Mise au point

MOLECULAR EVOLUTION INFERRED FROM SMALL SUBUNIT RRNA SEQUENCES:

WHAT DOES IT TELL US ABOUT PHYLOGENETIC RELATIONSHIPS AND T A X O N O M Y O F THE P A R A B A S A L I D S ?

VISCOGLIOSI E.*, EDGCOMB V.P.**, GERBOD D.*, NOËL C* & DELGADO-VISCOGLIOSI P.***

Summary:

The Parabasala are a primitive group of protists divided into two classes: the trichomonads and the hypermastigids. Until recently, phylogeny and taxonomy of parabasalids were mainly based on the comparative analysis of morphological characters primarily linked to the development of their cytoskeleton. Recent use of molecular markers, such as small subunit (SSU) rRNA has led to new insights into the systematics of the Parabasala and other groups of protists. An updated phylogeny based on S S U rRNA is provided and compared to that inferred from ultrastructural data.

The S S U rRNA phylogeny contradicts the dogma equating simple characters with primitive characters. Hypermastigids, possessing a hyperdeveloped cytoskeleton, exhibit the most basal emergence in the parabasalid lineage. Other observations emerge from the S S U rRNA analysis, such as the secondary loss of some cytoskeleton structures in all representatives of the Monocercomonadidae, the existence of secondarily free-living taxa (reversibility of parasitism) and the evidence against the co-evolution of the endobiotic parabasalids and their animal hosts. According to phylogenies based on S S U rRNA, all the trichomonad families are not monophyletic groups, putting into question the validity of current taxonomic assignments. The precise branching order of some taxa remains unclear, but this issue can possibly be addressed by the molecular analysis of additional parabasalids. The goal of such additional analyses would be to propose, in a near future, a revision of the taxonomy of this group of protists that takes into account both molecular and morphological data.

KEY WORDS : Parabasalids, trichomonads, small subunit rRNA, molecular phylogeny, evolution, taxonomy.

Résumé : ÉVOLUTION MOLÉCULAIRE BASÉE SUR LES SÉQUENCES DE L ' A R N R DE LA PETITE SOUS-UNITÉ : QUE NOUS PROPOSE-T-ELLE SUR LES RELATIONS PHYLOGÉNÉTIQUES ET LA TAXONOMIE DES PARABASALA ? Les Parabasala forment un groupe de protistes primitifs regroupant deux classes : les trichomonadines et les hypermastigines. Jusqu'à ces dernières années, les relations de parenté et la taxonomie des Parabasala étaient essentiellement basées sur l'analyse comparative des caractères morphologiques qui étaient principalement liés au développement de leur cytosquelette. L'utilisation récente d'indicateurs moléculaires comme l'ARN ribosomique de la petite sous-unité a largement modifié la systématique des Parabasala comme celle d'autres groupes de protistes. Un arbre basé sur les séquences des ARN ribosomiques de la petite sous-unité est proposé et comparé à celui établi d'après les données ultrastructurales. Cet arbre contredit le dogme qui était qu'un caractère dit simple ne pouvait être qu'un caractère primitif. En effet, les hypermastigines, dotées d'un

cytosquelette très développé, ont l'émergence la plus précoce dans la lignée des Parabasala. D'autres observations peuvent être tirées de cet arbre comme la perte secondaire de certaines structures cytosquelettiques chez tous les représentants de la famille des AAonocercomonadidae, l'existence de formes secondairement libres (réversibilité du parasitisme) et l'absence de co-évolution entre les Parabasala et leurs hôtes respectifs. Il remet aussi en cause la validité de la taxonomie actuelle puisque toutes les familles de

trichomonadines ne forment pas de groupes monophylétiques.

L'émergence précise de certains taxa reste encore indécise mais ceci devrait pouvoir être résolu par l'étude moléculaire de plusieurs espèces de Parabasala encore non étudiées. La finalité de ce travail serait de proposer, dans un avenir proche, une nouvelle taxonomie de ce groupe de protistes qui prendrait en compte à la fois les données moléculaires et les données morphologiques.

MOTS CLES : Parabasala, trichomonadines, ARNr de la petite sous-unité, phyiogénie moléculaire, évolution, taxonomie.

INTRODUCTION

T

h e P a r a b a s a l i d e a , c o m m o n l y p a r a b a s a l i d s o r Parabasala, are flagellated protists e q u i p p e d with a characteristic mastigont (Honigberg, 1 9 6 3 ; Bru-

* Laboratoire de Biologie Comparée des Protistes, UPRESA CNRS 6023, Avenue des Landais, Aubière, France.

** Josephine Bay Paul Center for Molecular Evolution, Marine Bio- logical Laboratory, Woods Hole, MA 02543, USA.

*** Laboratoire d'Oncologie Moléculaire, Centre Jean Perrin, Rue Mon- talembert, Clermont-Ferrand, France.

Correspondence: Dr Eric Viscogliosi.

Tel: (33) 4 73 40 74 72 - Fax: (33) 4 73 40 76 70.

E-mail: viscogliosi@cicsun.univ-bpclermont.fr

gerolle, 1976; H o n i g b e r g & Brugerolle, 1990). T h e y are distinguished b y t h e p r e s e n c e o f o n e o r m o r e para- b a s a l a p p a r a t u s c o m p r i s i n g a p a r a b a s a l b o d y ( G o l g i c o m p l e x ) a n d a p a r a b a s a l filament. It is from this structural attribute that t h e n a m e o f the s u p e r o r d e r ( P a r a b a s a l i d e a ) derives. T h e s e o r g a n i s m s are a n a e - robic, lack mitochondria, but p o s s e s s h y d r o g e n o s o m e s , s p e c i a l i s e d o r g a n e l l e s l a c k i n g n u c l e i c acids a n d sur- r o u n d e d b y a d o u b l e m e m b r a n e in w h i c h a n a e r o b i c m e t a b o l i s m t a k e s p l a c e . All parabasalid g e n e r a studied to date e x h i b i t a s p e c i a l type o f c l o s e d mitosis c a l l e d c r y p t o p l e u r o m i t o s i s , c h a r a c t e r i z e d b y t h e p e r s i s t e n c e o f the n u c l e a r e n v e l o p e a n d the p r e s e n c e o f an extra- n u c l e a r s p i n d l e ( B r u g e r o l l e , 1 9 7 5 a ) . T h e s e structural

Parasite, 1999, 6, 279-291

2 7 9

Mise au point

Article available athttp://www.parasite-journal.orgorhttp://dx.doi.org/10.1051/parasite/1999064279

characteristics allow the separation o f the well defined s u p e r o r d e r P a r a b a s a l i d e a f r o m o t h e r f l a g e l l a t e d lineages.

More than eighty parabasalid g e n e r a have b e e n iden­

tified to date, and classified into two orders: Tricho- monadida and Hypermastigida (Corliss, 1 9 8 4 ; Cavalier- Smith, 1993). Most trichomonads are found in association with the respiratory, digestive and reproductive systems o f vertebrates, including m a m m a l s and invertebrates.

T h e r e are only three s p e c i e s o f t r i c h o m o n a d s that infect humans, and they are extremely site-specific (Honigberg, 1 9 7 8 ) . Trichomonas tenax lives in the oral cavity, Pentatrichomonas hominis in the intestinal tract and Trichomonas vaginalis in the genitourinary tract. O n l y Trichomonas vaginalis is considered patho­

g e n i c (Graves & Gardner, 1 9 9 3 ) . This parasite is the causative agent o f h u m a n trichomoniasis w h i c h has e m e r g e d as the most prevalent nonviral sexually trans­

mitted disease worldwide in r e c e n t years. T h e "classic signs" o f trichomoniasis are vaginitis and exocervicitis but Trichomonas vaginalis has also b e e n implicated in the p a t h o g e n e s i s o f preterm labor, premature rupture o f m e m b r a n e s , u p p e r reproductive tract postsurgical infections ( H e i n e & M c G r e g o r , 1 9 9 3 ) , and m a y have an important role in the amplification o f HIV-1 trans­

mission in developing countries (Sorvillo & Kerndt, 1 9 9 8 ) . Considerable research has also b e e n d o n e o n the t r i c h o m o n a d s that infect animals o f e c o n o m i c importance to humans, such as Tritrichomonas foetus ( Y u l e et al, 1 9 8 9 ) . It is the causative agent o f b o v i n e trichomoniasis, colonizing both male and female repro­

ductive tracts. T h e infection in c o w s is serious and usually leads to infertility, abortion and pyometra, resulting in substantial e c o n o m i c losses. S o m e tricho- m o n a d s p e c i e s are also k n o w n to b e free-living and h a v e b e e n isolated from the sediments o f freshwater lakes (Farmer, 1993) and marine sediments ( E d g c o m b et al, 1998).

S p e c i e s b e l o n g i n g to the s e c o n d class o f parabasalids, the hypermastigids, are found in association with the digestive tracts o f s o m e insects (Yamin, 1 9 7 9 ) . T h e insect hosts are termites and wood-eating c o c k r o a c h e s that maintain c o m p l e x symbiotic communities o f b a c ­ teria and protists in their hindguts. T h e s e symbiotic hypermastigids have a possible role in the digestion o f w o o d c o m p o n e n t s and are required for the survival o f their w o o d - f e e d i n g hosts ( G r o s o v s k y & Margulis, 1 9 8 2 ; B r e z n a k , 1 9 8 2 ) . It is difficult to extricate and study hypermastigids from this intimate community that are refractory to culturing m e t h o d o l o g i e s and h e n c e , most m o l e c u l a r work.

Until recently, p h y l o g e n e t i c relationships a m o n g para­

basalids w e r e mainly e x a m i n e d b y the comparative analysis o f m o r p h o l o g i c a l characters. In 1 9 6 3 , Honig­

berg w a s the first to publish his view o f the evolutio-

nary relationships among the trichomonads. This scheme was based o n light-microscope studies o f living, or fixed and stained material. Thereafter, n u m e r o u s electron m i c r o s c o p e studies provided information confirming the original s c h e m e o f Honigberg (Brugerolle, 1 9 7 6 ) . C o m m o n features shared b e t w e e n hypermastigids and trichomonads led to the formal union o f these two classes o f protists into the parabasalids (Levine et al, 1 9 8 0 ) . T h e s e evolutionary s c h e m e s w e r e mainly b a s e d o n a restricted n u m b e r o f characters mostly linked to the structure and d e v e l o p m e n t o f the cytoskeleton. As with other protist groups, the p r e s u m e d evolution o f the parabasalids reflected the traditional view regarding polarization o f cytoskeletal d e v e l o p e m e n t from simple to c o m p l e x . Parabasalids exhibiting a rudimentary cytoskeleton w e r e thought to o c c u p y a basal position and to have given rise to the other more c o m p l e x para­

basalids. Despite the well k n o w n morphological c h a ­ racteristics o f most parabasalid taxa, their phylogenetic relationships w e r e often unclear. Indeed, identifying h o m o l o g o u s morphological features and defining pola­

rity o f character states is extremely difficult and fre­

quently arbitrary. T h e a b s e n c e o f a character can either b e ancestral or due to secondary loss, indistinguishable o u t c o m e s b y morphological analyses alone. Phyloge- nies based o n molecular analyses provided n e w insights into the reconstruction o f evolutionary relationships in parabasalids, as well as other protist groups.

Zuckerkandl & Pauling ( 1 9 6 5 ) w e r e the first to pro­

p o s e the use o f the g e n e s e q u e n c e s o f macromolecules as molecular indicators. Not all m o l e c u l e s are equally useful for such studies and the m o l e c u l e o f c h o i c e has b e e n the small subunit rRNA (SSU rRNA). This m o l e ­ cule is c o n s e r v e d throughout the living world (present in all k n o w n s p e c i e s as well as in mitochondria and chloroplasts), providing thousands o f h o m o l o g o u s cha­

racters, thus allowing in principle the construction o f phylogenies for any c h o s e n group ( P a c e et al, 1 9 8 6 ; W o e s e , 1 9 8 7 ; Olsen & W o e s e , 1993). Furthermore, SSU rRNA proved to b e an almost ideal phylogenetic marker given its functional conservation, the size o f the m o l e ­ cule, and its primary and s e c o n d a r y structures w h i c h contain regions o f variable conservation, thus aiding the alignment o f s e q u e n c e s b e t w e e n s p e c i e s and allo­

wing the c o m p a r i s o n o f b o t h closely- and distantly- related organisms. S e q u e n c i n g t e c h n i q u e s have b e e n s u b s e q u e n t l y streamlined, and several hundred SSU rRNA s e q u e n c e s from eucaryotes are n o w available in public databases. T h e use o f PCR primers targeting c o n s e r v e d regions o f the g e n e s e n c o d i n g SSU rRNAs have m a d e possible the cloning and s e q u e n c i n g o f unculturable species, s u c h as m a n y o f the parabasalid taxa (Medlin et al, 1 9 8 8 ) . In sum, SSU rRNA a p p e a r e d to b e the best tool available for determination o f rela­

tionships a m o n g the parabasalids.

280 Parasite, 1999, 6, 279-291

In this study, w e analyze a b r o a d p h y l o g e n y including SSU rRNA s e q u e n c e s from m o r e than 2 5 parabasalid s p e c i e s to c o m p a r e m o r p h o l o g i c a l and m o l e c u l a r phy­

togenies o f parabasalids. W e n o t e large d i s c r e p a n c i e s b e t w e e n ultrastructural a n d m o l e c u l a r data, a n d the­

refore p r o p o s e c h a n g e s in the p r e s u m e d p h y l o g e n e t i c relationships o f parabasalids a n d question the present validity o f t a x o n o m i c assignments in this group o f pro- tists. In addition, w e put forward h y p o t h e s e s o n the evolution o f the cytoskeleton o f these microorganisms.

CYTOSKELETON

OF THE TRICHOMONAD CELL

T

h e b a s i c structure o f a t r i c h o m o n a d cell is illus­

trated b y Figure 1 ( B r u g e r o l l e , 1 9 7 6 ; Honigberg

& Brugerolle, 1990). Trichomonads are e q u i p p e d with three to five anteriorly-directed flagella. In the hypermastigids, the n u m b e r o f anterior flagella multi­

plied, reaching as m a n y as a thousand. In t r i c h o m o -

Fig. 1. - Schematic diagram of a trichomonad cell. AF: anterior fla­

gellar PAX: preaxostylar fibers; UM: undulating membrane; N:

nucleus; G: parabasal apparatus (parabasal fibers plus dictyosomes);

C: costa; AX-PE: axostyle-pelta complex; RF: recurrent flagellum.

MOLECULAR PI IYLOGENY OF FARASALIDS

nads, the recurrent flagellum is either free or attached to the cell b o d y , forming the Lindulating m e m b r a n e . T h e undulating m e m b r a n e is c o m p o s e d o f the recur­

rent flagelhim a n d microfibrillar structures l o c a t e d b e l o w the m e m b r a n e o f the cell along the adherent z o n e with the recurrent flagellum. T h e flagella a n d undulating m e m b r a n e are the l o c o m o t i v e organelles.

T h e principal m e a n o f l o c o m o t i o n in t r i c h o m o n a d genera is the undulating m e m b r a n e , w h e r e a s the ante­

rior flagella mainly serve to c h a n g e the direction o f m o v e m e n t (Kulda et al., 1 9 8 6 ) . A b r o a d striated root, the costa, o c c u r s only in t r i c h o m o n a d s p o s s e s s i n g an undulating m e m b r a n e . It is a s s u m e d that the costa serves as a m e c h a n i c a l support for the undulating m e m b r a n e . I n d e e d , the c o s t a e x t e n d s immediately b e l o w it within the cytoplasm and in s o m e taxa, micro­

fibrillar layers c o n n e c t the costa to the undulating m e m b r a n e .

Several characters are c o m m o n to all parabasalid s p e ­ cies. T h e first o n e is the striated root k n o w n as the parabasal fiber, w h i c h polarizes the d i c t y o s o m e s for­

ming the parabasal apparatus. T h e parabasal apparatus is u n i q u e to the parabasalids. T h e r e is n o s u c h des­

c r i b e d relation b e t w e e n a striated root a n d the dic­

t y o s o m e s in other groups o f protists (Grain, 1 9 8 6 ) . In most o f the trichomonads, the root is c o m p o s e d o f two branches or parabasal fibers and e a c h branch may sup­

port a Golgi b o d y . In larger cells such as hypermasti­

gids, there are up to 5 0 parabasal fibers, e a c h sup­

porting a Golgi b o d y ( H o l l a n d e & Caruette-Valentin, 1 9 7 1 ) . A n o t h e r c o m m o n cytoskeletal structure is the microtubtilar axostyle-pelta c o m p l e x c o m p o s e d o f a

Annexe I. - Glossary A p o m o r p h y : A derived character state.

Bootstrapping: A statistical method based on repeated random sampling (up to 1,000 and more) with replace­

ment from an original sample to provide a collection of new estimates of some parameter, from which confi­

dence limits can be calculated (percentages).

Clade: A monophyletic taxon.

Monophyletic: A group that contains all of the descen­

dants of the most recent common ancestor of the consti­

tuent species.

Outgroup: One or more taxa assumed to be outside the group under consideration (ingroup) and used to help resolve the polarity of characters.

Plesiomorphy: An ancestral character state.

Polyphyletic: Term applied to a group of organisms which does not include the most recent common ancestor of those organisms.

T a x o n : Any named group of organisms, not neccessarily a clade.

Parasite, 1999, 6, 279-291

Mise au point

²⁸¹

sheet o f cross-linked microtubules. It constitutes the longitudinal axis o f the cell. In s o m e g e n e r a the a x o - style-pelta c o m p l e x is c o m p o s e d o f rolls o f microtu- bular rows giving a very b r o a d structure. T h e p r e a x o - stylar fibers are additional striated roots shared by all parabasalids. T h e y are c o n n e c t e d to the basal b o d i e s and adhere to the axostyle-pelta c o m p l e x . Previous observations have s h o w n that microtubules o f the a x o ­ style-pelta c o m p l e x b e g i n to a s s e m b l e in the vicinity o f the preaxostylar fibers, w h i c h s e e m to function as a microtubule organizing c e n t e r (Brugerolle, 1976).

TAXONOMY OF THE PARABASALIDS

A

s stated a b o v e , parabasalids exhibit diverse cytoskeletal arrangements, c o m p o s e d o f micro­

fibrillar and microtubular structures. S o m e o f these structures are c o m m o n to all the parabasalid taxa while others are g e n e r a - o r family-specific and, thus, represent significant t a x o n o m i c features. O n the basis o f cytological studies, the trichomonads were separated into five families (Honigberg, 1 9 6 3 ; Brugerolle, 1976;

Lee, 1 9 8 5 ; P e c k a et al, 1996): M o n o c e r c o m o n a d i d a e , T r i c h o m o n a d i d a e , C o c h l o s o m a t i d a e ( o n l y o n e genus recently identified ( P e c k a et al., 1996), Cochlosoma and thus not indicated in Figure 2 ) , D e v e s c o v i n i d a e and C a l o n y m p h i d a e (Fig. 2 ) . T h e M o n o c e r c o m o n a d i d a e have a rudimentary cytoskeleton. T h e y d o not have a costa, and if present, have a w e a k l y d e v e l o p e d undu­

lating m e m b r a n e . In the p r o p o s e d p h y l o g e n e t i c trees o f Honigberg (1963) and Brugerolle (1976), the M o n o ­ c e r c o m o n a d i d a e o c c u p i e d a basal position and gave rise to the T r i c h o m o n a d i d a e exhibiting a m o r e c o m ­ p l e x cytoskeleton. T r i c h o m o n a d i d a e , w h i c h p o s s e s s a costa and an undulating m e m b r a n e , are divided into t w o main subfamilies: the T r i c h o m o n a d i n a e and the Tritrichomonadinae, d e p e n d i n g in large m e a s u r e o n the types o f costa and undulating m e m b r a n e ( B r u g e ­ rolle, 1976). T w o major costa types can b e distingui­

s h e d b a s e d o n b a n d pattern, although e a c h type exhi­

bits the s a m e periodicity (about 42 n m ) . T h e first type ( C I - t y p e ) is present in Tritrichomonasfoetus and other Tritrichomonadinae, and the s e c o n d o n e (C-type) is found in Trichomonas vaginalis and o t h e r T r i c h o m o ­ nadinae. Tritrichomonadinae also possess a "rail-like"

undulating m e m b r a n e while T r i c h o m o n a d i n a e exhibit a "lamellar" o n e .

D e v e s c o v i n i d a e , Calonymphidae and hypermastigids p r o b a b l y e v o l v e d from Tritrichomonadinae ( B r u g e ­ rolle, 1976). In D e v e s c o v i n i d a e there is a specialized organelle that m a y b e h o m o l o g o u s with the undula­

ting m e m b r a n e o f T r i c h o m o n a d i d a e . T h e recurrent flagellum adheres to the plasma m e m b r a n e o f the cell, under w h i c h there is a c o m p l e x system o f fibers called

the cresta by Kirby ( 1 9 4 1 ) . T h e last trichomonad family, the Calonymphidae, is characterized b y the multipli­

cation o f nuclei and o f their associated mastigont sys­

tems. Finally certain characters have undergone an e x a g g e r a t e d d e v e l o p m e n t in hypermastigids, such as the multiplication o f flagella and parabasal fibers.

EARLY EMERGENCE OF THE PARABASALIDS AMONG THE EUKARYOTES

A

t the e n d o f the 1970's, Taylor p r o p o s e d phy­

logenetic relationships a m o n g eukaryotes o n the basis o f morphological characters. T h e s e data allow the identification o f numerous groups o f protists including the parabasalids, but fail to generate a phylogenetic framework linking these groups. In a detailed cytological study o f parabasalids, Brugerolle ( 1 9 7 6 ) did not identify p o s s i b l e relationships with other groups o f flagellated protists. T h e p h e n o t y p i c variation within protists far e x c e e d s that s e e n in other e u k a r y o t i c k i n g d o m s , m a k i n g the identification o f h o m o l o g o u s morphological characters o n a large e v o ­ lutionary scale, such as that e n c o m p a s s i n g all protists, difficult. As an alternative to traditional phenotypic mar­

kers, p h y l o g e n e t i c relationships can b e established through comparisons o f SSU rRNAs. S e q u e n c e analyses have revealed the e x i s t e n c e o f three primary lines o f descent: Eubacteria ( o r Bacteria), Archaebacteria (or Archaea) and Eukaryota ( P a c e et al, 1 9 8 6 ) . A phylo- gram, including most o f the protist groups, is s h o w n in Figure 3 (Sogin, 199D- Diplomonads, such as Giardia lamblia, represent the earliest diverging lineage, fol­

l o w e d by the microsporidia and parabasalids. Several studies (Leipe et al, 1993; Van Keulen et al, 1 9 9 3 ; Phi­

lippe & Adoutte, 1 9 9 5 ) have s h o w n that the relative branching order o f these three amitochondriate pro­

tist groups is unsettled. Indeed the order o f e m e r g e n c e o f diplomonads, microsporidia and parabasalids at the b a s e o f t h e e u k a r y o t i c t r e e m a y b e significantly influenced b y outlying prokaryotic taxa with different G + C compositions in their rRNA coding regions. Their

"primitive" position c o u l d also b e consistent with the lack o f mitochondria, suggesting that mitochondrial endosymbiosis did not o c c u r at the origin o f the first eukaryotic cell (Cavalier-Smith, 1993). However, recent molecular phylogenies b a s e d primarily o n heat s h o c k proteins suggest that the mitochondrial endosymbiotic event could have occurred before the e m e r g e n c e o f diplomonads ( R o g e r et al, 1 9 9 8 ) , microsporidia (Hirt et al, 1997; G e r m o t et al, 1997; Peyretaillade et al, 1998) and parabasalids ( B u i et al, 1996; G e r m o t et al, 1996; Roger et al., 1996) indicating that these groups may b e secondarily amitochondriate. In this case, if w e consider that the eukaryotic cell originated before the

282 Mise au point Parasite, 1999, 6, 279-291

Parasite, 1999, 6, 279-291

Mise au point

₂₈₃

M O I E C U L A R I M I Y L O G E N Y O F P A R A B A S A L I D .

Fig. 2. - Diagrammatic representation of the presumed evolution in parabasalids established from morphological data (redrawn from Honig- berg, 1963 and Brugerolle, 1976; modified from Dyer, 1989).

VISCOGLIOSI E., EDGCOMB V.P., GERBOD D„ NOEL (J. & UELOAUU-VliUJULlUM P.

Fig. 3- - Three kingdom unrooted tree inferred from the comparison of SSU rRNA sequences. Lengths of bran­

ches correspond to genetic distances. The earliest diver­

ging eukaryotic lineages are presented by amitochondriate protists: Diplomonads, Micro- sporidia and Trichomonads.

Trichomonads are underlined (modified from Sogin, 1991).

acquisition o f the mitochondrion e n d o s y m b i o n t , thus there may b e primitively amitochondriate eukaryotes remaining to b e discovered.

MOLECULAR PHYLOGENIES OF PARABASALIDS

T h e first m o l e c u l a r p h y l o g e n y o f parabasalids w a s b a s e d o n large subunit rRNA s e q u e n c e s (LSU rRNA) and w a s limited to the c o m p a r i s o n o f several d o m a i n s o f the m o l e c u l e from nine tricho- m o n a d s p e c i e s (Viscogliosi et al., 1 9 9 3 ) . More recently, o t h e r m o l e c u l e s have b e e n used as p h y l o g e n e t i c indi­

cators, such as iron-containing s u p e r o x i d e dismutase ( F e S O D ) and glyceraldehyde-3-phosphate dehydroge­

n a s e ( G A P D H ) in order to infer relationships within almost the s a m e sampling o f t r i c h o m o n a d taxa (Vis­

cogliosi et al., 1 9 9 6 , 1 9 9 8 ) . T h e s e molecular trees only included representatives o f the M o n o c e r c o m o n a d i d a e a n d T r i c h o m o n a d i d a e b e c a u s e o f the unavailability o f a x e n i c cultures o f representatives o f the other high level taxa. This restriction has b e e n circumvented by the use o f specific primers allowing the amplification o f SSU rRNA c o d i n g regions from s e l e c t e d eukaryotes.

T h e s e q u e n c e s o f these primers are conserved in para­

basalids and can b e used to amplify species which are difficult to culture (Mediln et al, 1988). Gunderson et al. ( 1 9 9 5 ) w e r e the first to publish a large phylogeny o f parabasalids inferred from SSU rRNA s e q u e n c e s obtained from four T r i c h o m o n a d i d a families. Several authors subsequently obtained additional s e q u e n c e s from trichomonads ( B e r c h t o l d & König, 1995; B e r c h - told et al, 1 9 9 5 ; Silberman et al, 1996; Fukura et al, 1996; Eclgcomb et al, 1998; Keeling et al, 1998; Del- gado-Viscogliosi et al, 1 9 9 9 ) and hypermastigids ( K e e ­ ling et al, 1 9 9 8 ; O h k u m a et al, 1998; D a c k s & Red- field, 1 9 9 8 ) that allow p h y l o g e n e t i c analysis o f 25 identified parabasalid taxa.

SALIENT POINTS IN THE MOLECULAR PHYLOGENY OF PARABASALIDS BASED ON SSU RRNA SEQUENCES

I

n this review, w e analyzed a broad p h y l o g e n y including all parabasalid s e q u e n c e s available in the data b a s e s to date. Parabasalid s e q u e n c e s w e r e ali­

g n e d with a set o f eukaryotic s e q u e n c e s , and phylo­

g e n e t i c inferences w e r e restricted to sites that could

284 M i s e au point Parasite, 1 9 9 9 , 6, 2 7 9 - 2 9 1

MOLECUIAR PHYLOGENY OF PARABASALIDS

b e u n a m b i g u o u s l y aligned according to conservation o f primary and s e c o n d a r y structures. T h e SSU rRNA s e q u e n c e s o f Oryza sativa, Oxytricha granulifera,

Toxoplasma gondii, Diaphonoeca grandis and Achlya bisexualis w e r e u s e d as outgroups for rooting the parabasalid tree. This data set included 1,403 positions w h i c h w e analyzed with distance matrix, m a x i m u m parsimony and m a x i m u m - l i k e l i h o o d (ML) methods.

ANALYSIS OF THE SSU RDNA SEQUENCES FROM PARABASALIDS

T h e amplification o f the SSU rDNA c o d i n g regions o f p a r a b a s a l i d s p e c i e s p r o d u c e d a DNA fragment o f around 1 5 0 0 b p in length with a G + C content o f 4 6 - 50 % e x c e p t for Dientamoeba fragilis w h i c h contains 1,600 b p in length and had a G + C content o f 3 8 % (Silberman et al, 1 9 9 6 ) . T h e l o w e r G + C content o f the Dientamoeba fragilis SSU rRNA is a c c o u n t e d for by e x t e n d e d A + T rich hypervariable regions. T h e nucleotide chain length o f parabasalid SSU rRNAs are several h u n d r e d residues shorter than the 1,800 posi­

tions typical in other eukaryotes and in correlation with previous data o n r i b o s o m e s ( C h a m p n e y et al, 1 9 9 2 ) . R i b o s o m e s from these microorganisms are atypical o f eukaryotic cells and h a v e several characteristics in c o m m o n with t h o s e o f prokaryotics. T h e y exhibit a sedimentation coefficient o f 70S w h i c h dissociated into subunits o f 50S and 30S. Moreover, ribosomal RNAs from parabasalids exhibit the m o l e c u l a r sizes typical o f 23S and l 6 S prokaryotic ribosomal RNAs (Noller, 1984; G u n d e r s o n & Sogin, 1 9 8 6 ) . T h e small size o f the ribosomal RNAs may b e consistent with the postulated early e v o l u t i o n a r y d i v e r g e n c e o f the p a r a b a s a l i d s ( C h a m p n e y et al, 1 9 9 2 ) .

BRANCHING PATTERN WITHIN THE PARABASALIDS In the ML tree s h o w n in Figure 4, parabasalids form a m o n o p h y l e t i c group as s u p p o s e d by morphological characters (Brugerolle, 1 9 7 6 ) . Within the parabasalids, w e identified four clades well supported by bootstrap values. In addition to these relatively robust groups, there are a l s o several discrete lineages comprising single s p e c i e s .

T h e first clade includes all the free-living g e n e r a o f the M o n o c e r c o m o n a d i d a e {Monotrichomonas, Ditricho-

monas and Pseudotrichomonas) and nearly all s p e c i e s from the subfamily T r i c h o m o n a d i n a e (Trichomonas vaginalis, Trichomonas tenax, Tetratrichomonas gal- linarum, Pentatrichomonas hominis, Trichomitus try- panoides and Pseudotrypanosoma giganteum). This

large g r o u p also contains the g e n u s Pentatrichomo- noides, usually p l a c e d in its o w n subfamily Pentatri- c h o m o n o i d i n a e (Brugerolle, 1 9 7 6 ) . Interestingly, Tri­

chomitus batrachorum, previously classified in the

Trichomonadinae, d o e s not b e l o n g to this large cluster.

This grouping o f the free-living M o n o c e r c o m o n a d i d a e with the T r i c h o m o n a d i n a e is in correlation with rela­

tionships inferred from morphological data. T h e Tri­

c h o m o n a d i n a e a n d t h e f r e e - l i v i n g s p e c i e s s h a r e c o m m o n morphological characters such as an undu­

lating m e m b r a n e . T h e undulating m e m b r a n e is a ribbon-like e x p a n s i o n containing a d e n s e fibre, not found in m e m b e r s o f the subfamily Tritrichomona- dinae, represented by Tritrichomonas foetus in our analysis. Among the free living m o n o c e r c o m o n a d s , Pseudotrichomonas is the most basal o f the three taxa.

T h e positioning o f Pseudotrichomas is consistent with a s t e p w i s e r e d u c t i o n o f flagellar n u m b e r . I n d e e d , Monotrichomonas possesses two flagella w h e r e a s three and four flagella o c c u r in Ditrichomonas and Pseudo­

trichomonas, respectively (Farmer, 1993; Simpson et al, unpublished observations). Within the T r i c h o m o n a ­ dinae, Pentatrichomonas exhibits the most basal posi­

tion. O n e characteristic o f this taxa is the possession o f five anterior flagella ( H o n i g b e r g et al, 1 9 6 8 ) , w h e ­ reas the o t h e r T r i c h o m o n a d i n a e s p e c i e s h a v e four anterior flagella. Pentatrichomonoides also p o s s e s s e s five anterior flagella, but the arrangement o f the basal b o d i e s is different to that o f Pentatrichomonas (Bru­

gerolle et al, 1 9 9 4 ) . According to our data, the pre­

s e n c e o f five anterior flagella is probably an a p o m o r - phic state in the T r i c h o m o n a d i n a e .

A s e c o n d clade unites the D e v e s c o v i n i d a e and the Calonymphidae, that is consistent with earlier t a x o - n o m i c divisions and evolutionary s c h e m e s b a s e d o n light- and e l e c t r o n - m i c r o s c o p i c studies ( H o n i g b e r g , 1963; Brugerolle, 1 9 7 6 ) . It also includes the unidenti­

fied Porotermes adamsoni gut symbiont s e q u e n c e as previously s h o w n (Keeling et al, 1 9 9 8 ) . A third clade contains the s p e c i e s Trichomitus batrachorum strains R105 and B U B and Hypotrichomonas acosta. LJnlike

Trichomitus which possesses a d e v e l o p e d costa, Hypo­

trichomonas is acostate but p o s s e s s e s a poorly d e v e ­ loped undulating m e m b r a n e . Several authors (Samuels, 1959; Honigberg, 1963; Mattern et al., 1969; Brugerolle, 1 9 7 1 ) have already p r o p o s e d a c o m m o n evolutionary history for these two taxa o n the basis o f ultrastruc- tural data that is in a g r e e m e n t with our molecular ana­

lysis. T h e fourth clade consists o f the three identified hypermastigid s p e c i e s from the g e n u s Trichonympha.

T h e p h y l o g e n e t i c position o f the trichomonads Tritri­

chomonas foetus, Monocercomonas sp. and Dienta­

moeba fragilis remained unresolved, as o b s e r v e d in p r e v i o u s p h y l o g e n e t i c a n a l y s e s u s i n g S S U rDNA s e q u e n c e s ( D a c k s & Redfield, 1 9 9 8 ; E d g c o m b et al, 1998; Keeling et al., 1998). With low bootstrap support, Tritrichomonas and Dientamoeba grouped together in our tree, while Monocercomonas sp. e m e r g e d inde­

pendently. Nevertheless, the grouping o f Tritricho- Parasite, 1999, 6, 279-291

Mise au point 2 8 5

Fig. 4. - Rooted phylogenetic tree of parabasalids based on the comparison of SSU rDNA sequences. The tree was constructed using the maximum likelihood method (ML) using PAUP* version 4.0.0d63. After parameter optimization, heuristic searches under ML found the best tree with InL = - 10920.34123. Bootstrap proportions are given as percentages near the individual nodes. Bootstrap values are computed by three different tree reconstruction methods: maximum likelihood (102 replicates), distance (1,000 replicates) and parsimony (1,000 repli­

cates). Asterisks designate nodes with bootstrap values below 50 %. The horizontal length of each branch is proportional to the estimated number of substitutions whereas the vertical length of each branch is arbitrary. The systematic assignment of each parabasalid taxon is indicated as established from morphological data: (M) = Monocercomonadidae; (T) = Trichomonadidae; (D) = Devescovinidae; (C) = Calo- nymphidae; (H) = Hypermastigid; (?) = Unidentified.

286 Mise au point Parasite, 1999, 6, 279-291

MOLECUIAR PHYLOGENY OF PARABASALIDS

monas with Monocercomonas sp. is found in phylo- g e n i e s b a s e d on LSU rRNA (Viscogliosi et al, 1 9 9 3 ) , F e S O D (Viscogliosi et al, 1996), GAPDH (Viscogliosi

& Muller, 1 9 9 8 ) and fumarase ( G e r b o d et al, unpu­

blished data) s e q u e n c e s , suggesting that these two genera m a y b e related. Interestingly, w e note that Tri-

trichomonas and Monocercomonas sp. e m e r g e d as sister groups o f the clade that includes Devescovinidae.

T h e D e v e s c o v i n i d a e share characters in c o m m o n with Tritrichomonas and Monocercomonas sp. from w h i c h they p r o b a b l y e v o l v e d (Honigberg, 1963; Brugerolle, 1976).

In summary, the principle clades identified in our m o l e c u l a r p h y l o g e n y , T r i c h o m o n a d i n a e , C a l o n y m - phidae, D e v e s c o v i n i d a e and the free-living M o n o c e r - c o m o n a d i d a e are consistent with classifications b a s e d on morphological analysis, although the relative bran­

ching order o f t h e s e groups and the p h y l o g e n e t i c p o s i t i o n s o f s o m e t a x a r e m a i n u n r e s o l v e d . T h i s c o n g r u e n c e b e t w e e n m o r p h o l o g i c a l and m o l e c u l a r data is only o b s e r v e d at s o m e intermediate and higher t a x o n o m i c levels. This partial c o n s i s t e n c y supports the validity o f the m o l e c u l a r a p p r o a c h and confirms the quality o f s o m e , but not all, o f the morphological cha­

racters that w e r e used. However, major surprises are also o b s e r v e d in the m o l e c u l a r analysis that provide an u n e x p e c t e d evolutionary scenario in parabasalids.

MONOCERCOMONADIDAE AND TRICHOMONADIDAE DO NOT FORM MONOPHYLETIC GROUPS

Molecular phylogenetic approaches can b e used to test t a x o n o m i c relationships originally inferred from mor­

phological data. A taxon is c o n s i d e r e d to b e natural w h e n it appears on ph yl og en et i c trees as a m o n o - phyletic clade. In our phylogenetic analysis neither the M o n o c e r c o m o n a d i d a e or the T r i c h o m o n a d i d a e are m o n o p h y l e t i c (sensu H e n n i g ) , at least o n e o f the two being polyphyletic. As stated above, the principle taxo­

n o m i c character separating these two families is the p r e s e n c e / a b s e n c e o f a costa. This fiber is present in T r i c h o m o n a d i d a e w h e r e a s it is absent in the M o n o ­ c e r c o m o n a d i d a e . Within the T r i c h o m o n a d i d a e , m e m ­ bers o f the Tritrichomonadinae such as Tritrichomonas foetus p o s s e s s C I - t y p e costa while most m e m b e r s o f

the T r i c h o m o n a d i n a e such as Trichomonas vaginalis p o s s e s s C-type costa. Only the T r i c h o m o n a d i n a e Tri- chomitus batrachorum exhibits a C l - t y p e costa. T h e C-type costa found is usually presumed to have derived from the "primitive" Cl-type costa b e c a u s e o f its higher ultrastructural c o m p l e x i t y (Viscogliosi & Brugerolle, 1994). This is supported by the earlier e m e r g e n c e o f Tritrichomonas foetus and Trichomitus batrachorum in o u r p h y l o g e n e t i c t r e e . A c c o r d i n g to b i o c h e m i c a l , i m m u n o l o g i c a l and preliminary molecular data (Vis­

cogliosi & Brugerolle, 1994; Viscogliosi & Delgado-Vis­

cogliosi, 1 9 9 6 ) , c o s t a proteins c o m p o s i n g the C- a n d C l - t y p e c o s t a e p r o b a b l y b e l o n g to the s a m e protein family. T h u s the polyphyly o f the T r i c h o m o n a d i d a e w h i c h w o u l d require multiple origins for the costa, is unlikely and we suggest the alternative o f polyphyly o f the M o n o c e r c o m o n a d i d a e . Since all M o n o c e r c o ­ m o n a d i d a e studied to date are sister g r o u p s o f taxa possessing a costa ( T r i c h o m o n a d i d a e ) , it a p p e a r s that M o n o c e r c o m o n a d i d a e are p r o b a b l y s e c o n d a r i l y a c o - state and that the a p p a r e n t simplicity o f their c y t o s - k e l e t o n w a s misleading b o t h as a primitive c h a r a c t e r and as a significant t a x o n o m i c criterion. Similarly, w e also n o t e in o u r analysis that b o t h D e v e s c o v i n i d a e and C a l o n y m p h i d a e do not c o m p o s e m o n o p h y l e t i c groups.

EARLY EMERGENCE OF HYPERMASTIGIDS

T h e early e m e r g e n c e o f the hypermastigids, the para­

basalids with a h y p e r d e v e l o p e d cytoskeleton, classi­

cally c o n s i d e r e d to b e a late-evolving lineage, c a u s e s us to reconsider the polarization o f characters in c o m ­ parative morphological analyses. In our SSU rDNA phylogeny, the hypermastigids o f the genus Tricho- nympha represent the deepest branch, followed b y the Reticulitermes flavipes'gyA symbiont 1. This root (or the most primitive taxon o f the g r o u p ) is well s u p p o r t e d by bootstrap resampling. Hypermastigids have tradi­

tionally b e e n c o n s i d e r e d to h a v e arisen from the D e v e s c o v i n i d a e (Brugerolle, 1 9 7 6 ) . H o w e v e r , they are not specifically related to this family in our analysis.

T h e i d e n t i f i c a t i o n o f t h e u n k n o w n p a r a b a s a l i d s e q u e n c e Reticulitermes flavipes gut symbiont 1 remains unresolved although the hypermastigids Spirotricho- nympha, Microjoenia or Holomastigotes are the most likely sources o f this s e q u e n c e ( Y a m i n , 1 9 7 9 ) . In pre­

viously published analyses ( G u n d e r s o n et al, 1 9 9 5 ; Dacks & Redfield, 1998; Silberman et al, 1996), the ear­

liest b r a n c h o f the parabasalids is the unidentified symbiont 1 from Reticulitermes flavipes, which is fol­

l o w e d by Trichonympha. Keeling et al. ( 1 9 9 8 ) have found that the root o f the parabasalid g r o u p w a s dif­

ficult to discern accurately but p r o p o s e d a root s o m e ­ w h e r e a m o n g the b r a n c h e s leading to Trichonympha and unidentified putative hypermastigids. Although the e x a c t branching pattern for the basal lineages is not clear yet, all are hypermastigids or putative hyper­

mastigids. In sum, the molecular data indicate that the cytoskeletal c o m p l e x i t y o f parabasalids, while useful as a t a x o n o m i c trait, is not a reliable phylogenetic indi­

cator, a n d that their traditional polarization, from

"simple" to " c o m p l e x " must b e turned almost upside- down. A parallel can b e drawn b e t w e e n these results and t h o s e obtained in other groups o f protists such as the ciliates ( B a r o i n - T o u r a n c h e a u et al, 1 9 9 2 ) , w h i c h suggest that a c o m p l e x oral apparatus may b e a ple-

Parasite, 1999, 6, 279-291

Mise au point 287

siomorphic state. C o m p l e x characters may not always b e a p o m o r p h i c states.

FREE-LIVING MONOCERCOMONADS ARE DESCENDED FROM ENDOBIOTIC ANCESTORS

Ditrichomonas, Pseudotrichomonas and Monotricho- monas are free-living organisms isolated from sediment samples. T h e s e free-living organisms w e r e logically thought to antedate t h o s e o f a parasitic or symbiotic habit. Until now, phylogenetic analyses o f parasitic pro- tist clades such as the diplomonads (Brugerolle, 1 9 7 5 b ) a p p e a r to c o r r o b o r a t e the generalization o f the irre­

versibility o f parasitism. Brugerolle & Taylor ( 1 9 7 7 ) and Vickerman ( 1 9 8 9 ) assume the free-living status to b e p l e s i o m o r p h i c for the d i p l o m o n a d s , in a c c o r d a n c e with the notion that parasitism c a n n o t b e reversed.

H o w e v e r Siddall et al. ( 1 9 9 3 ) , o n the basis o f phylo­

genetic trees inferred from ultrastructural characters, propose that parasitism is plesiomorphic for this group and invoke a reversal from parasitism to a free-living strategy in descendants to explain the o b s e r v e d pat­

terns. In our SSU rDNA phylogeny, the free-living parabasalid clade e m e r g e s as the sister group o f the T r i c h o m o n a d i n a e with high bootstrap support. In s h o ­ wing the free-living s p e c i e s to b e positioned "high" in the parabasalid tree, it strongly supports the hypothesis that they are d e s c e n d e d from e n d o b i o t i c ancestors a n d , t h u s , a p p e a r to b e s e c o n d a r i l y f r e e - l i v i n g . Moreover, the m o n o p h y l y o f the free-living parabasa- lids invokes a single reversal event from parasitism to a free-living strategy. It also m e a n s that s o m e paraba- salids are able to pursue a free-living m o d e in microae- robic sediments without mitochondria and that primi­

tively free-living parabasalids have not b e e n found yet.

POLYPHYLY OF THE GENUS TRICHOMITUS

T h e g e n u s Tricbomitus, and particularly the type s p e ­ cies Tricbomitus batracbomm (frog enclosymbiont), has b e e n extensively studied and characterized b y light and e l e c t r o n m i c r o s c o p y ( B r u g e r o l l e , 1 9 7 1 ; Honigberg et al, 1 9 7 2 ) . T h e strains R105 and B U B w e r e positively identified as different strains o f Tri­

cbomitus batrachorum. T h e SSU rDNA s e q u e n c e s o f these two strains only differ in two positions ( 5 2 7 and 1332 b a s e s from the 5' e n d ) and they group c o n s i s ­ tently with the m o n o c e r c o m o n a d Hypotrichomonas acosta as already predicted (Samuels, 1 9 5 9 ) . This phy­

logenetic positon differs markedly from that o f Tri­

cbomitus trypanoides (termite-gut s y m b i o n t ) w h i c h is the sister group o f a n o t h e r termite-dwelling species, Peiitatrichomonoides scroa. From this data it is clear that Tricbomitus batrachorum and Tricbomitus trypa­

noides are n o n - c o n g e n e r i c . T a k i n g these molecular data into consideration, it is possible to repeat a c o m ­ parative morphological analysis o f these two s p e c i e s

and find several ultrastnictural differences that appeared as minor features. For instance, w e note differences b e t w e e n these taxa in the length o f the costa, the form o f the parabasal b o d y and the possible p r e s e n c e o f a structure associated with the costa, the c o m b ( B r u g e ­ rolle, 1971; Honigberg et al, 1972; Boykin et al, 1986).

T h e s e differences reflect polyphyly o f the genus Tri­

cbomitus, and therefore it would b e logical to consider such characters in further comparative structural ana­

lyses. Interestingly w e note that the devescovinid genus Metadevescovina is also polyphyletic in our SSU rDNA analysis but the distance o b s e r v e d b e t w e e n the t w o Metadevescovina species is shorter than that calculated b e t w e e n the two Tricbomitus species.

Tricbomitus trypanoides was originally classified in the genus Trichomonas ( D u b o s c q & Grasse, 1924) but authors around that time usually assigned all paraba­

salids with an undulating m e m b r a n e to the genus Tri­

chomonas. Moreover, Tricbomitus trypanoides differs from Trichomonas b y several ultrastructural characters and, thus, it should b e classified as a n e w genus o f T r i c h o m o n a d i n a e . Pending revision o f the g e n u s Tri­

cbomitus, w e refer to the genus Tricbomitus as e n c o m ­ passing m e m b e r s m o r e closely related to Tricbomitus batrachorum (the type species). Finally it must also b e indicated that Tricbomitus batrachorum has b e e n pre­

viously classified in the T r i c h o m o n a d i n a e , in conflict with our phylogenetic tree. In fact Tricbomitus batra­

chorum exhibits characters from both T r i c h o m o n a ­ dinae (undulating m e m b r a n e ) and Tritrichomonadinae ( C l - t y p e costa). T h e early e m e r g e n c e o f Tricbomitus batrachorum at the b a s e o f the Trichomonadinae-Tri- trichomonadinae dichotomy is therefore not really sur­

prising and is in a g r e e m e n t with previous molecular phylogenies b a s e d o n LSU rRNA (Viscogliosi et al, 1 9 9 3 ) and GAPDH s e q u e n c e s (Viscogliosi & Miiller, 1 9 9 8 ) , emphasizing its crucial evolutionary position a m o n g the T r i c h o m o n a d i d a e .

IS THERE CORRELATION BETWEEN THE PHYLOGENY OF PARABASALIDS AND THEIR HOSTS?

W e also studied the phylogenetic position o f paraba­

salids isolated from various hosts in an attempt to gain further insights into the evolution o f parasitism a m o n g these protists. Our analysis provides n o e v i d e n c e for a long-term c o - e v o l u t i o n o f parabasalids and their hosts, either vertebrate or invertebrate. T h e s e o b s e r ­ vations suggest that evolution o f parabasalids w a s a c c o m p a n i e d b y multiple c h a n g e s o f hosts and habi­

tats. For instance, the early e m e r g e n c e o f Tritricho- monas foetus and Dientamoeba fragilis (mammalian parasites), together with the late e m e r g e n c e o f the insect gut inhabitants Tricbomitus trypanoides, Penta- trichomonoides scroa and Pseudotrypanosoma gigan- teum in our SSU rDNA phylogeny, is intriguing evi-

2 8 8 Mise au point Parasite, 1999, 6, 279-291

MOLECULAR PHYLOGENY OF PARABASALIDS

d e n c e against t h e theoretical co-evolution o f these organisms a n d their animal hosts as already suggested by Silberman et al. ( 1 9 9 6 ) . H o w e v e r o u r data c a n n o t e x c l u d e the possibility o f a limited (short-term) c o - e v o - lution o f hosts a n d parasites. Within the m o n o p h y l e t i c clade o f the Trichomonadinae, which includes mamma- lian ( T r i c h o m o n a s vaginalis, Trichomonas tenax a n d Pentatrichomonas hominis), bird (Tetratrichomonas gallinarum) a n d termite hosts (Trichomitus trypa-

noides, Pentatrichomonoides scroa a n d Pseudotrypa- nosoma giganteum), t h e parasitic parabasalid lineages o f this g r o u p e m e r g e in an order reflecting t h e e v o - lutionary history o f their hosts, with t h e e x c e p t i o n o f Pentatrichomonas hominis.

CONCLUSION

I

n recent molecular studies, there has been a ten- dency to believe that evolutionary reconstructions have resolved all major questions regarding para- basalid evolution. In this review we show that some uncertainties remain in the molecular phylogeny of parabasalids, such as the position of several taxa and the relative branching order of the major groups of parabasalids. The inclusion of several as yet untou- ched taxa in molecular phylogenetic analyses, parti- cularly termite symbionts, could help to address these issues. Molecular phylogenies as inferred from SSU rRNA sequences are providing some important and sometimes unexpected insights into the evolution of the parabasalid group. Despite some encouraging congruencies, the molecular data conflict with the established systematics based on a limited number of morphological characters and, broadly, a simple-to- complex polarisation of evolution. One of the best examples of these discrepancies is the probable poly- phyly o f the M o n o c e r c o m o n a d i d a e . M o l e c u l a r approaches allow us to follow the elaboration of the cytoskeleton in the parabasalid lineage, and in some taxa, to shed light on undetected secondary losses of cytoskeletal structures or to propose an increased

"weight" for some ultrastructural characters in further analyses.

It is true that many systematists have emphasized the conflicts and deemphasized the concordance between inferences based on molecular data and those based on morphology, although each approach has distinct virtues and pitfalls. However studies that incorporate both molecular and morphological data will provide much better descriptions and interpretations of biolo- gical diversity than those that focus on just one approach. The necessary revision o f the parabasalid taxonomy should take into account both molecular and morphological data in the near future.

Parasite, 1999, 6, 279-291

REFERENCES

BAROIN-TOURANCHEAU A., DELGADO P., PERASSO R. & ADOI.TTE A .

A broad molecular phylogeny of ciliates: identification of major evolutionary trends and radiations within the phylum. Proceedings of the National Academy of Sciences USA, 1992, 89, 9 7 6 4 - 9 7 6 8 .

BERCHTOLD M. & KÖNIG H . Phylogenetic position of the two uncultivated trichomonads Pentatrichomonoides scroa Kirby and Metadevescovina extranea Kirby from the hindgut of the termite Mastotermes darwiniensis Froggatt.

Systematic and Applied Microbiology, 1995, 18, 567-573.

BERCHTOLD M., BREUNIG A . & KÖNIG H . Culture and phyloge-

netic characterization of Trichomitus trypanoides Duboscq

& Grasse 1924, n. comb: a trichomonad flagellate isolated from the hindgut of the termite Reticulitermes santonensis Feytaud. Journal of 'Eukaryotic Microbiology, 1995, 42, 388- 391.

BOYKIN M.S., STOCKERT L., BUHSE H.E. & SMITH-SOMERVILLE H.E.

Trichomitus trypanoides (Trichomonadida) from the ter- mite Reticulotermes flavipes. II. Fine structure and identi- fication of the cloned flagellate. Transactions of the Ame- rican Microscopical Society, 1986, 105, 223-238.

BREZNAK J . A . Intestinal microbiota of termites and other xylo- phagous insects. Annual Review of Microbiology, 1982, 3 6 , 323-343.

BRUGEROLLE G. Ultrastructure du genre Trichomitus Swezy 1915, Zooflagellata, Trichomonadida. Protistologica, 1971, 7, 171-176.

BRUGEROLLE G. Étude de la cryptopleuromitose et de la mor- phogénèse de division chez Trichomonas vaginalis et chez plusieurs genres de Trichomonadines primitives. Pro- tistologica, 1975a, 11, 457-468.

BRUGEROLLE G. Contribution à l'étude phylétique des diplo- zoaires (Zoomustigophorea. Diplozoa. Dangeard. 1 9 1 0 ) . VI. Caractères généraux des diplozoaires. Protistologica, 1975b, 11, 111-118.

BRUGEROLLE G. Cytologie ultrastructurale, systématique et évolution des Trichomonadida. Annales de la Station Bio- logique de Besse-en-Chandesse, 1976, 10, 1-90.

BRUGEROLLE G. & TAYLOR F.J.R. Taxonomy, cytology and evo- lution of the Mastigophora, in: Proceedings of the Fifth International Congress of Protozoology, 1977, 14-28.

BRUGEROLLE G., BREUNIG A . & KÖNIG H . Ultrastructural study

of Pentatrichomonoides sp., a trichomonad flagellate from Mastotermes darwiniensis. European Journal of Protisto- logy, 1994, 30, 372-378.

Bui E.T.N., BRADLEY P . J . & JOHNSON P . J . A common evolutio- nary origin for mitochondria and hydrogenosomes. Pro- ceedings of the National Academy of Sciences USA, 1996, 93, 9651-9656.

CAVALIER-SMITH T. Kingdom Protozoa and its 18 phyla. Micro- biological Reviews, 1993, 57, 953-994.

CHAMPNEY W.S., CHITTUM H.S. & SAMUELS R. Ribosomes from

trichomonad protozoa have prokaryotic characteristics.

International Journal of Biochemistry, 1992, 24, 1125- 1133.

Mise au point 289

VISCOGLIOSI E., E D G C O M B V.P., G E R B O D D., NOEL C & D E L G A D O - V I S C O G L I O S I P.

CORLISS J.O. The kingdom Protista and its 45 phyla. BioSys- tems, 1984, 17, 87-126.

DACKS J . B . & REDFIELD R.J. Phylogenetic placement of Tri- chonympha. Journal of Eukaryotic Microbiology, 1998, 45, 445-447.

DELGADO-VISCOGLIOSI P., VISCOGLIOSI E . , GERBOD D., KULDAJ., SOGIN M.L. & EDGCOMB V.P. A comprehensive molecular phylogeny of parabasalids based on small subunit rRNA sequences : an unexpected evolutionary scenario. Journal of Eukaryotic Microbiology, 1999, in press.

DUUOSCQ O. & GRASSE P-P. Notes sur les protistes parasites des termites de France. I. Trichomonas trypanoides, n. sp.

Comptes-rendus des Séances de la Société de Biologie, Paris, 1924, 90, 547-550.

DYER B.D. Phylum Zoomastigina Class Parabasalia, in: Hand- book of Protoctista. Margulis L., Corliss J.O., Melkonian M. & Chapman D J . (eds), Jones & Bartlett, Boston, 1989, 252-258.

EDGCOMB V.P., VISCOGLIOSI E., SIMPSON A.G.B., DELGADO-VIS- COGLIOSI P., ROGER A J . & SOGIN M.L. New insights into the phylogeny of trichomonads inferred from small subunit rRNA sequences. Protist, 1998, 149, 359-366.

FARMER M.A. Ultrastructure of Ditrichomonas honigbergii n. g., n. sp. (Parabasalia) and its relationship to amitochondrial protists. Journal of Eukaryotic Microbiology, 1993, 40, 619- 626.

FUKURA K., YAMAMOTO A., HASHIMOTO T. & GOTO N. Nucleo- tide sequence of the SrRNA gene and phylogenetic ana- lysis of Trichomonas tenax. Microbiology and Immunology, 1996, 40, 183-188.

GERMOT A., PHILIPPE H. & LE GUYADER H. Presence of a mito- chondrial-type 70-kDa heat shock protein in Trichomonas vaginalis suggests a very early mitochondrial endosym- biosis in eukaryotes. Proceedings of the National Aca- demy of Sciences USA, 1996, 93, 14614-14617.

GERMOT A., PHILIPPE H. & LE GUYADER H. LOSS of mitochon- dria in microsporidia: evidence from a mitochondrial-type HSP-70 in Nosema locustae. Molecular and Biochemical Parasitology, 1997, 87, 159-168.

GRAIN J . The cytoskeleton in protists: nature, structure and function. International Review of Cytology, 1986, 104, 153- 249.

GRAVES A. & GARDNER W.A. Pathogenicity of Trichomonas vaginalis. Clinical Obstetrics and Gynecology, 1993, 36, 145-152.

GROSOVSKY B.D.D. & MARGULIS L. Termite microbial commu- nities, in: Experimental Microbial Ecology. Bums R . G &

Slater J.H. (eds), Oxford, Blackwell Scientific, 1982, 519- 532.

GUNDERSON J.H. & SOGIN M.L. Length variations in eukaryotic rRNAs: small subunit rRNAs from the protists Acantha- moeba castellani and Euglena gracilis. Gene, 1986, 44, 63- 70.

GUNDERSON J . , HINKLE G , LEIPK D., MORRISON H.G., STICKEL S.K., ODELSON D.A., BREZNAK J.A., NERAD T.A., MÏJLLER M. &

SOGIN M.L. Phylogeny of trichomonads inferred from small- subunit rRNA sequences. Journal of Eukaryotic Microbio- logy, 1995, 42, 411-415.

HEINE P. & MCGREGOR J.A. Trichomonas vaginalis: a ree- merging pathogen. Clinical Obstetrics and Gynecology, 1993, 36, 137-144.

HIRT R.P., HEALY B., VOSSBRINCK C.R., CANNING E.U. & EMBLEY T.M.

A mitochondrial HSP70 orthologue in Vairimorpha neca- trix. molecular evidence that microsporidia once contained mitochondria. Current Biology, 1997, 7, 995-998.

HOLLANDE A. & CARRUETTE-VALENTIN J . Les atractophores, l'induction du fuseau et la division cellulaire chez les Hypermastigines, étude infrastructurale et révision systé- matique des Trichonymphines et des Spirotrichonym- phines. Protistologica, 1971, 7, 5-100.

HONIGBERG B.M. Evolutionary and systematic relationships in the flagellate order Trichomonadida Kirby. Journal of Pro- tozoology, 1963, 10, 20-63.

HONIGBERG B.M., MATTERN C.F.T. & DANIEL W.A. Structure of Pentatrichomonas hominis as revealed by electron micro- scopy. Journal of Protozoology, 1968, 15, 419-430.

HONIGBERG B.M., DANIEL W.A. & MATTERN C.F.T. Fine struc- ture o f Trichomitus batrachorum (Petty). Journal of Pro- tozoology, 1972, 19, 446-453.

HONIGBERG B.M. Trichomonads of importance in human médecine, in. Parasitic Protozoa. Kreier J.P. (ed), Academic- Press, New York, 1978, vol. II, 275-454.

HONIGBERG B.M. & BRUGEROLLE G. Structure, in. Trichomonads parasitic in humans. Honigberg B.M. (ed), Springer-Verlag, New York, 1990, 5-35.

KEELING P.J., POULSON N . & MCFADDEN G.I. Phylogenetic diver- sity of parabasalian symbionts from termites, including the phylogenetic position of Pseudotrypanosoma and Tricho- nympha. Journal of Eukaryotic Microbiology, 1998, 45,643- 650.

KIRBY B. Devescovinid flagellates o f termites. I. The genus Devescovina. University of California Publications in Zoo- logy, 1941, 45, 1-92.

KULDAJ., NOHYNKOVA E. & LUDVIKJ. Basic structure and func- tion of the trichomonad cell. Acta Universitatis Carolinae- Biologica, 1986, 30, 181-210.

LEE J J . Order 5. Trichomonadida, in: Illustrated Guide to the Protozoa. Lee J . J . , Hutner S.H. & Bovee E.G. (eds), Allen Press, Lawrence, Kansas, 1985, 119-127.

LEIPE D.D., GUNDERSON J.H., NERAD T.A. & SOGIN M.L. Small subunit ribosomal RNA+ of Hexamita inflata and the quest for the first branch in the eukaryotic tree. Molecular and Biochemical Parasitology, 1993, 59, 41-48.

LEVINE N.D., CORLISS J . O , Cox F.E., DEROUX G , GRAIN J . , HONIG- BERG B.M., LEEDALE G.F., LOEBLICH A.R.D., LOM J . , LYNN D., MERINFELD E.G., PAGE F.C, POLJANSKY G , SPRAGUE V , VAVRA J .

& WALLACE F.G. A newly revised classification of the pro- tozoa. Journal of Protozoology, 1980, 27, 37-58.

MATTERN C.F.T., DANIEL W.A. & HONIGBERG B.M. Structure of Hypotrichomonas acosta (Moskowitz) (Monocercomona- didae, Trichomonadida) as revealed by electron micro- scopy. Journal of Protozoology, 1969, 16, 668-685.

MEDLIN L., ELWOOD H . J . , STICKEL S. & SOGIN M.L. The charac- terization of enzymatically amplified eukaryotic l6S-like rRNA coding regions. Gene, 1988, 71, 491-499.

290 Mise au point Parasite, 1 9 9 9 , 6, 2 7 9 - 2 9 1

NOLLER H.F. Structure of ribosomal RNA. Annual Review of Biochemistry, 1984, 53, 119-162.

OHKUMA M., OHTOKO K., GRÜNAU C , MORIYA S. & KUDO T .

Phylogenetic identification of the symbiotic hypermastigote Trichonympha agilis in the hindgut of the termite Reti- culitermes speratus based on small-subunit rRNA sequence.

Journal of Eukaryotic Microbiology, 1998, 45, 439-444.

OLSEN G.J. & WOF.SE C R . Ribosomal RNA: a key to phylo- geny. FASEBJournal, 1993, 7, 113-123.

PACE N.R., OLSEN G.J. & WOESE C R . Ribosomal RNA phyto­

geny and the primary lines of evolutionary descent. Cell, 1986, 45, 325-326.

PECKA Z., NOHYNKOVA E. & KULDA J . Ultrastructure of Cochlo- soma anatis Kotlan, 1923 and taxonomic position of the family Cochlosomatidae (Parabasala: Trichomonadida).

European Journal of Protistology, 1996, 52, 190-201.

PEYRETAILLADE E., BROUSSOLE V., PEYRET P., METENIER G., GOUY M.

& VIVARES C P . Microsporidia, amitochondriate protists, possess a 70-kDa heat shock protein gene of mitochon­

drial origin. Molecular Biology and Evolution, 1998, 15, 683-689.

PHILIPPE H. & ADOUTTE A. How reliable is our current view of eukaryotic phytogeny?, IN. Protistological Actualities, Proceedings of the Second European Congress of Protis­

tology. Brugerolle G. & Mignot J.P. (eds), 1995, 17-33.

ROGER A.J., CLARK C G . & DOOLITTLE W . F . A possible mito­

chondrial gene in the early-branching amitochondriate protist Trichomonas vaginalis. Proceedings of the National Academy of Sciences USA, 1996, 93, 14618-14622.

ROGER A.J., SVÄRD S.G., TOVAR J . , CLARK C.G., SMITH M.W., GILLIN F . D . & SOGIN M.L. A mitochondrial-like chape- ronin 60 gene in Giardia lamblia: Evidence that diplo- monads once harbored an endosymbiony related to the progenitor of mitochondria. Proceedings of the National Academy of Sciences PSA, 1998, 95, 229-234.

SAMUELS R. Studies of Tritrichomonas batrachorum. 3. Abnormal mitosis and morphogenesis. Transactions of the American Microscopical Society, 1959, 78, 49-65.

SIDDALL M.E., BROOKS D.R. & DESSER S.S. Phytogeny and the

reversibility of parasitism. Evolution, 1993, 47, 308-313.

SILBERMAN J.D., CLARK C G . & SOGIN M.L. Dientamoeba fra- gilis shares a recent common evolutionary history with the trichomonads. Molecular and Biochemical Parasitology, 1996, 76, 311-314.

SOGIN M.L. Early evolution and the origin of eukaryotes. Cur­

rent Opinion in Genetics and Development, 1991, 1, 457- 463.

SORVILLO F . & KERNDt P. Trichomonas vaginalis and amplifi­

cation of HIV-1 transmission. The Lancet, 1998, 351, 213- 214.

TAYLOR F.J.R. Problems in the development of an explicit hypothetical phytogeny of the lower eucaryotes. BioSys- tems, 1978, 10, 67-89.

VAN KEULEN H., GUTELL R.R., GATES M.A., CAMPBELL S.R., ERLANDSEN S.L., JARROLL EX., KULDA J . & MEYER E.A. Unique

phylogenetic position of Diplomonadida based on the complete small subunit ribosomal RNA sequence o f Giardia ardeae, G. muris, G. duodenalis and Hexamita sp. FASEB Journal, 1993, 7, 223-231.

VICKERMAN K. Phylum Zoomastigina Class Diplomonadida, in:

Handbook of Protoctista. Margulis L., Corliss J.O., Melko- nian M. & Chapman D.J. (eds), Jones & Bartlett, Boston, 1989, 200-210.

VISCOGLIOSI E., PHILIPPE H., BAROIN A., PERASSO R. & BRUGE­

ROLLE G. Phytogeny of trichomonads based on partial sequences of large subunit rRNA and on cladistic analysis of morphological data. Journal of Eukaryotic Microbiology, 1993, 40, 411-421.

VISCOGLIOSI E. & BRUGEROLLE G. Striated fibers in trichomo­

nads: costa proteins represent a new class of proteins for­

ming striated roots in protists. Cell Motility and the Cytos- keleton, 1994, 29, 82-93.

VISCOGLIOSI E. & DELGADO-VISCOGLIOSI P. Cytoskeleton, divi­

sion and evolution in a primitive group of protists, the tri­

chomonads. Année Biologique, 1996, 35, 146-162.

VISCOGLIOSI E., DURIEUX I., DELGADO-VISCOGLIOSI P., BAYLE D.

& DIVE D. Phylogenetic implication of iron-containing superoxide dismutase genes from trichomonad species.

Molecular and Biochemical Parasitology, 1996, 80, 209- 214.

VISCOGLIOSI E. & MÜLLER M. Phylogenetic relationships of the glycolytic enzyme, glyceraldehyde-3-phosphate dehydro­

genase, from parabasalid flagellates. Journal of Molecular Evolution, 1998, 47, 190-199.

WOESE CR. Bacterial evolution. Microbiological Reviews, 1987, 51, 221-271.

YAMIN M.A. Flagellates of the orders Trichomonadida Kirby, Oxymonadida Grasse, and Hypermastigida Grassi & Foà reported from lower termites (Isoptera families Mastoter- mitidae, Kalotermitidae, Hodotermitidae, Termopsidae, Rhinotermitidae, and Serritermitidae) and from the wood- feeding roach Cryptocercus (Dictyoptera: Cryptocercidae).

Sociobiology, 1979, 4, 5-119.

YULE A., SKIRROW S.Z. & BONDURANT R.H. Bovine Trichomo­

niasis. Parasitology Today, 1989, 5, 373-377.

ZUCKERKANDL E. & PAULING L. Molecules as documents of evo­

lutionary history. Journal of Theoretical Biology, 1965, 8, 357-366.

Reçu le 6 juillet 1999 Accepté le 22 octobre 1999

291

MOLECULAR PHYLOGENY OF PARABASALIDS

P a r a s i t e , 1 9 9 9 , 6, 2 7 9 - 2 9 1

Mise ou point