Contributions to Zoology, 70 (3) 127-138 (2001)

SPB Academic Publishing bv, The Hague

Soil-dwelling polychaetes: enigmatic as ever? Some hints on their

phylogenetic relationships as suggested by a maximum parsimony analysis of 18S rRNA gene sequences

³ Emilia Rota Patrick Martin² & Christer Erséus ¹,

1 di Dipartimento Biologia Evolutivei. Universitd di Siena, via P. A. Mattioli 4. IT-53100 Siena, Italy, e-mail:

2 Institut des Sciences naturelles de des [email protected]; royal Belgique, Biologic Eaux donees, 29 rue Vautier,

B-1000 e-mail: 3 Bruxelles, Belgium, [email protected]; Department of Invertebrate Zoology, Swedish Museum of Natural History, Box 50007, SE-104 05 Stockholm, Sweden, e-mail: [email protected]

Keywords: Terrestrial Polychaeta, Parergodrilus heideri, Stygocapitella subterranea, Hrabeiella

I8S rRNA periglandulata, gene, molecular phylogeny, rapid radiation

Abstract Collectionof new specimens 130

DNA extraction, amplification and sequencing 130

Alignment To re-evaluate 130 the various hypotheses on the systematic position of Phylogenetic analyses 130 Parergodrilus heideri Reisinger, 1925 and Hrabeiella Results 132 periglandulata Pizl & Chalupský, 1984,the sole truly terrestrial Discussion 132 non-clitellateannelidsknown to date, their phylogenetic relation- ships Acknowledgements 136 were investigated using a data set of new 18S rDNA References 136 of sequences these and other five relevant annelid taxa, including

an unknown of species Ctenodrilidae, as well as homologous

sequences available for 18 already polychaetes, one aphano- neuran, 11 clitellates, two pogonophorans, one echiuran, one Introduction

sipunculan, three molluscs and two arthropods. Two different

alignments were constructed, according to analgorithmic method terrestrial forms constitute (Clustal Truly a tiny minority W) and on the basis of a secondary structure model non-clitellate annelids, (DCSE), A maximum parsimony analysis was performed with among only represented by

arthropods asan unambiguous outgroup. With both alignments, Parergodrilus heideri Reisinger. 1925 and Hra- the resulting topology confirms the validity of groupingP. heideri beiella Pizl & periglandulata Chalupsky, 1984 over and Stygocapitella subterranea Knöllner, 1934 into the family a total of 12,000 species (estimate of described Parergodrilidae. Hrabeiellaperiglandulatanever clusters with them ‘Polychaeta’ acc. to K. Fauchald, in Minelli, 1993). and itsposition relative to this and other polychaete fami- lies is This and their unorthodox still obscure, but a close relationship with aphanoneurans disproportion morphol-

is suggested the most trees. All these have caused these two to be by parsimonious taxa appear ogy worms omitted

to be far fromthe Clitellata. Most relationships polychaetes from the among most zoology textbooks. In case of P. ar e not supported by significant bootstrap and Bremer values. 75 heideri, known years ago, such an omis- These already polytomies are corroborated by independent evidence sion is also and partly explained by its ever ambigu- are interpreted as from ancient and resulting an emergence a ous Rouse & Fauchald rapid radiation ofPolychaeta. phylogenetic position. (1997: 141) recently stated: “The morphological simplic-

ity and the lack of essential information for this

Contents group [the Parergodrilidae] mean that their place-

ment is unresolvable at present”. Indeed, owing to

Abstract their unusual habitat (inland 127 soils, mainly in for- Introduction 127 ests), restricted biogeography (middle Europe) and Material and methods 129 meiofaunal body dimensions (1-2 mm, with 8 to Selection oftaxa 129 14 chaetigerous segments), the biology and mor-

is the of paper dedicatedto memory Professor Tor Gustav Karling, 1909-98 128 E. Rota, P. Martin & C. Erseus - Phytogeny of soil-dwelling polychaetes

the phology of these two taxa were rarely elaborated sistency in supposedly typical ‘archiannelid’ first-hand by polychaete specialists. The bulk of characters {Stygocapitella has an internal nerve cord, information was instead produced and discussed well developed circular muscles, no locomotory

terricolous ciliation of and no adhesive by students of different worm groups. epidermis, pygidial of the Erich Reisinger, a specialist of terrestrial flat- glands) precluded a close relationship two

first discovered P. heideri in beech for- thus concluded that worms, taxa. Karling Stygocapitella

should indeed be the ests in Stciermark, Austria. Reisinger (1925, 1929) placed among Polychaeta

in initially assigned this worm (erroneously interpreted Sedentaria Drilomorpha (which Hatschek’s sys-

the as hermaphroditic) to a monotypic family (Parergo- tem included Cirratulidae, Arenicolidae, drilidae) in the Archiannelida, suborder “Rotatorio- Capitellidae, Maldanidae, Ctenodrilidae and Sterna-

but gona”. In the latter, he included the Nerillidae, spidae; see Fauchald & Rouse, 1997), segre-

Histriobdellidae and Dinophilidae, which also (fol- gated in a monotypic family, the Stygocapitellidae, lowing Heider, 1922) he considered to be anatomi- close to the Ctenodrilidae and the Capitellidae. cally intermediate between the rotifers and the Whether Parergodhlus should be accommodated annelids. Reisinger’s work was scathingly criticized in the same family Karling left to Reisinger to

the by Meyer (1927), who presented a different analysis decide. In meanwhile, Reisinger had discovered of the structure of Parergodrilus suggesting that the males of Parergodrilus and was preparing an

of the it is a reduced enchytraeid oligochaete. The worm updated comparative study reproductive

30 of the was not reinvestigated during the following systems two species. Reisinger (1960) ac-

but the of authorities (Michaelsen, to years, majority cepted Karling’s suggestion unify Stygocapitel-

lidae 1928; Stephenson, 1930; Cernosvitov, 1937; Du and Parergodrilidae but correctly gave prior-

Bois-Reymond Marcus, 1948: 8, footnote) favoured ity to the latter name, and the redefined family,

Meyer's view. following the dismemberment of the taxon Archian-

In the meantime, another minute annelid, Stygo- nelida (Beklemischev, 1958), he now placed un-

had been the capitella subterranea Knollner, 1934, der Polychaeta Sedentaria, at an intermediate discovered from littoral subsoil waters in the Kiel level between the Drilomorpha and the Nerillidae.

Bay, Germany. In spite of its homonomous seg- Interestingly, in the Grasse volume on Annelida, mentation, the lack of dorsal chaetal bundles and Parergodrilus was ‘for practical reasons’ still re- the aberrant ventral location of its sexual opening, ferred to the archiannelids (Beauchamps, 1959),

the it was attributed to polychaete family Capitel- but treated also under the enchytraeids (Avel, 1959).

lidae, In 1955, Tor Karling collectedS. subterranea in spite of subsequent research, a more precise

has on the coast of Scania, Sweden, and immediately classification of the Parergodrilidae not as yet perceived its intriguing resemblance to P. heideri. been possible. Dales (1962, 1963), in consideration

Karling (1958) carried out a detailed anatomical of the structure of the pharyngeal organ, suggested

ndn- to study of the two taxa, adding support to the the ‘curious little stygocapitellids’ be an early clilellate nature of Parergodrilus and substantiat- offshoot from the ancestral stock of which the ing its affinities with Stygocapitella. He regarded ‘archiannelids’ of today are specialized survivors.

maintainedtheir but the position of S. subterranea in the Capitellidae He position as uncertain some-

untenable its how close the and as and pointed out similarity (in gen- to Cirratulidae Ctenodrilidae

the order eral appearance, gut anatomy, possession of a car- (the three families were grouped in diac body) with the Ctenodrilidae, although a close Cirratulida). Fauchald (1977), followed by Peltibone kinship with the latter was ruled out because of (1982), George & Hartmann-Schroder (1985) and the exclusively sexual and gonochoristic reproduc- Barnes & Harrison (1992), classified the Cteno- tion drilidae and in the order ofStygocapitella. Karling found even greater Parergodrilidae together

of similarities (in segmentation, arrangement me- Ctenodrilida, but, influenced by Wilfcrt (1973), senteries, circulatory system, alimentary canal, male assigned the Cirratulidae to a different order. To and female reproductive organs) with the ‘archian- the Ctenodrilida Fauchald (1977) gave no particu- nelid’ the lack of family Nerillidae, although con- lar position, but recognized them as simple bod- Contributions to Zoology, 70 (3) - 2001 129

in of the animal led forms like those included the Orbiniida, description nor indicated the local-

Psammodrilida, Cossurida, Spionida, Capitellida ity of the finding. The species was independently and Opheliida. He refused to group these orders discovered in South Bohemia and thoroughly stud-

under the old concept Drilomorpha as “these forms ied by light microscopy by Pizl & Chalupsky (1984).

are about as far apart as any other grouping of Contrary to Parergodrilus and Stygocapitella,

polychaetes that might be proposed, judging from Hrabeiella has a dorsal pharyngeal organ and is

differences in tagmatization, parapodial develop- hermaphroditic, in these respects resembling an

its lack ment and setal distribution” (op.cit.: 9). oligochaete. However, of a clitellum and

Bunke noted that the ventral its overall of the internal (1967) pharyngeal arrangement organs pre-

the annelidPotamodrilus clude the Pizl organ of freshwater flu- a position among Clitellata. &

viatilis Lastochkin, 1935, is similar to that of the Chalupsky (1984) suggested that it should be either

‘aberrant polychaetes Parergodrilus and Stygo- allocated to a new annelid class or to the Parergo-

be capitella’. He believed this similarity to super- drilidae. Ultrastructural studies of the fine mor-

ficial and maintained the Potamodrilidae and the phology of the body wall, chaetae, pharynx and

allied in the closely Aeolosomatidae Oligochaeta. nervous system (Rota & Lupetti, 1996; Rota, 1998;

first However, an opposite view was taking root, Purschke, 1999) have revealed a strange combi- expressed by Brinkhurst (1971: 177), thenby Riser nation of characters which indeed precludes this

also from (1980), Giere & Riser (1981) and Timm (1981): worm any previously recognized non-

Not the the different pharyngeal construction and the lack clitellate family. even similarities between

of a clitellum exclusion of the Pota- the modified of Hrabeiella and the true suggested highly sperm

modrilidae and Aeolosomatidaefrom the Clitellata, dimorphic spermatozoaofProtodrilidae (von Nord-

and they were formally placed in a separate taxon hcim, 1989; Rota & Lupetti, 1997) seem to represent

Aphanoneura (Timm, 1981). Subsequent ultrastruc- an obvious synapomorphy, although the structures

tural investigations of the spermatozoa ofPotamo- involved are fundamentally homologous (Rota &

drilus and Aeolosoma have indeed revealed no Lupetti, 1997).

The aim of the synapomorphic correspondence between either present study was to re-evaluate

taxon and the Clitellata (Bunke, 1985, 1986). all the older interpretations reviewedabove by using

Purschke out of the 18S RNA (1987) carried an accurate compara- complete sequences ribosomal

tive of the ultrastructure of the better elucidate the study ventral buc- gene, to phylogenetic relation-

cal organs of Parergodrilus and Stygocapitella, ships of these obscure animals. Except forArenicola

providing evidence for their monophyly. Purschke marina, all material sequenced for this analysis

(1988a) also found that the pharynx ofthe Parergo- comes from Mediterranean latitudes and includes

drilidae does not share of its with the first record of subterranea any apomorphies Stygocapitella and

the other Nerillidae or former archiannelids, nor the second of Parergodrilus heideri for Italy. As

with the the Ctcnodrilidae, but rather represents representative of the Ctcnodrilidae, we used a new

most advanced stage of evolution of a muscular genus and species, gonochoristic and sexually di- bulb which is widespread in several non-'archian- morphic, recently discovered at the Elba Island,

nelid’ families such the which will polychaete as Orbiniidae, be described in a separate publication. Ctcnodrilidae, and Spionidae. Although noting

‘great differences’ between the pharynges of Pota-

modrilus and Parergodrilus he did not exclude Material and methods ,

the that close exist possibility a relationship may (Purschke, 1987). Selection of taxa

The first known record of Hrabeiella peri-

glandulata is the German Ulfert by oligochaetologist The data set analysed in this study consists of:

Graefe, who named it Adenodrilus the determined 18S rDNA punctulatus newly sequences of [nomennudum and nudum]discussed its ] briefly polycha- Hrabeiella periglandulata, Parergodrilus heideri, ete but neither nature (Graefe, 1977) provided a Stygocapitella subterranea, Aeolosomahemprichi, - 130 E. Rota, P. Martin & C. Erseus Phylogeny ofsoil-dwelling polychaetes

Protodrilus purpureus, Ctenodrilidae n. gen. et sp., DNA extraction, amplification and sequencing

the and Arenicola marina; a compilation of all

of this available in EMBL for DNA of A. H. and S. sequences gene poly- hemprichi, periglandulata

chaetes and other taxa assumed to be closely re- subterranea was extracted according to a standard

lated to, or members of, ‘non-clitellate Annelida’, Chelex™ procedure (Singer-Sam et ah, 1989; Hillis

namely, Sipuncula, Echiura, Pogonophora (includ- et ah, 1996). DNA extraction of P. heideri and A.

& made Pure PCR ing Vestimentifera) and Aphanoneura (Rouse marina was using “High Template

Fauchald, 1995); and another eleven 18S rDNA Preparation” from Boehringer-Mannheim Bio- sequences, also taken from EMBL, representing chemicals, and that of Ctenodrilidae n. gen et sp. the acanthobdelli- and Mini Kit” oligochaete, branchiobdcllidan, P. purpureus using “QIAamp DNA dan and euhirudinean Clitcllata (Table 1). from Quiagen, following the instructions of the

The choice of an outgroup to Polychaeta is prob- manufacturers. 18S rRNA gene fragments of A.

lematic because recent molecular studies failed to hemprichi and //. periglandulata were amplified

and Martin recover the monophyly of this taxon (McHugh, sequenced according to (2001), using

1997; Kojima, 1998; Westheide et al., 1999) and the 16 primers of Winnepenninckx et al. (1994), because putatively valid outgroups, such as Mol- kindly granted us by the authors. Specimens of the

annelids lusca, proved to be scattered among or other taxa were analysed following the protocol even appeared as an ingroup within polychaetes described in Erseus et al. (2000).

(Winnepenninckx et al., 1995, 1998; Siddall et al.,

1998). For this reason, not only three Mollusca

included in the data but also two Arthropoda were Alignment

set. Arthropods were designated as an unambigu-

in accordance with their it has been shown that the ous outgroup (Tabic I), Since repeatedly se-

molecular the the influence the placement among Ecdysozoa, quence alignment may phylogenetic sister of the to which all group Lophotrochozoa relationships inferred from ribosomal genes (Wiigele others taxa herewith considered belong (Aguinaldo & Stanjek, 1995; Winnepenninckx & Backeljau,

Adoutte et ah, 1997; et al., 2000). 1996; Erseus et al., 2000; Martin et al., 2000), we

studied 18S rDNA data in two sequence aligned

different ways: (1) using DCSE (De Rijk & De

Collection of new specimens Wachter, 1993; De Rijk, 1995), which considers a

secondary structure model; (2) using Clustal W

The newly sequenced material was collected at the (Thompson et al., 1994), default settings, without

following localities (abbreviations used in Table manual corrections. Whatever the alignment method

° 43°20'N 11 of 1): (MO) Montalbuccio, 15'E, Siena, used, some hyper-variable regions the gene

Italy (in 1998; coll. E. Rota); (CA) Camugnano| (domain 23 in particular; Van de Peer et al., 1996)

° 44°10'N 11 and dis- 10'E, near Bologna, Italy (in 1999; coll. were virtually impossible to align were

C. Jacomini and E. Rota); (SA) Costa Paradise, carded from the final alignment.

41°03'N 8°55'E, Sardinia (in 1999; coll. E. Rota);

(DU) Monkstown, 53°18'N 6°10'E, Dublin, Ireland

(in 1999; coll. O. Schmidt); (EL) Capo S. Andrea, Phylogenetic analyses

Elba Island, 42°48'N 10°09'E, Tyrrhenian Sea (in

All the material 2000; coll. E. Rota and C. Erseus). Maximum parsimony analyses of the resulting align- fixed and in 95-99 % alcohol. was preserved ments (EMBL accession numbers ALIGN-000074

for Clustal W and ALIGN-000096 for DCSE) were

performed using PAUP*, version 4.0b4a (Swofford,

1998) with the following settings: unweighted

characters, heuristic search, random addition of

with 100 tree-bisection- sequences replicates, Contributions to Zoology, 70 (3) - 2001 131

Table I. List of used in the species 1 8S rDNA analysis (binomial name, abbreviation ofthe collecting locality of new material, EMBL accession number).

arthropoda

Chelicerata Aphonopelma sp. {-‘Eurypelma(=“Eurypelma californica”)californica ”) X13457

Insecta Tenebrio molitor 1758 Linnaeus, X07801

MOLLUSCA

Pectinidae Placopecten magellanicus (Gmelin, 1791) X53899

Achatinidae Limicolariakambeul (Bruguiere, 1789)1789) X66374

Chitonidae AcanlhopleuraAcanthopleurajaponica (Lischke, 1873) X70210 S1PUNCULA

Phascolosomatidae Phascolosoma granulation Leuckart, 1828 X79874 echiuraECHIURA

Ochetostoma Echiuridae Ochelostoma erythrogrammon RuppellRiippell & Leuckart,Leuckart, 1830 X79875 POGONOPIIORAPOGONOPHORA

19631963 Frenulata Siboglinum fwrdicumfiordicum Webb, X79876

Vestimentifera Ridgeia piscesae Jones, 1985 X79877 POLYCHAETA

?7 Hrabeiella periglandulata Pizl & Chalupsky, 1984 MO AJ31050IAJ310501

aculeala 1761 Aphroditidae Aphrodita aculeata Linnaeus, 1761 Z83749

Arenicolidae Arenicola marina (Linnaeus, 1758)1758) DU AJ310502

Capitellidae Capitella capitata (Fabricius, 1780) U67323

Chaetopteridae ChaelopterusChaetoptems variopedatusvariopedatus (Renier, 1804) U67324

Cirratulidae Dodecaceria concharum Orsted, 1843 U50967

Ctenodrilidae n. gen. et sp. EL AJ310503

Glyceridae Glycera americana Leidy,Leidy, 1855 U19519

Magelonidae Magelonamirabilis (Johnston,(Johnston, 1865) U50969

Myzostomidae Myzostoma sp.sp. AF1169I6

Nephtyidae Nephtys hombergii Savigny, 18181818 U50970

Ncreididae Neanthes succineasuccinea (Frey & Leuckart, 1847) (=”Nereis(="Nereis limbata”) U36270

NeanthesNeanlhes virens (Sars, 1835) (=”Nereis(=”Nereis virensvirens”)”) Z83754 Orbiniidae Scoloplos armiger (Miiller,(Miillcr, 1776) U50972

Parergodrilidae Parergodrilus heideriheideri Reisinger, 1925 CA AJ310504

?Parergodrilidae Stygocapitella subterranea Knollner, 1934 SA AJ310505

PolynoidaePolynoidae Harmothoe impar (Johnston, 1839)1839) U50968

Protodrilidae Protodrilus EL purpureuspurpureas (Schneider, 1868) EL AJ3I0506AJ310506

Questidae Questapaucibranchiata Giere & Erseus, 1998 AF209464

Sabellidac 1822 Sabellidae Sabellapavonina Savigny, 1822 U67144

Protula Scrpulidae sp.sp. U67142

Spionidae Polydora ciliata (Johnston, 1838) U5097IU50971

Pygospio elegans Claparede, 1863 U67143 Terebellidae Lanice conchilega (Pallas, 1766) X79873 aphanoneuraAPHANONEURA

Aeolosomatidae Aeolosoma sp. Z83748

Aeolosoma hemprichi Ehrenberg, 1828 MO AJ3I0500AJ310500 OLIGOCHAETA

1 Enchytracidae Enchytraeus sp. 1 Z83750

2 EnchytraeusEnchylraeus sp. U95948 Lumbricidae Lumbricus rubellus Hoffmeister, 18431843 Z83753

Eisenia felida (Savigny, 1826) X79872 Naididae Stylaria sp. U95946 Tubificidae Tubifex sp. U67145

BRANCHIOBDELLIDABranch 1 015 dell i da

Branchiobdellidae Branchiobdellidae Xironogiton victoriensis Gelder & Hall, 1990 AF115977

holtihold 1963 Cambarincola Hoffman, 1963 AF115975API 15975 ACANTHOBDELLIDA

Acanthobdellidae AcanthobdellaAcanlhobdellapeledina Grube, 1851 AF099948 euhirudinea

Glossiphoniidae Glossiphonia complanata (Linnaeus, 1758) AF099943

Hirudinidae Hirudo medicinalis 1758 Hirudo medicinalisLinnaeus,Linnaeus, Z83752Z83752 132 E. Rota, P. Martin & C. Erseus - Phylogeny of soil-dwelling polychaetes

reconnection (TBR) branch-swapping algorithm, (Frenulata) and Ridgeia piscesae (Vestimentifera).

in treated miss- As far as are ‘Multrees’ option effect, gaps as soil-dwelling polychaetes concerned,

the sister of ing. Bootstrap analyses were performed using a Parergodrilus heideri is taxon Stygo-

of the heuristic search and TBR branch-swapping on 100 capitella subterranea and the monophyly replicates. The Bremer support index was calcu- Parergodrilidae is consistently supported in both

lated with PAUP* in connection with AutoDecay trees. Hrabeiellaperiglandulata links to the trees

4.0 (Eriksson, 1998). far from the parergodrilids and as a sister taxon to

Aphanoneura, but the latter relationship has no

bootstrap support, whatever the alignment.

Results In both trees, Capitella capitata joins the serpulid

Protula, with the ClustalW alignment giving strong

this association to 2). a The alignment based on a secondary structure model support (Fig. Lastly, strong

446 is also obtained for a sister produced a data set of2216 sites, from which support relationship between and Scolo- were excluded, resulting in a final alignment of Questa paucibranchiata

517 which in turn form a sister to 1770 sites, of which 918 were constant and plos armiger, group but the latter parsimony-informative. ClustalW considered fewer Magelona mirabilis, relationship re-

ceives weak or 2) indels, yielding a final alignment of 1718 included (Fig. 1) insignificant support (Fig. under sites (364 excluded characters), 851 constant and our alignments.

526 parsimony-informative characters. The DCSE and ClustalW alignments yielded 4 and 3 equally

Discussion most-parsimonious trees (MPTs; the consensus trees

are shown in Figs 1, 2), respectively.

of If Arthropoda can be considered as a valid The lack resolution is the most distinctive fea-

outgroup to Polychaeta, then Mollusca, as well as ture of the polychaetc relationships suggested by This be Sipuncula, Echiura and Pogonophora, arc dispersed the present study. may due to conflicting

in studies. contained in the ISSrRNA among polychaetes as noticed previous phylogenetic signals

of their asserted Abouheif al. which Both trees arc largely resolved, but most gene, as by et (1998),

nodes have low bootstrap and Bremer support makes the molecule unsuitable for reconstructing

values. Only 14 nodes are supported by bootstrap the evolutionary history of metazoa phyla. Alter-

50 in values over in both trees; such nodes are 17 natively, as polychaetes certainly were present by each of the ‘DCSE’ and ‘Clustal’ trees, with eight the Middle-Cambrian (Fauchald & Rouse, 1997), of them concerning clitellate relationships. Within the impossibility of resolving the branching order

the between ‘leech- in this well other of Clitellata, a close relationship group, as as major types meta-

like’ worms (Branchiobdellida) and the polycha- zoans having radiated during this period, can be

ctes Aphrodita aculeata + Neanthes virens, is sug- interpreted as the result of an explosive radiation

gested by both alignments, which would render (Philippe et al., 1994; Balavoine & Adoutte, 1998;

clitellatcs the is paraphylctic. Adoutte et al., 2000). While gene suppos-

Trivial associations between taxa belonging to edly suited to solve relationships at this level, clades

the same have fast and near in time to family (e.g. Aeolosomatidae, Spionidae) may emerged too too

are recovered and strongly supported, as expected. enable the accumulation of mutations on short

This is not always true, as exemplified by the two branches corresponding to this event (Philippe et

Neanthes species, N. succinea and N. virens, which al., 1994).

are assumed to belong to the same family (Nerei- Considering that only 20 polychaete families of

but into 80 & didae), are here consistently segregated a total of about (Fauchald Rouse, 1997) are

be two distant clusters, each with a different species studied at present, the poor resolution may also of scale worms (Aphrodita aculeata and Harmothoe due to insufficient taxonomic sampling. It is well-

impar) as sister taxon. A sister relationship is known that with small numbers of taxa, the choice

strongly supported between Siboglinum fiordicum of species can profoundly affect the phylogenetic Contributions to Zoology, 70 (3) - 2001 133

Consensus of 4 most trees ofannelid parsimonious relationships, including some other metazoan taxa (see Table 1), inferred '° m l lc of 18S alignment rRNA based a ' gene sequences upon secondary structure model (DCSE). Length of trees: 3301 steps, Cl

excluding uninformative characters) = = 0.3344, RI 0.3849. Numbers above and below internodes indicate bootstrap and Bremer values, Bremer respectively. Only supports greater than I are shown (ACA: Acanthobdellida; APH: Aphanoneura; ART: >t iropoda; BRA: ECU: EU1I: Branchiobdellida; Echiura; Euhirudinea; ERE: Frenulata; MOL: Mollusca; OLE Oligochaeta; POL; 0 ychaeta; SIP: Sipuncula; VES: Vestimentifera). 134 E. Rota, P. Martin & C. Erseus - Phytogeny ofsoil-dwelling polychaetes

Table 1), inferred Fig.2. Consensus of 3 most parsimonious trees of annelid relationships, including some other metazoan taxa (see

from the ClustalW of 18S rRNA of Cl uninformative = alignment gene sequences. Length trees: 3245 steps, (excluding characters)

0.3460, RI = 0.3946. Numbers above and below internodes indicate bootstrap and Bremer supports, respectively. Only Bremer

supports greater than 1 are shown (abbreviations; see Fig. 1). Contributions to Zoology, 70 (3) - 2001 135

reconstruction (Lecointre et al., 1993). A careful at least four different times (Purschke, 1988b)

Brown al. study of et (1999), based on three dif- Despite the lack of resolution, our trees give ferent genes, did not significantly improve the evidence for confirming or ruling out some hypo- resolution, despite a broad sampling of polycha- theses about soil-dwelling polychaetes relationships.

less than cte diversity. In addition, no five genes The validity of grouping P. heideri and S. subter- have been studied far to elucidate so polychaete ranea into the family Parergodrilidae is here con-

of them relationships but none has convincingly firmed. In contrast, H. periglandulata never clus- recovered groupings (McHugh, 1997; Giribet & tered with them and its position relative to this

Ribera, 1998; Kojima, 1998; Siddall et ah, 1998; and other polychaete families still remains obscure. Brown et ah, 1999; Erseus et ah, 2000). The bio- It is also strongly suggested that all of these poly-

of these hence chaetes far from logical reality polytomies appears are clitellates, which means, for corroborated by independent evidence, which sup- instance, that the hypothetical relationship between

the P. ports hypothesis of an ancient emergence and heideri and the oligochaete family Enchytraeidae, explosive radiation of polychaetes. once suggested by Meyer (1927), is dismissed.

The performance ofmaximum parsimony is usu- Similarly, a possible close affinity between Pa- ally improved by weighting substitution types but rergodrilidae and Aphanoneura noted by Bunke the choice of a weighting scheme is not a trivial (1967), the reality of which was not excluded by task and involves that important assumptions are Purschke (1987), receives no confirmation from difficult to prove (Milinkovitch et al., 1996). Given molecular data. The hypothesis of Parergodrilidae the radiation’ such close ‘explosive hypothesis, a pro- being to Ctenodrilidae is neither supported cedure will supposedly not improve the resolution nor refuted by our study. Interestingly, under both of polychaete relationships and is probably not alignments used in this study, the most parsimoni- justified. ous hypotheses place H. periglandulata close to

The idea that the differentiation the of polycha- Aphanoneura. The support for such a phylogenetic cte families has followed a pattern of rapid radia- position, however, is still too weak. tion is not new. Fauchald (1974) considered this Other relationships are worthy of note, while hypothesis to explain “the confusion in structure not directly related to soil-dwelling polychaetes. a >id numbers of in As anterior appendages the differ- already noticed by Erseus et al. (2000), a close ent families, the odd distribution of nephridial struc- relationship between Questidae and Orbiniidae is tures and varied the development of the nervous here confirmed, in accordance with morphological system”. Fauchald’s phylogeny implied parallel data (Rouse & Fauchald, 1997), rendering irrel- evolution of these and other in the evant about the organ systems conjectures oligochaetoid nature elass and, indeed, multiple and parallel evolution- of some of their morphological and development ary modifications in number other & are observed a of features (Giere Riser, 1981). Further, our study Polychaete attributes (e.g. chaetae, foregut, male does not contradict the monophyletic nature of a gametes, larval type), which further complicates clade constituted by Frenulata and Vestimentifera attempts at reconstructing phylogeny at supra-fa- pogonophorans suggested by Winnepenninckx et unlial level. Some about the al. authorities now questions homology (1995, 1998) (many treatpogono- °1 these features, formerly considered to involve phorans as Siboglinidae, a polychaete family; many different families, are gradually being solved McHugh, 2000). The Aphanoneura are not closer hy studying their differentiation and fine details to the Clitellata than other any annelid grouping. al the ultrastructural level. Thus, for instance, Conflicting evidence generated by earliermolecular

Polychaete to have evolved in- introsperm seem studies (Moon et al., 1996; Winnepenninckx et al.,

times & dependently many (Jamieson Rouse, 1989), 1998) probably resulted from too small a taxon s Pecialized chaetae such as hooded hooks have sampling. Se parately evolved in Eunicida and in capitellids The fact that some closely related polychaetes and spionids (Bartolomaeus, 1998), and ventral do not cluster together in our trees is all the more

Pharyngeal were invented since either the organs by polychaetes puzzling, they belong to same genus - 136 E. Rota, P. Martin & C. Ersens Phytogeny ofsoil-dwelling polychaetes

to DNA and made (the two Neanthes species) or the same superfam- primers necessary amplification sequencing

by him. We all thank Dr Mari Kallersjo, Mr Rasmus Hovmoller ily (the scale worms A. aculeata and H. impar; and Ms Maria Ekman (all at the Molecular Systematics Aphroditacea; Fauchald, 1977). Clearly, re-sequenc- Laboratory, Swedish Museum of Natural History, Stockholm) (he I8S of these taxa is in ing gene imperative for assistance with the other This work new sequences. was these future studies to establish whether anomalies supported by funds from the Italian National Program “Aspetti

della biodiversita animale in Italia” result from biases identifi- descrittivi e metodologici (bad sequence, wrong

(MURST grant no. 9905317452); and the Swedish Natural cation, contamination) or have true biological Science Research Council. meaning. In the latter case, this would be a further warning that the absence of suggestions of rela- tionships between, for instance. Hrabeiella and References Parcrgodrilidae, or between Aphanoneura and Cli-

tellata, is at the most indicative and that sequenc- Abouheif E, Zardoya R, Meyer A. 1998. Limitations of of other conservative is needed. ing genes badly 18S rRNA data: for metazoan sequence Implications re- The phylogenetic position of Myzostomida has constructing a phylogeny of the animal kingdom and in- been much debated over the years, although most ferring the reality of the Cambrian explosion. ./, Mol.

Evol. 47: 394-405. often considered to be close to, or even within,

Adoutte for A, Balavoine G, Lartillot N, Lespinet O, the annelids (see Eeckhaut et ah, 2000, a re-

Prud’homme 15, de Rosa R. 2000. The new animal view). On the basis of analyses oftwo nucleargenes phylogeny: Reliability and implications. Proc. natn. Acad. subunit ribosomal RNA and fac- (small elongation Sci. U.S.A. 97: 4453-4456.

1 Eeckhaut al. concluded tor- a), however, et (2000) Aguinaldo AMA, Turbeville JM, Linford LS, Rivera MC,

but that myzostomids are not annelids, more likely Garey JR, Raff RA, Lake JA. 1997. Evidence for a cladc of nematodes, arthropods and other moulting ani- close flatworms. In our of the a group to study mals. Nature 387: 489-493. 18S rRNA gene, Myzostoma sp. clusters among Avel M. 1959. Classe des Annelides OUgochetes (Oligochaeta the polychaetes in all of the most parsimonious Huxley, 1875). In: Grasse PP, ed. Trade de Zoologie, I. but there is trees (Figs. 1,2), no bootstrap support V, fuse. I. Paris: Masson, 224-470.

for this position. Moreover, as no flatworm taxa Balavoine G, Adoutte A. 1998. One or three Cambrian

radiations? Science 280: 397-398. were included in our analysis, the results of Barnes RD, Harrison FW. 1992. Introduction. In: Harrison Eeckhaut ct al. (2000) are here neither supported FW, Gardiner SL, eds. Microscopic anatomy of inverte- nor contradicted. brates, vol. 7, Annelida. New York: Wiley-Liss, 1-9. as a of relevance for clitellate re- Lastly, point Bartolomaeus T. 1998. Chaetogenesis in polychaetous

lationships, the odd location of Branchiobdcllida Annelida - Significance for annelid systematics and the

of the 100; 348-364. close to two polychaetes was recently shown to position Pogonophora. Zoology Beauchamps P de. 1959. Archiannelides (Archiannelida result from a spurious attraction (Martin, 2001), Hatschek, IS93). In: Grasse PP, cd. Trade de Zoologie, that the so monophyly ofClitellata cannot be ques- t. V. fuse. I. Paris: Masson, 197-223.

tioned so far. Beklemischev VN. 1958. Die Grundlagen der vergleichenden

Anatomic der Wirhellosen Bd. I [German translation!, ,

Berlin: Deutscher Verlag Wissenschaften, 1- 441.

Brinkluirst RO. 1971. Phytogeny and classification. Part Acknowledgements I. In: Brinkhurst RO, Jamieson BGM. Aquatic Oligochaeta

of the world. Edinburgh: Oliver & Boyd, 165-177, is with the collection ER grateful to a number offriends for help Brown S, Rouse G, Hutchings P, Colgan I). 1999. Assess- of material: Mr Ariano Buracchi assisted duringfield sampling ing the usefulness of histone H3, U2 snRNA and 28S of H. periglandulata and A. hemprichi. Prof. Marco Curini- rDNA in analyses of polychaete relationships. Ausl. J. Gallctti offered facilities for the collection ofS. subterranea in Zool. 47: 499-516.

Sardinia, Dr Carlo Jacomini provided forest soil samples Bunkc D. 1967. Zur Morphologic und Systematik der containing P. heideri from his PhD site. Dr Olaf Schmidt sent Aeolosomatidae Beddard 1895 und Potamodrilidae nov. material ofA. marina from Ireland, Drs Frank Thiermann and fam. (Oligochaeta). Zool. Jb. Syst. 94: 187-368. Christian Borowski assisted with collecting sublittoral samples Bunke D. 1985. Ultrastructure of the spermatozoon and Island. his by SCUBA diving at Elba PM expresses gratitude to spermiogenesis in the interstitial annelid Potamodrilus Drs Thierry Backeljau and Birgitta Winnepenninckx (Royal fluviatilis. /. Morphol. 185: 203-216, Belgian Institute of Natural Sciences, for giving all Brussels) Bunke I). 1986. Ultrastructural of the investigations sper- Contributions 70 - to Zoology, (3) 2001 137

matozoon and its in Aeolosoma litorale genesis with Polychaeta (Annelida): an ultrastructural review. Biol. Rev.

considerations on the phylogenetic implications for the Cambridge 64: 93-157.

Aeolosomatidae (Annelida). J. Ultrastr. mol. Str. Res. Karling TG. 1958. Zur Kenntnis von Stygocapitella

95: 113-130. subterranea Knollner und Parergodrilus heideri Reisinger Cernosvitov L. 1937. System der Enchylraeiden. Bull. Assoc. (Annelida). Ark. Zool. (s. 2) 11: 307-342.

Russe Rech. sci. 5: 263-295. Prague Knollner F. 1934. Die Tiere des Kiislengrundwassers bei Dales RP. 1962. The polychaete stomodaeum and the inter- Schilksee (Kieler Bucht). 5. Stygocapitella subterranea

of the families of Proc. Zoo!. Schr. relationships Polychaeta. nov. gen. nov. spec. naturw. Ver. Schleswig-Hol- Soc. London 139: 389-428. stein 20: 468-472.

Dales RP. 1963. Annelids. London: Hutchinson University Kojinia S. 1998. Paraphyletic status of Polychaeta suggested Library, 1-200. by phylogenetic analysis based on the amino acid se-

De P. 1995. a database ribosomal of factor-1 Rijk Optimisation of for quences elongation a. Mol. Phyl. Evol. 9: 255- RNA structure and application in structural and evolu- 261.

research. PhD of tionary thesis, University Antwerp. Leeointre G, Philippe H, Le HLV, Lc Guyadcr H. 1993. Belgium. Species sampling has a major impact on phylogenetic De Rijk P, De Wachter R. 1993. DCSE, an interactive tool inference. Mol. Phyl. Evol. 2: 205-224.

for sequence alignment and secondary structure research. Martin P. 2001. On the origin of Hirudinea and the demise

Biosci. 9: 735-740. Comput. Appl. of the Oligochaeta. Proc. R. Soc. Loud. B, 268: 1089-

Du Bois-Reymond Marcus E. 1948. Further archiannelids 1098.

from Brazil, Com. Zool. Mus. Hist. Nat. Montevideo 2(48): Martin P, Kaygorodova I, Sherbakov DY, Vcrheycn E. 1-27, 2000. Rapidly evolving lineages impede the resolution of Eeckhaut I, I), Tiedemann McHugh Marclulyn P, R, phylogenetic relationships among Clitellata (Annelida).

Monteyne D, Jangoux M, Milinkovitch MC. 2000. Mol. Phyl. Evol. 15: 355-368.

link between Myzostomida: a trochozoans and flatworms? McHugh D. 1997. Molecular evidence that echiurans and

Proc. R. Soc. Land. B 267: 1383-1392. pogonophorans..are derived annelids. Proc. Natl. Acad. Eriksson T. 1998. AutoDecay ver. 4.0. (program distributed Sci. USA 94: 8006-8009.

by the authors). Department of Botany, Stockholm Uni- McHugh I). 2000. Molecular phylogeny of the annelids.

versity, Stockholm, Sweden. Can. J. Zool. 78: 1873-1884,

Erseus C, Prestegaard T, Kiillcrsjd M. 2000. Phylogenetic Meyer A. 1927. 1st Parergodrilus heideri (Reisinger) ein of Tubificidae (Annelida, analysis Clitellata) based on Archiannelide? Zool. Anz. 72: 19-35.

I8S rDNA sequences. Mol. Phyl. Evol. 15: 381-389. Michaelsen W. 1928. Oligochaeta. In: Ktikenthal W, Krum-

Fauchald K. 1974. Polychaete phylogeny: a problem in proto- bach T, eds. Handbuch der Zoologie, 2. Berlin & Leipzig,

stome evolution. Syst. Zool. 23: 493-506. 1-118.

Fauchald K. 1977. The polychaete worms. Definitions and Milinkovitch MC, LeDuc RG, Adachi J, Farnir F, Georges

the and Nat. keys to orders, families genera. Hist. Mus. M, Hasegawa M. 1996. Effects of character weighting

Los Sci. Serv. 28: Angel. Cty 1-188, and on species sampling phylogeny reconstruction: a case Fauchald K, Rouse GW. 1997. based on DNA data in Polychaete systematics: past study sequence Cetaceans. Genet-

and present. Zool. Scr. 26: 71-138. ics 144: 1817-1833.

Hartmann-Schroder G. 1985. Minelli A. George JD, Polychaetes: 1993. Biological systematics: The state of the British Amphinomida, Spintherida and Eunicida. Keys and art. London: Chapman & Hall, 1-387.

notes the the of the Moon Kim for identification of species. Synopses SY, CB, Gelder SR, Kim W. 1996. Phyloge-

British Fauna, 32, London: Brill and 1-221. netic of the aberrant Backhuys, positions branchiobdellidans and

Gicre O, Riser NW. 1981. Questidae - with Polychaetes aphanoneurans within the Annelida as derived from I8S

oligochaetoid and Zool. Scr. ribosomal RNA 324: 229- morphology development. gene sequences. Hydrobiologia 10: 95-103. 236.

Giribet G, Ribera C. 1998. The position of arthropods in Nordheim H von. 1989. Vergleichende Ultrastrukturunter-

the animal search for reliable kingdom: a a outgroup for suchungen der Eu- und Paraspermien von 13Protodrilus-

internal Mol. Evol. 9: 481- arthropod phylogeny. Phyl. Arten (Polychaeta, Annelida) und ihre taxonomische und 488. phylogenetische Bedeulung. Helgol Meeresunt. 43: 113- Gracfe U. 1977. Adenodrilus Graefe (in press) and Parer- 156.

godrilus two aberrant Annelida in forest Reisinger soils. Pettibone MH. 1982. Annelida. In: Parker SP, ed. Synopsis

Publ. Centro Biol. 9: 25. and piren. exp. classification of the living organisms, 2. New York: Heider K. 1922. Ubcr Archianneliden. 1-43. Sitzungsb. Kgl. preuss. McGraw-Hill,

Akad. Wiss., phys.-math. Ki. Berlin 6: 39-44. Philippe H, Chenuil A, Adoutte A. 1994. Can the Cam

Hillis DM, Mable BK, Larson A, Davis SK, Zimmer EA. brian explosion be inferred through molecular phylog

1996. Nucleic acids IV: and In: Hillis ? sequencing cloning. eny Development 1994 Suppl.: 15-25.

DM, Moritz C, Mable BK, eds. Molecular Pizl V, J. 1984. Hrabeiella systematics. Chalupsky periglandulata gen.

Sunderland, USA: Sinauer 321-381, et Associates, - curious from sp, n. (Annelida) a worm Czechoslova- Jamieson BGM, Rouse GW. 1989. The spermatozoa of the kia. Vest. Cs. Spolec. Zool. 48; 291-295. 138 E. Rota, P. Martin & C. Erseus - Phytogeny of soil-dwelling polychaetes

Purschke G. 1987. Anatomy and ultrastructure of ventral Stephenson J. 1930. The Oligochaeta. Oxford: Clarendon

and their in 1-978. pharyngeal organs phylogenetic importance Press.

Polychaeta (Annelida). 111. The pharynx of the Parergo- Swofford DL. 1998. PAUP*. Phylogenetic Analysis Using

drilidae. Zool. Jb., Abt. Anat. Ontog. Tiere 115: 331-362. Parsimony (*and other Methods). Version 4. Sunderland,

Purschke G. 1988a. Anatomy and ultrastructure of ventral USA: Sinauer Associates.

and their in Gibson 1994. Clustal W: pharyngeal organs phylogenetic importance Thompson JD, Higgins DG, TJ.

V. The of the Cteno- the of Polychaeta (Annelida). pharynges Improving sensitivity progressive multiple sequence

drilidac and Orbiniidae. 108: 119-135. Zoomorphology alignment through sequence weighting,positions-specific

Purschke G. 19881). In: Westheide Hermans and matrix choice. Nucleic Acids Pharynx. W, gap penalties weight

CO, eds. The ultrastructure of Polychaeta. Microfauna Res. 22: 4673-4680.

marina 4: 177-197. Timm T. 1981. On the origin and evolution of aquatic

Purschke G. 1999. Terrestrial polychaetes - models for the Oligochaeta. Eesti NSV Tead. Akad. Toimet. Biol. 30:

evolution of the Clitellata (Annelida)? Hydrobiologia 406: 174-181.

87-99. Van de Peer Y, S, De Wachtcr R. 1996. A Chapellc quan-

E. 1925. Ein landbewohnender Archiannelide titative of nucleotide substitution in bacterial Reisinger map rates

(Zugleich ein Beilrag zur Systemalik der Archianneliden). rRNA. Nucleic Acids Res. 24: 3381-3391.

Z, Morphol. Okol Tiere 3: 197-254. Wiigelc JW, Stanjek G. 1995. Arthropod phylogeny in-

from Reisinger E. 1929. Die systematische Stellung von ferred partial I2S rRNA revisited: monophyly of

heideri Zool. 80: 12-20. the Tracheata Zool. Parergodrilus Reisinger. Anz. depends on sequence alignment. J.

Reisinger E. I960. Die Posting des Parergodrilus- Problems. Syst. Evol. Research 33: 75-80.

Z. Morphol. Okol. Tiere 48: 517-544. Westheide W, McHugh I), Purschke G, Rouse GW. 1999.

Riser NW. 1980. The aberrant polychaet Stygocapitella from Systematization of the Annelida: different approaches.

some American beaches. Wasmann J. Biol. 38: 10-17. Hydrohiologia 402: 291-307.

Rota E. 1998. Morphology and adaptations of Parergodrilus Wilfert M. 1973. Ein Beitrag zur Morphologic, Biologie

Reisinger and Hrabeiella Pizl & Chalupsky, two enig- und systematischen Stellung des Polychaeten Ctenodrilus

matic soil-dwelling annelids. Ilal. J. Zool. 65: 75-84. serratus. Helg. wiss. Meeresunters. 25: 332-346.

Rota E, Lupctti P. 1996. An ultrastructural investigation of Winnepenninckx B, Backeljau T. 1996. 18S rRNA align-

Hrabeiella Pizl & Chalupsky. 1984 (Annelida). 1. Chaetae ments derived from different secondary structure models

and body wall organization. Hydrobiologia 334: 229-239. can produce alternative phylogenies. J. Zool. Syst. Evol.

Rota E, Eupctti P. 1997. An ultrastructural investigation of Res. 34: 135-143.

Hraheiella Pizl & Chalupsky, 1984 (Annelida). II. The Winnepenninckx B, Backeljau T, De Wachter R. 1994.

spermatozoon. Tissue & Cell 29: 603-609. Small ribosomal subunit RNA and the phylogeny of

Rouse liW, Fauchald K. 1995. The articulation of anne- Mollusca. The Nautilus , Suppl. 2: 98-110.

lids. Zool. Scr. 24; 269-301. Winnepenninckx B, Backeljau T, De Wachter R. 1995.

Rouse GW, Fauchald K. 1997. Cladistics and polychaetes. Phylogeny of protostome worms derived from 18S rRNA

26: Mol. Biol. 12: Zool. Scr. 139-204. sequences. Evol. 641-649.

Siddall ME, Fitzhugh K, Coates KA. 1998. Problems de- Winnepenninckx BMH, Van de Peer Y, Backeljau T. 1998.

termining the phylogenetic position of Echiurans and Metazoan relationships on the basis of I8S rRNA se-

with limited data. Cladistics 14: 401-410. few later. Zool. Pogonophorans quences: a years Amer. 38: 888-906.

Siuger-Sam Tanguay RC, Riggs AI). 1989. Use of Chelcx

to improve the PCR signal from a small number of cells, i Received: 7 September 2000

Amplifications 3:11.