+ All Categories
Home > Documents > Evolution of Anaerobic Ciliates from the Gastrointestinal Tract

Evolution of Anaerobic Ciliates from the Gastrointestinal Tract

Date post: 09-Feb-2022
Category:
Upload: others
View: 3 times
Download: 0 times
Share this document with a friend
12
1195 Mol. Biol. Evol. 15(9):1195–1206. 1998 q 1998 by the Society for Molecular Biology and Evolution. ISSN: 0737-4038 Evolution of Anaerobic Ciliates from the Gastrointestinal Tract: Phylogenetic Analysis of the Ribosomal Repeat from Nyctotherus ovalis and its Relatives Angela H. A. M. van Hoek, Theo A. van Alen, Vera S. I. Sprakel, Johannes H. P. Hackstein, and Godfried D. Vogels Department of Microbiology and Evolutionary Biology, Faculty of Science, University of Nijmegen, The Netherlands The 18S and 5.8S rDNA genes and the internal transcribed spacers ITS-1 and ITS-2 of ciliates living in the hindgut of frogs, millipedes, and cockroaches were analyzed in order to study the evolution of intestinal protists. All ciliates studied here belong to the genus Nyctotherus. Phylogenetic analysis revealed that these ciliates form a monophyletic group that includes the distantly related anaerobic free-living heterotrichous ciliates Metopus palaeformis and Me- topus contortus. The intestinal ciliates from the different vertebrate and invertebrate hosts are clearly divergent at the level of their rDNA repeats. This argues for the antiquity of the associations and a predominantly vertical transmission. This mode of transmission seems to be controlled primarily by the behavior of the host. The different degrees of divergence between ciliates living in different strains of one and the same cockroach species most likely reflect the different geographical origins of the hosts. In addition, host switches must have occured during the evolution of cockroaches, since identical ciliates were found only in distantly related hosts. These phenomena prevent the reconstruction of potential cospeciation events. Introduction About 7,500 species of ciliated protozoa have al- ready been described, and extrapolations suggest that there might be more than 10,000 species of free-living ciliates thriving in the most divergent ecological niches (Grell 1973; Hawksworth and Kalin-Arroyo 1995; Fin- lay et al. 1996; Hausmann and Hu ¨lsmann 1996). Ciliates even live in the intestinal tracts of animals: they repre- sent a substantial fraction of the complex microbiota that populate the rumina and the hindguts of many herbiv- orous mammals (Hungate 1966; Williams 1986; Wil- liams and Coleman 1991). Potentially, the digestive tracts of hundreds of herbivorous mammalian species provide a wealth of anaerobic ecological niches. Such environments should favor the coevolution of the intes- tinal microbiota with their hosts, and one might expect that the intestinal ciliates which have been described to date represent only a minor fraction of a hitherto un- known number of anaerobic species (Hackstein 1997). Anaerobic ciliated protozoa are also found in the guts of many arthropods (Lucas 1927; Kudo 1931; Wil- lis and Roth 1960; Hoyte 1961a; Breznak 1982; Cruden and Markovetz 1987; Gijzen et al. 1991). A recent sur- vey of 20 higher taxa of arthropods revealed that per- manent associations between ciliates and their arthropod hosts were nonrandom: only millipedes and cockroaches were found to host substantial numbers of ciliated pro- tozoa (Hackstein and Stumm 1994). Systematic screens Abbreviations: ARDRA, amplified ribosomal DNA restriction analysis; ITS, internal transcribed spacer; SEM, scanning electron mi- croscope. Key words: Nyctotherus ovalis, anaerobic intestinal ciliates, rDNA, internal transcribed spacer, phylogenetic reconstruction, cock- roaches, evolution. Address for correspondence and reprints: Johannes H. P. Hack- stein, Department of Microbiology and Evolutionary Biology, Toer- nooiveld 1, NL-6525 ED Nijmegen, the Netherlands. E-mail: [email protected]. among 40 species of cockroaches belonging to various genera confirmed that at least 17 cockroach species host intestinal ciliates (Hackstein 1997). The majority of these ciliates seem to belong to the genus Nyctotherus (Leidy 1849, 1853). Ciliates belonging to this taxon are also known to occur in the intestinal tract of frogs and reptiles (Bhatia and Gulati 1927; Lucas 1927; Wichter- man 1937; McKean 1972). However, the taxonomic re- lationships between the intestinal ciliates of the various hosts and their phylogenetic positions with respect to their potential free-living relatives and ancestors has re- mained largely unclear until now (Schlegel 1991; Em- bley et al. 1995; Hirt et al. 1995; Hammerschmidt et al. 1996). Here, we describe the phylogenetic analysis of the 5.8S and 18S rDNA genes and the internal transcribed spacers (ITS-1 and ITS-2) of Nyctotherus ovalis ciliates from the hindgut of the cockroaches Periplaneta amer- icana and Blaberus sp. In addition, we show that these ciliates are closely related to Nyctotherus cordiformis and similar ciliates from the intestinal tracts of frogs and millipedes. They form a monophyletic clade that also includes the anaerobic free-living heterotrichous ciliates Metopus palaeformis and Metopus contortus. Lastly, we present evidence for substantial DNA sequence diver- gence not only between ciliates living in different host species, but also between ciliates that live in different, isolated strains of one and the same cockroach species. We show that horizontal transfers between host species did occur, and we discuss the significance of the DNA sequence divergence for the antiquity of the associations between intestinal ciliates and their hosts. Materials and Methods Sources The cockroach strains Periplaneta americana var. Amsterdam, P. americana var. Dar es Salaam, and P. americana var. Nijmegen were isolated from free-living populations. The cockroach strains P. americana var. Downloaded from https://academic.oup.com/mbe/article/15/9/1195/1412734 by guest on 07 December 2021
Transcript

1195

Mol. Biol. Evol. 15(9):1195–1206. 1998q 1998 by the Society for Molecular Biology and Evolution. ISSN: 0737-4038

Evolution of Anaerobic Ciliates from the Gastrointestinal Tract:Phylogenetic Analysis of the Ribosomal Repeat fromNyctotherus ovalis and its Relatives

Angela H. A. M. van Hoek, Theo A. van Alen, Vera S. I. Sprakel,Johannes H. P. Hackstein, and Godfried D. VogelsDepartment of Microbiology and Evolutionary Biology, Faculty of Science, University of Nijmegen, The Netherlands

The 18S and 5.8S rDNA genes and the internal transcribed spacers ITS-1 and ITS-2 of ciliates living in the hindgutof frogs, millipedes, and cockroaches were analyzed in order to study the evolution of intestinal protists. All ciliatesstudied here belong to the genus Nyctotherus. Phylogenetic analysis revealed that these ciliates form a monophyleticgroup that includes the distantly related anaerobic free-living heterotrichous ciliates Metopus palaeformis and Me-topus contortus. The intestinal ciliates from the different vertebrate and invertebrate hosts are clearly divergent atthe level of their rDNA repeats. This argues for the antiquity of the associations and a predominantly verticaltransmission. This mode of transmission seems to be controlled primarily by the behavior of the host. The differentdegrees of divergence between ciliates living in different strains of one and the same cockroach species most likelyreflect the different geographical origins of the hosts. In addition, host switches must have occured during theevolution of cockroaches, since identical ciliates were found only in distantly related hosts. These phenomenaprevent the reconstruction of potential cospeciation events.

Introduction

About 7,500 species of ciliated protozoa have al-ready been described, and extrapolations suggest thatthere might be more than 10,000 species of free-livingciliates thriving in the most divergent ecological niches(Grell 1973; Hawksworth and Kalin-Arroyo 1995; Fin-lay et al. 1996; Hausmann and Hulsmann 1996). Ciliateseven live in the intestinal tracts of animals: they repre-sent a substantial fraction of the complex microbiota thatpopulate the rumina and the hindguts of many herbiv-orous mammals (Hungate 1966; Williams 1986; Wil-liams and Coleman 1991). Potentially, the digestivetracts of hundreds of herbivorous mammalian speciesprovide a wealth of anaerobic ecological niches. Suchenvironments should favor the coevolution of the intes-tinal microbiota with their hosts, and one might expectthat the intestinal ciliates which have been described todate represent only a minor fraction of a hitherto un-known number of anaerobic species (Hackstein 1997).

Anaerobic ciliated protozoa are also found in theguts of many arthropods (Lucas 1927; Kudo 1931; Wil-lis and Roth 1960; Hoyte 1961a; Breznak 1982; Crudenand Markovetz 1987; Gijzen et al. 1991). A recent sur-vey of 20 higher taxa of arthropods revealed that per-manent associations between ciliates and their arthropodhosts were nonrandom: only millipedes and cockroacheswere found to host substantial numbers of ciliated pro-tozoa (Hackstein and Stumm 1994). Systematic screens

Abbreviations: ARDRA, amplified ribosomal DNA restrictionanalysis; ITS, internal transcribed spacer; SEM, scanning electron mi-croscope.

Key words: Nyctotherus ovalis, anaerobic intestinal ciliates,rDNA, internal transcribed spacer, phylogenetic reconstruction, cock-roaches, evolution.

Address for correspondence and reprints: Johannes H. P. Hack-stein, Department of Microbiology and Evolutionary Biology, Toer-nooiveld 1, NL-6525 ED Nijmegen, the Netherlands. E-mail:[email protected].

among 40 species of cockroaches belonging to variousgenera confirmed that at least 17 cockroach species hostintestinal ciliates (Hackstein 1997). The majority ofthese ciliates seem to belong to the genus Nyctotherus(Leidy 1849, 1853). Ciliates belonging to this taxon arealso known to occur in the intestinal tract of frogs andreptiles (Bhatia and Gulati 1927; Lucas 1927; Wichter-man 1937; McKean 1972). However, the taxonomic re-lationships between the intestinal ciliates of the varioushosts and their phylogenetic positions with respect totheir potential free-living relatives and ancestors has re-mained largely unclear until now (Schlegel 1991; Em-bley et al. 1995; Hirt et al. 1995; Hammerschmidt et al.1996).

Here, we describe the phylogenetic analysis of the5.8S and 18S rDNA genes and the internal transcribedspacers (ITS-1 and ITS-2) of Nyctotherus ovalis ciliatesfrom the hindgut of the cockroaches Periplaneta amer-icana and Blaberus sp. In addition, we show that theseciliates are closely related to Nyctotherus cordiformisand similar ciliates from the intestinal tracts of frogs andmillipedes. They form a monophyletic clade that alsoincludes the anaerobic free-living heterotrichous ciliatesMetopus palaeformis and Metopus contortus. Lastly, wepresent evidence for substantial DNA sequence diver-gence not only between ciliates living in different hostspecies, but also between ciliates that live in different,isolated strains of one and the same cockroach species.We show that horizontal transfers between host speciesdid occur, and we discuss the significance of the DNAsequence divergence for the antiquity of the associationsbetween intestinal ciliates and their hosts.

Materials and MethodsSources

The cockroach strains Periplaneta americana var.Amsterdam, P. americana var. Dar es Salaam, and P.americana var. Nijmegen were isolated from free-livingpopulations. The cockroach strains P. americana var.

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021

1196 van Hoek et al.

Table 1Primers Used for PCR and Sequencing

Primer Sequence (59 to 39) Positions on Eukaryotic rDNA

Euk-forward . . . . . . . . . .Euk300F . . . . . . . . . . . . .Euk300R . . . . . . . . . . . . .Euk528F . . . . . . . . . . . . .

AATCTGGTTGATCCTGCCAGTAGGGTTCGATTCCGGAGCTCCGGAATCGAACCCTCGGTAATTCCAGCTCC

18S18S18S18S

1–21360–376376–360568–583

Euk528R . . . . . . . . . . . .Euk690F . . . . . . . . . . . .Euk690R . . . . . . . . . . . .Euk1055F . . . . . . . . . . .Euk1055R . . . . . . . . . . .

GGAGCTGGAATTACCGAGAGGTGAAATTCTAGAATTTCACCTCTGGGTGGTGCATGGCCGCGGCCATGCACCACC

18S18S18S18S18S

583–568875–888888–874

1238–12521252–1238

Euk1200F . . . . . . . . . . .Euk1200R . . . . . . . . . . .Euk-reverse . . . . . . . . . .M13-forward . . . . . . . . .M13-reverse . . . . . . . . .

CAGGTCTGTGATGCCCGGGCATCACAGACCTGTGATCCTTCTGCAGGTTCACCTACGTAAAACGACGGCCAGTCAGGAAACAGCTATGAC

18S18S18S

1396–14111411–13961737–1726

18SF . . . . . . . . . . . . . . . .5.8SF . . . . . . . . . . . . . . .5.8SR . . . . . . . . . . . . . . .28SR . . . . . . . . . . . . . . . .

CACACCGCCCGTCGCTACTACCGATTGGCGAGTCATCAGATCTTTGATCAAAGATCTGATGACTCGCAATATGCTTAAGTTCAGCGG

18S5.8S5.8S

28S

1605–162985–104

104–8544–15

NOTE.—Positions are based on the rDNA genes of Tetrahymena thermophila (X54512).

FIG. 1.—Scanning electron micrograph (SEM) of Nyctotherusovalis in the hindgut of Periplaneta americana var. Amsterdam. Bar5 20 mm.

Bayer, Blaberus sp. var. Nijmegen, Blaberus sp. var.Amsterdam, and Blaberus sp. var. Dusseldorf originatedfrom laboratory cultures. Blaberus sp. var. Nijmegen isidentical to Blaberus sp. var. Amsterdam, but has beenkept for 6 years in Nijmegen as a separate line. Thecockroaches were cultured at a temperature of 218C.They were fed apple, potato, commercial pelleted foodfor rabbits, and water ad libitum.

The julid millipede, listed as ‘‘Unidentified A’’ inHackstein and Stumm (1994) was cultured in the labo-ratory at 218C and fed potato.

Frogs and tadpoles were obtained from Dr. H. Strij-bosch, Department of Environmental Biology, Univer-sity of Nijmegen.

Isolation of Ciliates

Cockroaches were selected for high methane pro-duction (.90 nmol CH4/g cockroach/h, as describedearlier [Hackstein and Stumm 1994]), since such cock-roaches were likely to host large ciliate populations.Cockroaches were anesthetized with CO2 and dissectedunder a dissecting microscope. The hindgut was re-moved and placed in an electromigration device. Afterapplication of an electric field (20 V), the ciliates movedtoward the anode, where they were picked up with aPasteur pipette (Wagener, Stumm, and Vogels 1986).They were centrifuged for 5 min at 2,000 rpm andwashed three times in sterile electromigration buffer (2.7mM K2HPO4, 1.8 mM KH2PO4, 21.5 mM KCl, 20 mMNaCl, 6.1 mM MgSO4·7H2O, 0.5 mM L-cysteine, 0.5mM CaCl2·2H2O, 0.5 mM titanium citrate [Zehnder andWuhrmann 1976], and 1 mM NaHCO3 [pH 7.5]). A so-lution of 100 ml 5% Chelex-100 (Walsh, Metzger, andHiguchi 1991) was added to the pellet. The sample wasfrozen in liquid nitrogen and stored at 2208C.

Alternatively, single ciliate cells were picked upwith a drawn out Pasteur pipette and washed and cen-trifuged three times in a fresh droplet of sterile elec-tromigration buffer. Individual cells were transferredinto Eppendorf tubes. After addition of 50 ml 5% Chel-ex-100 the cells were frozen and stored at 2208C.

The galvanotactic behavior of the ciliates was alsoused to isolate the organisms from the intestinal tract ofthe julid millipede (‘‘Unidentified A,’’ described inHackstein and Stumm 1994), from the hindguts of tad-poles of Rana ridibunda, and from young adults of Ranatemporaria.

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021

Evolution of the Intestinal Ciliates 1197

FIG. 2.—Histogram of the size distribution of Nyctotherus ovalis from the different cockroach hosts. Samples of 250 ciliates from a singlehindgut were analyzed. The few individuals of a size above 160 mm most likely represent zygotes.

FIG. 3.—Riboprints of the 18S rDNA genes of single ciliates of two different cockroach hosts after digestion with the enzyme DdeI. Lanes1–6: Nyctotherus ovalis from Periplaneta americana var. Bayer. Lane 7: Biozym low ladder (1 kb–100 bp). Lanes 8–9: N. ovalis from P.americana var. Nijmegen.

Length Measurements

All ciliates from the hindgut of one specimen ofeach of the seven different cockroach strains were iso-lated by electromigration and transferred to a Sedgewickrafter (Graticules LTD, Tonbridge Kent, England).Swimming, untreated ciliates were photographed at 403magnification with a Leitz photomicroscope. The sizesof 250 ciliates from each of the seven strains were mea-sured on enlarged photographic prints.

Scanning Electron Microscopy (SEM)

Scanning electron microscopy of the hindgut of P.americana var. Amsterdam was performed as describedin Cazemier et al. (1997).

DNA Isolation, PCR Amplification, and SequencingFour or eight microliters of proteinase K (10 mg/

ml) was added to the frozen single ciliates and pellets,respectively. Pellets were homogenized with a sealedPasteur pipette. Single-cell samples were vortexed forhalf a minute. After incubation for 3-4 h at 568C, thehomogenates were heated at 958C for 10 min in orderto inactivate the proteinase K, chilled on ice, and cen-trifuged at 13,000 rpm for 10 min. Aliquots of the su-pernatants were used for the amplification of the 18Sand 5.8S rDNA genes and the internal transcribed spac-ers (ITS-1 and ITS-2) of the ciliates.

PCR amplification of the total 18S rDNA genes ofthe ciliates from cockroaches was performed using the

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021

1198 van Hoek et al.

FIG. 4.—Neighbor-joining tree (Saitou and Nei 1987) inferred from approximately 1,100 positions of the small subunit rDNA sequencesusing the Jukes and Cantor (1969) algorithm. The distance data were bootstrap resampled 100 times (Felsenstein 1985). Only bootstrap valuesabove 90 are displayed. The stramophile Ochromonas danica was used as an outgroup.

primers euk-forward and euk-reverse in 25-ml reactionvolumes (table 1). The 5.8S rDNA and the ITS regionswere amplified with the primers 18SF and 28SR in 25-ml reaction volumes. PCR amplification of the 18SrDNA genes of the ciliates from a millepede and theanurans was performed using the euk300F and euk-re-verse primers in 25-ml reaction volumes. DNA concen-trations were estimated after electrophoresis on a 6%PAGE gel and staining with silver nitrate (Bassam, Cae-tano-Anolles, and Gresshoff 1991).

The cloning of the 18S rDNA gene of N. ovalisfrom P. americana var. Amsterdam was performed witha blunt-end ligation of the gene using the SureCloneLigation Kit (Pharmacia Biotech) and Escherichia coliJM 109 competent cells. The FlexiPrep Kit (Pharmacia

Biotech) was used to harvest and purify the plasmidDNA from recombinant clones for sequencing. Twoclones with the 18S rDNA gene were sequenced withfive forward and five reverse eukaryotic SSU rDNAprimers (table 1, cf. Elwood, Olsen, and Sogin 1985)and the M13-forward and -reverse primers.

PCR products of the 18S rDNA, ITS-1, 5.8SrDNA, and ITS-2 of one individual ciliate from each ofthe various hosts were sequenced directly. The PCRproducts were separated on a 1% agarose gel, and thedesired band was electroeluted. DNA was precipitatedwith 0.1 volumes 7.5 M ammonium acetate (pH 5.2)and 2.5 volumes 96% ethanol for 1 h at 2208C. Finally,the PCR fragments were purified with the FlexiPrep Kit(Pharmacia Biotech) and used for sequencing with the

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021

Evolution of the Intestinal Ciliates 1199

FIG. 5.—Riboprints of the 18S rDNA genes of single ciliates of the seven different cockroach hosts after digestion with the enzyme DdeI.Lane 1: Undigested 18S rDNA PCR fragment. Lane 2: Biozym low ladder (1 kb–100 bp). Lane 3: Nyctotherus ovalis from Periplaneta americanavar. Amsterdam. Lane 4: N. ovalis from P. americana var. Bayer. Lane 5: N. ovalis from Blaberus sp. var. Amsterdam. Lane 6: N. ovalis sp.from Blaberus sp. var. Nijmegen. Lane 7: N. ovalis from Blaberus sp. var. Dusseldorf. Lane 8: N. ovalis from P. americana var. Dar es Salaam.Lane 9: N. ovalis from P. americana var. Nijmegen.

Table 2Restriction Fragment Analysis (ARDRA) of the 18S rDNA Genes of the Intestinal Ciliates of the DifferentCockroaches with the Restriction Enzymes DdeI, HaeIII, HinfI, and SspI

Species 2 3 4 5 6 7 Total

Total number of restriction fragments and shared fragments between the different ciliate ribotypes1. Nyctotherus ovalis from Periplaneta americana var. Amster-

dam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2. N. ovalis from P. americana var. Bayer . . . . . . . . . . . . . . . . . . . . . . .3. N. ovalis from P. americana var. Dar es Salaam . . . . . . . . . . . . . . .4. N. ovalis from P. americana var. Nijmegen . . . . . . . . . . . . . . . . . . . .5. N. ovalis from Blaberus sp. var. Amsterdam . . . . . . . . . . . . . . . . . . .6. N. ovalis from Blaberus sp. var. Dusseldorf . . . . . . . . . . . . . . . . . . .7. N. ovalis from Blaberus sp. var. Nijmegen . . . . . . . . . . . . . . . . . . . .

9 1013

91010

9191310

159989

919131019

9

17191818191719

Distance matrix1. N. ovalis from P. americana var. Amsterdam . . . . . . . . . . . . . . . . . .2. N. ovalis from P. americana var. Bayer . . . . . . . . . . . . . . . . . . . . . . .3. N. ovalis from P. americana var. Dar es Salaam . . . . . . . . . . . . . . .4. N. ovalis from P. americana var. Nijmegen . . . . . . . . . . . . . . . . . . . .5. N. ovalis from Blaberus sp. var. Amsterdam . . . . . . . . . . . . . . . . . . .6. N. ovalis from Blaberus sp. var. Dusseldorf . . . . . . . . . . . . . . . . . . .7. N. ovalis from Blaberus sp. var. Nijmegen . . . . . . . . . . . . . . . . . . . .

0.135 0.1120.071

0.1350.1230.155

0.13900.0710.123

0.0260.1390.1350.1550.139

0.13900.0710.12300.139

NOTE.—Distance matrix was calculated according to Swofford and Olsen (1990).

primers displayed in table 1. The rDNA genes and theinternal transcribed spacers were sequenced in both di-rections using an ABI Prism 310 Automated DNA se-quencer (Perkin Elmer) and an ABI PRISM Dye Ter-minator Cycle Sequencing Ready Reaction Kit (PerkinElmer).

Amplified Ribosomal DNA Restriction Analysis(‘‘Ribotyping’’)

Amplified 18S rDNA of the various ciliates fromthe seven cockroach strains was subjected to amplifiedribosomal DNA restriction analysis (ARDRA) in orderto confirm the results obtained by DNA sequence anal-ysis. From each of the seven cockroach strains, 10 to20 single ciliates (derived from at least five differentcockroach hindguts) were analyzed. In addition, DNAextracted from ciliate pellets isolated from the four P.

americana strains was also subjected to ARDRA. Theamplified 18S rDNA was digested with 5 U of a restric-tion enzyme and the corresponding enzyme buffer in atotal volume of 15 ml. The following restriction enzymeswere used: AccI, AluI, ApaI, AsnI, BamHI, BglII, CfoI,DdeI, DraI, EcoRI, HaeIII, HindIII, HinfI, MspI, MvaI,NciI, NcoI, NotI, PstI, PvuI, RsaI, Sau3AI, ScrFI, SmaI,SspI, SwaI, and TaqI. Digestions were performed over-night at the recommended incubation temperature forthe endonuclease. The digested products were heated for10 min at 958C, chilled on ice, and separated on a 6%PAGE gel. The gels were stained with silver nitrate.

The informative patterns, i.e., the number of bandsand the number of shared fragments, obtained with therestriction enzymes DdeI, HaeIII, HinfI, and SspI wereused to calculate a d-matrix (Swofford and Olsen 1990).

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021

1200 van Hoek et al.

Table 3Size and Nucleotide Composition Data for the ITS-1, 5.8S, and ITS-2 Regions in Nyctotherus ovalis from theDifferent Cockroach Hosts

ITS-1

Length(bp) %A %T %G %C

5.8S

Length(bp) %A %T %G %C

ITS-2

Length(bp) %A %T %G %C

Nyctotherus ovalis from Periplaneta americanavar. Amsterdam . . . . . . . . . .var. Bayer . . . . . . . . . . . . . .var. Dar es Salaam . . . . . . .var. Nijmegen . . . . . . . . . . .

93939393

36.637.637.635.5

26.929.028.029.0

9.79.79.79.7

26.923.724.725.8

158158158158

27.827.227.227.8

24.124.724.724.1

23.424.124.124.1

24.724.124.124.1

136137137137

32.435.035.034.3

21.322.622.623.4

21.320.420.420.4

25.021.921.921.9

N. ovalis from Blaberus sp.var. Amsterdam . . . . . . . . . .var. Dusseldorf . . . . . . . . . .var. Nijmegen . . . . . . . . . . .

939393

37.637.637.6

29.025.829.0

9.79.79.7

23.726.923.7

158158158

27.827.827.8

24.124.124.1

24.123.424.1

24.124.724.1

137137137

35.032.835.0

22.621.922.6

20.421.220.4

21.924.121.9

Cryptocaryon irritansa . . . . .Ichthyophthirius multifiliisa . .Tetrahymena thermophilab. . .

133146134

43.638.435.8

39.834.933.6

5.39.6

13.4

11.317.117.2 153 30.1 24.2 27.5 18.3 178 30.3 25.8 20.8 23.0

a Parasitic ciliates, from Diggles and Adlard (1997).b Free-living ciliate, from Engberg and Nielsen (1990).

This matrix was used to construct a tree with the neigh-bor-joining method (Saitou and Nei 1987).

Phylogenetic Analysis

18S rDNA sequences from 41 ciliates, 2 apicom-plexa, 2 dinoflagellates, and 1 stramenopile were ob-tained from the GenBank and EMBL databases. Thesesequences, together with the 18S rDNA sequences of N.ovalis from P. americana var. Amsterdam, N. cordifor-mis from R. ridibunda and R. temporaria, and N. veloxfrom the julid millipede ‘‘Unidentified A’’ were alignedusing the PileUp program of the Wisconsin package,version 9.1. Hypervariable regions were removed fromthe alignment. For distance and parsimony analyses, thealignment was reduced to approximately 1,100 posi-tions.

EDNADIST (modified PHYLIP version 3.572c ofDNADIST by Felsenstein [1993]) was used to calculatethe sequence similarity and evolutionary distances usingthe Jukes and Cantor (1969) and Kimura (1980) nucle-otide substitution models. A distance matrix tree wasconstructed using the neighbor-joining method (Saitouand Nei 1987). The distance data were bootstrap resam-pled 100 times (Felsenstein 1985).

EDNAPARS (modified PHYLIP version 3.572c ofDNADIST by Felsenstein [1993]) was used to performa parsimony analysis.

ResultsCiliates from the Hindguts of Cockroaches

In the hindguts of full-grown cockroaches, hun-dreds to thousands of Nyctotherus ciliates were found(fig. 1). A conspicuous size polymorphism was charac-teristic for these ciliates. Ciliates of largely varying sizeswere present in a single hindgut. In addition, ciliatesfrom the hindguts of the different Blaberus spp. ap-peared to be smaller than those from the P. americanastrains. However, a biometric analysis showed that cil-iate sizes were normally distributed (fig. 2). Since the

ciliates were also morphologically similar, it is likelythat all ciliates in the hindgut of a single cockroach be-long to one population. The modes of the length distri-butions of the ciliates were host-specific, and they con-firm our observation that the smallest ciliates live in thelargest host (Blaberus sp.; fig. 2). Only rarely (K1%)were double-sized ciliates observed in all hosts; thesemorphs had been interpreted as macrogamonts or zy-gotes by McKean (1972).

Restriction analysis with 28 different enzymes ofthe amplified 18S rDNA genes was used to confirm thatthere is only one ribotype of ciliates present in the hind-gut of an individual cockroach. The restriction patternsobtained from 10–20 individual ciliates isolated from atleast 4 different cockroaches from each of the 7 hoststrains were analyzed. Consistently, the restriction pat-terns revealed that only one ciliate ribotype was presentin all of the cockroaches belonging to one and the samestrain (fig. 3).

Phylogenetic Analysis of the Ribosomal SmallSubunit rDNA

The complete 18S rDNA gene of N. ovalis fromthe hindgut of the cockroach P. americana var. Amster-dam has a length of 1,707 nucleotides (accession num-ber AJ222678). There were no indications of structuralpeculiarities or fragmentation of the coding region. TheG/C content of the coding region of the small subunitis 43.8%. Also, partial sequences (61,340 nt) of the 18SrDNA of Nyctotherus ciliates from a millipede and twofrog species were obtained. Phylogenetic analysis of thesmall subunit rDNA genes was performed by distancematrix analysis using the Jukes and Cantor (1969) al-gorithm for 45 ciliate, 2 apicomplexan, and 2 dinofla-gellate species, with an alga as outgroup. The resultsstrongly suggested that the intestinal Nyctotherus ciliatesfrom the various hosts form a monophyletic group to-gether with the free-living anaerobic heterotrichous cil-iates M. palaeformis and M. contortus. High bootstrap

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021

Evolution of the Intestinal Ciliates 1201

values for the Nyctotherus-Metopus cluster strongly sup-port the close relationship and confirm the postulatedbranching order (fig. 4).

DNA Sequence Divergence Among Ciliates fromDifferent Cockroaches

Restriction analysis of the 18S rDNA of Nyctother-us ciliates revealed that 4 (DdeI, HaeIII, HinfI, and SspI)of the 28 restriction enzymes generated different restric-tion patterns for 5 of the 7 ciliate samples from thedifferent cockroach strains (fig. 5). Only ciliates fromthe cockroach species Blaberus sp. var. Nijmegen, Bla-berus sp. var. Amsterdam, and P. americana var. Bayerexhibited identical restriction patterns (table 2). The in-formative patterns generated by these four enzymeswere used to construct a distance matrix tree (see be-low). It became evident that the five different ciliatesamples were closely related but occupied clearly dif-ferent positions in the phenogram. This divergence wasconfirmed by DNA sequence analysis of the 5.8S rDNAgenes and the adjacent internal transcribed spacers (ITS-1 and ITS-2). The ITS-1 and ITS-2 regions are remark-ably short (93 nt and 136–137 nt, respectively). The G/C contents of these regions are 34.6 6 1.4% and 43.36 1.7%, respectively (table 3). Alignment of the ITSregions and the 5.8S rDNA gene revealed no obviousgaps or length differences within the Nyctotherus group(fig. 6). The pairwise comparison of the 18S, 5.8S andITS sequences from ciliates living in the various cock-roach strains revealed a divergence of up to 5% (table5). Phylogenetic analysis of the 5.8S rDNA genes andthe ITS-1 and ITS-2 sequences from the ribosomal re-peats allowed the construction of a distance matrix tree(fig. 7) that is similar to the tree obtained by ARDRAof the amplified 18S rDNA genes (not shown).

DiscussionPhylogenetic Aspects

Ciliates represent a large group of morphologicallyvery diverse protists that are characterized by a nucleardimorphism, sexual reproduction in form of conjugation,and a complex infraciliature (Small and Lynn 1981;Foissner and Foissner 1988; Finlay et al. 1996; Haus-mann and Hulsmann 1996). DNA sequencing data haveconfirmed that ciliates form a monophyletic taxon whichradiated early after their common ancestor separatedfrom the other eukaryotes (Sogin and Elwood 1986;Schlegel 1991; Embley et al. 1995; Hirt et al. 1995;Hammerschmidt et al. 1996).

The phylogenetic analysis of 18S rDNA sequenc-ing data of N. ovalis from the hindgut of the cockroachP. americana var. Amsterdam, N. velox from the hindgutof the julid millipede ‘‘Unidentified A,’’ and N. cordi-formis from the intestinal tract of two anurans placesthem with high bootstrap values in a clade with the dis-tantly related free-living anaerobic heterotrichs M. pa-laeformis and M. contortus (fig. 4). This position is notsubject to change when different distance and tree-build-ing algorithms (i.e., Jukes and Cantor 1969; Kimura1980; parsimony) are used. Therefore, the phylogenetic

analysis of the 18S rDNA genes of the various Nycto-therus species confirms that anaerobic heterotrichousciliates cluster clearly distinctly from the aerobic het-erotrichs (De Puytorac, Grain, and Legendre 1994; Hirtet al. 1995).

Sequence Divergence

The DNA sequence divergence between the ribo-somal genes of the different Nyctotherus species is sub-stantial (e.g., approximately 5% between N. ovalis andN. cordiformis; table 4). However, a divergence up to5% can also be observed between N. ovalis from thevarious cockroach strains (table 5). The divergence withrespect to their closest free-living relatives, i.e., Metopusspecies, exceeds 10%. Unexpectedly, the divergence forthe (noncoding) ITS regions is not significantly higherthan that for the coding parts of the 18S rDNA genes.However, the ITS-1 and ITS-2 of the ribosomal repeatsare short (93 and 136–137 nt, respectively; see table 3);they are the smallest among eukaryotes (cf. Odorico andMiller 1997). Dot-plot analysis did not provide any ev-idence for the presence of repeated DNA sequences (notshown). It seems reasonable to assume that only thoseparts of the ITS that are under functional constraintshave been retained in all the ciliates of the Nyctotheruscluster.

Mode of Transmission

Phylogenetic analysis of the 18S rDNA genes ofthe different Nyctotherus species from frogs, millipedes,and cockroaches reveals that these ciliates representwell-separated evolutionary lines, although the branch-ing order of the millipede and the frog symbionts cannotbe resolved (fig. 4). The remarkable divergence is ratherunexpected since the gastrointestinal tract in animals isan open ecosystem. This raises the question of how, un-der these conditions, a long-lasting genetic isolation isachieved that is rigid enough to allow the evolution ofsuch a divergence. The intestinal tracts of animals mustbe colonized by a complex microbiota, including Nyc-totherus, after hatching from the eggs, molting, andmetamorphosis. Cockroaches, for example, molt severaltimes before reaching the adult stage. Periplaneta amer-icana, for instance, molts approximately 12 times overa period of 1 year (Guthrie and Tindall 1968). Duringevery molt, the intestinal biota of the hindgut are shed,together with the cuticle covering this part of the intes-tinal tract. Freshly molted cockroaches eat their own me-conium or those from members of their local population.This behavior allows the recolonization of the intestinaltract with a microbial biota that is characteristic for thehost population (cf. Nalepa and Bell 1997). Consequent-ly, the behavior of the hosts, at least in cockroaches,might favor a vertical transfer. However, ciliates of thegenus Nyctotherus form cysts as resting and propagationstages. Therefore, a horizontal transfer to other host spe-cies (or strains) should be possible during the (re-)in-fection phases. The presence of ciliates with identical18S rDNA in two only distantly related cockroach spe-cies, i.e., P. americana (Blattellidae) and Blaberus sp.(Blaberidae), proves that host switches can occur under

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021

1202 van Hoek et al.

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021

Evolution of the Intestinal Ciliates 1203

FIG. 7.—Unrooted neighbor-joining tree (Saitou and Nei 1987) based on 5.8S rDNA, ITS-1, and ITS-2 sequences (fig. 6) of the intestinalciliates from the different cockroach hosts using the Jukes and Cantor (1969) algorithm. Distance data were bootstrap resampled 100 times(Felsenstein 1985). The scale bar indicates the distances.

FIG. 6.—Alignment of the 5.8S rDNA, ITS-1, and ITS-2 sequences, including parts of the 18S and 28S rDNA genes of Nyctotherus ovalisfrom the different cockroach hosts. The alignment includes sequences from two parasitic ciliates (#Diggles and Adlard 1997) and a free-livingciliate (*Engberg and Nielsen 1990). Points indicate identical nucleotides. Dashes mark gaps in the alignment.

natural conditions (fig. 7). However, such switches seemto occur only rarely. Also in zoological gardens, wherelarge populations of free-living Blattela germanica, Pyc-nocelus surinamensis, P. australasia, and P. americanacoexist with Blaberus sp. cultures, we did not observeany transfers of intestinal ciliates (Hackstein 1997). Wealso did not find any evidence for an acquisition of ‘‘for-eign’’ ciliates by Blaberus sp. var. Amsterdam over aperiod of 4 years.

Experimental studies revealed that transfers be-tween different species are hampered by several con-straints. For example, transfers are only possible be-tween certain host species, and they can occur only ifthe recipient is freshly molted (Hoyte 1961a, 1961b; un-published data). DNA sequence data also do not provideany evidence that cockroaches acquired intestinal cili-ates from millipedes or anurans. Moreover, the ciliatesthat live in the intestinal tracts of the various vertebratesand arthropods exhibit distinct morphological traits, and,consequently, they have been described as different spe-cies. Our study confirms the assumption that the ciliatesliving in the intestinal tracts of animals exhibit a re-markable host- and strain-specificity (Bhatia and Gulati1927; Doflein and Reichenow 1953; Hoyte 1961a,1961b). Spatial and behavioral isolation of their hostsmight be responsible for a predominantly vertical trans-

mission of the ciliates described in this study. But inaddition, a horizontal spread of intestinal ciliates mightalso be hampered by a genetically controlled characterof the host, since intestinal protists are distributed in anonrandom manner (Hackstein and Stumm 1994; Hack-stein 1997). Such a phenomenon has also been observedfor other arthropods and their intestinal protists, and ithas been speculated as to whether host taxonomy is ofsignificance for the presence of intestinal symbionts(Hennig 1981; Nalepa 1991; Thorne 1991; Hacksteinand Stumm 1994; Grancolas and Deleporte 1996). Con-sequently, one of the main criteria for differentiation ofthe various Nyctotherus species has been their associa-tion with particular hosts (Hoyte 1961a). Since we havedemonstrated that host switches can occur, this concept,at least for cockroaches, is of limited value. Moreover,the substantial DNA sequence divergence (up to 5%; seetables 4 and 5) between ciliates living in different strainsof one and the same host species might be used as anargument to assign these ciliates to different species(Vandamme et al. 1996). However, despite the substan-tial DNA sequence divergence, ciliates from cockroach-es look rather similar. Nyctotherus species from milli-pedes and frogs, on the other hand, exhibit rather char-acteristic traits that allow an easy discrimination fromthe ciliates from cockroaches. Therefore, we suggest the

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021

1204 van Hoek et al.

Table 4Pairwise Comparison of the 18Sa rDNA Sequences of Anaerobic Heterotrichs

Length(nt) 2 3 4 5 6

1. Nyctotherus ovalis from Periplaneta americana var. Amsterdam . . .2. Nyctotherus cordiformis from Rana ridibunda . . . . . . . . . . . . . . . . . . .3. Nyctotherus cordiformis from Rana temporaria . . . . . . . . . . . . . . . . . .4. Nyctotherus velox from the julid millipede ‘‘Unidentified A’’ . . . . . .5. Metopus palaeformis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6. Metopus contortus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1,3481,3431,3391,3461,3361,329

95.1 95.499.1

95.795.495.0

89.089.288.989.2

85.386.286.288.390.4

NOTE.—Numbers represent percentage identity calculated with the Bestfit program of the Wisconsin package, version 9.1.a Partial sequence (675%). Metopus palaeformis, M86385; M. contortus, Z29516.

maintenance of the arbitrary description of N. ovalis forciliates living in cockroaches, N. velox for ciliates frommillipedes, and N. cordiformis for ciliates living in an-uran hosts until additional morphological, physiological,and genetic data support a more differentiated classifi-cation. The additional indication of the particular hoststrain might be sufficient to identify the various ciliatesand to avoid discussions about their classification as sep-arate species or subspecies.

Antiquity of the Symbiotic Associations

The phylogenies of the intestinal ciliates and thesignificant degree of DNA sequence divergence betweenthe different Nyctotherus species and variants argue forthe antiquity of the associations between ciliates andtheir arthropod and vertebrate hosts. Notwithstanding,highly divergent ciliates are also found in differentstrains of one and the same host species, and becauseof host switches, the phylogenies of intestinal ciliatesand their cockroach hosts do not match. Such a situationis clearly different from the coevolution of the myce-tome-inhabiting flavobacteria and their cockroacheshosts, for which a complete match between the phylog-enies of the symbionts and their hosts has been dem-onstrated (Bandi et al. 1995). A comparable coevolutionhas also been described for Buchnera and their aphidhosts (Moran and Baumann 1994; Baumann et al. 1995).Such a coevolution is the result of a strictly verticaltransmission of the symbionts that is facilitated by anegg-mediated transfer. Also, Wolbachia, a rickettsia-likegerm line parasite of many arthropods (Werren 1997),is transmitted by eggs from the mother to the filial gen-eration. However, switches of Wolbachia between hoststhat belong to different higher taxa have occured (Schilt-huizen and Stouthamer 1997) similar to the situationobserved for the intestinal ciliates of cockroaches. Suchhost switches hamper the reconstruction of potentialcospeciation events, but fossilized insects might provideclues for a calibration of the molecular clock. Fossilsthat exhibit very suggestive cockroach traits are as oldas 250–300 Myr (Hennig 1981). The morphology of thechewing appendages of many of these insects did notchange substantially over at least 250 Myr, suggestingthat the feeding habits of fossil cockroach-like insectswere similar to those of extant cockroaches. Therefore,one might speculate as to whether the ancestors of cock-roaches hosted ciliates similar to those of their present-

day relatives. A more detailed analysis of the intestinaltract of fossilized cockroaches for the presence of ciliatecysts should allow the identification of potential Nyc-totherus species and thereby provide direct evidence forthe antiquity of these symbiotic associations.

Acknowledgments

We would like to thank Dr. Jorg Rosenberg, Bo-chum, Germany, for the SEM photograph displayed infigure 1. We are indebted to Prof. Dr. Martin Schlegel,Leipzig, Germany, for his support and advice concern-ing the cloning of the ribosomal genes. We thank himand Dr. Richard Stouthamer, Wageningen, The Nether-lands, for critical reading of the manuscript. The re-peated advice of Jack Leunissen (Caos-Camm, Nijme-gen, the Netherlands) is gratefully acknowledged. Thegenerous gift of the amphibs by Dr. H. Strijbosch wasessential for this study. This work was supported by agrant from The Netherlands Organization for ScientificResearch (NWO) to A.H.A.M.v.H.

Sequence Availibility

The sequence data from the Nyctotherus speciesdescribed here have been deposited in the EMBL data-base: 18S rDNA gene of N. ovalis from P. americanavar. Amsterdam (AJ222678), 18S rDNA gene (partial)of N. cordiformis from R. ridibunda (AJ006711), 18SrDNA gene (partial) of N. cordiformis from R. tempor-aria (AJ006712), 18S rDNA gene (partial) of N. veloxfrom the julid millipede ‘‘Unidentified A’’ (AJ006713),5.8S rDNA gene and flanking ITS sequences of N. oval-is from P. americana var. Amsterdam (AJ006714), 5.8SrDNA gene and flanking ITS sequences of N. ovalisfrom P. americana var. Bayer (AJ006715), 5.8S rDNAgene and flanking ITS sequences of N. ovalis from P.americana var. Dar es Salaam (AJ006716), 5.8S rDNAgene and flanking ITS sequences of N. ovalis from P.americana var. Nijmegen (AJ006717), 5.8S rDNA geneand flanking ITS sequences of N. ovalis from Blaberussp. var. Nijmegen (AJ006718), 5.8S rDNA gene andflanking ITS sequences of N. ovalis from Blaberus sp.var. Amsterdam (AJ006719), and 5.8S rDNA gene andflanking ITS sequences of N. ovalis from Blaberus sp.var. Dusseldorf (AJ006720).

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021

Evolution of the Intestinal Ciliates 1205

Tab

le5

Pai

rwis

eC

omp

aris

onof

the

18S

aan

d5.

8SrD

NA

,an

dth

eIT

S-1

and

ITS

-2S

equ

ence

sof

Nyc

toth

eru

sov

alis

from

the

Dif

fere

nt

Coc

kro

ach

Hos

ts

Len

gth

(nt)

18S

23

45

67

5.8S

23

45

67

ITS

-1

23

45

67

ITS

-2

23

45

67

N.

oval

isfr

omP

erip

lane

taam

eric

ana

1.va

r.A

mst

erda

m..

.2.

var.

Bay

er..

....

..3.

var.

Dar

esS

alaa

m..

4.va

r.N

ijm

egen

....

.

820

823

824

825

95.5

95.0

98.8

96.0

97.8

97.0

95.5

100 98

.897

.8

98.9

95.3

94.6

95.5

95.5

100 98

.897

.8

97.5

96.2

98.7

96.8

99.4

99.4

97.5

100 98

.799

.4

100 97

.596

.296

.8

97.5

100 98

.799

.4

96.8

97.8

98.9

96.8

96.8

97.8

96.8

100 98.9

96.8

98.9

95.7

96.8

95.7

96.8

100 98.9

96.8

95.6

95.6

100

95.6

97.1

97.1

95.6

100

100 97.1

97.8

94.9

94.9

96.4

95.6

100

100 97.1

N.

oval

isfr

omB

labe

rus

sp.

5.va

r.A

mst

erda

m..

.6.

var.

Dus

seld

orf

....

7.va

r.N

ijm

egen

....

.

823

822

823

95.3

100 95

.397

.510

0 97.5

95.7

100 95

.794

.910

0 94.9

NO

TE.—

Num

bers

repr

esen

tpe

rcen

tage

iden

tity

calc

ulat

edw

ith

the

Bes

tfit

prog

ram

ofth

eW

isco

nsin

pack

age,

vers

ion

9.1.

aP

arti

alse

quen

ce(6

50%

).F

orle

ngth

sof

5.8S

rDN

A,

ITS

-1,

and

ITS

-2,

see

tabl

e3.

LITERATURE CITED

BANDI, C., M. SIRONI, G. DAMIANI, L. MAGRASSI, C. A. NA-LEPA, U. LAUDANI, and L. SACCHI. 1995. The establishmentof intracellular symbiosis in an ancestor of cockroaches andtermites. Proc. R. Soc. Lond. B Biol. Sci. 259:293–299.

BASSAM, B. J., G. CAETANO-ANOLLES, and P. M. GRESSHOFF.1991. Fast and sensitive silver staining of DNA in poly-acrylamide gels. Anal. Biochem. 196:80–83.

BAUMANN, P., L. BAUMANN, C.-Y. LAI, D. ROUHBAKHSH, N.A. MORAN, and M. A. CLARK. 1995. Genetics, physiology,and evolutionary relationships of the genus Buchnera: in-tracellular symbionts of aphids. Annu. Rev. Microbiol. 49:55–94.

BHATIA, B. L., and A. N. GULATI. 1927. On some parasiticciliates from Indian frogs, toads, earthworms and cock-roaches. Arch. Protistenkd. 57:85–120.

BREZNAK, J. A. 1982. Intestinal microbiota of termites and oth-er xylophagous insects. Annu. Rev. Microbiol. 36:323–343.

CAZEMIER, A. E., J. H. P. HACKSTEIN, H. J. M. OP DEN CAMP,J. ROSENBERG, and C. VAN DER DRIFT. 1997. Bacteria inthe intestinal tract of different species of arthropods. Mi-crob. Ecol. 33:189–197.

CRUDEN, D. L., and A. J. MARKOVETZ. 1987. Microbial ecol-ogy of the cockroach gut. Annu. Rev. Microbiol. 41:617–643.

DE PUYTORAC, P., J. GRAIN, and P. LEGENDRE. 1994. An at-tempt at reconstructing a phylogenetic tree of the ciliophorausing parsimony methods. Eur. J. Protistol. 30:1–17.

DIGGLES, B. K., and R. D. ADLARD. 1997. Intraspecific vari-ation in Cryptocaryon irritans. J. Eukaryot. Microbiol. 44:25–32.

DOFLEIN, F., and E. REICHENOW. 1953. Lehrbuch der Proto-zoenkunde. VEB Gustav Fischer Verlag, Jena, Germany.

ELWOOD, H. J., G. J. OLSEN, and M. L. SOGIN. 1985. The smallsubunit ribosomal RNA gene sequences from the hypotri-chous ciliates Oxytricha nova and Stylonychia pustulata.Mol. Biol. Evol. 2:399–410.

EMBLEY, T. M., B. J. FINLAY, P. L. DYAL, R. P. HIRT, M. WIL-KINSON, and A. G. WILLIAMS. 1995. Multiple origins ofanaerobic ciliates with hydrogenosomes within the radiationof aerobic ciliates. Proc. R. Soc. Lond. B Biol. Sci. 262:87–93.

ENGBERG, J., and H. NIELSEN. 1990. Complete sequence of theextrachromosomal rDNA molecule from the ciliate Tetra-hymena thermophila strain B1868VII. Nucleic Acids Res.18:6915–6919.

FELSENSTEIN, J. 1985. Confidence limits on phylogenies: anapproach using the bootstrap. Evolution 39:783–791.

. 1993. PHYLIP (phylogeny inference package). Ver-sion 3.5c. Distributed by the author, Department of Genet-ics, University of Washington, Seattle.

FINLAY, B. J., J. O. CORLISS, G. ESTEBAN, and T. FENCHEL.1996. Biodiversity at the microbial level: the number offree-living ciliates in the biosphere. Q. Rev. Biol. 71:221–235.

FOISSNER, W., and I. FOISSNER. 1988. Ciliophora. Verlag derOsterreichischen Akademie der Wissenschaften, Wien, Aus-tria.

GIJZEN, H. J., C. A. M. BROERS, M. BARUGHARE, and C. K.STUMM. 1991. Methanogenic bacteria as endosymbionts ofthe ciliate Nyctotherus ovalis in the cockroach hindgut.Appl. Environ. Microbiol. 57:1630–1634.

GRANDCOLAS, P., and P. DELEPORTE. 1996. The origin of pro-tistan symbionts in termites and cockroaches: a phyloge-netic perspective. Cladistics 12:93–98.

GRELL, K. G. 1973. Protozoology. Springer-Verlag, Berlin.

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021

1206 van Hoek et al.

GUTHRIE, D. M., and A. R. TINDALL. 1968. The biology of thecockroach. Edward Arnold, London.

HACKSTEIN, J. H. P. 1997. Eukaryotic molecular biodiversity:systematic approaches for the assessment of symbiotic as-sociations. Antonie Van Leeuwenhoek 72:63–76.

HACKSTEIN, J. H. P., and C. K. STUMM. 1994. Methane pro-duction in terrestrial arthropods. Proc. Natl. Acad. Sci. USA91:5441–5445.

HAMMERSCHMIDT, B., M. SCHLEGEL, D. H. LYNN, D. D. LEIPE,M. L. SOGIN, and I. B. RAIKOV. 1996. Insights into theevolution of nuclear dualism in the ciliates revealed by phy-logenetic analysis of rRNA sequences. J. Eukaryot. Micro-biol. 43:225–230.

HAUSMANN, K., and N. HULSMANN. 1996. Protozoology.Thieme Verlag, Stuttgart, Germany.

HAWKSWORTH, D. L., and M. T. KALIN-ARROYO. 1995. Mag-nitude and distribution of biodiversity. Pp. 107–138 in V.H. HEYWOOD, ed. Global biodiversity assessment. Cam-bridge University Press, Cambridge, England.

HENNIG, W. 1981. Insect phylogeny. John Wiley and Sons,Chichester, England.

HIRT, R. P., P. L. DYAL, M. WILKINSON, B. J. FINLAY, D. M.ROBERTS, and T. M. EMBLEY. 1995. Phylogenetic relation-ships among karyorelictids and heterotrichs inferred fromsmall subunit rRNA sequences: resolution at the base of theciliate tree. Mol. Phylogenet. Evol. 4:77–87.

HOYTE, H. M. D. 1961a. The protozoa occurring in the hind-gut of cockroaches. II. Morphology of Nyctotherus ovalis.Parasitology 51:437–463.

. 1961b. The protozoa occurring in the hind-gut ofcockroaches. III. Factors affecting the dispersion of Nyc-totherus ovalis. Parasitology 51:465–495.

HUNGATE, R. E. 1966. The rumen and its microbes. AcademicPress, New York.

JUKES, T. H., AND C. R. CANTOR. 1969. Evolution of proteinmolecules. Pp. 21–132 in H. N. MUNRO, ed. Mammalianprotein metabolism. Academic Press, New York.

KIMURA, M. 1980. A simple model for estimating evolutionaryrates of base substitutions through comparative studies ofnucleotide sequences. J. Mol. Evol. 16:111–120.

KUDO, R. R. 1931. Handbook of protozoology. Bailliere, Tin-dall and Cox, London.

LEIDY, J. 1849. On entophyta in living animals. New speciesof entozoa. On glandulae odoriferae. Proc. Acad. Natl. Sci.Philadelphia 4:225–236.

. 1853. Some observations on Nematoidae imperfecta,and descriptions of three parasitic Infusoriae. Trans. Amer.Phil. Soc. 10:241–244.

LUCAS, C. L. T. 1927. A study of excystation in Nyctotherusovalis. J. Parasitol. 14:161–175.

MCKEAN, G. 1972. A study of the ciliated protozoon Nycto-therus ovalis. Morphology: general and ultrastructural, lifecycle: anisogamonty and division. Ph.D. thesis, Universityof Illinois at Urbana–Champaign. UMI Dissertation Infor-mation Service, Ann Arbor, Mich.

MORAN, N., and P. BAUMANN. 1994. Phylogenetics of cyto-plasmically inherited micro-organisms of arthropods.Trends Ecol. Evol. 9:15–20.

NALEPA, C. A. 1991. Ancestral transfer of symbionts betweencockroaches and termites: an unlikely scenario. Proc. R.Soc. Lond. B Biol. Sci. 246:185–189.

NALEPA, C. A., and W. J. BELL. 1997. Postovulation parentalinvestment and parental care in cockroaches. Pp. 26–51 inJ. A. CHOE and B. J. CRESPI, eds. The evolution of socialbehavior in insects and arachnids. Cambridge UniversityPress, Cambridge, England.

ODORICO, D. M., and D. J. MILLER. 1997. Variation in theribosomal internal transcribed spacers and 5.8S rDNAamong five species of Acropora (Cnidaria; Scleractinia):patterns of variation consistent with reticulate evolution.Mol. Biol. Evol. 14:465–473.

SAITOU, N., and M. NEI. 1987. The neighbor-joining method:a new method for reconstructing phylogenetic trees. Mol.Biol. Evol. 4:406–425.

SCHILTHUIZEN, M., and R. STOUTHAMER. 1997. Horizontaltransmission of parthenogenesis-inducing microbes inTrichogramma wasps. Proc. R. Soc. Lond. B Biol. Sci. 264:361–366.

SCHLEGEL, M. 1991. Protist evolution and phylogeny as dis-cerned from small subunit ribosomal RNA sequence com-parisons. Eur. J. Protistol. 27:207–219.

SMALL, E. B., and D. H. LYNN. 1981. A new macrosystem forthe phylum Ciliophora Doflein, 1901. Biosystems 14:387–401.

SOGIN, M. L., and H. J. ELWOOD. 1986. Primary structure ofthe Paramecium tetraurelia small subunit rRNA coding re-gion: phylogenetic relationships within the ciliophora. J.Mol. Evol. 23:53–60.

SWOFFORD, D. L., and G. J. OLSEN. 1990. Phylogeny recon-struction. Pp. 411–501 in D. M. HILLIS and C. MORITZ, eds.Molecular systematics. Sinauer, Sunderland, Mass.

THORNE, B. L. 1991. Ancestral transfer of symbionts betweencockroaches and termites: an alternative hypothesis. Proc.R. Soc. Lond. B Biol. Sci. 246:191–195.

VANDAMME P., B. POT, M. GILLIS, P. DE VOS, K. KERSTERS,and J. SWINGS. 1996. Polyphasic taxonomy, a consensusapproach to bacterial systematics. Microbiol. Rev. 60:407–438.

WAGENER, S., C. K. STUMM, and G. D. VOGELS. 1986. Elec-tromigration, a tool for studies on anaerobic ciliates. FEMSMicrobiol. Ecol. 38:197–203.

WALSH, P. S., D. A. METZGER, and R. HIGUCHI. 1991. Chelex-100 as a medium for simple extraction of DNA for PCR-based typing from forensic material. BioTechniques 10:506–513.

WERREN, J. H. 1997. Wolbachia run amok. Proc. Natl. Acad.Sci. USA 94:11154–11155.

WICHTERMAN, R. 1937. Division and conjugation in Nyctothe-rus cordiformis (Ehr.) Stein (Protozoa, Ciliata) with specialreference to the nuclear phenomena. J. Morphol. 60:563–611.

WILLIAMS, A. G. 1986. Rumen holotrich ciliate protozoa. Mi-crobiol. Rev. 50:25–49.

WILLIAMS, A. G., and G. S. COLEMAN. 1991. The rumen pro-tozoa. Springer-Verlag, New York.

WILLIS, L. M., and E. R. ROTH. 1960. The biotic associationsof cockroaches. Smithsonian Institution, Washington, D.C.

ZEHNDER, A. J. B., and K. WUHRMANN. 1976. Titanium(III)citrate as a nontoxic oxidation-reduction buffering systemfor the culture of obligate anaerobes. Science 194:1165–1166.

PEKKA PAMILO, reviewing editor

Accepted May 28, 1998

Dow

nloaded from https://academ

ic.oup.com/m

be/article/15/9/1195/1412734 by guest on 07 Decem

ber 2021


Recommended