free of PCR inhibitors.
four pupae) (Table 1). An overall prevalence of 65% (26/40)
and 64.6% (62/96) was observed in young and adult sheep, 3.3. Molecular identification of Bartonella spp. in M.
respectively (Table 1). An overall number of 229 M. ovi- ovinus using gltA and rpoB genes
nus specimens (138 f, 86 m and five pupae) and 554 M.
ovinus specimens (272 f, 282 m and zero pupae) were col- To identify Bartonella spp., we amplified a 200- to
lected from young and adult sheep, respectively (Table 1). 250-bp fragment of the gltA gene and an 800- to 900-
A statistically significant difference was not noted in the bp fragment of the rpoB gene from each of the 50 DNA
Author's personal copy
B. Kumsa et al. / Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 69–76 73
Table 1
Overall percentage of M. ovinus infestations by sex and age of sheep at three sites in northern Ethiopia.
Kimbibit 31/41 (75.6%) 45/59 (76.3%) 21/27 (77.8%) 54/72 (75%) 76/100 (76%)
(225: 120 f, 104 m, 1 p) (438: 214 f, 222 m, 2 p) (152: 85 f, 64 m, 3 p) (511: 249 f, 262 m, 0 p) (663: 334 f, 326 m, 3 p)
Chacha 3/6 (50%) 5/11 (45.4%) 3/5 (60%) 5/12 (41.7%) 8/17 (47%)
(10: 5 f, 5 m, 0 p) (88: 58 f, 28 m, 2 p) (76: 52 f, 22 m, 2 p) (22: 11 f, 11 m, 0 p) (98: 63 f, 33 m, 2 p)
Shano 2/10 (20%) 2/9 (22.2%) 1/7 (14.3%) 3/12 (25%) 4/17 (23.5%)
(17: 10 f, 7 m) (5: 3f, 2 m) (1: 1 f, 0 m) (21: 12 f, 9 m) (22: 13 f, 9 m)
Total 36/57 (63.1%) 52/79 (65.8%) 26/40 (65%) 62/96 (64.6%) 88/134 (65.7%)
(252: 135 f, 116 m, 1 p) (531: 275 f, 252 m, 4 p) (229: 138 f, 86 m, 5 p) (554: 272 f, 282 m, 0 p) (783: 410 f, 368 m, 5 p)
Table 2
Prevalence of Bartonella spp. using ITS gene by qPCR analysis of M. ovinus collected from sheep in northern Oromia.
Site M. ovinus sex Female host Male host Young host Adult host
No. positive M. No. positive M. No. positive M. No. positive M.
ovinus/No. tested M. ovinus/No. tested M. ovinus/No. tested M. ovinus/No. tested M.
ovinus (%) ovinus (%) ovinus (%) ovinus (%)
Kimbibit F 92/120 (76.7%) 203/214 (94.8%) 84/85 (98.8%) 211/249 (84.7%) 295/334 (88.3%)
M 76/104 (73.1%) 213/222 (95.9%) 32/35 (91.4%) 228/262 (87%) 289/326 (88.6%)
Pupa 0/1 (0) 2/2 (100%) 2/3 (66.7%) 0/0 (0) 2/3 (66.7%)
Total 168/225 (74.7%) 418/438 (95.4%) 147/152 (96.7%) 439/511 (85.9%) 586/663 (88.4%)
Chacha F 4/5 (80%) 57/58 (98.3%) 51/52 (98.1%) 10/11 (90.9%) 61/63 (96.8%)
M 4/5 (80%) 27/28 (96.4%) 21/22 (95.4%) 10/11 (90%) 31/33 (93.9%)
Pupa 0/0 (0) ½ (50%) ½ (50%) 0/0 (0) ½ (50%)
Total 8/10 (80%) 85/88 (96.6%) 73/76 (96%) 20/22 (90.9%) 93/98 (94.9%)
Shano F 7/12 (58.3%) 2/3 (66.7%) 1/1 (100%) 8/14 (57.1%) 9/15 (60%)
M 4/5 (80%) 2/2 (100%) 1/1 (100%) 5/6 (83.3%) 6/7 (85.7%)
Total 11/17 (64.7%) 4/5 (80%) 2/2 (100%) 13/20 (65%) 15/22 (68.2%)
Overall 187/252 (74.2%) 507/531 (95.5%) 222/230 (96.5%) 472/553 (85.3%) 694/783 (88.6%)
samples randomly selected to represent M. ovinus col- for all positive-control samples in both qPCR and con-
lected from the three sites and sheep of different sex and ventional PCR assays validated our molecular findings
age groups. A BLAST analysis of the gltA gene sequence and confirmed that our working conditions precluded any
from all 50 M. ovinus specimens showed 99.6–100% accidental cross-contamination of the samples during the
(240/241–241/241) nucleotide homology with the Gen- study period.
Bank reference sequence of the candidate B. melophagi A molecular identification study of Bartonella spp. in M.
gltA gene obtained from M. ovinus found on sheep in ovinus using the rpoB and gltA genes confirmed the pres-
the USA (GenBank accession number: AY692475). Sim- ence of B. melophagi in our samples. This finding supports
ilarly, a BLAST analysis of the rpoB gene sequence of the earlier argument that the rpoB and gltA genes have been
Bartonella spp. obtained from all 50 M. ovinus speci- described as more potent than all other genes for the differ-
mens showed 100% (800/800) nucleotide identity with the entiation of closely related Bartonella spp. [35]. B. melophagi
GenBank reference sequence B. melophagi rpoB obtained was isolated from sheep blood in the USA [37], and recently,
from M. ovinus found on sheep in the USA (GenBank the same bacteria were isolated from the blood of two
accession number: EF605288); however, published arti- female patients with pericarditis and skin lesions in the
cle is not available on B. melophagi in M. ovinus of USA [38]. Furthermore, have been deposited the rpoB gene
sheep. sequence of B. melophagi amplified from M. ovinus in Gen-
Bank (accession number: EF605288), but these data are
4. Discussion still unpublished and not accessible using PubMed search.
In this study, the observation of negative results With all of these previously established facts in mind, we
for all negative-control samples and positive findings hypothesize that M. ovinus may play an important role in
Author's personal copy
74 B. Kumsa et al. / Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 69–76
Fig. 3. Map of Ethiopia showing the study sites and the rate of Bartonella melophagi infection in Melophagus ovinus, sheep ked by site.
the transmission of B. melophagi from sheep to sheep in The detection of B. melophagi in the non-feeding pupae
Ethiopia. To our knowledge, this study is the first to report of M. ovinus suggest the possibility of vertical transmission
B. melophagi in M. ovinus collected from sheep in Ethiopia. of B. melophagi in M. ovinus, as has been argued before [7].
We believe that additional in-depth studies are neces- The overall prevalence of 65.7% for M. ovinus recorded
sary to determine the precise role of M. ovinus, sheep and in our study indicates the great importance of M. ovinus
humans in B. melophagi transmission in the highlands of in sheep at the study sites in northern Oromia. This high
Ethiopia. overall prevalence is most likely attributable to several
An overall prevalence of 88.6% for B. melophagi DNA important factors, including favorable climatic conditions
in M. ovinus collected from sheep was recorded in our important for the survival, reproduction and development
study and is in agreement with the previously reported of all stages of M. ovinus; malnutrition and poor husbandry
prevalence of Bartonella spp. in other species of arthro- systems; the poor awareness of farmers; and inadequate
pods in other countries [7]. However, overall prevalence veterinary services at the study sites. This finding is in
values that are lower [39,24] and higher than our results line with previous reports from many sites in Ethiopia
for Bartonella spp. have been reported [7], variation that that showed that M. ovinus is restricted to highland areas
is most likely attributable to differences in geographical, with cool climates, where sheep are commonly raised, but
animal and vector factors. The absence of a significant dif- is not found in lowland areas in Ethiopia [31]. However,
ference in the overall prevalence of B. melophagi in M. this finding contrasts with the situation in most developed
ovinus from the three sites and between M. ovinus col- European countries and other countries of the world where
lected from female compared with male sheep and young M. ovinus is eradicated with effective pesticides [2].
compared with adult sheep in this study suggests that A significantly higher prevalence of M. ovinus in Kim-
the study site and the sex and age of the sheep do not bibit than at both the Chacha and the Shano sites in our
affect the prevalence of B. melophagi in M. ovinus. This study possibly reflects that the Kimbibit site, with an alti-
observation supports the role of M. ovinus as a poten- tude of 3000 m.a.s.l., is a more suitable environment for
tial vector of B. melophagi, as has been suggested before reproduction of M. ovinus than the other two sites, which
[38]. are at lower altitudes [31]. The absence of a statistically
The absence of a significant difference in the overall significant difference in the overall prevalence of M. ovinus
prevalence of B. melophagi between female and male M. among sheep of different sex and age groups in our study
ovinus collected from sheep in this study suggests that the is in line with earlier observations that suggest that differ-
sex of M. ovinus does not affect the organism’s infection ences in the prevalence of M. ovinus infestation in sheep
with B. melophagi, which is in line with a previous report of are associated with variations in altitude and environmen-
100% prevalence of Bartonella spp. in M. ovinus in the USA tal factors more than with host-dependent factors due to
[7]. the high likelihood of the transfer of M. ovinus by direct
Author's personal copy
B. Kumsa et al. / Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 69–76 75
contact among sheep, including transfer from ewes to [12] Anderson BE, Neuman MA. Bartonella spp. as emerging human
lambs during suckling [2]. pathogens. Clin Microbiol Rev 1997;10(2):203–19.
[13] Chomel BB, Kasten RW. Bartonellosis, an increasingly recognized
zoonosis. J Appl Microbiol 2010;109(3):743–50.
5. Conclusion [14] Saisongkorh W, Rolain JM, Suputtamongkol Y, Raoult D. Emerging
Bartonella in humans and animals in Asia and Australia. J Med Assoc
Thai 2009;92(5):707–31.
This study provides molecular evidence of B. melophagi [15] Chomel BB, Boulouis HJ, Maruyama S, Breitschwerdt EB. Bar-
infection in M. ovinus collected from sheep in Ethiopia. tonella spp. in pets and effect on human health. Emerg Infect Dis
Given the high prevalence of M. ovinus infestations in sheep 2006;12(3):389–94.
[16] Trataris AN, Rossouw J, Arntzen L, Karstaedt A, Frean J. Bartonella
and the very high prevalence of B. melophagi infection in M.
spp. in human and animal populations in Gauteng, South Africa, from
ovinus in our study, together with the previous isolation of 2007 to 2009. Onderstepoort J Vet Res 2012;79(2):E1–8.
the same bacteria from human and sheep blood in the USA, [17] Tsai YL, Chang CC, Chuang ST, Chomel BB. Bartonella species and their
ectoparasites: selective host adaptation or strain selection between
M. ovinus may be more important in Ethiopia than previ-
the vector and the mammalian host? Comp Immunol Microbiol Infect
ously believed. This study highlights that B. melophagi in Dis 2011;34(4):299–314.
M. ovinus on sheep in highlands in Ethiopia may have cer- [18] Maillard R, Riegel P, Barrat F, Bouillin C, Thibault D, Gandoin C,
tain zoonotic importance. Overall, our present data warrant et al. Bartonella chomelii sp. nov., isolated from French domes-
tic cattle (Bos taurus). Int J Syst Evol Microbiol 2004;54(Pt 1):
detailed future studies on ruminant-associated Bartonella 215–20.
spp. in Ethiopia, a country for which there is no informa- [19] Maillard R, Grimard B, Chastant-Maillard S, Chomel B, Delcroix T,
tion about Bartonella spp. and associated diseases in either Gandoin C, et al. Effects of cow age and pregnancy on Bartonella
infection in a herd of dairy cattle. J Clin Microbiol 2006;44(1):
humans or animals. In addition, future studies are neces- 42–6.
sary to address the isolation and culture of this bacterium [20] Cherry NA, Maggi RG, Cannedy AL, Breitschwerdt EB. PCR detection
from human and sheep blood and its public health signifi- of Bartonella bovis and Bartonella henselae in the blood of beef cattle.
Vet Microbiol 2009;135(3–4):308–12.
cance in Ethiopia. [21] Mediannikov O, Davoust B, Cabre O, Rolain JM, Raoult D. Bartonellae
in animals and vectors in New Caledonia. Comp Immunol Microbiol
Acknowledgments Infect Dis 2011;34(6):497–501.
[22] Tsai YL, Chomel BB, Chang CC, Kass PH, Conrad PA, Chuang ST. Bar-
tonella and Babesia infections in cattle and their ticks in Taiwan. Comp
The authors greatly acknowledge the sheep owners at Immunol Microbiol Infect Dis 2011;34(2):179–87.
the study sites for allowing us to collect M. ovinus from [23] Welc-Faleciak R, Grono K. The first cases of Bartonella bovis infec-
tion in cattle from Central Europe. Vet Microbiol 2013;162(2–4):
their sheep. We would also like to thank the laboratory
954–6.
technicians from URMITE, Marseille, and particularly Jean- [24] Chung CY, Kasten RW, Paff SM, Van Horn BA, Vayssier-Taussat M,
Michel Bérenger, and Veronique Brice for their technical Boulouis HJ, et al. Bartonella spp. DNA associated with biting flies
support. from California. Emerg Infect Dis 2004;10(7):1311–3.
[25] Billeter SA, Levy MG, Chomel BB, Breitschwerdt EB. Vector transmis-
sion of Bartonella species with emphasis on the potential for tick
References transmission. Med Vet Entomol 2008;22(1):1–15.
[26] Breitschwerdt EB, Kordick DL. Bartonella infection in animals:
[1] Gibson W, Pilkington JG, Pemberton JM. Trypanosoma melophagium carriership, reservoir potential, pathogenicity, and zoonotic
from the sheep ked Melophagus ovinus on the island of St Kilda. Para- potential for human infection. Clin Microbiol Rev 2000;13(3):
sitology 2010;137(12):1799–804. 428–38.
[2] Small RW. A review of Melophagus ovinus (L.), the sheep ked. Vet [27] Kumsa B, Parola P, Raoult D, Socolovschi C. Molecular detection of
Parasitol 2005;130(1–2):141–55. Rickettsia felis and Bartonella henselae in dog and cat fleas in Central
[3] Pfadt RE. Sheep ked populations on a small farm. J Econ Entomol Oromia, Ethiopia. Am J Trop Med Hyg 2013.
1976;69(3):313–6. [28] Tiao N, Darrington C, Molla B, Saville WJ, Tilahun G, Kwok OC,
[4] Martinkovic F, Matanovic K, Rodrigues AC, Garcia HA, Teixeira et al. An investigation into the seroprevalence of Toxoplasma gondii,
MM. Trypanosoma (Megatrypanum) melophagium in the sheep ked Bartonella spp., feline immunodeficiency virus (FIV), and feline
Melophagus ovinus from organic farms in Croatia: phylogenetic leukaemia virus (FeLV) in cats in Addis Ababa, Ethiopia. Epidemiol
inferences support restriction to sheep and sheep keds and close rela- Infect 2013;141(5):1029–33.
tionship with trypanosomes from other ruminant species. J Eukaryot [29] Meheretu Y, Leirs H, Welegerima K, Breno M, Tomas Z, Kidane
Microbiol 2012;59(2):134–44. D, et al. Bartonella prevalence and genetic diversity in small
[5] Nelson WA, Slen SB. Weight gains and wool growth in sheep mammals from Ethiopia. Vector Borne Zoonotic Dis 2013;13(3):
infested with the sheep ked Melophagus ovinus. Exp Parasitol 164–75.
1968;22(2):223–6. [30] Kumsa B, Socolovschi C, Parola P, Rolain JM, Raoult D. Molec-
[6] Luedke AJ, Jochim MM, Bowne JG. Preliminary bluetongue transmis- ular detection of Acinetobacter species in lice and keds of
sion with the sheep ked Melophagus ovinus (L.). Can J Comp Med Vet domestic animals in Oromia Regional State, Ethiopia. PLoS ONE
Sci 1965;29(9):229–31. 2012;7(12):e52377.
[7] Halos L, Jamal T, Maillard R, Girard B, Guillot J, Chomel B, et al. [31] Kumsa B, Beyecha K, Geloye M. Ectoparasites of sheep in three agro-
Role of Hippoboscidae flies as potential vectors of Bartonella spp. ecological zones in central Oromia, Ethiopia. Onderstepoort J Vet Res
infecting wild and domestic ruminants. Appl Environ Microbiol 2012;79(1):E1–7.
2004;70(10):6302–5. [32] CSA. Ethiopian agricultural sample survey report on livestock and
[8] Hornok S, de la Fuente J, Biro N, Fernandez de Mera IG, Meli ML, Elek livestock characteristics, vol. II, 2007/08. Addis Ababa, Ethiopia: CSA;
V, et al. First molecular evidence of Anaplasma ovis and Rickettsia 2008.
spp. in keds (Diptera: Hippoboscidae) of sheep and wild ruminants. [33] Rolain JM, Rousset E, La Scola B, Duquesnel R, Raoult D. Bartonella
Vector Borne Zoonotic Dis 2011;11(10):1319–21. schoenbuchensis isolated from the blood of a French cow. Ann N Y
[9] Chu CY, Jiang BG, Qiu EC, Zhang F, Zuo SQ, Yang H, et al. Borrelia Acad Sci 2003;990:236–8.
burgdorferi sensu lato in sheep keds (Melophagus ovinus), Tibet, China. [34] Rolain JM, Franc M, Davoust B, Raoult D. Molecular detection of Bar-
Vet Microbiol 2011;149(3–4):526–9. tonella quintana, B. koehlerae, B. henselae, B. clarridgeiae, Rickettsia
[10] Breitschwerdt EB, Maggi RG, Chomel BB, Lappin MR. Bartonellosis: felis, and Wolbachia pipientis in cat fleas, France. Emerg Infect Dis
an emerging infectious disease of zoonotic importance to animals 2003;9(3):338–42.
and human beings. J Vet Emerg Crit Care (San Antonio) 2010;20(1): [35] La Scola B, Zeaiter Z, Khamis A, Raoult D. Gene-sequence-based crite-
8–30. ria for species definition in bacteriology: the Bartonella paradigm.
[11] Guptill L. Bartonellosis. Vet Microbiol 2010;140(3-4):347–59. Trends Microbiol 2003;11(7):318–21.
Author's personal copy
76 B. Kumsa et al. / Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 69–76
[36] Maillard R, Vayssier-Taussat M, Bouillin C, Gandoin C, Halos L, [38] Maggi RG, Kosoy M, Mintzer M, Breitschwerdt EB. Isolation of Can-
Chomel B, et al. Identification of Bartonella strains isolated from wild didatus Bartonella melophagi from human blood. Emerg Infect Dis
and domestic ruminants by a single-step PCR analysis of the 16S-23S 2009;15(1):66–8.
intergenic spacer region. Vet Microbiol 2004;98(1):63–9. [39] Reeves WK, Nelder MP, Cobb KD, Dasch GA. Bartonella spp. in deer
[37] Bemis DA, Kania SA. Isolation of Bartonella sp. from sheep blood. keds, Lipoptena mazamae (Diptera: Hippoboscidae), from Georgia
Emerg Infect Dis 2007;13(10):1565–7. and South Carolina, USA. J Wildl Dis 2006;42(2):391–6.
6. CONCLUSIONS
275
276
6.1. Conclusion and perspectives
277
small size ticks, eggs, larvae and nymphs of ticks other body
parts without abdomen seems more applicable.
Our study on fleas provided molecular evidence that R. felis and
B. henselae in the fleas collected from dogs and cats in Ethiopia.
Likewise, our study on lice and sheep ked documented the
presence of Acinetobacter spp. in the lice and sheep ked
(Melophagus ovinus) collected from domestic animals in
Ethiopia and high prevalence of Bartonella melophagi in
M. ovinus of sheep in highlands of Ethiopia.
Overall, our findings alert physicians managing patients with
fever of unknown aetiology in Ethiopia and those who care for
travellers from Ethiopia to consider the presence of several
species of zoonotic vector-borne bacteria such as SFG
rickettsiae, C. burnetii, R. felis, B. henselae and B. melophagi as
potential causative agents in their differential diagnosis. In
addition, future studies are needed to address the isolation,
culture and genotypes of these zoonotic bacteria from different
arthropod species collected from animals and vegetation
(example questing ticks), and in blood samples from humans,
domestic and other reservoir animals and determine the public
health significance of these bacteria in Ethiopia.
Future studies are needed to address the vector competence of
Rhipicephalus (Boophilus) decoloratus and Am. gemma for
278
R. africae by demonstrating its ability to acquire the pathogen
from an infected host and transmit it to nave susceptible animals
during the subsequent feeding. Similarly, the vector competence
of M. ovinus for Bartonella melophagi, Amblyomma spp. ticks
for the new Borrelia spp. remains to be confirmed by future
studies.
Other important questions awaiting future studies include
determination to get clear information on the comparative role
of different species of domestic and wild animals, the role of
different species of arthropods and identification of reservoir
animals and description of diseases caused by reported bacteria
in human patients. Much works is needed to increase the society
and professional awareness about vector-borne bacterial
diseases so as to mitigate the negative effects on public health.
Future studies are expected to develop revised MALDI-TOF MS
protein profiling method for tick identification by providing
standardized methods of sample preparation, protein
differentiation, parameters of spectra quality control and spectra
analysis and interpretation methods so that this technique will be
widely used uniformly around the world. In addition,
standardized MALDI-TOF MS method for identification of the
small size ticks, eggs, larvae and nymphs of ticks need to be
developed.
279
280
Bibliography
281
282
References
283
6. Breitschwerdt, E. B., R. G. Maggi, B. B. Chomel, and
M. R. Lappin. 2010. Bartonellosis: an emerging
infectious disease of zoonotic importance to animals
and human beings. J.Vet.Emerg.Crit Care
(San.Antonio.) 20:8-30. doi:VEC496
[pii];10.1111/j.1476-4431.2009.00496.x [doi].
284
12. Hoogstraal, H. 1956. African Ixodidae. 1. Ticks of the
Sudan (with special reference to equatorial province
and with preliminary reviews of the genera Boophilus,
Margaropus and Hyalomma). Department of the Navy,
Bureau of the Medicine and Surgery, United States
Government Printing Office, Washington D. C.
285
henselae in dog and cat fleas in Central Oromia,
Ethiopia. Am.J.Trop.Med.Hyg. 90:457-462.
286
Trans.R.Soc.Trop.Med.Hyg. 102:945-949. doi:S0035-
9203(08)00113-2 [pii];10.1016/j.trstmh.2008.03.015
[doi].
287
29. Philip, C. B., D. B. Lackman, E. J. Bell, and L. E.
Hughes. 1966. Laboratory identification of typhus
isolated by Reiss-Gutfreund from Ethiopian livestock
ticks. Am.J.Trop.Med.Hyg. 15:950-953.
288
36. Sangare, A. K., A. Boutellis, R. Drali, C. Socolovschi,
S. C. Barker, G. Diatta, C. Rogier, M. M. Olive, O. K.
Doumbo, and D. Raoult. 2014. Detection of Bartonella
quintana in African body and head lice.
Am.J.Trop.Med.Hyg. 91:294-301. doi:ajtmh.13-0707
[pii];10.4269/ajtmh.13-0707 [doi].
289
to Identification of species. Bioscience Reports,
Edinburgh Scotland,U.K.
43. Walker J.B., Keirans J.E., Horak I.G., 2000. The Genus
Rhipicephalus (Acari, Ixodidae): A Guide to the Brown
Ticks of the world. Cambridge Univeristy Press,
Cambridge.
290