Professional Documents
Culture Documents
In summary, it is plausible that microbial interactions are further underlined by animal experiments showing that death occurred
multifactorial and involve a complex interplay between multiple in 35% and 15% of mice infected with either influenza virus or
host factors and bacterial characteristics that may have important pneumococcus, respectively, whereas 100% of mice infected with both
consequences for both the composition and stability of the pathogens simultaneously succumbed to infection within one day [39].
microbial community itself and susceptibility to disease [37]. Besides synergism between influenza virus and S. pneumoniae, other
interactions between viral and bacterial species have been described in
Viral–Bacterial Interaction the literature, as shown in Table 1 [39–69].
The mechanisms by which viruses influence bacterial coloniza-
Interactions between viruses and bacteria in the pathogenesis of
tion and invasion are very diverse. We have summarized the
respiratory infections have been extensively reported in the literature.
known mechanisms in Figure 2a and 2b and will discuss each of
Perhaps the most well-known viral–bacterial interaction is the
these mechanisms below.
synergism between influenza virus and S. pneumoniae [38]. Although
an influenza virus infection alone can be fatal, mortality increases
dramatically when a bacterial super-infection occurs, as in the case of Viral Predisposition to Bacterial Adherence
the ‘‘Spanish flu’’ pandemic in 1918–1919 when millions of people Since attachment of a pathogen to mucosal surfaces is the first
died, most from secondary pneumococcal pneumonia [38]. This is step towards respiratory disease, and viral infection alters the
Human Studies
Asymptomatic
Virus Bacterium Association Children Animal Studies In Vitro Studies
Virus (column one) and respective bacterium (column two) for which interactions were observed (column three), and source of evidence: from human studies (column
four), animal studies (column five), or in vitro studies (column six) showing type of epithelium tested.
NA, data not available from literature.
doi:10.1371/journal.ppat.1003057.t001
defense of the host epithelium in general [70], it has been postulated Disruption of the Epithelium Barrier
that viral presence may render the epithelium more susceptible to The epithelial layer of the respiratory tract mucosa is the first
bacterial colonization [10]. Mouse studies have shown that viral line of defense against a bacterial invader: loss of barrier function
predisposition to bacterial attachment not only occurs in case of a could therefore lead to entry of pathogens. Viruses generally
simultaneous infection, but also up to a week after initial viral replicate intracellularly and can subsequently disarrange cellular
infection [46,48,55] or even after full recovery from influenza processes or kill infected cells through metabolic exhaustion or
infection [71]. Moreover, Hakansson et al. [53] demonstrated that direct lysis. Induced cell death may in turn lead to denudation of
not all viral types and bacterial species and strains interact to the the epithelial layer [59,65], exposing the basement membrane. S.
same extent; only pneumococcal strains with high adhesive capacity pneumoniae was found to bind strongly to fibronectin, which is
were able to adhere to human respiratory epithelium infected with prominently exposed at the basement membrane after denudation
adenovirus, and this effect was restricted to types of adenoviruses of epithelium [72]. Similarly, S. aureus [73] and M. catarrhalis [74]
generally able to cause respiratory disease in humans. have been shown to bind to extracellular matrix proteins,
suggesting that these species might also benefit from virus-induced Production of Viral Factors
damage to epithelium. Furthermore, the binding capacities of Influenza virus is thought to increase bacterial adherence by
bacteria to fibronectin appear to be strongly influenced by the alternative mechanisms, such as the ability to produce neuramin-
amount of exposed fibronectin and exposure duration [72], and idase (NA). NA produced by influenza and para-influenza viruses
since viral presence may directly induce upregulation of fibronec- creates an entry point for bacteria into host cells by cleaving sialic
tin expression, as has been shown for rhinovirus, this will acids residues, thereby exposing bacterial receptors on the surface
additionally enhance pathobiont binding [40]. of the upper respiratory tract [79–81]. This is supported by several
Another consequence of disrupted epithelium is the loss of in vitro and animal studies, including studies on the effects of NA
epithelial integrity and decreased inhibition of bacterial translo- inhibitors [58,81,82]. Although some bacteria such as S. pneumoniae
cation. This has been clearly shown for rhinovirus-induced naturally express NA [83], the contribution of bacterial NA to viral
paracellular migration of H. influenzae [75]. Viruses may also replication seems to be negligible, most likely due to poor
induce damage to ciliated cells, resulting in decreased mucociliar enzymatic activity and stringent binding requirements of bacterial
velocity and impaired bacterial clearance [61,65]. NA compared to viral NA [84].
RSV, on the other hand, does not produce NA. Instead,
Upregulation of Adhesion Proteins adherence of bacteria to RSV-infected cells is thought to be
Viral presence in the infected cells may alter the expression of directly mediated by expression of RSV-protein G [46,47,51].
antimicrobial peptides, also known as defensins [76], secreted in the Blockade of G-protein activity, however, does not completely
respiratory mucosa [56], which are key innate immune components reduce excess bacterial colonization in RSV-infected cells in vitro
that directly eliminate pathogenic bacteria [76]. Viral infection also [51]. This implies that other mechanisms might be involved during
triggers a pro-inflammatory response that leads to upregulation of viral–bacterial co-occurrence, such as upregulation of additional
adhesion proteins in a range of cells, including epithelial cells. These receptors like ICAM-1 and PAFr [48,51] or other indirect
adhesion proteins act as receptors that allow immune cells to bind to pathways.
virus-infected cells and combat the viral invader. This is illustrated
by the upregulation of eukaryotic cell surface receptors such as Dysfunction of Immune System Components
intracellular adhesion molecule 1 (ICAM-1), outer membrane As described above, viral-induced expression of adhesion
protein P5-homologous fimbriae (P5 fimbriae), carcinoembryonic molecules may support adhesion of neutrophils, monocytes,
adhesion molecule-1 (CEACAM-1), and platelet-activating factor and other immune cells to virus-infected cells. This may
receptor (PAFr) in different cell types upon infection with a virus increase recruitment and activation of pro-inflammatory
such as respiratory syncytial virus (RSV) or para-influenza virus immune cells. However, respiratory viruses may also directly
[48,52]. Several bacterial species are able to adhere to a diverse affect the immune system, for example by impairment of
group of these adhesion proteins on the surface of host cells neutrophil function, decreased oxidative burst [55,85], and
[28,40,41,48,77]. For example, rhinovirus is able to induce enhanced neutrophil apoptosis, thereby increasing susceptibil-
upregulation of ICAM-1 needed for its own invasion as well as for ity to bacterial superinfection [85–87]. Additionally, some
adhesion of H. influenzae [43,48]. Moreover, some strains of S. strains of influenza virus infection may predispose to superin-
pneumoniae and H. influenzae express the natural PAFr-ligand fection by S. aureus due to ineffective natural killer (NK) cell
phosphorylcholine, which also allows them to attach to, and invade, recruitment and activation [64]. Viral infection may also alter
host cells. Increased PAFr expression in reaction to a viral infection monocyte function, resulting in lower surface expression of CD
may therefore facilitate adherence of both S. pneumoniae and H. receptors [50]. In addition, viral presence also affects the
influenzae [40,41,48]. However, influenza viruses might form an production and biological activity of cytokines [54]. For
exception, as in vitro studies have found that influenza virus did not example, virus-induced interferon (IFN)-a and IFN-b induce
alter the expression of several receptors, including ICAM-, impaired neutrophils responses due to inadequate production
CAECAM, and PAFr [48]. In particular, conflicting data have of neutrophil chemoattractants [88]. In addition, IFN-c
been reported for a potential role of PAFr in the protection against downregulates the activity of macrophages [89], thus impair-
influenza-related bacterial superinfection in mouse models [39,78], ing bacterial clearance in its initial phase. It has also been
though this might be explained by strain-related differences as well shown that blockage of IFN-c decreases susceptibility to
as the timing and order of viral and bacterial exposure [78]. secondary bacterial pneumonia in mice [89]. Moreover, tumor
n (%)
Polyoma
Picorna AdV HBoV RSV hMPV CoV IV PIV Viruses
HRV EV
2011 [67] Autumn, winter, spring 66 6 m–3 y Healthy 28 (42%) 5 (7.6%) 4 (6.1%) 13 (20%) 6 (9.1%) 1 (1.5%) 5 (6.1%) 0 (0%) 6 (9.1%) WU 9
(14%); KI 1
(1.5%)
2011 [99] All year 34 ,1 y Healthy 8 (24%)
7
2010 [97] All year 57 ,12 y Rural 2 (4%)
2010 [100] All year 425 ,3 y At risk 140 (33%) 68 (16%) 18 (4%) 29 (7%) 15 (4%) 3 (1%) 13 (3%)
2009 [101] Autumn, winter, spring 65 ,7 y Healthy 14 (22%) 2 (3%) 0 (0%) 0 (0%) 0 (0%) 5 (8%) 0 (0%)
2008 [102] All year 116 ,14 y Healthy 11 (9.5%) 5 (4.3%) 2 (1.7%) 1 (0.8%) 1 (0?8%) 0 (0%) 0 (0%) 0 (0%)
2008 [112] Autumn, winter, spring 100 #3 y Healthy 43 (43%)
2007 [103] Autumn, winter, spring 269 1.5–9.3 y Healthy 29 (11%) 2 (1%) 2 (1%) 1 (0%) 1 (0%)
2006 [96] All year 456 ,1 y High risk of atopy 52 (11%) 2 (0%) 24 (5%) 1 (0%) 20 (4%) 0 (0%) 4 (1%)
2006 [104] All yearc 410 1–9 y Healthy 37 (9%)d
2004 [94] NS 70 5m Healthy 12 (17%) 2 (3%) 3 (4%) 1 (1%) 1 (1%)
64 1y Healthy 18 (28%) 0 (0%) 0 (0%) 0 (0%) 0 (0%)
38 1.5 y Healthy 10 (26%) 0 (0%) 0 (0%) 0 (0%) 0 (0%)
49 2y Healthy 7 (14%) 0 (0%) 1 (2%) 0 (0%) 0 (0%)
a
Related to geographical area.
b
Number of samples tested.
c
Stratified for season.
d
Picornavirus general.
M, months of age; Y, years of age; HRV, human rhinoviruses; EV, entero viruses; AdV, adeno viruses; HBoV, human bocavirus; RSV, respiratory syncytial virus; hMPV, human metapneumovirus; CoV, corona viruses; IV, influenza
viruses; PIV, para-influenza viruses; NS, not specified.
doi:10.1371/journal.ppat.1003057.t002
necrosis factor (TNF)-a production is downregulated during viral Asymptomatic Presence of Viruses In Vivo
infection, which may also lead to increased susceptibility to
secondary bacterial infections [50]. Respiratory viruses can also The impact of viral presence could be far more extensive than
interact with toll-like receptor (TLR) pathways, preventing appro- currently thought. In addition to bacterial commensals, viruses are
priate routing of immune responses [90]. This is, for example, also commonly found in the nasopharynx of asymptomatic
illustrated by data obtained from a co-infection model with individuals. With the introduction of viral PCR techniques, it
influenza virus and S. pneumoniae in mice, where excessive has become feasible to detect and distinguish between respiratory
immunosuppressive interleukin (IL)-10 production following co- viruses in larger epidemiological studies. A concise review showed
infection has been observed, which was associated with enhanced that up to 68% of respiratory samples from asymptomatic
bacterial colonization and increased mortality [71]. individuals were positive for respiratory viruses [93]. When
specifying these numbers for symptom-free children, studies have
Unidirectional or Bidirectional Synergism reported presence rates of 16%–33% in developed communities
Most studies point towards a unidirectional viral predisposition [68,94–96] and 4%–52% in developing communities [68,97–100].
to bacterial colonization. However, there are some clues that a Interestingly, children in some native populations, such as
preceding bacterial infection may also increase susceptibility to a Australian Aboriginals and Alaska Yup’ik Eskimos, are known to
consecutive viral infection. For example, Sajjan et al. [42] showed be more susceptible to diseases caused by respiratory pathogens,
that H. influenzae is able to stimulate expression of ICAM-1 and and also seem to more frequently carry respiratory viruses during
TLR-3 on human airway epithelial cells, providing an entry point healthy periods [68,100]. A detailed overview of data on the
for rhinovirus. Another report suggested that human bronchial asymptomatic presence of viruses is presented in Table 2.
epithelial cells pre-incubated with pneumococcus, but not with H. Differences between studies are likely to be explained by
influenzae, M. catarrhalis, or S. aureus, were more susceptible to inclusion criteria and heterogeneity of populations due to
human metapneumovirus [45]. Moreover, it might also be possible differences in age, sample size, genetic background, season of
that microbial interactions may disturb the equilibrium of the sampling, lifestyle, and environmental circumstances, as well as
microbiota, creating an opportunity for viral invasion and health status and registration of respiratory symptoms prior to or
transmission. This was recently underlined by Kuss et al. [91], following sampling.
who showed that transmission of an enteric virus was less The interpretation of viral presence in human respiratory
successful when the intestinal microbiota of mice were disbalanced samples is therefore becoming increasingly complex. In children,
by antibiotic treatment. Importantly, viruses might even be Singleton et al. [100] proposed dividing respiratory viruses into
capable of using their microbial environment to escape immune two groups, depending on their viral contribution to disease. The
clearance [92]. Little information exists, however, regarding contributing factor to illness of a given viral pathogen was
bacterial predisposition to viral disease, and further research is estimated by the proportion of all hospitalized cases related to this
needed to unravel the extent to which bacteria enhance viral virus divided by its presence rate in asymptomatic children. Group
presence. 1 includes viruses with a significantly greater contribution to
References
1. World Health Organization Geneva, Health Statistics and Informatics 8. Spor A, Koren O, Ley R (2011) Unravelling the effects of the environment and
Department (2011) Causes of death 2008. In: World health statistics 2011. host genotype on the gut microbiome. Nat Rev Microbiol 9(4): 279–290.
170 pages, ISBN 978 92 4 156419 9 (NLM classification: WA 900.1). pp. 57– 9. Grice E, Segre J (2011) The skin microbiome. Nat Rev Microbiol 9(4): 244–253.
76. Available: http://www.who.int/gho/publications/world_health_statistics/ 10. Bogaert D, De Groot R, Hermans PW (2004) Streptococcus pneumoniae
EN_WHS2011_Full.pdf. Accessed 7 December 2012. colonisation: The key to pneumococcal disease. Lancet Infect Dis 4: 144–154.
2. Black RE, Cousens S, Johnson HL, Lawn JE, Rudan I, et al. (2010) Global, 11. Garcia-Rodriguez JA, Fresnadillo Martinez MJ (2002) Dynamics of nasopha-
regional, and national causes of child mortality in 2008: A systematic analysis. ryngeal colonization by potential respiratory pathogens. J Antimicrob Che-
child health epidemiology reference group of WHO and UNICEF. Lancet mother 50(suppl_3): 59–74. 10.1093/jac/dkf506.
375(9730): 1969–1987. 12. Margolis E, Yates A, Levin B (2010) The ecology of nasal colonization of
3. Vergison A, Dagan R, Arguedas A, Bonhoeffer J, Cohen R, et al. (2010) Otitis streptococcus pneumoniae, haemophilus influenzae and staphylococcus aureus:
media and its consequences: Beyond the earache. Lancet Infect Dis 10(3): 195– The role of competition and interactions with host’s immune response. BMC
203. 10.1016/S1473-3099(10)70012-8. Microbiology 10(1): 59.
4. Watson KBH, Carville KMA, Bowman JBA, Jacoby P, Riley TV, et al. (2006) 13. Chesson P (2000) General theory of competitive coexistence in spatially-varying
Upper respiratory tract bacterial carriage in aboriginal and non-aboriginal environments. Theor Popul Biol 58(3): 211–237. 10.1006/tpbi.2000.1486.
children in a semi-arid area of western Australia. Pediatr Infect Dis J 25(9): 14. Brogden KA, Guthmiller JM, Taylor CE (2005) Human polymicrobial
782–790. 10.1097/01.inf.0000232705.49634.68. infections. Lancet 365(9455): 253–255. 10.1016/S0140-6736(05)17745-9.
5. Blaser MJ, Falkow S (2009) What are the consequences of the disappearing 15. Kwambana B, Barer M, Bottomley C, Adegbola R, Antonio M (2011) Early
human microbiota? Nat Rev Microbiol 7(12): 887–894. 10.1038/nrmicro2245. acquisition and high nasopharyngeal co-colonisation by streptococcus pneu-
6. Murphy T, Bakaletz L, Smeesters P (2009) Microbial interactions in the moniae and three respiratory pathogens amongst Gambian new-borns and
respiratory tract. Pediatr Infect Dis J 28(10): S121–S126. infants. BMC Infect Dis 11(1): 175.
7. Hardin G (1960) The competitive exclusion principle. Science 131(3409): 16. Jourdain S, Smeesters PR, Denis O, Dramaix M, Sputael V, et al. (2011)
1292–1297. Differences in nasopharyngeal bacterial carriage in preschool children from