Molecular Ecology (1995)4,499-503

SHORT COMMUNICATION

Use of the PCR and fluorescent probes to recover SSU

rRNA gene sequences from single cells of the ciliate
protozoon Spathidiurn
P. L. DYAL, S. HOPE, D. M. R O B E R T S and T. M. EMBLEY
Microbiology Group, Department of Zoology, The Natural History Museum, Cromwell Road, London S W 7 SBD, UK

Abstract
A two-stage heminested PCR approach was developed to amplify small subunit (SSU)
rDNA sequences, via two overlapping fragments, fmm single cells of microbial eucary-
otes. The method was evaluated using the ciliate protozoon Spathidium when PCR prod-
ucts were obtained from nine of 10 cells tested. Southern blotting demonstrated that all
fragments contained the same sequence in a region of SSU rDNA which is normally high-
. ly variable between species. A fluorescent oligonucleotide probe was used to demonstrate
that this sequence also occurred in fixed cells of Spathidium. Fixatives containing mer-
curic salts preserved cell shape and allowed probe binding with little background auto-
fluorescence. The Spathidium sequence is closely related to that from the haptorid
Homalozoon vmniculare.
Keywords: ciliates, fluorescentprobes, PCR, phylogeny, single cells, SSU rDNA
Received 16 Nmembn 1994; mision uccepted 30 March 1995

ent groups of microbial eucaryotes which live in the

Introduction
Microbial eucaryotes (pmtists) play important ecological
roles in primary production and decomposition (Fenchel
1987; Sleigh 1991).Identification of microbialeucaryotesin
natural samples is normally based upon morphology,
since most have not been cultured. Some protists show
much morphological detail so an expert can identify them
and study their distributions and ecology, other groups are
poor in morphological detail and are taxonomically diffi-
cult (Fenchell993).Morphologicaland ultrastructural fea-
tures have also been the main characters used for making
phylogenetic inferences about microbial eucaryotes, and it
is upon these foundations that hypotheses about the dis-
tributions and origins of ecologically interesting traits
have been formulated.
Over the past 10 years or so, it has become possible to
i d e n w microbial eucaryotes, and to study their phyloge-
netic relationships using small subunit (SSU) ribosomd
RNA sequences (reviewed in Sogin 1991). The results of
these studies have provided more reliable phylogenetic
trees for microbial eucaryotes and new insights into the
evolution of phenotypic features. For example, it is now
apparent that there are at least two fundamentally differ-
Correspondence: T. Martin Embley. Tel.: 071 9388760. Fax:
071 9388754.E-mail: tme@nhm.ac.uk
absence of oxygen (Cavalier-Smith 1987). Eucaryotes such
as Giardia and relatives are anaerobic because they
diverged from the eucaryotic tree before the acquisition of
the bacterial symbionts which gave rise to mitochondria
(Gray 1992; Sogin et al. 1989). In contrast, anaerobic ciliates
and fungi are probably secondary adaptations fmm aero-
bic species in order to colonize anoxic habitats (Dore &
Stahl1991; Embley & Finlay 1994b).
The invention of the PCR reaction (Mullis et al. 1986)
and the recognition that eucaryote SSU rRNA sequences
contain strongly conserved primer sequences (Elwood et
al. 1985), means it is now possible to recover SSU rDNA
sequence information from small amounts of eucaryote
biomass (Medlin et al. 1988) and potentially single cells. A
reliable single cell technique would effectively eliminate
the constraints imposed by the need to cultivate microbial
eucaryotes, making any interesting species accessible to
study. In the present investigation we have combined and
modified methods based upon PCR and whole cell
hybridization with fluorescent probes @elong ef d.1989;
-ley et ul. 1992; Stahl & Amann 1991), to amplify and
sequence the SSU rDNA from single cells of the large car-
nivorous ciliate Spathidium. This species is a representative
of the Haptorida, a group of mainly raptorial aliates
whose members are commonly encountered in envimn-
mental samples (Corliss 1979), but for which only a single
500 P. L . DYAL etal.
SSU rDNA sequence is currently available (Leipe et al.
1994). Ciliates are particularly amenable to a single cell
amplification a p p m c h since they often contain many
copies of their SSU rRNA gene sequences (Prescott 1994).
The method we used in the present study incorporates a
two stage, heminested amplification procedure (Li et al.
1990), because PCR using a single set of primers to ampli-
fy the complete SSU rDNA has not proved to be reliable
for single ciliates in our laboratory. Nested or heminested
PCR strategies are commonly used for single cell amplifi-
cations to increase reliability, yield and specificity (Li et al.
1991). We used universal eucaryote primers to amplify
the SSU rDNA gene and fluorescent probing of intact cells
to confirm that the sequence came from cells of Spafhidiiirn
(Amann et al. 1991; Delong et nl. 1989; Elwood et 01. 1985;
Embley et al. 1992). We felt that it was important to be able
to link a sequence to a particular morphotype, because
identification of ciliates is traditionally based upon mor-
phology, and because PCR reactions using single cells
taken from mixtures may be prone to contamination with
other microbial eucaryotes of unknown phylogenetic posi-
tion.
Methods
Sputhidium sp. was collected from a water sample taken
from Phuket, Thailand (a gift from Mr A. Spencer). It was
maintained in 1% (wfv) Cerophyl (Agri-Tech, USA) with
food protozoa from the same sample. Ten live cells of
Sputhidium were manipulated by hand using fine capillar-
ies through three 20-mL washes of distilled water, to elim-
inate other eukaryotes. Each cell was transferred into a
separate mimfuge tube and allowed to air dry at room
temperature. PCR reactions were carried out in these
tubes. Negative controls comprised 10 pL of wash solu-
tion, and positive controls contained 100 ng of Tetrahpena
DNA.
Small subunit ribosomal RNA genes were amplified
using a two step heminested PCR protocol. Primary
amplifications of full-length SSU rRNA genes used pub-
lished PCR conditions and the primers 5fw AYCTG-
GTTGATYYTGCCAC and reverse 3rv TGATC-
CATCTGCAGGTTCACCT (Embley et al. 1992). The pri-
mary amplifications were spin-dialysed to remove the first
set of primers using Microcon-100 concentrators (Amicon)
against sterile water (100 pL, three times). One microlitre
of the retentate was used for each of two (per sample) sec-
ondary nested amplifications containing one of the follow-
ing primer pairs: 5fw & 1705rv ACGGGCGGTGTGTRC,
or 377fw CCGGAGARGAAGCCTGA& 3rv (Elwood et al.
1985).
To investigate if all detectable PCR products contained
the same sequence, an oligonucleotide probe was
designed based upon the sequence of one PCR product
and used to probe Southern blots of all nested PCR prod-
ucts. Material from one successful secondary amplification
(tube 5 in Fig. 1A) was PEG precipitated and sequenced
directly (Embley 1991) using intemal primers (Elwood et
al. 1985). An oligonucleotide probe 5-ACCCTCACTA-
CAATCGCGAT was designed to one of the most variable
regions of SSU rRNA sequence (helix 18 in region V3,
Neefs et al. 1993). The oligonucleotide was end-labelled
using [-p32P] ATF and used under stingent hybridization
conditions to probe the Southern blots.
To confirm that the amplified sequences originated
from Spathidium, the oligonucleotide was labelled with
tetramethylrhodamine and used to probe a mixture of
Spathidium, food ciliates and prokaryotes (Delong et al.
1989; Embley et al. 1992). A positive control probe which
binds to a l l eucaryote SSU RNAs (Stahl & Amann 1991)
was labelled with carboxyfluorescein. All of the probes
were synthesised using Amino-link 2 (Applied
Biosystems), coupled to the dye (Applied Biosystems)
using the manufacturers protocols, and purified as
described previously (Embley & Finlay 1994a). In initial
experiments, it was found that paraformaldehyde (tested
Fig. 1 Results of nested PCR reac-
tions to amplify complementary
overlapping fragments of SSU rDNA
sequence from single cells Of the cili-
ate Spathidium. The top half of the
figure shows two sets of nested PCR
reactions using different pairs of
primers with starting template from
individual (1-10) primary amplifica-
tions.The bottom shows a Southern
blot of the same fragments, probed
under stringent conditionswith an
oligonudeotide designed from a
variable stretch of fragment 5 A.
Lane T contains amplified product
from Tetrahymena.
 rDNA SEQUENCES FROM SINGLE PROTOZOA
Pig. 2 Micrographs showing the result
of probing a mixture of microorgan-
isms with the same probe labelled with
tetramethylrhodamine.The cells were
fixed with Zenkers Fluid. On the left is
a phase contrast image showing
Spathidim (S)surrounded by other
microorganisms. On the right is an epi-
fluorescence image of the same field
following excitation for the tetramethyl-
rhodamine labelled probe. There is a
small amount of autofluoresdng mater-
ial in the field but the positive probe
response to Spathidium is clear. The
scale bar is 50 pm.
in a range of 0.5-lo%, w/v), which is the standard fixative
for whole cell hybridization (Stahl & Amann 1991), lysed
Spathidium (but interestingly not the food ciliates). A vari-
ety of other simple fixatives commonly used in protistol-
ogy were therefore evaluated, the aim being to identify
one which preserved the morphology of all the ciliates in
the mixture and allowed probe binding while limiting aut-
ofluorescence. All of the fixative solutions were prepared
as described in Roberts & Causton (1988) and comprised
the following: gluteraldehyde (2.5%), Susas fltid,
Zenkers fluid, Hellys fluid, ZFA (basically Zenkers plus
acetic acid and formalin), Schaudinns fluid, Bouins fluid,
Parduczs fluid, osmium tetroxide (2%) and Champys
fluid. Cell suspensions in glass tubes were treated with
each fixative (3:l v/v; fixative: cell suspension) at room
temperature for 3 min, gently centrifuged (700 g) and
washed (twice) with 20 mL of distilled water. The cells
were then postfixed using the standard 4% ( w / v )
paraformaldehyde procedure, re-suspended in 100 pL of
0.1% Nonidet P-40 in distilled water, and 10 pL was placed
onto a gelatin coated slide and allowed to air dry.
Hybridization experiments were carried out on ethanol
dehydrated cell smears at 48 C for 4-6 h (Stahl & Amann
1991; Embley et al. 1992).
Results and discussion
None of the 10 primary amplifications yielded a PCR
product of the correct length which could be detected
using ethidium bromide fluorescence or by oligonu-
cleotide probing of Southern blots (data not shown). The
primary amplifications were spin-dialysed to remove the
first set of primers and two,(for each cell) heminested PCR
reactions were carried out each using 1 pL of the dialysed
retentate from the first reaction as a starting template.
Removal of the first set of primers was found to be neces-
sary to avoid obtaining a smear, rather than a discrete
band, in the second set of reactions. Spin dialysis may also
help to remove water soluble inhibitory material which
501
may contribute to poor PCR yields. The primer pairs cho-
sen for the second reactions generated overlapping prod-
ucts which in combination encompassed the entire length
of eucaryotic SSU rDNA. The overlap region between the
two products contains a number of highly variable
sequence regions (e.g. regions V3 & V4, Neefs et al. 1993)
analysis of which can be used to confirm that pairs of
products were from the same full-length starting template.
Eighteen (i.e. nine pairs) out of 20 nested PCR reactions
gave a single band which was detectable using gel elec-
trophoresis and ethidium bromide fluorescence, but the
yield varied considerably between reactions (Fig. 1). The
PCR products from tubes containing Spathidim template
were slightly smaller (estimated total length of 1.6 kbp)
than the product for the positive control Tetruhymena. The
smaller size is consistent with the reported length (1640
bases) of the single published haptorid SSU rRNA
sequence of Homalozoon vermiculare (Leipe et al. 1994).
Nested products from one cell were precipitated and
sequenced directly using a linear PCR reaction. The entire
sequence was obtained from these reactions thus avoiding
the need to clone PCR products.
All of the PCR products gave a positive hybridization
signal with the probe designed to bind to a stretch of SSU
rDNA sequence which normally demonstrates variation
between even closely related organisms (region V3, Neefs
et al. 1993). Two of the weakest PCR reactions (7 and 8,
Fig. 1A) gave detectable probe signals (not shown) only
after extended incubation. The results strongly suggest
that all of the nested PCR products contain the same vari-
able stretch of sequence and all are likely to originate from
the same source.
The next part of thc experiment was to confirm that the
sequence originated from cells of Spathidim. In initial
experiments it was observed that published (Stahl &
Amann 1991)whole cell probing methods were unsuitable
for Spathidium. We therefore evaluated a variety of the
standard fixatives used in protistology, for their utility in
whole cell probing of this cell and the other eucaryotes in
502 P. L . DYAL e t a l .
the mixture. Six of the 11 fixatives tested gave good preser-
vation of cell shape combined with strong probe fluores-
cence and low autofluorescene.With these fixativesit was
observed tha4 the probe bound to sequences in Spathidim
but not to any of the other m i a ~ ~ r g a n i s m(e.g.
s Figure 2).
Five of the successful fixatives (Susas,Zenkers, Hellys, method.
ZFA and Schaudinns fluids)contained mercuric salts as a
Common featum, and various combinations of acid, for-
malin or ethanol as additional fixatives or to aid penetra-
tion. There were no obvious differences observed between
the performance of thesefive fixatives for probing purpos-
es and all worked well with Sputhidium and the ciliates in
the mix. Barins fluid, which contains picric acid and for-
malin, also worked well. Osmium based fixatives (osmium
tetroxide, Parduczs and Champys fluids) preserved cell
shape, but the osmium precipitate prevented the detection
of a spefihc signal. Autofluorescence of protists was very
strong with gluteraldehyde at 2.5% and it was not possible
to discriminate any specific probe response. Fluorescent
probing to identify the origin of a sequence, or to identify
and count cells in environmental samples, is a robust tech-
nology and it can be done on crude preparations (Stahl&
Amann 1991). It holds much promise for the identification
and enumeration of taxonomically difficult microorgan-
isms in environmental samples. However, as our results
demonstrate, some experimentation with different fixa-
tives may be necessary to avoid lysis of some cells when
dealing with mixtures of microbial eucaryotes.
A full length SSU rDNA sequence for Spathidium was
obtained by directly sequencing pooled overlapping PCR
fragments and it has been deposited in Eh4BL as 222931.
Phylogenetic analysis of this sequence (not shown) reveals
that S p a t h i d i m forms a sister group relationship with the
other sequenced haptorid H o d o z o o n vermiculure, but the
precise phylogenetic position of this dade within the radi-
ation of ciliates is not m l v e d at present (Leipeet al. 1994;
Hirt et al. 1995).
Ciiiates are important consumexs in a range of environ-
ments and they display a variety of ecologically interesting
features (Sleigh 1991). Some ciliates are anaerobic, some
form symbioses with biochemically different prokaryotes,
and some have a complex feeding apparatus while others
are very simple (Fenchel & Finlay 1989;Fenchel6r Finlay
1991; Sleigh 1991). Phylogenetic trees based upon SSU
rDNA sequences can provide a framework for E o n -
strutting the historical pattern of origins and diversifica-
tion of these and other characters (Brooks Q McLennan
1991; Wanntorp et al. 1990). The combination of methods
described in this paper have proven to be very reliable for
recovering SSU rDNA sequences from single or small
numbers of ciliate protozoa. We have used it to amplify
SSU rDNA from phylogenetidy diverse aliates indud-
ing single cells of the anaerobic ciliates Strombidium and
Saprodinium. Typically the percentage success from single
cells is greater than 50%, so if two cells are isolated and
processed, there is an excellent chance of recovering a
sequence from one of them. This means that any interest-
ing ciliate which can be separated from a sample can be
investigated for its SSU rDNA sequence using this
Acknowledgements
The authors would like to thank Dr Robert Hirt for useful discus-
sions. This work is partly funded by the Natural Environment
Research Council.
References
Amann R. Springer N, Ludwig W, Gortz HD, Schleifer KH (1991)
Identification in situ and phylogeny of uncultured bacterial
endosymbionts. Nuture, 351,1614
Brooks DR, McLennan DA (1992) Phylogeny, Ecology and Behavior.
University of Chicago Press.
Cavalier-SmithT (1987)Eukaryotes with no mitochondria. Nature,
326,332-3.
Corliss JO (1979)The Ciliated Protozoa. Pergamon Press. Oxford
New York.
Delong EF, Wickham GS, Pace NR (1989) Phylogenetic stains:
Ribosomal RNA-based probes for the identification of single
cells. Science, 243,1360-1363.
Dore J, Stahl DA (1991) Phylogeny of anaerobic rumen chytrid-
iomycetes inferred from small subunit ribosomal RNA
sequence comparisons. Canadian Iournal Of Botany, 69,
1964-1971.
Elwood HJ, Olsen GJ, Sogin ML (1985) The small-subunit r i b
ma1 RNA gene sequences from the hypohichous ciliates
Oxytncha nma and Stylonychia pustulafa. Molecular Biology and
Evolution,2,399-410.
Embley TM (1991) The linear PCR reaction: a simple and robust
method for sequencing amplified rRNA genes. Letters In
Applud Microbiology, 13,171-4.
Embley TM, Finlay BJ (1994a) Systematic and morphological
diversity of endosymbiotic methanogens in anaerobic aliates.
Antonie van Irtuwenhoek, 64,261-271.
Embley TM. Finlay BJ (1994b) The use of small subunit rRNA
sequences to unravel the relationships between anaerobic cili-
ates and their methanogen endosymbionts. Microbiology, 140,
225-35.
Embley TM. Finlay BJ, Thomas RH,Dyal PL (1992) The use of
rRNA sequences and fluorescent probes to investigate the
phylogenetic positions of the anaerobic ciliate Metogus p a k -
fornris and its archaeobacterial endosymbiont. journal of
General Micmbwlogy, 138,1479-87.
Fenchel T (1987) Ecology of Protozoa. Science Technical.Madison.
Fenchel T (1993) There are more small than large spedeS? Oikos,
68.375-378.
Fenchel T,Finlay BJ (1989)Kentrophoros:A mouthless ciliate with
a symbiotic kitchen garden Opheliu, 30,75-94.
Fenchel T, Finlay BJ (1991)The biology of free-living anaerobic cil-
iates. Eumpeun ]ournd of Protistology, 26,201-215.
Gray MW (1992) The Endosymbiont Hypothesis Revisited.
International Review of Cytology, 141,23%357.
Hirt Rp, Dyal PL, Wilkinson M, Finlay Bj, McL Roberts D, Embley
 rDNA S E Q U E N C E S F R O M SINGLE P R O T O Z O A
TM (1995) Phylogenetic relationships among karyorelictids
and heterotrichs inferred from small subunit rRNA sequences:
resolution at the base of the aliate tree. Molecular Phylogenetics
and Evolution, 4, in press.
Leipe DD, Bemhard D, W e g e l M, Sogin ML (1994) Evolution of
16Slike ribosomal RNA genes in the ciliophoran taxa
Litostomatea and Phyllopharyngea. European Journal of
Profistology, 30, 354-361.
Li H, Cui X, Amheim N (1990) Direct electrophoretic detection of
the allelic state of single DNA molecules in human sperm by
using the polymerase chain reaction. Proceedings of the National
Academy of Sciences USA, 87,4580-4584.
Li H, Cui X, Amheim N (1991) Analysis of DNA sequence varia-
tion in single cells. Methods: A Companion to Methods in
Enzymoiogy, 2,49-59.
Medlin L, Elwood HJ, Stickel S, Sogin ML (1988) The characteri-
zation of enzymatically amplified eukaryotic 16Slike rRNA-
coding regions. Gene, Tl, 491-9.
Mullis K, Faloona F, Scharf S, Saiki R, Horn G, Erlich H (1986)
Specific enzymatic amplification of DNA in viho: the poly-
merase chain reaction. Quantitutiw Biology, LI, 263-273.
Neefs J-M, Peer W d , Rijik PD, Chapelle S, Wachter RD (1993)
Compilation of small ribosomal subunit RNA structures.
Nucleic Acids Research, 21,302!53049.
Prescott D (1994)The DNA of ciliated protozoa. Microbiological
Rm'ews,58,233-267.
503
Roberts DM, Causton H (1988)Silver nitrate impregnation of cil-
iated protozoa. Arclriwsfur Protistenkunde, 135,299-318.
Sleigh M (1991)P ~ ~ Q ZandO Qother protists. Cambridge University
Press.Cambridge.
Sogin ML (1991) Early evolution and the origin of eukaryotes.
Curent Opinion in Genetics and Deuelopment, 1,457463.
Sogin ML, Gunderson JH, Elwood HJ, Alonso RA, Peattie DA
(1989) Phylogenetic meaning of the kingdom concepc: An
unusual ribosomal RNA from Giardia lamblia. Science, 243,
75-77.
Stahl DA, Amann R (1991) Development and application of
nudeic acid probes. In: Nucleic acid techniques in bacterial sys-
tematics (eds Stackebrandt E, Goodfellow M), pp. 205-248.
Chichester: John Wiley & Sons.
Wanntorp H-E, Brooks DR, Nilsson T, Nylin S, Ronquist F, Steams
SC, Wedell N (1990) Phylogenetic approaches in ecology.
Oikos, 57, 11!9-131.
This paper is the result of work aiming to develop simple and
robust methods for studying microbial eucaryotes. Martin
Embley and Pat Dyal are working in collaboration with Dr Bland
Finlay (Institute of Freshwater Ecology, England) to study ciliate
phylogeny and ecology, with particular emphasis on the evolu-
tion of anaerobic ciliates. Dave Roberts and Sue Hope study cili-
ate morphology and ultrastructure.