COST

European cooperation in the field of scientific and technical research

European Commission

COST 819
Entomopathogenic nematodes

Taxonomy, phylogeny and gnotobiological studies of entomopathogenic nematode bacterium complexes

EUR 18832 EN

Cover illustration: Electron micrograph of a fastfreeze fixation and freeze substitution of a culture on nutrient agar in stationary growth of Photorhab dus sp. (type strain ATCC 29304) showing typical inclusions in the cell protoplasms. The section was contrasted with uranyl acetate and lead citrate. Scale bar = 0.5 . Courtesy of Dr Nol Boemare, Laboratoire de pathologie compare, Universit Montpellier II, France.

European Commission Directorate-General Science, Research and Development

COST 819
Entomopathogenic nematodes

Taxonomy, phylogeny and gnotobiological studies of entomopathogenic nematode bacterium complexes

Proceedings of the workshop held at Horticulture Research International, Wellesbourne, Warwick, United Kingdom, on 22 and 23 April 1998

Edited by Nol Boemare, Paul Richardson and Franoise Coudert

European Commission Rue de la Loi/Wetstraat 200 B-1049 Brussels

1999

EUR 18832 EN

LEGAL NOTICE Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use which might be made of the following information.

A great deal of additional information on the European Union is available on the Internet. It can be accessed through the Europa server (http://europa.eu.int). Cataloguing data can be found at the end of this publication. Luxembourg: Office for Official Publications of the European Communities, 1999 ISBN 92-828-5823-5 European Communities, 1999 Reproduction is authorised provided the source is acknowledged. Printed in Belgium
PRINTED ON WHITE CHLORINE-FREE BLEACHED PAPER

CONTENTS
CONTENTS 3 FOREWORD 5 PARTICIPANT LIST (22-24 Apri! 1998) 7 SUMMARY AND PROGRAM 9 THE HIDDEN PROKARYOTIC DIVERSITY : UNRAVELLING THE EXTENT OF SYMBIOTIC ASSOCIATIONS by Erko Stackebrandi 13 MOLECULAR PHYLOGENY OF NITROGEN-FIXING BACTERIA IN SYMBIOSIS WITH PLANT ROOTS by Philippe NORMAND 29 COMPARATIVE AND PHYLOGENETIC ASPECTS OF TWO INSECT ENDOSYMBIOSES : APHIDS AND WEEVILS by Abdelaziz Heddi, Hubert Charles, Yvan Rahb 37 BIOSYSTEMATICS, PHYLOGENY AND POPULATION GENETICS OF ENTOMOPATHOGENIC NEMATODES by W.M. Hominick, D.J. Hunt, A.P. Reid, B.R. Briscoe and D.A. Bohan' 45 PCR-RFLP OF 16S rDNA, AS A FAST METHOD TO IDENTIFY XENORHABDUS AND PHOTORHABDUS. by B. Brunei ', M. Fisher-Le Saux 2, A. Givaudan 2, A. Lanois 2, and N. E. Boemare 2 55 APPLICATION OF PCR-RFLP OF 16S rDNA TO STUDY THE SYMBIONTS OF TROPICAL EPNs by Marion Fischer-Le Saux', H. Maulon2, B. Brunei3 andN. Boemare' 63 PHYLOGENETIC ANALYSIS OF SYMBIOSIS BETWEEN ENTOMOPATHOGENIC NEMATODES AND THEIR BACTERIA by Tie Liu, Ralph Berry, and George Poinar 73 IDENTIFICATION OF PHOTORHABDUS LUMINESCENS SUBSPECIES by Ralf-Udo Ehlers 83 MOLECULAR PHYLOGENETICS AND TAXONOMIC HETEROGENEITY OF PHOTORHABDUS LUMINESCENS : OLD AND NEW AT A by E. Szallas, Andrs Fodor, G. Kovacs, C. Koch, J. Burghart, O. Buss ', K. Nealson 2, and E. Stackebrandt 385 IDENTIFICATION AND CHARACTERISATION OF INSECTICIDAL TOXIN GENES AND BACTERIA ASSOCIATED WITH ENTOMOPATHOGENIC NE MATODE S by Morgan, J. A. W.1, P. Jarre2, D.J.E llis2, M.A.Ousley' 87 NUTRITIONAL SIGNIFICANCE OF PHOTORHABDUS BACTERIA FOR THE GROWTH AND REPRODUCTION OF HETERORHABDIUS NEMATODES byRichouHan and Ralf-Udo Ehlers 89 HETERORHABDITIS SPP. AND PHOTORHABDUS SPP. : CO-EVOLUTION AND GNOTOBIOLOGY by O. Oravecz, E. Szlls, . Panjav, Andras Fodor ', . Adams 2, and . Stackebrandt3 97 TOWARDS THE PHYSICAL MAP OF THE GENOPHOR (CHROMOSOME) OF PHOTORHABDUS LUMINESCENS by Zsuzsanna Kiss', Cs. P. Ortutay', A. Fodor', M. Rott2, and K.H. Nealson3 99 THE CONCEPT OF MONOXENIC ASSOCIATIONS IN EPNs: LIMITS OF THEIR NATURAL OCCURRENCE AND OF GNOTOBIOLOGICAL INTERPRETATIONS by Nol Boemare, E. Bonifassi, M. Fischer Le Saux ', and G. Smart Jr 2 101 OBSERVATIONS ON THE BACTERIAL SYMBIONT ASSOCIATED WITH THE NEMATODE, STEINERNEMA ABBASI (STEINERNEMATIDAE : NEMATODA) by S. Elawad, R. Robson and Nigel Hague 105

FOREWORD
Working group 4 (WG4) of COST 819 decided to organize, in Wellesbourne in April 1998, a workshop on the bacterial symbionts of Entomopathogenic nematodes (EPNs). The decision was the result of recent progress in the taxonomy and phylogeny of these special bacteria. The purpose of the workshop was to clarify the relationships of the bacteria with their nematode hosts. A decision was also taken to examine the gnotobiological data that has accumulated from 10 years of research, in various laboratories, in order to define the level of specificity between the symbiotic association and the usefulness of gnotobiology as a tool for taxonomie purposes. April 1998 proved an opportune time for such a workshop ! During the meeting, Dr. Alun Morgan described the occurrence of toxin production from Xenorhabdus spp. (p. 88) and, just two months later, the importance of the bacterial symbionts of EPNs was further emphasized. In June a paper in Science announced the characterization of Photorhabdus toxins that were effective per os (Bowen et al., 1998) ; this followed other significant papers (Blackburn et al., 1998; Bowen and Ensign, 1998) and a patent (Ensign et al, 1997). Our understanding of toxins which were also dealt with in two other paptents, one on Xenorhabdus (Akhurst and Smigielski, 1995) and another on Photorhabdus (Ragni et ai, 1996), highlights the role of bacteria in the pathogenic action of the nematode bacterium complexes. One day, selection and biotechnological manipulation of appropriate bacterial strain genes may lead to the development of new bio-insecticides and/or transgenic plants lethal to insect pests. Because of the increased interest in mechanisms of pathogenicity, WG2 and WG4 decided to organise another meeting on toxins and symbiont by-products for the next COSTIOBC meeting in Vienna in March 1999, as a part of a comprehensive overview on the toxic activity of nematode bacterium complexes. World experts will be invited, by IOBC and COST, to focus on our present state of knowledge in this important field. Returning to our meeting in Wellesbourne, on behalf of all our members, I would like to take this opportunity to thank very much the kind welcome of our English hosts. We were very grateful for the words of welcome from Professor C.C. Payne, Chief executive of Horticulture Research International. Special thanks must be addressed to Dr. Paul Richardson, Dr. Steve Long, Deena Welch and Michelle Kind for the local organization of the meeting and to Microbio Ltd. for sponsoring the participants' document wallets. We spent a lovely time in Stratford upon Avon, Shakespeare's birthplace. We enjoyed the town with its well-preserved 15th century buildings. We also had a memorable banquet in the medieval surroundings of Warwick castle with prestigious meals and nice entertainment ! So, I again address our gratitude to the British delegation of COST 819 who organized everything so well. Montpellier, 06/10/98 Nol Boemare Chairman

Blackbum, M., Golubeva, E., Bowen, D., and rench-Constant, R. H. (1998) A novel insecticidal toxin from Photorhabdus luminescens, toxin complex a (Tea), and its histopathological effects on the midgut of Manduca sexta. Appi Environ. Microbiol. 64, 3036-3041. Bowen, D., Rocheleau, . ., Blackburn, M., Andreev, O., Golubeva, E., Bhartia, R., and rench-Constant, R. H. (1998). Insecticidal toxins from the bacterium Photorhabdus luminescens. Science 280, 2129-2132. Bowen, D. J., and Ensign, J. C. (1998). Purification and characterization of a high-molecular-weight insecticidal protein complex produced by the entomopathogenic bacterium Photorhabdus luminescens. Appi Environ. Microbiol. 64, 3029-3035.

PARTICIPANT LIST
Robin BE DDING Mark BLIGHT Noel BOE MARE Mari BOFF John BROWNE Brigitte BRUNE L AnnBURNELL Shulong CHE N Jozef COOSE MANS Franoise COUDE RT John CROWE Ralf-Udo E HLE RS Marion FISCHE R LE SAUX Paul FITTE RS Andrs FODOR Bertold FRJDLE NDE R Karin GE RBE R Lonne GE RRITSE N Itamar GLAZE R David GLE N Christine GRIFFIN Jerg GRUNDE R Nigel HAGUE Richou HAN Birgit HASS Abdelaziz HE DDI Barry HOLLAND Kjell-Ove HOLMSTROM Bill HOMTNICK Magdalena JAWORSKA Zsuzsa KISS Elisabeth KOSCHIE R Igor KRAME R Jie LIU Maurice MOE NS Alun MORGAN Otto NIELSEN Philippe NORMAND Orsolya ORAVE CZ Hori PANJAV Mavji PATE L Rolo PE RRY

(2224 April 1998)1

CSLRO, Australia University of Paris, France University of Montpellier, France -DLO, The Netherlands National University of Ireland INRA, France National University of Ireland Rijksstation Nematologie E ntomologie, Belgium Katholieke University Leuven, Belgium COST Secretariat, Brussels University of California, USA Institut fur Phytopathologie, Kiel, Germany University of Montpellier, France National University of Ireland Etvs University, Hungary Bio Integrated Technology, Italy Intitut fur Phytomedizin, Austria -DLO, Netherlands Volcani Centre, Israel IACR, Bristol, UK National University of Ireland Federal Research Station, Switzerland University of Reading, UK Christian-Albert University of Kiel, Germany Berlin, Germany INSA, Lyon, France University of Paris, France Uppsala Genetic Centre, Sweden CAB International, UK Academy of Agriculture, Poland Etvs University, Hungary University of Agricultural Sciences, Austria Swiss Federal Research Station, Switzerland Oregon State University, USA Rijksstation Nematologie Entomologie, Belgium HRI Wellesbourne, UK Royal Vet. & Agrie. University, Denmark INSA, Lyon, France Etvs University, Hungary EtvsUniversity, Hungary Imperial College, London, UK IACR, Rothamsted, UK

' Addresses of the participants are detailed in the Annual reports (1997, 1998), and addresses of the invited speakers are mentioned at the top of each repon of this volume

Ame PETERS Holger PHILIPSEN Andrea PIMENTA Lihong QIU Alex REID Nick RENN Manuele RICCI Paul RICHARDSON Solveig SALINAS-HAUKELAND Cndido SANTIAGO-ALVAREZ Danny SEGAL Nelson SLMOES Aharon SOLOMON Erko STACKEBRANDT Marek TOMALAK Karolien VAN DEN BERGH Rick VAN DER PAS Tibor VELLAI Denis WRIGHT

Institute for Phytopathologie, Germany Royal Vet. & Agrie. University, Denmark University of Paris, France CSIRO, Australia CAB International, UK Central Science Laboratory, York, UK Bio Integrated Technology, Italy HRI Wellesbourne, UK Crop Research Institute, Norway Universidad de Crdoba, Spain Tel-Aviv University, Israel Universidade dos Aores, Portugal University of Jerusalem, Israel DSM, Germany Institute of Plant Protection, Poland Katholieke University Leuven, Belgium Koppert B.V., The Netherlands Etvs University, Hungary Imperial College, London, UK

SUMMARY AND PROGRAM

The workshop on bacterial symbionts was divided in three sessions. The morning session of 22 April, chaired by Professor Erko Stackebrandt, Director of the German collection of microorganisms and Editor-in-Chief of the International Journal of Systematic Bacteriology, was devoted to the taxonomy and the phylogeny of symbioses other than those involving EPNs in order to consider some valid comparisons in other research fields. In addition, as an introduction to the debate on the taxonomy of the bacteria, Dr. Bill Hominick (UK), Director of CABI and Convenor of WG1 of COST 819, addressed our current knowledge of the taxonomy of EPNs. The afternoon session, on the taxonomy and phylogeny of EPN symbioses , was chaired by Dr. Philippe Normand, Director of the French laboratory of Microbial Ecology of the Soil, and specialist in phylogeny. On the morning session of 23 April the session entitled Gnotobiological studies on EPN symbioses , was chaired by Dr. Noel Boemare, Chairman of COST 819 and Convenor of WG4. The programme was as follows :

WEDNESDAY 22 APRIL Session 1A: Taxonomy and Phylogeny of Bacterial Symbioses Chairman: Dr Erko Stackebrandt 9.50 The hidden prokaryotic diversity: unravelling the extent of symbiotic associations. Erko Stackebrandt (DSMZ, Braunschweig, Germany)

10.20 Molecular phylogeny of nitrogen-fixing bacteria in symbiosis with plant roots. Philippe Normand (CNRS, Universit Lyon, France) 10.50 Break 11.10 Comparative and phylogenetic aspects of two insect endocytobioses: aphids and weevils. Abdelaziz Heddi, H. Charles and Y. Rahb (INSA-INRA, Lyon, France) 11.40 Present knowledge of the taxonomy of the nematode hosts. Bill Hominick (CABI, International Institute of Parasitology, St Albans, UK) 12.10 General discussion 12.40 Lunch

10

Session IB: Taxonomy and Phylogeny of EPN Symbionts Chairman: Dr Philippe Normand 14.00 PCR-RFLP of 16S rDNA, as a fast method to identify Xenorhabdus and Photorhabdus. Brigitte Brunei, M. Fischer Le Saux, A. Lanois and N. Boemare (INRA, France) 14.20 Application of PCR-RFLP of 16S rDNA to study the symbionts of tropical EPNs. Marion Fischer Le Saux, H. Maulon, . Brunei and N. Boemare (INRA, France) 14.40 Phylogenetic analysis of symbiosis between entomopathogenic nematodes and their bacteria. Jie Liu, R. Berry and G. Poinar (Oregon State University, USA) 15.10 Identification of Photorhabdus luminescens sub-species. Ralf-Udo Ehlers (University of Kiel, Germany) 15.40 Break 16.00 Molecular phylogenetics and taxonomie heterogeneity of Photorhabdus luminescens: old and new data. E. Szallas, Andrs Fodor, G. Kovacs, C. Koch, J. Burghart, O. Buss, (University of Budapest, Hungary), K. Nealson (University of Wisconsin, Milwaukee, USA) and E . Stackebrandt (DSMZ, Braunschweig, Germany) 16.30 Identification and characterisation of insecticidal toxin genes and bacteria associated with entomopathogenic nematodes. Alun Morgan, P. Jarrett, D. Ellis and M. Ousley (Horticulture Research International, Wellesbourne, UK) 17.00 Observations on the bacterial symbiont associated with the nematode, Steinernema abbasi (Steinernematidae : Nematoda). S. Elawad, R. Robson and Nigel Hague (University of Reading, UK) 17.30 Discussion

11

THURSDAY 23 APRIL Session 1C: Gnotobiological Studies on EPN Symbioses Chairman: Dr Nol Boemare 9.30 Nutritional significance of Photorhabdus bacteria for the growth and reproduction of Heterorhabditis nematodes. Richou Han, and R.-U. Ehlers (University of Kiel, Germany)

70.00 Heterorhabditis spp. and Photorltabdus spp.: co-evolution and gnotobiology. O. Oravecz, E. Szallas, H. Panjav, Andrs Fodor (University of Budapest, Hungary), B. Adams (University of Wisconsin, USA) and E. Stackebrandt (DSMZ, Braunschweig, Germany) 10.30 Break 11.00 Toward the physical map of Photorhabdus luminescens. Zsuzsanna Kiss, C. Ortutay, A. Fodor, (University of Budapest, Hungary ) and K. Nealson (University of Wisconsin, USA) 11.30 The concept of monoxenic associations in EPNs: limits of their natural occurrence and of gnotobiological interpretations. Nol Boemare, E. Bonifassi, M. Fischer Le Saux (INRA Montpellier-Antibes, France) and G. Smart Jr (University of Florida, Gainesville, USA) 12.00 Closing remarks

13

THE HIDDEN PROKARYOTIC DIVERSITY : UNRAVELLING THE EXTENT OF SYMBIOTIC ASSOCIATIONS

by Erko Stackebrandt

DSMZ-Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Mascheroder Weg 1b, 38124 Braunschweig, Germany

Abstract
It has been estimated (Haywood, 1995) that the number of validly described species of animals, plants and lower eukaryotes is about 500 times larger than those of prokaryotic species (in numbers 2.000.000 : 4.000). This small fraction of eubacteria and archaebacteria (or, as named today, Bacteria and Archaea) is surprising considering that piokaryotic species evolved more than 3.5 billion years ago, exploring and occupying any niche that have been investigated for the presence of prokaryotic organisms. Insects, on the other hand, comprising more than a million species evolved during the Cambium less than 600 million years ago. Recent progress in the elucidation of phylogenetic relatioships of prokaryotes which are associated with eukaryotic cells reveal that the majority of prokaryotic species may not be free-living but associated with the eukaryotic cells.

Introduction
In the pre-molecular era, endosymbiotic asssociations, recognized more than a century ago, concentrated mainly on the elucidation of the origin of chloroplasts and mitochondria. Later, the ability of bacteria of the genera Rhizobium, Frankia and Anabaena to associate with legumes, Casuarina and water ferns, respectively, caught the attention of physiologists. However, most endosymbiotic relationships between microorganisms and their eukaryotic hosts were mainly descriptive and the taxonomie affiliation of the vast majority of uncultured microorganisms remained virtually unknown. In contrast, microbial partners participating in non-obligate symbiotic relationships were identified long before the molecular era; examples have been described for the ectosymbiotic microbiota of the rumen, gut and skin. It was not before 1982 that Prochloron didemni, the endosymbiont of the ascidian Lissoclinum patellum, was affiliated to the cyanobacteria through the analysis of partial 16S rRNA analysis (Seewaldt and Stackebrandt 1982). Ten years later, a wealth of information is available on the molecular phylogeny of symbionts and endosymbionts from a broad spectrum of eukaryotic hosts, ranging from protozoa to vertebrates. The availability of a database, consisting of thousands of 16S rDNA sequences of free living prokaryotic species, facilitates the search for the branching point of host-associated bacteria in the phylogenetic tree. Fundamental questions about the identity of the prokaryotic partner, the evolution of symbiotic relationships, and the mechanisms of symbiont transmission can now be addressed.

14

Elucidation of phylogenetic relationships

As the evolutionary history of contemporary organisms can be inferred from the analysis of the primary structure of homologous molecules which accompanied the history of the organisms from early evolution on (Zuckerkandl and Pauling, 1965) they can be ranked according to their evolutionary history. Number and nature of sequence differences among proteins and genes coding for rRNA and proteins reflect their phylogenies and consequently allow to recognise pairs or groups of organisms which evolved from a common ancestor. It might even be possible to measure the time elapsed since the divergence of the most recent ancestor. However, this assumption is true only if changes are selectively neutral and independent of the overlying phenotype and the selection pressure dictated by its function. While the molecular clock of the majority of the about 8.000 available 16S rDNA sequences runs by and large isochronically (i.e. with constant rate), bradytelically (slower than isochronic) and tachytelically evolving (faster than isochronic) organism are known. As pointed out below, there is molecular evidence that many bacteria change their tempo of molecular evolution during or after their manifestation as obligate endosymbionts within the eukaryotic host. Several algorithms are available for inferring phylogenetic trees from 16S rDNA data: pairwise distance, maximum parsimony, and maximum likelihood methods (Li and Graur, 1991). Trees derived from different analyses agree in general; most deviations are found in the branching points of deeply rooting branches. Results of comparative analyses of other conservative molecules responsible for central functions such as elongation factors, subunits of ATPase, and RNA polymerases in general do support the rRNA data (Schleifer and Ludwig, 1989). As numerous factors influence the topology of a branching pattern the taxonomist is in a difficult position to judge about the quality of the phylogenetic tree he has generated. It is well documented that a phylogenetic tree is a dynamic construct which will change with any new sequence included (Stackebrandt, 1991). The branching points as well as the length of branches are influenced, among other factors, by the base ratio of the rDNA molecule. In contrast to the broad spectrum of mol% G+C content of DNA of prokaryotic species (ranging from 27 to 76 mol%) that of the rRNA gene of the vast majority of cultivated prokaryotic species, as well as those of many endosymbionts (e.g. those of mealybugs, Solemya, Lucinoma) range between 52 and 56 mol%. While many thermophilic species (but not exclusively these) show a higher G+C content of up to 60 mol% (Rainey et al., 1993), many endosymbionts have a rDNA G+C content of less that 50 mol%. To give only a few examples: endosymbionts of Besimia tabaci, 48,7 mol %; Mindarus vicroria 47,8 mol %; Glossina brevipalpis, 49,2 mol %, and Camponotus floridanus, 47,8 mol %). Thorough phylogenetic analysis (e.g. transversion analysis) is needed in order to generate phylogenetic patterns which reflect the course of the evolution of the analyzed molecule most closely.

Table 1.E xamples of symbiotic Bacteria and Archaea and their phylogenetic affiliation.
(lost rimyema compressa Metopus contortus l'lagiopyla nasuta Mastotermes darwiniensis Muscidifurax uiniraptor Bangasternus orientalis Rhinocyllus canictis Trichogramma dei on Nasonia vitripennis Dysmicoccus neobrivipes Solemya velum Heiiothis virescens Bathymodiolus thermophilus Calyptogena elongata Calyptogena magnifica Vesicomya chordata Anodonlia phillipiana Codakia orbicularis 7 'hyasira exuosa Lucina oridanus Riftia pachyptila Siphoninus phyllireae Trialeurodes vaporariorum Anomalops katoptrnn Kryplopharanon alfredi Cryptosaras coitesi Sitophilus zeamais Acyrthosiphon sp. Camponotus floridamis. (lossina paiidipes Bemisia abaci Phylogenetic group Anaerobic ciliate Anaerobic ciliate Anaerobic ciliate Australian termite Pteromalid parasioid of house fly Coleptera, weevil Coleptera, weevil Parasite wasp of Lepidoptcra Parasitic wasp Homoptera, mealy bug, Bivalve Moth Bivalve Bivalve Bivalve Bivalve Bivalve Bivalve Bivalve Bivalve Tube worm Ash whitefly Greehouse whitefly Deep sea anglerfish Flash light fish Deep sea anglerfish Coleptera, maize weevil Pea aphid Giant ants Tsetse Sweet potato whitefly Phylogenetic affiliation of symbiont Melhanocorpusculum Methonocorpusculum Melhanomicrobiaceae Spirochaeta -subclass -subclass -subclass -subclass -subclass -subclass -subclass -subclass -symbiont group I -symbiont group I -symbiont group I -symbiont group I -symbiont group II -symbiont group II -symbiont group II -symbiont group II -symbiont group II -subclass -subclass -subclass -subclass -subclass Enlerobacleriaceae Enlerobacleriaceae Enlerobacleriaceae Enlerobacleriaceae Enlerobacleriaceae Reference/EMBL accession number Z16412 Embleyetal., 1992 Z29437 Berchtold et al., 1994 Stouthamer et al., 1993 M85266 M85267 Stouthamer et al., 1993 Ghemaetal., 1991 Munson et al., 1992 M90415 L22481 Disteletal., 1988 L25719 Disteletal., 1988 L25713 L25711 Disteletal., 1988 Distel and Wood, 1992 L25707 Disteletal., 1988 ARB ARB Z19081 ARB Haygood et al., 1992 M85270 Untermanetal., 1989 Schrder et al., 1996 Beard et al., 1993 ARB

16

Identification of the symbiont

Most endosymbiotic bacteria are defined as "as yet uncultured" organisms which have not yet been cultivated on artificial growth media in the laboratory- although serious attempts to cultivate them are not routinely performed. Hence, genes coding for rRNA cannot be isolated selectively but must be identified within clone libraries of PCR amplified rDNA using prokaryote-specific PCR primers and total DNA that was extracted from the host-symbiont association (Jacobi et al., 1995). If the host contains a single symbiotic partner only, the clone library will consist exclusively of the one unique rDNA insert. If the association is more complex the clone library is likely to contain phylogenetically different inserts. In order to exclude the posssibility that the 16S rDNA sequence originate from a contaminating prokaryote and to verify the phylogenetic identity and the location of the putative endosymbiont within the hosts' tissue, the symbiont should be detected within the natural sample by in situ hybridization (see below). The 16S rDNA sequence will in most cases provide sufficient unique nucleotide stretches to allow generation of symbiont-specific oligonucleotide probes (Embley and Finlay, 1994; Jacobi et al., 1997). Key physiological properties, the presence of which may be derived from the symbionts' closest phylogenetic neighbors, may help to elucidate its ecological role, e.g the uptake of H2 from hydrogenosomes of ciliates by their archaeal methanogenic endosymbionts (Embley and Finlay, 1994).

The evolution of symbionts

Based upon the analysis of the many thousand 16S/18S rDNA sequences the phylogenetic tree of lifes presently comprises three domains, i.e. primary lineages, two of which are composed of prokaryotes (Archaea, Bacteria), one of eukaryotes (Eukarya). Most endosymbionts investigated so far originate from the domain Bacteria where they are found in a few main line of descent (Fig. 1). Here, the majority of endo-symbionts are moderately related to free living organisms. The ability to thrive in certain anaerobic protozoa of the genera Metopus, Plagiopyla and Trimyema appears to be widespread among methanogenic Archaea (kingdom Crenarchaeota) (Embley and Finlay, 1994). It is not surprising - but worth noting - that no symbionts have as yet been described to share a common ancestry with those prokaryotes which define the most deeply branching lineages, such as the Crenarchaeota, Aquifexales, and other branches comprising thermophilic and photototrophic organisms. As symbionts and non-symbionts share more than 80% 16S rDNA sequence similarity between with each other it can be derived from the plot shown in Fig. 2, that the invasion of the eukaryotic host by prokaryotic cells must have occurred less than 2 Gy ago. Figure 2 plots geological time with 16S rDNA similarity values determined for the emergence of key physiological types in evolution (cyanobacteria, facultative anaerobic bacteria, aerobic bacteria). Thus, the origin of endosymbionts correlates with the origin of the eukaryotic cell and endosymbiosis has occurred repeatedly in (perhaps all) newly evolving eukaryotic lineage.

17

In order to study the long-term history of symbiotic associations the phylogenetic trees of hosts and their symbionts should be compared. Few data only are available, the most convincing study being the one on the endosymbionts of aphids. The topology of the symbiont (Buchnera aphidicold) tree is completely concordant with host phylogeny, based upon morphology (Moran et al., 1993). The fossil and biogeographic time points for the host phylogeny has been used to calibrate the 16S rDNA of the closely related endosymbionts (> 8% similarity). The value of 1% fixed substitutions per 25 to 50 milllion years (Moran et al., 1993) is similar to the value of 1% per 50 million years determined by Ochman and Wilson (1987) on the basis of a broader range of non-obligate symbiotic relationships (e.g. Rhizobium/legumes; Photobacterium/fish; enterobacteria/mammals), and 1% per 60 million years suggested by Stackebrandt (1995) for the past 500 years on the basis of the correlation graph shown in Fig. 2. The hosts' advantage of the association has been unraveled in a few cases only such as the removal of hydrogen produced from hydrogenosomes of ciliates (see above), provision of nutritional carbon to the host bivalves by sulfur-oxidizing gill symbionts (Dando and Southward, 1986), or of essential amino acids to aphids by their endosymbionts (Baumann et al., 1998). Molecular tools have also been used to elucidate the transmission route of symbionts in tropical lunicid bivalves (Gros et al., 1998) and deep sea bivalves (Krueger et al., 1996) by application of the PCR techniques in ovaries, testis and gill tissue.

Symbionts of the Proteobacteria

As the majority of symbionts of invertebrates and vertebrates are clustering phylogenetically most closely with Gram-negative free-living bacteria it can be deduced from the molecular data (Figures 3 to 5), that Gram-negative bacteria are the most successful candidates for forming symbiosis, including obligate endosymbiotic associations. Most symbionts are found to cluster with members of the class Proteobacteria (Fig. 1), but not exclusively so. Bacterial symbionts of wood-eating cockroaches and termites, presently assigned to the genea Hollandia, Pillotina, Clevelandia and Diplocalyx, are large Gramnegative spirochetes. A symbiont of the Australian termite Mastotermes darA'iniensis has been described to be moderately related to Spirochaeta stenostrepta (Berchtold et al., 1994) (Fig. 3). Sphingobacterium comitans, a member of the Bacteroides Flav ob act er ium phylum, was shown to co-evolve as a symbiont in the myxobacterium Chondromyces crocatus (Jacobi et al., 1997). In-situ hybridisation with fluorescently labelled rRNA-directed probes and scanning confocal laser microscopy confirmed the symbiotic relationship between these two organisms as Sphingobacterium comitans was found within the sporangioles of the fruiting body o Chondromyces crocatus.. A few Gram-positive host-associated bacteria have been identified so far, such as the microparasite Pasteuria penetrans (in root-knot nematodes), Epulopiscium (fish symbiont) and Frankia (nitrogen fixing on Casuarina and relatives).

18
Proteobacteria, gamma Proteobacteria, beta Proteobacteria, alpha Proteobacteria, epsilon Proteobacteria, delta Synerglstts Flexispss Nitrospira Acidobacteriales Vemicomtcmbiales ffbmbacter Spirochaetales Cytophagales Chlorobiales dosiridia/bacjlli Actinobacteria cyanobacteria A +*-

Deinococcus/Thermus Chloroflexales Ccprcthermobacterium Pianctomycetales Thermotogales Dyctioglomus

Euryarchaeota Coryarchaeota Korarchaeota

-4-

Eukarya

10%

Fig. 1. Main lines of descent, showing the three domains Bacteria, Archaea and Eukarya as revealed by parsimony analysis of 16S rDNA. Source of sequences was the database ARB (ARBOR, ludwig @mbitum2.biol.chemie.tu-muenchen.de). Only relative distances are indicated. The presence of main lineages from which symbionts evolved are indicated by arrows. The scale bar indicates 10 estimated changes per nucleotide position.

19

Evolution of facultative anaerobic - aerobic prokaryotes

Geological time (Gy)

O O, content 2% 0 2 content 0.2% (Pasteur point) Origin of eukaryotes
8!g8!&SS&gggSg&&gg

Origin of mammals main radiation of aerobic bacteria Origin of vertebrates occurrence of aerobic bacteria

jaagaaiglaiai

radiation of cyanobacteria

3 -

occurrence of Fe" oxidizing phototrophes Origin of Life on Earth 80 90 1OO

% 16S r D N A similarity

Fig. 2. Correlation between the occurrence of certain bacterial groups during evolution as derived from paleochemical record (>3 Gy) and 16S rRNA sequence similarities between the most unrelated members of these groups (>80% similarity). The approximate origin of eukaryotes, vertebrates and mammals is indicated Modified from Stackebrandt (1995).

20

Burkholderia gladioli beta - Proteobacteria r DysmlCOCCUS neoblivipes endosymbiont " Pset/dOCOCCUS/ong/sp/nus emkjsymbiont PseudocOCCUS maritimus endosymbiont r Ehriichia chaffeen sis r- Ehrlichia mun s L- Cowdria rumin an tium Trichogramms corduben sis parthenogenesis bacterium Trichogramme deion parthenogenesis bacterium Trichogramme pretiOSUm parthenogenesis bacterium Wolbachia pipien tis Ban gastemus on en talis endosymbiont Nasoniavitripennis incompatibly bacterium aso nia longCOmlS incompatibility bacterium Nasonia giraulti incompatibility bacterium Nasonia Vitripennis incompatibility bacterium Muscifdurax un raptor parthenogenesis bacterium I Rhinocyllus con icus endosymbiont Ehrlichia sen n etsu "L Ehrlichia risticii Caedibacter caryophilus Hoiospora obtusa - Rickettsia n ca ade n sis Orientia tsutsugamushi

alpha Proteobacteria

Triticum 1 Secale S Zea Spimchaetales s

Spirochaeta stenostrepta Mastotermes darwiniensis endosymbiont

Fig. 3. Dendrogram of 16S rDNA relatedness between some symbi oti c members of the a- and subclasses of the class Proteobacteri a as revealed by parsi mony analysi s. Host-associ ated bacteri a in black, culti vated and descri bed speci es in red. Source of sequences was the database ARB. The scale bar indicates 10 estimated changes per nucleotide position.

21

From the viewpoint of physiological and morphological diversity and the number of described genera the class Proteobacteria constitutes the most diverse higher taxon within the domain Bacteria. Morphological diversity range from simple spherical forms, to myxobacteria forming fruiting bodies. Physiologies include chemolitho-autotrophy, photosynthesis, fermentation, anaerobic respiration, respiration, nitrogen fixation, and other types; Proteobacteria embrace organisms known to have close associations with eukaryotic hosts as pathogens, parthenogenesis bacteria (Fig. 3), incompatibility bacteria (Fig. 3), symbionts, and organelles (such as the mitochondria of plants which originated from aproteobacterial ancestors; Fig. 3). E ndosymbionts (primary symbiont) and non-obligately associates (secondary symbionts) of the same host belong to different phylogenetic groups (e.g. symbionts of Acyrthosiphon or the son-killer trait of Nasonia vitripennis). This finding indicates that the same host is susceptible to more than a single invasion process and that not all symbiotic relationships result in obligate endosymbiosis. -subclass of Proteobacteria. This subclass contains a wide spectrum of genera, the members of which are closely associated with eukaryotic cells. Prime examples are the nitrogen-fixing genera Rhizobium, Sinorhizobium, Bradyrhizobium, Mesorhizobium and Azorhizobium. The plant pathogen Agrobacterium tumefaciens can be considered a Tl-plasmid containing Rhizobium species. Pathogens include, among others, Afipia, Brucella, Bartonella, Rickettsia, Ehrlichia, Orienta, and Anaplasma. A highly related cluster within this subclass is formed by Wolbachia pipientis, the symbionts of the weevil Bangasternus orientalis, as well as the cytoplasmatically inherited bacteria such as the parthenogenesis bacteria (PB) of Trichogramma and Muscifidurax uniraptor, and the cytoplasmatic incompatibility bacteria (CIB) of Nasonia (Fig. 3). Other members of this group (not shown) are the PB and CIP of Culex pipiens, Drosophila simulons and Ephestia cautelld). As compared to the degree of 16S rDNA similarity among cultured strains they could be considered members of the same species. These relationships demonstrate that the common ancestor of Wolbachia and relatives invaded a broad spectrum of insect hosts which, as shown by the rather high similarity values of up to 99% similarity, must have ocuured rather recently in evolution. -subclass of Proteobacteria. Members of -Proteobacteria include a wide range of mainly pathogenic plant- and animal-associated bacteria, such as members of Burkholderia, Azoarcus, Ralstonia, Bordetella, K ingella, Eikenella, and Neisseria. Also included in this subclass is the kinetoplast of Crithidia (Du and Chang, 1994). The endosymbionts of the mealy bugs Dysmicoccus and Pseudococcus (Munson et al., 1992) are moderately related members of the genera Ralstonia and Burkholderia (Fig. 3).

22

-subclass of Proteobacteria. By far the highest number of endosymbionts of insects and vertebrates are members of this subclass, which contains a wide spectrum of animal and human pathogens, such as members of the families Enterobacteriaceae, Legionellaceae, Pasteurellaceae, Vibrionaceae, and of the genera Pseudomonas (sensu stricto), and Acinetobacter. The symbionts of entomopathogenic nematodes of the genera Steinernema and Heterorhabditis, i.e. Photorhabdus and Xenorhabdus, are members of the E nterobacteriaceae. E ndosymbionts of different origin are found to be related to different groups of free living bacteria. The endosymbionts cluster according to the phylogenic rank of their hosts which may be indicative of coevolution events: (i) the primary endosymbionts of giant ants, aphids, tese-tse, and the sweet potato white fly Bemisia abaci cluster within the radiation of enterobacteria (Fig. 4). These organisms have a 16S rDNA G+C content that is somewhat lower than that those of their cultured relatives (see above) but the application of algorithms that compensate for the bias does not change their membership to the enterobacteria; (ii) the symbionts of fish with light organs, such as the deep sea angler fish and the flash light fish, cluster with different members of Vibrio; (iii) two moderately related subcluster consist of the gill symbionts of bivalves. These organisms are remotely related to methylorrophic bacteria (Fig. 5) and Thiomicrospira (not shown). While the host of the first subgroup are deep sea animals, members of the second group are animals of coastal waters. All of their symbionts are sulfuroxidizing organisms which provide their hosts with nutritional carbon (Distel and Wood, 1992); (iv) the S-symbiont of the whitefly Bemisia tabaci and symbionts of other whiteflies form a separate phylogenetic cluster, as does Heliothis virescens testis endosymbiont and the symbiont of the bivalve Solemya velum symbiont (Fig. 5).

The taxonomy and naming of symbionts

The ease at which molecular data is generated has increased the involvement of sequencing technology in the characterization of prokaryotes that are difficult to cultivate and can only be described in limited terms. However, microbiologists are reluctant to accept a sequence-only description of a uncultured bacterial species, although the International Code of Nomenclature of Bacteria permits the naming of taxa for organisms that cannot at present be maintained in the laboratory as a pure culture, provided the presence of a description, preserved specimen, or illustration is provided as a type (Rule 16a of the Code). Examples are Prochloron dtdemni, Planctomyces bekefti, Pasteuria penetrans, as well as Budinera aphicola and Holospora obstura. Stackebrandt and Murray (1995) proposed the use the provisional taxon "Candidatus" for the characterization of uncultured organisms which, in addition to one or more gene sequences, should include structural, metabolic, and reproductive features, together with the natural environment in which the organism can be identified by in-sttu hybridization or other similar techniques for cell identification. The following properties and their combinations have been proposed to be inadequate for formal naming (Murray and Stackebrandt, 1995):

23

Klebsiella pneumoniae Erwinia carotovora Acyrthosiphon pisum ssymbiont RhopalOSiphum padl endoeymbiont I Acyrthosiphon pisum P endosymbiont Melaphis rhois endosymbiont Camponotus herculaneus endoeymbiont CamponotUS floridanus endoeymbiont Glossina palpalis Pendosymblont J Glossina tachinoides Pendoeymblont I Glossina austeni Pendoeymblont Glossina morSans Pendosymblont Glossina b revipalpis pendosymblont Besimia tab aci symbiont Photob acterium luminescens Xenorhabdus nematophilus Proteus vulgaris Arsenophonus nasoniae Budvicia aquatica Sitophilus zeamais endoeymbiont Glossina palidipes endoeymbiont Sitophilus oryzae endoeymbiont Ruminobacter amylophilus Vibrio nigripulchritudo Vibrio orientalis r Photob lepharon steinetzi symbiont "L Photoblepharon palpeb rals symbiont Photob lepharon palpeb rals symbiont Kryptopharanon alf redi symbiont Anomalops katoptron symbiont Vibrio mediterranei Vibrio fischeri Melanocetus johnsoni symbiont Cryptosaras couesi symbiont Vibrio aspartigenicus

rd

Bemisia tab aci symbiont Siphoninus phillyreae symbiont Trialeurodes vaporariorum symbiont

Fig. 4. Dendrogram of 16S rDNA relatedness of symbiotic members of the -subclass of the class Proteobacteria which are related to the families Enterobacteriaceae and Vibrionaceae. See legend of Fig.3 for further information.

24

Bemisia tabaci symbiont

1

Siphoninus phil l yreae symbiont

Tria l eurodes vaporariorum eymbtont Zymobacter pa l mae Marinobacter hydrocarbonoc l asticus Heliothis Virescens testis endoeymbiont Solemya vel um symbiont Thiothrix nivea Methylomicrobium a l bum Methylobacter l uteus Methylomonas methanica Louisiana Slope mussel symbiont Thiomicrospira crunogena Cal yptogena el ongate gin symbiont ' Ca/ypfogena Oregon gill symbiont r- Vesicomya chordata gin eymbtont ^ Cal yptogena magnifica symbiont Bathymodiolus thermophil us gin symbiont

r-CLucina fl oridana

Codakia costata gm symbiont giti symbiont Lucinoma aequizonata gin symbiont L Lucinoma annuiate gin symbiont Codakia orbicul aris gin symbiont f?/7/a pachyptila trophosome symbiont Thyasira fl exuosa gin symbiont Anodontia phil l ipiana gul symbiont

1 0 %

Fig. 5. Dendrogram of 16S rDNA relatedness of symbiotic members of the -subclass of the class Proteobacteria which form individual lineages. See legend of Fig.3 for further information.

25

(i) Information based exclusively on the analysis of clones generated from DNA that has been isolated directly from a natural sample. (ii) The authenticity of the genetic material has been verified in the environment by reamplification of DNA using sequence-specific PCR primers but the presence of cells containing that DNA has not been confirmed by microscopy or isolation. Most examples are available from molecular ecological studies using 16S rDNA for the assessment of microbial diversity. (iii) The origin of the genetic material from a living cell has been detected within the natural sample but the authenciticy of the bacterial cell within the host is not proven by in situ hybridization, such as in the case of the agent of bacillary angiomatosis and Whipple's disease bacillus. (iv) The origin of the genetic material from a living cell has been detected within the natural sample by in situ hybridization but information on properties other than its phylogenetic position and morphology is lacking. Examples for the use of the "Candidatus" concept, which may be considered an intermediate status between a mere sequence and a validly described species are "Candidatus Arsenophonus triatominarum", "Candidatus Phytoplasma australiensis", "Candidates Sphingobacterium comitans" and "Candidatus Microthrix parvicellum". The latter two examples refer to non-obligate symbiotic relationships. The "Candidatus" concept may encourage macrobiologists to contact microbiologists (and vice versa) in order to initiate communicate about the species concepts in bacteriology which does not follow the biological species concept. Proper classification, identification and nomenclature, i.e. proper taxonomy, is the clue to biology, and a prerequisit to facilitate understanding of evolutionary processes, including symbiosis. This field has received immense interest in the past in studies on biodiversity and conservation. The fact that a new phylogenetic lineage of symbionts and pathogens emerge with every new host investigated may lead to the exciting conclusion that the number of as yet uncultured taxa, worthy of species rank, may soon extend the level of cultivated bacterial species.

References
Baumann, P., Baumann, L., Clark, M.A., and Thao, M.L. (1998). Buchnera aphidicola: the endosymbiont of aphids. ASM News 64:203-209. Beard, C.B., O'Neill, S.L., Mason, P., Mandelco, L., Woese, CR., Tesh, R.B., Richards, F.F. and Aksoy, S. (1993). Genentic transformation and phylogeny of bacterial symbionts from tsetse. Insect Mol. Biol. 1:123-131. Berchtold, M., Ludwig, W., and Knig, H. (1994) 16S rDNA sequence and phylogenetic position of an uncultivated spirochete from the hindgut of the termite Mastotermes darwmiensis Froggatt. FEMS Microbiol. Lett. 123:269-274. Dando, P.R., and Southward, A.J. (1986). Chemoautotrophy in bivalve mollusks of the genus Thyasira. J. Mar. Biol. Assoc. UK 66:915-929. Distel, D.L. Sulfur-oxidizing bacterial endosymbionts: analysis of phylogeny and specificity by 16S rRNA sequences. J. Bacteriol. 170:2506-2510.

Distel, D.L., Lane, D.J., Olsen, G.J., Giovannoni, S.J., Pace, B., Pace, N.R., Stahl, D.A., and Feibeck, . (1988). Sulfur-oxidizing bacterial endosymbionts: analysis of phylogeny and specificity by 16S rRNA sequences. J. Bacteriol. 170:2506-2510. Distel, D.L., and Wood, A.P. (1992). Characterization of the gill symbiont of Thyasira flexuosa (Thyasiridae: Bivalvia) by use of polymerase chain reaction and 176S rRNA sequence analysis. J. bacteriol. 174:6317-6320. Du Y., and Chang K.P. (1994). Phylogenetic heterogeneity of three Crithidia spp. vs. Crithidia fasciculata. Mol. Biochem. Parasitol. 66:171-174. Embley, T.M., and Finlay, B.J. (1994). The use of small subunit rRNA sequences to unravel the relationships between anaerobic ciliates and their methanogen endosymbionts. Microbiology (Reading) 140:225-235. Embley, T.M., Finlay, B.J., and Brown, S. (1992) RNA sequence analysis shows that the symbiont of the ciliate Metopus contortus are polymorphs of a single methanogen species. FEMS Microbiol. Lett. 97:57-62. Gerna, R.L., Werren, J.H., Weisburg, W., Cote, R., Woese, CR., Mandeleo, L., and Brenner, D.J. (1991). Arsenophonus nasoniae gen. nov., sp. nov., the causative agent of the sonkiller trait in the parasitic wasp Nasonia vitripennis. Int. J. Syst. Bacteriol. 41:563-565. Gros, O., De Wulf-Durand, P., Frenkiel, L., and Moueza, M. (1998). Putative environmental transmission of sulfur-oxidizing bacterial symbionts in tropical lucinid bivalves inhabiting various environments. FEMS Microbiol. Lett. 160:257-262. Haygood, M.G., Sidtel, D.L., and Herring, P.J. (1992). Polymerase chain reaction and 16S rRNA gene sequences from the luminous bacterial symbionts of two deep-sea anglerfish. J. Mar. Biol. Ass. 72:149-159. Heywood, V.H. (1995). Global biodiversity assessment, pp. 107-192. University Press, Cambridge. Jacobi, CA., Amus, ., Reichenbach, ., and Stackebrandt, E . (1997). Molecular evidence for association between the Sphingobacterium-Uke organism "Candidatus comitans" and the myxobacterium Chondromyces crocatus. Appi. E nviron. Microbiol. 63:719723. Krueger, D.M., Gustavson, R.G., and Cavanaugh, CM. (1996). Vertical transmission of chemoautotrophic symbionts of the bivalve Solemya velum (Bivalvia: Protobranchia). Biol. Bull. 190:195-202. Li, W.-H., and Graur, D. (1991). Fundamentals of molecular evolution. Sinauer Associates, Inc., Sunderland, Massachusetts. Moran, .., Munson, M.A., Baumann, P., and Ishikawa, H. (1993). A molecular clock in endosymbiotic bacteria is calibrated using the insect host. Proc. Royal Soc. London Ser. B 235:167-171.

27

Munson, M.A., Baumann, ,, and Moran, N.A.(1992). Phylogenetic relationships of the endosymbionts of the mealybugs (Homoptera:Pseudococcidae) based on 16S rDNA sequences. Mol. Phylogenet. Evol. 1: 26-30. Murray, R.G.E ., and Stackebrandt, E . (1995). Implementation of the provisional status Candidatus for incompletely described procaryotes. Int. J. Syst. Bacteriol. 45: 186-187. Ochman, H., and Wilson, A.C. (1987). E volution in bacteria: evidence for a universal substitution rate in celluar genomes. J. Mol. Evol. 26:74-86. Rainey, F.A., Ward, N.L, Morgan, H.W. Toalster, R., and Stackebrandt, E . (1993) Phylogenetic analysis of anaerobic thermophilic bacteria: aid for their reclassification. J. Bacteriol. 175:4772-4779. Schleifer KH, and Ludwig W. (1989). Phylogenetic relationships among bacteria In: The Hierarchy of Life, ed. Fernholm, ., Bremer, . and Jrnwall, H., pp. 103-117. Elsevier, Amsterdam. Schrder, D., Deppisch, ., Obermeyer, M., Krohne, G., Stackebrandt, E ., Hoelldobler, ., Gbel, W., and Gross, R. (1996). Intracellular endisymbiotic bacteria of Camponotus species (carpenter ants): systematics, evolution and ultrastructural characterization. Mol. Microbiol. 21:479-489. Seewaldt, E., and Stackebrandt, E. (1982). Partial sequence of 16S rRNA and the phylogeny of Prochloron. Nature 295:618-620. Stackebrandt, E. (1992) Unifying phylogeny and phenotypic properties. In: The Prokaryotes (2nd ed.)., ed. Balows, ., Trper, H.G., Dworkin, M., Harder W., and Schleifer K.H., pp. 19-47, Springer-Verlag, New York. Stackebrandt, E . (1995) Bacterial phylogeny. In: Molecular Basis of Virus E volution, ed. Gibbs, A. J., Calisher, CH. and Garcia-Arenal, F., pp. 15-28. Academic Press, London Stouthamer, R., Breeuwer, J.A.J., Luck, R.F., and Werren, J.H. (1993). Molecular identification of microorganisms associated with parthenogenesis. Nature 361: 66-68. Unterman, B.M., Baumann, P., and McLean, D.L. (1989). Pea aphid symbiont relationships established by analysis of 16S rRNAs. J. Bacteriol. 171:2970-2974. Zuckerkandl E , and L. Pauling. 1965. Molecules as documents of evolutionary history. J. Theor. Biol. 8:357-66

29

MOLECULAR PHYLOGENY OF NITROGEN-FIXING BACTERIA IN SYMBIOSIS WITH PLANT ROOTS

by Philippe NORMAND

Laboratoire dEcologie Microbienne du Sol, OMR CNRS 5557, UCB Lyon], 69622 Villeurbanne Cedex, France. Tel. 33 4 7243 1377. Fax: 33 4 7243 1223. normand@biomserv.univ-lyonl.fr. Plants roots develop and thrive in the soil, an environment containing millions of bacteria, fungi and other micro-organisms per gram. The taxonomie diversity of these microbes is hard to assess but a few thousands are described in the Bergey's Manual (1989), the ultimate reference in determinative microbiology. More recently, our perception of this diversity has been modified by the realization that microbial groupings based on biochemical characteristics were misleading (Woese, 1987) and that several bacteria existed in water (Ward et a l , 1990), and in the soil (Stackebrandt et al., 1993) that were not grown in pure culture. The quasi-exponential increase in the precision of DNA-identification of microorganisms is illustrated by Fig. 1 which shows the yearly (blue) and cumulative (red ribbon) number of 16S rRNA sequences in Genbank. Given also that the soil is a repository of bacteria that are brought there by the wind, by rain, by excrements, by dying animal hosts and plant debris, it can be postulated that plant roots are in intimate contact with millions of bacterial taxa, many of which will have negative or positive effects on the growth of plants. Extant plant taxa will thus have evolved to coexist with this rich array and even benefit from it. Furthermore, some microbial and plant taxa have gone a step further and developped intricate relationships that have permitted the plants to gain access to the two most limiting elements in the soil: nitrogen and phosphorus (Brady, 1974). Fungi for instance have developped symbiotic structures called mycorrhizae which purpose it is to provide plants with an extended soil exploring capacity, these will not be considered further in this discussion. There are different types of root symbioses involving bacteria, the two best known being the nitrogen-fixing Rhizobiaceae and Frankia, which permit the host plants to grow in areas where nitrogen is limiting. Nitrogen-limiting environments today are sand dunes, volcanic lava, glacial moraines, burned fields and forests, mine spoils, steep mountainous slopes and these environments are indeed those where pioneer Legumes and actinorhizals thrive (Dixon and Wheeler, 1986). The taxonomy of these two groups of plants shows a contrasted situation between the tight grouping of plants in symbiosis with Rhizobiaceae (with Parasponia being of course outlying) and the spread of actinorhizal plants (Fig. 2). It has nevertheless been proposed that there was a measure of grouping that could be construed as showing an underlying propensity of a group of plants to establish root symbiosis (Soltis et al., 1995), but the arguments to support this are not overwhelming. In both types of symbiosis, the bacteria colonize and multiply in the rhizosphere, penetrate either through root hair or intercellular cracks, and induce the synthesis of adventitious roots that are subsequently colonized by the bacteria that differentiate into specialized structures, bacteroids in the case of Rhizobiaceae and diazovesicles in the case of Frankia.

30

1 1 L

I I ^^t^^n^ <^&^6000 ^tVSfifp*"^ mt^f*^ f'""" L^Li^Sg ^^71^,yei^e^iin^-' \ | 2 0 0 0

^^Br

y&fr

> ^ ^ 160" ^ I 140C0 yay 1 1 2 0 0 C

hoooo

Fig. 1. Number of 16S rRNA (rrs gene) sequences present in Genbank. The blue ribbon is the number of sequences deposited each year while the red ribon is the resulting cumulative total which was 15086 in April of 1997.

Pro talcs - Elaeagnaccae: - Elaeagnus - Hippophae - Shepherdia

Rhamnales - Rhamnaceaee. g.:

Ceanothus

Colietia
Discaria

\
Fabales - Leguminosa e

Cas ua rales
- Casuarinaceae:

Rosales - Rosaceae e. g.: - Cercocarpus - Cowan io - Dryas Furshia ROSIDAE

Allocasuarina
Casuahna Gvmnostoma

Fgales - Betulaccae: - Anu s HAMAMELEDAE

Urticales - Ulmaceac: Farasponia

Hamameldales

Myricales - Myricaceae: Comptonia - Kvnca

\
CARYOPHYLLIDAE
Ranunculales - Conanaccac: Cortara

ASTERIDAE

Fig. 2. Phylogenetic relationships between plant taxa in symbiosis with nitrogen-fixing bacteria according to the classification of Cronquist (1988) The red arrows indicate the taxa in symbiosis with Rhizobiaceae

Thousands of strains of both bacteria have already been isolated and characterized from both a biological and a taxonomical point of view. One major characteristic that emerges from such studies is the lack of congruence between infectivity and phylogeny of the

31

bacteria. The taxonomy and phylogeny of these bacteria has been studied with phenotypic methods and later on, with the advent of PCR (Mullis and Faloona, 1987) with genomic methods. The 16S rRNA (rrs) gene has been the most thoroughly studied and surprising affiliations have been found. In the case of Rhizobium, the tangled trees (Fig. 3) comprises interspersed Agrobacterium species and this suggests that numerous separate genetic events have taken place and lateral transfer of symbiotic genes is thus postulated. It also comprises other telluric and animal pathogens bacterial genera.
Nitrobacter "**" Rhodopseudomonas Blupini 100 Bjaponicum Afipia Phyllobacterium ' - ^ Mtianshanense 92 90 Mamorphae Mhuakuii 99 Mmediterraneum Mloti Mycoplana Steranga ^ 99 Sfredii Smeliloti Blastobacter Avitis 100 Atumefaciens Arubi 92 Rgalegae Rgiardinii 100 Arbizogenes ' Rtropici Rleguminosamm Retli 94 100 - Rgallicum Rmongolense Rhainanensis Xanthobacter -_, Acaulinodans ^ ^ _

, 0.01 s/s ,

100

-c

100

I '

Fig. 3. Phylogenetic relationships established through comparison of rrs genes between alpha Proteobacteria in nitrogen-fixing symbiosis with plants and their close neighbours. The Rhizobacteriaceae species names are preceded by a first letter that stands for the five genera (B= Bradyrhizobium, M=Mesorhizobium, S= Sinorhizobium, R= Rliizobium and A= Azorhizobium). The blue arrows indicate the Agrobacterium spp. and the red arrows those other taxa that are not symbiotic with Legumes. The numbers at nodes indicate those groupings supported by bootstrap analysis (Felsenstein, 1985). The bar stands for a genetic distance of 0.01 substitution per site.

The particularly long (600-1200 nt) ribosomal IGS can be used for high resolution phylogeny in this group of alpha Proteobacteria (Normand et al., 1996) or simply used for rapid large-scale strain discrimination through PCR/RFLP (Khbaya et al., in preparation). Host specificity in this group of bacteria depends mainly on the presence nod genes, the products of which sense molecules such as flavonoids exsuded by the plant roots (nodD) and others (nodABCE...) that participate in the synthesis of Nod factors that are specific for certain host plants. Some bacteria have several copies of the sensor gene nodD and the resulting phylogeny shows a high rate of evolution and a reticulate pattern. These illustrates

32

the poor value of this gene as a phylogenetic marker because of its pivotal role in the symbiosis and the resulting complex evolutionnary pressures.
0.1 100

92

100

r 78 60

99

Bradyrhizobium-Parasponium BelkaniiDl BjaponicumDl BelkaniiD2 BjaponicumD2 RtropiciD2 RtropiciDl Rleguminosarum-viciaeDl SmelilotiD2 - SmelilotiDl RgalegaeDl Rleguminosarum-trifoliiD RgalegaeD2 SmehlotiD3 SfrediiD2 Rleguminosarum-phaseoliD2 SfrediiDl R]eguminosarum-phaseoliD3

100

Azorhizobium Fig. 4. Phylogenetic relationships established through comparison of nodD genes between Rhizobacteriaceae species. The numbers at nodes indicate those groupings supported by bootstrap analysis (Felsenstein, 1985) The bar stands for a genetic distance of 0.1 substitution per site.

Nitrogen fixation is brought about by an enzymatic complex coded for by genes designated as nif The complete sequence of those genes in Klebsiella pneumoniae is now known (Arnold et al., 1988), there are about 20 genes and one of the most conserved of these is nifH (Normand et al., 1988) that codes for the dinitrogenase reductase protein. This gene is the one which sequence has been determined for the largest number of microorganisms, both symbiotic and non-symbiotic, both aerobic and anaerobic. The resulting phylogeny is puzzling because in general it reflects the one obtained by analysis of the rrs gene except that the beta Proteobacteria nifH sequences are split into two clusters, that alternative vanadium and ironbased nitrogenases (anf) are close to those of anaerobic bacteria, but most importantly because Azoarcus sequences are very much split, some close to the Azotobacter, others close to Thiobacillus and Azorhizobium (Fig. 5). This illustrates that nifH is also not a good phylogenetic marker because it has been subject to differential selection pressure (anaerobic vs aerobic environments) but being often plasmid-borne such as in Rhizobium, it is prone to being exchanged through conjugation. Nevertheless, the available data with several genes indicate that what was once considered a coherent taxon named Rhizobiaceae is now not much more than a convenient catch-all term for a group of bacteria that has had a complex evolutionnary history. This group of strains is now undergoing a deep remodelling with new species appearing in almost all issues of the IJSB and new genera proposed now and then to accomodate the wealth of data generated, especially now that molecular techniques make it easy to get rapidly reliable data and that focus is made on species-rich tropical genera.

33

Klebsiella gamma/beta Vibrio Alcaligenes ^"~ Azoarcus-communis Azotobacter 89 Nostoc 53 Anabaena Cyanobacteria Lyngbya c 56 Plectonema 64 Frankia Herbaspirillum Burkholderia Bradyrhizobium Azorhizobium alpha/beta Thiobacillus A**"" Azoarcus-6a3 Rhodobacter Rhizobium Azospirillum Sinorhizobium Paenibacillus Clostridium-pastl 89 Desulfovibrio Chromatium 85 Clostridium-past3 88 Azoto-anH3 M-ivanovi Fig. 5. Phylogenetic relationships through comparison nifH genes between bacterial species. The numbers at nodes indicate those groupings supported by bootstrap analysis (Felsenstein, 1985). The bar stands for a genetic distance of 0.1 substitution per site. Blue arrows indicate the Rhizobiaceae spp. and red arrows Azoarcus spp. sequences that are widely separated, indicating a ancient lateral transfer (Hurek et al., 1997). 0.1 s/s The situation of genus Frankia is different given the large number of uncultivated strains. This has promoted work focussed on the direct characterization of strains in nodules and in the rhizosphere. From that type of approach, the closest phyletic neighbor to Frankia was also found not to be the soil actinomycete Geodermatophilus with a comparable morphology (Hahn et a l , 1989) but the thermal spring cellulolytic rod Acidothermus (Fig. 6). Genus Frankia appears to be much more coherent (i.e. there are no non-symbiotic taxa interspersed with symbiotic ones) than the Rhizobiaceae species (Normand et al., 1996). It may be that it is due to the fact that actinomycetes in general are harder than alpha Proteobacteria to cultivate as it is known from work targetting antibiotic producers. This bias may be offset by direct characterization of the soil community using specific primers (Normand and Chapelon, 1997).

34

A/nus gen sp. 0.0072 s/s Alnus gen sp. 2, 3 Uryas* un-isolated Elaeagnusgen. sp.4-8 wCTI^~""~*""1

Frankia genus

non N?-fixcrs

/ww-rhizophere sequences

Acidothermus cellulolyticus

Geodermatophilus group Sporichthya polymorpha Actinoplanes group (incl. sequences mistakenly described in GenBank asFranfa'aG48 & L27) Streptomyces ambofaciens Dermatophilus congolensis Fig. 6. Phylogenetic relationships established through comparison of rrs genes between Frankia species (coloured) and neighbours. The bar stands for a genetic distance of 0.0072 substitution per site. In conclusion, the wealth of data pouring into D NA databases, the completely sequenced bacterial genomes and the new detection tools becoming available will modify the way we look at symbioses. In particular, knowledge of the genes involved in the symbiotic processes and the comparison of their sequences may provide insights on the way evolution has taken place in these two groups of symbiotic nitrogen-fixing bacteria that have permitted these groups of plants to evolve into irreplaceable pionneer plants. References Arnold,W., R u m p A , Klipp,W., Priefer,U.B. and Puhler, A. 1988. Nucleotide sequence of a 24,206-base-pair D NA fragment carrying the entire nitrogen fixation gene cluster of Klebsiella pneumoniae. J. Mol. Biol. 203: 715-738. Brady, N.C. 1974. The nature and properties of soils, 8th ed. Macmillan Pubi. Co. Inc. NY. NY. 639p. Cronquist A. 1988. The evolution and classification of the flowering plants, 2nd edition. Allen Lawrence, Kansas. Dixon R.O.D . and Wheeler C.T. 1986. Nitrogen Fixation in Plants. Blackie & sons ltd, Glasgow. 157p.

35

Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-791. Hahn, D., M. P. Lechevalier, A. Fischer, and E. Stackebrandt. 1989. Evidence for a close phylogenetic relationship between members of the genera Frankia, Geodermalophilus, and "Blastococcus" and emendation of the family Frankiaceae. Syst. Appi. Microbiol. 11: 236242. Hurek, T., Egener, T., and Rheinhold-Hurek . 1997. Divrgence in nitrogenases of Azoarcus spp., Proteobacteria of the Beta subclass. J. Bacteriol. 179: 4172-4178. Khbaya B., Zerhari K., Neyra M. Normand P. and Filali-Maltouf, A.K. Diversity of Rhizobia nodulating Acacia spp. in Morocco through the analysis of rRNA genes. In preparation . Mullis, K. B., and F. A. Faloona. 1987. Specific synthesis of DNA in vitro via a polymerasecatalized chain reaction. Methods Enzymol. 155: 335-350. Normand et Chapelon C. 1997. Direct characterization of Frankia and of close phyletic neighbors from an Alnus viridis rhizosphere. Physiologia Plantarum 99 : 722-731. Normand , Orso S, Cournoyer B, Jeannin , Chapelon C, Dawson J, Evtushenko L, and Misra AK. 1996. Molecular phylogeny of the genus Frankia and related genera and emendation of Family Frankiaceae. Int. J. Syst. Bacteriol. 46: 1-9. Normand P., Ponsonnet C, Nesme X., Neyra . and Simonet P. 1996. ITS analysis of procaryotes. In : Molecular Microbial Ecology Manual", A.D.L. Akkermans, J.D. van Elsas and F.J. De Bruijn (eds.), Kluwer acad. pubi., Dordrecht section 3.4.5, pp. 1-12. Normand P., Simonet P. and Bardin, R. 1988. Conservation of nif sequences in Frankia. Molecular and General Genetics 213 : 238-246. Soltis, D.E., Soltis, P.S., Morgan, D.R., Swensen, S.M., Mullin, B.C., Dowd, J.M. and Martin, P.G. 1995. Chloroplast gene sequence data suggest a single origin of the predisposition for symbiotic nitrogen fixation in angiosperms. PNAS 92: 2647-2651. Stackebrandt, E., Liesack W. and Goebel B.M. 1993. Bacterial diversity in a soil sample from a subtropical Australian environment as determined by 16S rDNA analysis. FASEB J. 7: 232236. Ward, D.M. Weiler, R. and Bateson, M.M. 1990. 16S rRNA sequences reveal numerous uncultured microorganisms in a natural community. Nature 344, 63-65. Williams, ST., M.E. Sharpe and J.B. Holt (eds.). 1989. Bergey's Manual of Systematic Bacteriology, Williams and Wilkins, Baltimore. Woese CR. 1987. Bacterial evolution. Microbiol. Rev. 51, 221-271.

37

COMPARATIVE AND PHYLOGENETIC ASPECTS OF TWO INSECT ENDOSYMBIOSES : APHIDS AND WEEVILS
by Abdelaziz Heddi, Hubert Charles, Yvan Rahb

1NSA-1NRA, Biologie applique 406, 20 ave. A. Einstein 69621 Villeurbanne, France.

Summary
Intracellular symbiosis is currently well documented regarding the role of the endocytobiotes on the metabolism and physiology of their host. Most insects feeding on an unbalanced diet such as phloem sap, seeds, or blood, harbor intracellular prokaryotes which supply the eukaryotic host cells with essential metabolites such as vitamins or amino acids. In the cereal weevils Sitophilus oryzae, these bacteria are intimately implicated in the energetic metabolism of their host by synthesizing pantothenic acid, riboflavin, and biotin; and increasing thereby mitochondrial enzymatic activities. In aphids, the intracellular bacteria Buchnera shows unusual molecular features allowing potentially increased synthesis of tryptophan and leucine. Some of the genes of the respective biosynthetic pathways are indeed located on plasmids. Phylogenetic analyses based on the small subunit rD NA (16S) are routinely used to infer evolutionary relationships between bacteria. Such analyses placed both the weevil endocytobiote and Buchnera within the gamma group of proteobacteria. However, what remains unclear and questionable is first, the relative positions of these bacteria (and more generally intracellular bacteria) among the nearby range of extracellular bacteria and second, the precise phyletic origins of the known endocytobiotes, because intra- and extra-cellular bacteria were shown to undergo different evolutionary processes. It is for example generally assumed that the G+C content of obligate intracellular bacteria is relatively low, resembling that of cytosolic organelles like mitochondria. This particular aspect has made doubtful the precise phylogenetic positioning of intracellular bacteria, either between each other or within the extracellular bacteria, since the phylogenetic tree profiles are changing with the phylogenetic methods used for tree construction.

Introduction
Intracellular symbiosis (or endocytobiosis) represents the most sophisticated relationship between eukaryotic cells and intracellular microorganisms. This association implies the coordination between associated genomes forming the symbiocosm. The weevil symbiocosm {Sitophilus oryzae) involves as much as four interacting genomes: nuclear, mitochondrial, the principal endocytobiote (-proteobacteria), and the Wolbachia (aproteobacteria), a non integrated bacterium infecting many strains of the weevils. The aphid symbiocosm is devoid of Wolbachia, but harbours a primary endocytobiote called Buchnera (Baumann et al, 1995), a secondary endocytobiote in some species and even an additional rickettsia in some populations of the pea aphid Acyrthosiphon pisum (Chen et al., 1997). In most insects feeding on unbalanced diets, integrated endocytobiosis often displays the differentiation of specialised host cells or bacteriocytes, packed with bacteria. Bacteriocytes themselves form a tissue (bacteriome), arranged as a packsaddle around the beginning of midgut of the weevil larvae, or localised in the haemolymph of the aphid larvae. In both

38

weevils and aphids, endocytobiotes are transmitted to the progeny exclusively by the oocytes, and are controlled for their number and localisation by the host chromosome, as it was shown in weevils (Nardon et al., 1998). The elimination of Buchnera from the aphid results in a drastic decrease of the insect weight and fecundity, but adult viability is not depressed (Rahb et al., 1993). In weevils, bacterial elimination does not kill the insects either, but impairs many physiological traits. Development time increases by 25-30%, fertility decreases by 1525% (Nardon, 1973), and adult loose their flying abilities (Grenier et al., 1994). All these physiological disruptions were shown to be related to the metabolic role of Sitophilus oryzae principal endocytobiote (or SOPE), which participates for example in the energetic metabolism of its host by increasing mitochondrial enzyme activities (Heddi et al., 1993). In aphids, Buchnera is mainly involved in amino acid synthesis and supply to the host (Febvay et al., 1995; Liadouze et al., 1996; Douglas, 1998). This bacterium shows unusual molecular features that should allow increased synthesis of two amino acids, tryptophan and leucine (Baumann et al., 1995). Anthranilate synthase, which is the first and rate-limiting enzyme of the tryptophan pathway, is located on plasmids as multiple tandem repeats of around 3.6Kb in most aphid species (Lai et al., 1996). For leucine, the whole leucine operon is located on a 7.8Kb plasmid (Bracho et al., 1995). Table 1 summarizes the main differences between the weevil and aphid endocytobioses. Table 1 : Compared features of two insect intracellular symbioses
(grain) SOPE host insect range weevil symbiosis aphid Buchnera symbiosis

Coleptera, Curculionidae unchecked outside genus Sitophilus co- unknown (aged <135 My)

Hemiptera, Aphidomorpha all Aphidoidea, except Phylloxeridae, Adelgidae strictly vertical, down to the species level (aged 150-200 My) maternal, cyclic reinfection of embryos (eggs) at blastod. stage nymph: peri-intestinal cell clusters adult: Id.+ embryos or eggs no stable line available (antibiotic feeding)

host/ symbiont evolution transmission

maternal, permanent infection of germ line larvae: fore/midgut junction adult: gastric caeca, ovariole apex availability of permanent lines (heat-cured or antibiotictreated) Proteobacteria, 3 subdivision Enterobacteriaceae

location of bacteri ome

experimental aposymbiosis

microbial lineage

Proteobacteria, 3 subdivision

GC content, % (global) 5 4 % GC content, % (16S 5 4 % gene)

ca. 30 % 50%

39

(grain) SOPE number identified of

weevil

symbiosis aphid Buchnera >100

symbiosis

genes 3 (16S, groEL, groES)

gene expression studies overexpression of groEL (bacterial genes) protein extrachromosomal elements 138 kb e.c. element

overexpression of groEL protein rpoH, dnaK, dnaJ, thrB, argA leu, tip (+ 1 heat-shock) plasmids

In conclusion, the metabolic and the physiological roles of intracellular bacteria are quite well investigated not only in weevils and aphids, but also in other endocytobiotic models such as ants, cockroaches and glossine flies. However, although molecular phylogeny recently resolved the global taxonomie positioning of most non-cultivable endosymbionts, some specific features of their molecular evolution (see below) render their fine tuning positioning in bacterial systematics somehow questionable. Important questions in term of evolutionary history of such symbioses still lack unambiguous answers (are some bacterial lineages more likely to have produced insect or invertebrate endosymbionts ? what is the closest free-living relative to a given endosymbiotic microorganism ?... ) The aim of the last part of this paper is to shed light on peculiar evolutionary aspects related to the life history of intracellular bacteria that may influence their phylogenetic positioning by standard molecular methods.

Molecular characteristics of intracellular bacteria

Recent studies have shown that intracellular bacteria undergo different evolutionary processes from the free living bacteria (Moran, 1996; Heddi et al., 1998). 1- Faster Evolutionary rates By comparing divergence of homologous regions of ribosomal DNA from aphids and Buchnera, substitution rates were shown to be 36 times greater for Buchnera than for their aphids hosts (Moran et al., 1995). Furthermore, 16S rDNA evolves about twice as fast in Buchnera as in related free-living bacterial lineages (Moran, 1996). One of the driving causes of such high evolutionary rates is related to their intracellular life history. Endocytobiotic bacteria are limited to their host space, a situation preventing recombination with other bacterial populations, which is a crucial phenomenon in prokaryotic evolution. Furthermore, they are submitted to a bottleneck effect during their transmission from one generation to the other, with only a small bacterial population infecting the embryo and giving rise to the new symbiotic population at each generation. 2- Mutational bias toward A~ content It is well documented that most of intracellular bacteria are A+T biased when compared to the extracellular bacteria, as recently reviewed (Heddi et al., 1998). Figure 1 shows that genomic G+C content of intracellular bacteria ranges from 26 to 60%, while the

40

overall range of intra- and extracellular bacteria is about 25-77%. The mean genomic G+C content is 39.5% for the intracellular bacteria and 56.4% for the extracellular bacteria. In this context, Buchnera (30% G+C) and SOPE (54% G+C) have different backgrounds. We may interpret this difference as related to the respective age of these endocytobioses (200-250 My for Buchnera while probably less than 135 My for SOPE), but peculiar molecular evolution of SOPE can not be excluded. In free-living prokaryotes with large and/or recombining populations, mutational bias has little effect on DNA compositions because it is overridden by selection, even at sites that are close to neutral. In intracellular conditions, selection may be less effective due to environmental stability. This, combined with a Muller's rchet in such asexual populations, may lead to a mutational bias affecting substitutions at both neutral and weakly selected sites. An analysis of the evolution of trp genes in Buchnera and E. coli actually confirmed one prediction to such situation, namely that elevated rates of nonsynonymous substitutions occur in Buchnera (Moran, 1996). AT bias seems to occur in most insect endosymbioses, with some notable exceptions such as the symbioses of weevils and mealybugs (Moran, 1996; Heddi et al., 1998).
25 20 15 10 5 0
-O

Extracellular
n = 202
56.4

A
Intracellular n = 17 Buchnera
39.5

5" 4" 3" 2" 1 " 0

iL.

In
40

,1
50

SOPE

li] , , , ,
10 20 30

Genomic G+C

60

70

80

90

100

Figure 1 : G + C content of genomic DNA of intra- and extracellular bacteria, from Heddi et al (1998)

3- Phylogenetic consequences Such evolutionary features, characteristic of the life history of intracellular symbioses, pose methodological problems to molecular systematics since they interfere with all basic assumptions underlying phylogeny inference'. Biases are probably limited when analysing evolutionary relationships within groups, which may be assumed to rely on similar rules of molecular evolution {e.g. host-symbiont coevolution in aphids); however, unravelling evolutionary relationships between endocytobiotes and their free-living relatives, or between different groups of symbiotic bacteria, may be much more sensitive to such biases. An
' which require analogous evolution rates among taxa, among sites, absence of mutational bias .

41

example of such uncertainties is given by the changes in relative positions of symbiont groups within the -proteobacteria, comprising symbionts from nematodes, whiteflies, aphids, weevils and ants (Figure 2). In the global tree extracted from the Ribosomal Database Project (not shown), E. coli is given as the closest neighbour to Buchnera (among the selected taxa presented in Figure 2, except all ant sequences and one SOPE sequence, absent from the RDP). However, using the same alignment and this restricted set of organisms, phylogeny reconstruction by standard methods results in slightly modified relationships (Fig. 2A), where aphid and ant symbionts are clustered together; finally, choosing a distance more adapted to sequences varying in their GC content (Galtier & Gouy, 1995, 1996) results in another tree topology, which places ant symbionts close to the SOPE lineage (Fig. 2B). Although both trees contain many weak bootstrap sections, their topologies are somehow different (see also whitefly symbiont positioning), stressing the need for a specific study of methodological biases when phylogenies include intracellular bacteria. Although the 16S gene seems well adapted to the analysis of this taxonomie level, information on other sequences may also be quite helpful. Within the protein data available for Buchnera for example, some genes might be used for such a purpose in that the inferred phylogenies are compatible with 16S based trees {groEL or ftsZ proteins), while others show drastic divergence from the standard position {rnpA -ribonuclease P- gene, unpublished results). Finally, non-sequence-based information will be needed to strenghten the phylogenies resulting from such expected refined phylogenies (genome organisation, operon structures, gene clustering and gene fusion events). Only such synthetic approaches will allow deciphering the evolutionary histories which led to the establishment of pathogenesis and/or symbiosis from parental free living organisms, and the growing availability of whole genome information is a crucial step towards such a synthesis.

Acknowledgements
We would like to thank Dr M. Gouy for his help in phylogeny reconstruction.

42

Xanthomonas campesina -Pseudomonas aeruginosa 96

0.02 whteHies _ Pasteureila cants ; Haemophilus influenzae Pasteureila haemolytica

100

; .

L rtl *

apiitdsl

Bucanera

100

1|

. P. vulgaris

60.Ple.shigel Yer.pesti3 1 Haf.alvGi

o l l 6 1 ZZZS.carotov I rtT--E r

Ha

marees

Sitophilus - Er. heroico

weevils

Ln_
3

rZ

.Glospa

-Xanthomonas campestris

0.02

' Pseudomonas aeruginosa 100 100

Pasteuretla canis

-czE
12 .''XehoriUbdtJA

Haemophilus influenzae ' Pasteuretla haomolytica

404

mtmatoaes

100 P.vulgaris

aphWe( Buchnera )

..

Sitophilus

WMVilt

:<'&^:5:;^:-^:;,:#;

Figure 2 : Neighbour-Joining trees, constructed with 1157 sites in a 16S gene alignment extracted from the ribosomal database (Madiak et al., 1997), showing the variation in positioning of invertebrate endosymbiont taxa within the -Proteobacteria, when using either a Kimura distance (A: upper tree) or a Galtier & Gouy distance (B: lower tree); bootstrap values indicated (n=500). Sequences for ant symbionts and SOPE were added to the alignment available at the RDP web site (march 1998).

43

References
Baumann P., Baumann L., Lai CY., Rouhbakhsh D., Moran NA., Clark MA. (1995) Genetics, physiology, and evolutionary relationships of the genus Buchnera: intracellular symbionts of aphids. Annu. Rev. Microbiol. 49, 55-94. Bracho AM., Martinez-Torres D ., Moya ., Latorre A. (1995) D iscovery and molecular characterization of a plasmid localized in Buchnera sp. bacterial endosymbiont of the aphid Rhopalosiphumpadi. J. Mol. Evol. 41, 67-73. Chen DQ., Campbell BC, Purcell AH. (1996) A new rickettsia from a herbivorous insect, the pea aphid Acyrthosiphon pisum (Harris). Curr. Microbiol. 33(2), 123-128. Douglas AE. (1998) Nutritional interactions in insect-microbial symbioses: Aphids and their symbiotic bacteria Buchnera. Annu. Rev. Entomol. 43, 17-37. Febvay G., Liadouze I., Guillaud J., Bonnot G. (1995) Analysis of energetic amino acid metabolism in Acyrthosiphon pisum: a multidimentional approach to amino acid metabolism studies in aphids. Arch. Insect Biochem. Physiol. 29, 45-69. Galtier N., Gouy M., Gautier C. (1995) Inferring phylogenies from D NA sequences of unequal base compositions. Proc. nati. Acad. Sci. USA 92, 11317-11321. Galtier N., Gouy M., Gautier C. (1996) SEA VIEW and PHYLO_WIN: two graphic tools for sequence alignment and molecular phylogeny. Comput. Applic. Biosci. 12, 543-548. Grenier AM., Nardon C, Nardon P. (1993) The role of symbiotes in flight activity of Sitophilus weevils. Entomol. exp. appi. 70, 201-208. Heddi ., Charles H, Khatchadourian C, Bonnot G., Nardon P. (1998) Molecular characterization of the principal symbiotic bacteria of the weevil Sitophilus oryzae: a peculiar G+C content of an endocytobiotic DNA. J. Mol. Evol. in press. Heddi ., Lefebvre F., Nardon P. (1993) Effet of endocytobiotic bacteria on mitochondrial enzymatic activities in the weevil Sitophilus oryzae (Coleptera, Curculionidae). Insect Biochem. Molec. Biol. 23, 403-411. Lai CY., Baumann P., Moran . (1996) The endosymbiont (Buchnera sp.) of the aphid Diuraphis noxia contains plasmids consisting of trpEG and tandem repeats of trpEG pseudogenes. Appi. Environ. Microbiol. 62, 332-339.

AA

Liadouze I., Febvay G., Guillaud J., Bonnot G. (1996) Metabolie fate of energetic amino acids in aposymbiotic pea aphid Acyrthosiphon pisum (Harris) (Homoptera: Aphididae). Symbiosis 21, 115-127. Maidak B. L., Olsen G. J., Larsen N., Overbeek R., Mccaughey M. J., Woese C. R. (1997) The RDP (Ribosomal Database Project). Nucl. Ac. Res. 25(1), 109-111. Moran N. (1996) Accelerated evolution and Muller's rchet in endosymbiotic bacteria. Proc. Natl. Acad. Sci. USA. 93, 287-2878. Moran N., von Dohlen CD., Baumann P. (1995) Faster evolutionary rates in endosymbiotic bacteria than in cospeciating insect hosts. J. Mol. Evol. 41, 727-731. Nardon P. (1973) Obtention d'une souche aposymbiotique chez le charanon Sitophilus sasakii Tak. : diffrentes mthodes et comparaison avec la souche symbiotique d'origine. C. R. Acad. Sci. 277D, 981-984. Nardon P., Grenier AM., Heddi A. (1998) Endocytobiote control by the host in the weevil Sitophilus oryzae, Coleptera, Curculionidae. Symbiosis, in press. Rahb Y., Delobel B., Febvay G., Nardon C, Nardon P. (1993) Are aphid bacterial endosymbionts involved in the biosynthesis of aphid peculiar triglycerides?. In: Endocytobiology V, Sato S., Ishida M., Ishikawa H. (Eds), Tubingen University Press, 139144.

45

BIOSYSTEMATICS, PHYLOGENY AND POPULATION GENETICS OF ENTOMOPATHOGENIC NEMATODES

by W.M. Hominick, D.J. Hunt, A.P. Reid, B.R. Briscoe and D.A. Bohan1

CABI Bioscience UK Centre (Egham), Bakeham Lane, Egham, Surrey, TW20 9TY, UK Long Ashton Research Station, Department of Agricultural Science, University of Bristol, LongAshton, BS18 9AF, UK

Introduction
This paper provides an overview of entomopathogenic nematode nomenclature and phylogeny, so that the most recent taxonomie status of the symbiotic partners of the bacteria can be appreciated. It is essential that the species and relationships of the nematode hosts are understood if any sense is to be made of the bacterial symbionts. Similarly, aspects of the biodiversity and population genetics of the nematodes are mentioned briefly, to provide a background against which the bacterial partners can be assessed.

Entomopathogenic Nematodes in the Phylum Nematoda

The higher order phylogeny of nematodes (i.e. at the level of family and higher) has always been controversial and fluid because of the morphological conservatism of the group. However, the powerful tools of molecular biology can provide new insights. Blaxter et al. (1998) recently produced the first attempt at a phylogenetic classification of nematodes, based on small subunit ribosomal DNA sequences from 53 species chosen to represent all the major taxa including animal and plant parasites as well as free-living ones. The results indicate that convergent morphological evolution has been extensive and that present higherlevel classification will need revision. They suggest that animal parasitism arose independently at least four times and plant parasitism three times. Their analysis included Steinernema carpocapsae and Heterorhabditis bacteriophora. Both share the same strategy of utilising bacterial symbionts to kill insects in which they then reproduce but do not share a common ancestry. This confirms Poinar's (1993) assessment based on morphology. However, he concluded that both belong to the group Rhabditida, while Blaxter et al. (1998) show that they are more distantly related than this would imply. This is because the order Rhabditida is paraphyletic, with members in two strongly supported clades, each representing a trophically diverse assemblage. Thus, Heterorhabditis is grouped with C. elegans and other members of the suborder Rhabditina and is particularly associated with members of the order Strongylida, which are small intestinal parasites of vertebrates (including hookworms). Steinernema groups with rhabditid families but also with the vertebrate parasitic genus Strongyloides and the plant parasitic orders Tylenchida and Aphelenchida. These groupings strongly support the notion that Heterorhabditis and Steinernema have independently evolved similar life strategies, and hence it is not surprising that their bacterial symbionts belong to two different genera, Photorhabdus and Xenorhabdus, respectively.

Two recent papers address the molecular phylogeny of steinernematid and heterorhabditid species (Reid, Hominick and Briscoe, 1997; Adams, Bumell and Powers, 1998). Before assessing these, it is necessary to consider entomopathogenic nematode nomenclature so that currently recognised species are apparent.

Nomenclature of Entomopathogenic Nematodes

A workshop convened by Working Group 1 (Isolation and Identification) of COST 819 produced a list of currently accepted species and a set of keys. These have been published and should be consulted for details and references (Hominick et al., 1997). The following list summarises the current situation, with S. abbasi added as it was described after the manuscript from the workshop was submitted. Family Steinern erna t ida e Chitwood & Chitwood, 1937 syn. Neoaplectanidae Sobolev, 1953 Type genus: Steinernema Travassos, 1927 Other genus: Neosteinernema Nguyen & Smart, 1994 Genus Steinernema Travassos, 1927 syn. Steineria Travassos, 1927 nee Steineria Micoletzky, 1922 Neoaplectana Steiner, 1929 Type species: Steinernema kraussei (Steiner, 1923) Travassos, 1927 syn. Aplectana kraussei Steiner, 1923 Steineria kraussei (Steiner, 1923) Travassos, 1927 Oxysomatium kraussei (Steiner, 1923) Skrjabin, Shikhobalova &Mozgovoi, 1951 Other species: 5. abbasi Elawad, Ahmad & Reid, 1997 S. arenarium (Artyukhovsky, 1967) Wouts, Mracek, Gerdin & Bedding, 1982 syn. Neoaplectana arenaria Artyukhovsky, 1967 Neoaplectana anomali Kozodoi, 1984 Steinernema anomalae (Kozodoi, 1984) Curran, 1989 S. affine (Bovien, 1937) Wouts, Mracek, Gerdin & Bedding, 1982 syn. Neoaplectana affinis Bovien, 1937 S. bicornutum Tallosi, Peters & Ehlers, 1995 S. carpocapsae (Weiser, 1955) Wouts, Mracek, Gerdin & Bedding, 1982 syn. Neoaplectana carpocapsae Weiser, 1955 Neoaplectana feltiae sensu Stanuszek, 1974, nee. Filipjev, 1934 Neoaplectana felliae pier darum Stanuszek, 1974 Steinernema feltiae pieridarum (Stanuszek, 1974) Wouts, Mracek, Gerdin & Bedding, 1982

47

Neoaplectana carpocapsae pieridarum Stanuszek, 1974 Neoaplectana dutkyi Turco, Thames & Hopkins, 1971 Steinernema dutkyi (Turco, Thames & Hopkins, 1971) Wouts, Mracek, Gerdin & Bedding, 1982 S. caudatum Xu, Wang & Li, 1991 S. ceratophorum Jian, Reid & Hunt, 1997 S. cubanum Mracek, Hernandez & Boemare, 1994 S.feltiae (Filipjev, 1934) Wouts, Mracek, Gerdin & Bedding, 1982 syn. Neoaplectana feltiae Filipjev, 1934 Neoaplectana bibionis Bovien, 1937 Steinernema bibionis (Bovien, 1937) Wouts, Mracek, Gerdin & Bedding, 1982 Neoaplectana leucaniae Hoy, 1954 Steinernema leucaniae (Hoy, 1954) Wouts, Mracek, Gerdin & Bedding, 1982 S. glaseri (Steiner, 1929) Wouts, Mracek, Gerdin & Bedding, 1982 syn. Neoaplectana glaseri Steiner, 1929 S. intermedium (Poinar, 1985)Mamiya, 1988 syn. Neoaplectana intermedia Poinar, 1985 S. karii Waturu, Hunt & Reid, 1997 S. kushidai Mamiya, 1988 S. longicaudum Shen & Wang, 1992 S. monticolum Stock, Choo & Kaya, 1997 S. neocurtillae Nguyen & Smart, 1992 S. oregonense Liu & Berry, 1996 S. puertoricense Romn & Figueroa, 1994 S. rarum (de Doucet, 1986) Mamiya, 1988 syn. Neoaplectana rara de Doucet, 1986 S. riobrave Cabanillas, Poinar & Raulston, 1994 S. ritter i de Doucet & Doucet, 1990 S. scapterisci Nguyen & Smart, 1990 syn. Neoaplectana carpocapsae 'Uruguay strain' of Nguyen & Smart, 1988 Species inquirendae: Neoaplectana agriotos Veremchuk, 1969 (? = S. carpocapsae) Neoaplectana belorussica Veremchuk, 1969 (? = S. carpocapsae) Neoaplectana bothynoderi Kirjanova & Putschkova, 1955 {? = S.feltiae) Neoaplectana gergica Kakuliya & Veremchuk, 1965 (? = S.feltiae) Neoaplectana janickii Weiser & Khler, 1955 Neoaplectana kirjanovae Veremchuk, 1969 (? = S.feltiae) Neoaplectana melolonthae Weiser, 1958 Neoaplectana menozzii Travassos, 1932 (? = S. affinefeltiae)

4S

Neoaplectana semiothisae Veremchuk & Litvinchuk, 1971 (? = S. carpocapsae) Neoaplectana tabanivora Rubstov & Polevik, 1979 Nomina nuda: Neoaplectana brevilarvalis pier idarum Sandner & Stanuszek, 1972 Neoaplectana chresima Steiner in Glaser, McCoy & Girth, 1942 (? = 5. carpocapsae) Neoaplectana dutkii Welch, 1963 (syn. of N dutkyi Turco et al, 1971 = S. carpocapsae) Neoaplectana elateridicola Veremchuk, 1970 (? = S. carpocapsae) Neoaplectana titovi Veremchuk, 1966 (syn. of T v " , elateridicola) Steinernema serratum Liu, 1992 Genus Neosteinernema Nguyen & Smart, 1994 Type species: N. longicurvicauda Nguyen & Smart, 1994 Other species: None described. Family Heterorhabditidae Poinar, 1976 Type and only genus: Heterorhabditis Poinar, 1976 Genus Heterorhabditis Poinar, 1976 syn. Chromonema Khan, Brooks & Hirschmann, 1976 Type species: Heterorhabditis bacteriophora Poinar, 1976 syn. Chromonema heliothidis Khan, Brooks & Hirschmann, 1976 Heterorhabditis heliothidis (Khan, Brooks & Hirschmann, 1976) Poinar, Thomas & Hess, 1977 Other species: H. H. H. H. H argentinensis Stock, 1993 brevicaudis Liu, 1994 hawaiiensis Gardner, Stock & Kaya, 1994 indica Poinar, Karunakar & David, 1992 marelata Liu & Berry, 1996 syn. H. hepialius Stock, Strong & Gardner, 1996

49

H. megidis Poinar, Jackson & Klein, 1987 H. zealandica Poinar, 1990 Species inquirendae: H. hoptha (Turco, 1970) Poinar, 1979 syn. Neoaplectana hoptha Turco, 1970 H. hambletoni (Pereira, 1937) Poinar, 1976 syn. Rhabditis hambletoni Pereira, 1937 Note: H. hawaiiensis and H. argentinensis may be junior synonyms of H. indica and H. bacteriophora, respectively (see Adams et al, 1998). With the exception of the steinernematids S. caudatum, S. neocurtillae and S. ritteri and the heterorhabditids H. brevicaudis and H. marelata, all of the described species have been characterised in our laboratory by molecular methods detailed in Hominick et al (1997). We are aware of several new species that are being described, but the names and details cannot be revealed until they are officially published. There is also a large number of undescribed species in laboratories around the world. For example, our collection at CABI Bioscience contains some 28 steinernematids and one heterorhabditid (from Malaysia) new to science. The geographical source and collectors are detailed in TABLE I. The isolates have been characterised by the same molecular methods used to characterise the named species. Indeed, over 500 isolates have been characterised in our laboratory. This established a data base for assessing variation between and within species, allowing us to deduce that the 28 species in TABLE I warrant specific status. Unfortunately, not all are still available in a viable state.

Molecular Phylogeny
A word of caution needs to be injected at this point. As increasing numbers of isolates are obtained and reared in laboratories, there are more chances for mix-ups, contamination and consequent incorrect identifications. It is essential to verify all identifications and to be sure that the source of DNA has been authoritatively certified. Molecular information attached to incorrect species identifications is becoming an increasing concern. Unfortunately, molecular data are sometimes given a status of scientific credibility and accuracy beyond the actual value. It must be linked to an authoritative name or it will be useless or even misleading. Adams et al (1998) assessed the phylogenetic relationships amongst described heterorhabditids using DNA sequences of the ITS1 region of the ribosomal tandem repeating unit. They examined 9 isolates, representing 8 described and one putative species. The species were those listed in the nomenclature section, except for H. brevicaudis, and the putative species was the "Irish" isolate K122 which cross-breeding and RFLP analyses suggest is a separate species. The analyses supported the existence of 3 closely related sister taxa, H. marelata - H. hepialius, H. indica - H. hawaiiensis and H. argentinensis - H. bacteriophora, with the authors cautiously suggesting that each may be conspecific (i.e. synonymous). Indeed, Stock (1997) has now decided that H. hepialius is a junior synonym of H. marelata. Work in our laboratory supports this and also that H. hawaiiensis is a junior synonym of//, indica and that H. argentinensis is a junior synonym of//, bacteriophora.

50

Table I. Steinernematids characterised by molecular methods at CABI BIOSCIENCE and assessed as being new to Science. Species 1 2 Reference code Several isolates Distribution characterised? B2 Yes UK & Cont. Europe B3 Yes UK & Cont. Europe CI Pi El Fl MY2 MY3 MY4 MY5 MY6 Macau Mt. Jiri T87 Ml Ruhuna SSL1 SSL2 Veni PC2 NC513 CM S2 Zimbabwe JM2 Vieti MC2 UH36 Yes Yes Yes Yes Yes Yes No No No No Yes Yes No No Yes Yes Yes No No No No Yes Yes No Yes Yes UK & Cont. Europe UK UK & Cont. Europe UK, Japan Japan Japan Japan Japan Japan China S. Korea Indonesia Malaysia Sri Lanka Sri Lanka Sri Lanka Venezuela Costa Rica USA Spain Spain Zimbabwe Malaysia Vietnam Colombia fCenya Collected by: Briscoe & Hominick; W. Steiner Briscoe & Hominick; P. Smits; J. Miduturi; D. Sturhan Briscoe & Hominick Briscoe & Hominick P. Richardson & R. Gwynn; W. Steiner Briscoe & Reid; Richardson & Gwynn; M. Yoshida M. Yoshida M. Yoshida M. Yoshida M. Yoshida M. Yoshida D. Fallon P. Stock D. Fallon E. Kondo D. Premachandra D. Amarasinghe D. Amarasinghe X. Fan C. Smith W. Brooks F. del Pino F. del Pino B. Briscoe & D. Kutywayo J. Mason S. Spiridonov J. C. Lpez-Nuez S. Mwaniki

"1

J)

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

Hence, it would appear that there are even fewer species of Heterorhabditis than thought, and that the ones that remain valid are spread widely over the globe. By contrast, steinernematids are much more species rich. Reid el al (1997) used RFLP analysis of the ITS region of the ribosomal DNA repeat unit to study 19 isolates belonging to 17 species. Since that paper was submitted for publication, the Kenya isolate has been described as S. karii and S. anomalae is now 5. arenarium. In any case, a tree was

51

constructed based on band sharing and this resulted in clusters of species which exhibited similar morphological or behavioural characters. In particular: S. scapterisci, S. carpocapsae, SSL2 and the Malaysia isolate all have infective juveniles (Us) less than 550 micrometers and all show nictation behaviour; The S. feltiae and 5. kraussei isolates grouped on the tree and these species are morphologically similar and sometimes confused; Species with large Us (5. cubanum, NC513, S. glaseri, S. arenarium and S. karii) grouped together; S. scapterisci and S. carpocapsae are morphologically similar but negative cross-breeding proved them to be different; S. arenarium and S. glaseri are morphologically similar and sometimes mistaken one for the other as are S. affine and 5. intermedium. It is reassuring that the molecular data apparently reflect biological groupings and that there is a basis for linking genetics with morphology. Hence steinernematid taxonomy seems to have a strong morphological basis.

Biodiversity
It is intriguing that there are so many species o Steinernema and comparatively few of Heterorhabditis. Bearing in mind how little of the globe has been surveyed (Hominick et al., 1996), and extrapolating from TABLE I, there must be a considerable number of steinernematid species yet to be discovered. It is unlikely that this is either a distortion induced by sampling methods or because there are fewer taxonomically useful morphological characters for heterorhabditids. Even if apparently unique morphological details are used to erect new heterorhabditid species, the molecular data seem to contradict this approach (Adams et al, 1998) and the conservative status quo prevails. Hence, there appear to be only a few valid species of heterorhabditids and the ones that do exist seem to have a global distribution. This phenomenon may be related to their unusual method of reproduction, whereby the first generation is hermaphroditic. Thus, successful reproduction in the host requires only one infective juvenile to establish with consequent reduction in genetic mixing. Hominick et al (1996) assessed the biodiversity and geographical distribution of entomopathogenic nematodes on a continent and country basis. If H. argentinensis is regarded as a synonym of H. bacteriophora, the latter species has been isolated from all continents, although so far not recorded from India and southeast Asia. Heterorhabditis indica is widely distributed in the tropics and H. megidis enjoys a wide distribution in temperate regions. The steinernematids, on the other hand, are much more species rich. Some, such as S. carpocapsae and 5. feltiae, are common and widespread whilst others {S. affine and S. riobrave, for example) appear to be more restricted in distribution. This may merely be an artefact caused by limited sampling and the relative prevalence of species - large numbers of samples must be taken in order to detect organisms with a low prevalence. For example, 5. longicaudum was isolated and described from China. It is, however, widely distributed in Australia (Bedding, pers. comm.) and has now been recovered from three sites in the USA and also from Korea (Stock, pers. comm.), a result confirmed by using molecular techniques.

52

Much of the world remains to be surveyed for entomopathogenic nematodes and understanding of their biogeography is in its infancy. There can be little doubt that the hand of Man has played an important part in their spread. Historically, soil has been moved around the world as a byproduct of trade, thus inadvertently transmitting nematodes. One welldocumented case concerns the potato, which originated in the high Andes and was introduced to Europe by the Spanish Conquistadores in the 16th century. At the same time, potato cyst nematodes were introduced on infested material. They have subsequently spread world wide as the crop gained in popularity, and infested material was disseminated with no regard for quarantine measures (Sharma et al, 1997). Another example is Radopholus similis, distributed with its banana hosts as infected planting material was transported worldwide to establish plantations (Sharma et al, 1997). There is no reason to suppose that entomopathogenic nematodes were not similarly dispersed in soil associated with planting material. In very recent times, entomopathogenic nematodes have been deliberately distributed as they were released for biological control programmes. Such influences of Man need to be acknowledged as we strive to understand biogeography and evolutionary relationships in the bacteria/nematode complex. As data accumulate on the biodiversity and biogeography of entomopathogenic nematodes, the information may well afford deeper biological insights. What are the important biological characters that allow H. indica, H. bacteriophora and H. megidis to disperse and establish globally? What is the gene flow between these populations? Why do steinernematids apparently speciate more than heterorhabditids? It is known that entomopathogenic nematodes are aggregated in distribution. Do steinernematids tend to form discrete local populations, so that the founder effect results in isolated gene pools and the potential evolution of new species? Until now, questions regarding population genetics for entomopathogenic nematodes have been purely speculative. Recently, a molecular technique called AFLP (Amplified Fragment Length Polymorphism) has been developed which allows the characterisation of populations within a species. Using AFLP, it is theoretically possible to ask whether different isolates of a species belong to one or more populations. A simplified AFLP protocol (Mueller et al, 1996) was modified slightly in our laboratory (Bohan, Reid and Hudson, in prep.) and used to assess the genetic diversity of steinernematids from a paired set of fields, designated Field 53 and 54, at IACR-Long Ashton Research Station (LARS), Bristol, UK. Both fields had previously been under winter wheat and at the time of sampling in early November 1997, both were bare of cover following recent cultivation. The sites were sampled on nested spatial scales 16m, 4m and lm apart. Approximately 200ml of soil was taken at each sampling point and baited with three Galleria larvae on two separate occasions three weeks apart. Individual infected larvae were placed in White traps and the nematode progeny collected. DNA was extracted from the nematode progeny of each infected larva and identified to species level using the ITS RFLP technique (Reid et al, 1997). Three species were recovered from the two sites: S.feltiae (20 isolates on the first baiting and 9 on the second), S. affine (5 isolates on the first baiting and 3 on the second) and Steinernema n. sp. Fl (one isolate only). The S. feltiae isolates were analysed using the simplified AFLP protocol. Some sites that yielded more than one infected Galleria resulted in AFLP profiles that were noticeably different from one another. In contrast, other sites yielded nematodes from all three larvae, the three sets of infectives producing identical patterns. Indeed, one site had three infections at both baiting times and the AFLP patterns for these six isolates were virtually identical (five identical and the sixth differing by only three bands). The AFLP patterns were analysed using GelCompar

53

package and a phylogenetic tree was constructed (using PHYLIP) to show relationships between the isolates. The populations grouped into three main branches. There was no correlation, however, between either spatial or temporal distribution, leading us to conclude that the nematodes in the two field sites were in effect acting as a metapopulation. This implies that there is considerable gene flow on the scale of the sites, at least for S. feltiae. However, these results are only preliminary and more extensive sampling to allow larger sample sizes is required. There are two important outcomes of this study: 1) the evidence for metapopulation dynamics within steinernematid populations and 2) that a practical and reliable technique is available allowing us to ask basic questions about population genetics of entomopathogenic nematodes .

Conclusion
This workshop will begin to assess the diversity of the bacterial symbionts of entomopathogenic nematodes. One challenge will be to test the species concept for the bacteria. It is essential to identify the nematodes accurately in any study of the bacteria, as the nematodes provide a living environment upon which the bacterial symbionts depend and in which they are protected and evolve. As knowledge of systematics and phylogeny increases for both groups of organisms and molecular data become readily available, a picture of evolutionary trends will emerge. It will be fascinating to compare the patterns for the bacteria within the two convergent nematode groups. A fundamental question, which this workshop will begin to address, must be : " What is the consequence of nematode gene flow, or lack of it, for the bacterial symbionts?" Put another way, "Are the bacteria co-evolving with their nematode hosts, or evolving independently?"

54

References Adams, B.J., Burnell, A.M. and Powers, T.O. 1998. A phylogenetic analysis of Heterorhabditis (Nemata : Rhabditidae) based on internal transcribed spacer 1 DNA sequence data. Journal of Nematology 30: 22-39. Blaxter, M.L., De Ley, P, Garey, J.R., Liu, L.X., Scheldeman, P., Vierstraete, ., Vanfleteren, J.R., Mackey, L.Y., Dorris, M., Frisse, L.M., Vida, J.T., and Thomas, W.K. 1998. A molecular evolutionary framework for the phylum Nematoda. Nature 392: 71-75. Elawad S., Ahmad, W. and Reid, A. P. 1997. Steinernema abbasi sp. n. (Nematoda: Steinemematidae) from the Sultanate of Oman. Fundamental and Applied Nematology 20: 435-442. Hominick, W.M., Briscoe, B.R., del Pino, F.G., Heng, J., Hunt, D.J., Kozodoi, E., Mracek, Z., Nguyen, K.B., Reid, A.P., Spiridonov, S., Stock, P., Sturhan, D., Waturu, C. and Yoshida, M. (1997) Biosystematics of entomopathogenic nematodes: current status, protocols and definitions. Journal of Helminthology 71: 271-298 Hominick, W.M., Reid, A.P., Bohan, D.A. & Briscoe, B.R. 1996. Entomopathogenic nematodes : Biodiversity, geographical distribution and the Convention on Biological Diversity. Biocontrol Science and Technology 6: 317-331. Mueller, U. G., Lipari, S. E. and Milgroom, M. G. 1996. Amplified Fragment Length Polymorphism (AFLP) fingerprinting of symbiotic fungi cultured by the fungus-growing ant Cyphomyrmex minutus. Molecular Ecology 5: 119-122. Poinar, G.O. Jr 1993. Origins and phylogenetic relationships of the entomophilic rhabditids, Heterorhabditis ana Steinernema. Fundamental and Applied Nematology 16: 332-338 Reid, A.P., Hominick, W.M. and Briscoe, B.R. 1997. Molecular taxonomy and phylogeny of entomopathogenic nematode species (Rhabditida : Steinemematidae) by RFLP analysis of the ITS region of the ribosomal DNA repeat unit. Systematic Parasitology 37: 187-193 Sharma, S.B., Price, N.S. and Bridge, J. 1997. The past, present and future of plant nematology in International Agricultural Research Centres. Nematological Abstracts 66: 119-142.

55

PCR-RFLP OF 16S rDNA, AS A FAST METHOD TO IDENTIFY XENORHABDUS AND PHOTORHABDUS.

by B. Brunei1, M. Fischer-Le Saux 2, A. Givaudan Boemare 2
2

, A. Lanois 2, and N. E.

Ecole Nationale Suprieure Agronomique de Montpellier, Institut National de Recherche Agronomique, Laboratoire de Recherche sur les Symbiotes des Racines and Unit de Recherche et de Formation de Science du Sol, 2, place P. Viala, 34 060 Montpellier Cedex 2 ', and Institut National de Recherche Agronomique, Laboratoire de Pathologie compare, C.P. 101, Universit Montpellier II, 34095 Montpellier Cedex 5 *, France.

Summary
Methods based on PCR-analysis of prokaryotic small-subunit ribosomal RNAs are a fast mean of typing bacteria that are especially difficult to identify by biochemical investigations. Simple DNA extraction or bacterial cell suspension, followed by DNAamplification and electrophoresis make these approaches technically non demanding and rapid. 16S ribosomal DNAs of 27 representatives of Xenorhabdus and Photorhabdus including type strains of the five recognized species were studied by performing a restriction fragment length polymorphism (RFLP) analysis on amplified 16S rRNA genes. Amplified 16S rDNAs were digested by eight tetracutter restriction enzymes. The resulting banding patterns of the strains were compared pairwise and the similarities were a measure of their relatedness that was represented in UPGMA tree form. The two genera were separated and branch off from other Enterobacteria representatives, with P. vulgaris being the closest species of Xenorhabdus and Photorhabdus. To identify valid species, three restriction enzymes are required: Cfol, Alui and Haelll. A relative high diversity was detected as 18 genotypes were found, suggesting several undefined species. The results showed a good correlation with previously published data based on DNA-DNA hybridization and sequence analysis of the 16S rRNA genes, allowing identification of strains to the level of genetic species. As further isolations are made, new genotypes may be added and this technique would be appropriate to obtain indicative phylogenetic and taxonomie information on symbionts of entomopathogenic nematodes. This information can be used to select strains for further detailed taxonomie studies and can be helpful for ecological studies addressing environmental adaptation.

Introduction
An important factor for a better understanding of Xenorhabdus and Photorhabdus bacteria and their different roles in symbiosis is the ability to identify species and track their host range and geographic distribution. This cannot be answered reasonable without a robust system of classification. Xenorhabdus and Photorhabdus are usually recognised by some phenotypic characters, such as catalase and bioluminescence tests but many classical identification tests are not sufficient to study population diversity as most strains are phenotypically very similar and fail to give positive results. The characterisation

56

was turned to DNA-based molecular biological techniques such as DNA/DNA hybridisation or 16S rDNA sequencing which showed differences (Rainey et al, 1995; Akhurst et al, 1996). PCR-based genomic typing provide a much more rapid way of screening large numbers of isolates. Simple DNA extraction or bacterial cell suspension, followed by DNA amplification and electrophoresis make these approaches technically non demanding and rapid. We chose to type bacteria by the Restriction Fragment Length Polymorphism (RFLP) analysis based on the 16S rRNA genes obtained after amplification (PCR). The ribosomal region is highly conserved among bacteria and it is now widely accepted that the ribosomal DNA can be used in molecular systematic studies. Amplified 16S rDNA restriction analysis (also named 16S ARDRA) has been shown to differentiate various bacterial species (Heyndricke/a/., 1996). In the work reported here, we have been analysing the 16S rDNA by PCR-RFLP from a selection of Xenorhabdus and Photorhabdus strains including all the available validated species.

Materials and methods

Bacterial strains A total of 27 bacterial strains, including 13 Xenorhabdus and 14 Photorhabdus, were used in this study. Three type strains of three other Enterobacteriaceae were included, Serratia marcescens, Proteus vulgaris, and Escherichia coli provided by the Collection of Institut Pasteur (Paris, France). Phase I and II were obtained as previously described (Boemare and Akhurst, 1988). Bacterial cultures and DNA extraction Bacterial cells for DNA extraction were cultivated on plates for 48h at 28C. Bacteria were pelleted by centri fugati on and washed twice. Either whole cells or purified DNA was used for PCR. The pellet was suspended in sterile Milli-Q (Millipore) water and was used directly as template for the PCR as described below. Otherwise DNA was extracted after a freezing-boiling cycle followed by a phenol extraction. Cells or DNA were stored at -20C until use. PCR-RFLP experiments The 16S rDNA region was amplified as described by Brunei et al. (1997). The universal primer pairs, designed by Weisburg et al. (1991) allows the amplification of the almost complete 16S rDNA (Fig.l). PCRs were performed in a PTC-100 Programmable Thermal Controller (MJ Research Inc.) in reaction mixtures of 50 contained 1 of cells or DNA (as described above), each deoxynucleoside triphosphate at 0.2 mM, IX Taq buffer, 2.5 U of Taq polymerase, and each primer at 0.2 . RFLP analyses were performed by restriction enzyme digests of 5-12 of unpurified PCR products. The eight following enzymes were used according to the recommendations of the manufacturers (Gibco BRL, Boerhinger): Cfo I, Hae III, Msp I, Alu I, Rsa I, Dde I, Htnf I and Sau3A I. Restriction fragments were separated on 3% NuSieve GTC agarose gels in TBE buffer. The UVilluminated gels were visualized and recorded with an video system (Appligene).

57

168 rDNA

ICS

2 3 8 rDNA

15S0 bp

Figure 1: Schematic representation of the general organization of a procaryotic rrn operon, indicating the relative position of primers used in the present study. IGS: inter genie sequence.

Xenorhabdus

Photorhabdus

1.6 Kb 1Kb 0.*Kb

Figure 2: Ethidium-bromide-stained 0.8% agarose gel displaying 16S amplification products. PCR amplification (A): from DNA of Xenorhabdus (lanes 1 to 3) and Photorhabdus (lanes 4 to 6), and (B): from whole cells after an initial heating at 95C for 3 minutes (lanes 1 and 5), 5 minutes (lanes 2 and 6), 7 minutes (lanes 3 and 7), and 10 minutes (lanes 4 and 8). L: ladder 123 bp (Gibco BRL); M: ladder 1 Kbp (Gibco BRL).

5S

PCR-RFLP data analysis The restriction patterns were compared in pairs for each enzyme digest. Fragments with the same estimated size were considered as fragments in common. The proportion of shared restriction fragments between two 16S rDNA sequences was calculated by the coefficient of Dice which was used to estimate the genetic divergence between patterns. A dendrogram was constructed from the similarity matrix by performing the UPGMA clustering algorithm, showing the clustering of the strains based on the similarities of their combined PCR-RFLP patterns.

Results and Discussion

The results of amplification from DNA extraction and from whole cell suspensions, are shown in Fig. 2. When the amplification was successful a band of about 1.6 Kb was obtained as expected. Direct amplification from whole cells could be easily obtained from Photorhabdus cultures (Fig. 2B). To lyse Xenorhabdus cells, several conditions of an initial heating, before the amplification program, were tested but no DNA was revealed, whereas for Photorhabdus in the same time conditions, a strong band was obtained. Thus, cells of Xenorhabdus appear more difficult to disrupt by heating than Photorhabdus ones, and Xenorhabdus cells should be firstly lysed by freezing-thawing cycles before DNA amplification. Then PCR products were cleaved by eight tetrameric restriction enzymes. Two were dimorphic {Rsal and Sau3M) only when compared to the three other Enterobacteriaceae, the six others showed polymorphism among Xenorhabdus and Photorhabdus strains. By combining all restriction patterns, polymorphism divided the 27 strains into 18 RFLP types: 9 types were detected for Xenorhabdus. Each species was distinct and even polymorphism could be seen inside species; for example there were three RFLP types inside Xenorhabdus bovienii (Brunei et al, 1997). No difference was detected between DNAs extracted from cultures in phase I and phase II. For Photorhabdus, nine 16S rDNA types were identified. In some cases, an extra band was detected: this can be deduced when you add the size of the restriction fragments and that you find a size which is higher than 1.6 Kb. This can be a drawback of this technique but it was easy to cope with by comparison with the closest patterns and because just one band was involved in few cases. To summarise, for detecting all of the diversity among Xenorhabdus, four enzymes should be used {Cfo I, Msp I, Hinf I and [Alu I or Hae III or Dde I]). Three enzymes {Cfo 1, Alu I and Hae III) are required to generate all the genotypes of Photorhabdus. To identify only the four Xenorhabdus species and all Photorhabdus types, three enzymes are necessary {Cfo I, Alu I and Hae III).

59

1 X nematophilus1 224, F1 ZXpoinanV, NC33 4K77 5XbeddingiF 6 X bovien'iF __7S 3 F3, SK2, F5 9SaV 10 P. luminescens', Hm 11C8406

ila

J-12IS5 "L27D1
13K80, HP88 1413*,79,-211,81 15 Meg 16 C1 17NZH 18 P. vulgaris'

IIb

19 coir
20 S. marcescens1
1

0.049

0.028

0.021 0.024 genetic distance

0.007

Figure 3 : UPGMA dendrogram based on PCR-RFLP data. A total of 93 bands was used.

60

Then the bands were scored for each RFLP type and a coefficient of similarity was calculated. This coefficient is used to generate the calculation of genetic distances, then the genetic distances were clustered by the unweighted pair-group method using arithmetic averages. The dendrogram obtained is shown in Fig. 3. Xenorhabdus strains are grouped together as shown in blue, and form the group called I. Each Xenorhabdus species was separated. The divergent strain SaV (16S rDNA type 9) branches far away from all the other Xenorhabdus but remains closer to Xenorhabdus as expected according to its phenotypic traits. The Photorhabdus strains shown in red, clustered together and formed the group called II. Group II can be divided in two distinct subgroups Ha and IIb. Proteus vulgaris is the nearest neighbour of these pathogenic nematode symbionts. These results were compared with the 16S rDNA sequences published by Rainey et al (1995). From these sequences, we deduced the restriction patterns and we built a dendrogram by using the same methods, this tree was in agreement with those generated by the complete sequences of the 16S rDNA (Bruneletal, 1997). Our results were compared with DNA/DNA hybridisation data (Akhurst et al, 1996). Among the Xenorhabdus species, the separation between species is clear and in full agreement, as within the species hybridisation values are more than 60%, and less than 35% between the species. In contrast, separation among Photorhabdus is not so easy according the DNA/DNA data: we have defined two subgroups Ha and IIb, and it is difficult to corroborate this separation by the DNA hybridisation studies because most values of DNA/DNA are around 40% to 60%: in the case o Photorhabdus the 16S rDNA typing appears surprisingly more precise than the DNA/DNA technique. To validate the status of species or subspecies among the genus of Photorhabdus, measures of ATm completing the DNA/DNA hybridisation values, should be performed. In conclusion, we obtained a good agreement with the previous DNA/DNA hybridisation and 16S rDNA sequence data. Therefore the restriction of PCR-amplified 16S rDNA is an accurate technique and symbionts can be identified at the level of species or subspecies. Furthermore, this method remains easy and rapid to apply: not including the bacterial culture, results could be read within two days. The restriction patterns are simple to interpret as they are composed of less than 10 fragments. With such a method, we planned to work with a larger number of strains (Fisher-Le Saux et al, 1998). It is important to work with a good sampling which represents a high level of diversity. Therefore, we will be able to develop a good picture of the diversity and the structure of these bacteria in relation with their host nematodes and their geographic origins. After such molecular typing, strains to be studied, can be selected for performing further detailed taxonomie studies such as DNA/DNA hybridisation and complete sequencing of the 16S rDNA to clarify taxonomie and phylogenetic studies.

Acknowledgements
We thank Dr. Laumond (INRA, Antibes, France) and Dr. Itamar Glazer (Volcani Center, Israel) for providing symbiont strains.

61

References
Akhurst, R. J., R. G. Mourant, L. Baud, and N. E. Boemare. 1996. Phenotypic and D NA relatedness study between nematode-symbiotic and clinical strains of the genus Photorhabdus (Enterobacteriaceae). Int. J. Syst. Bacteriol. 46 :1034-1041. Boemare, N. E., and R. J. Akhurst. 1988. Biochemical and physiological characterization of colony form variants m Xenorhabdus spp. (Enterobacteriaceae). J. Gen. Microbiol. 134:751761. Brunei, B., A. Givaudan, A. Lanois, R J . Akhurst, N.E. Boemare. 1997. Fast and Accurate identification of Xenorhabdus and Photorhabdus species by Restriction Analysis of PCRAmplified 16S rDNA. Appi Environ. Microbiol. 63:574-580. Fisher-Le Saux, M., H. Maulon, P. Constant, B. Brunei, N.E. Boemare. 1998. PCRribotyping of Xenorhabdus and Photorhabdus isolates in the Caribbean region in relation with the taxonomy and geographic distribution of their nematode hosts. Appi Environ. Microbiol. 64 (11) : in press. Heyndrickx, M., L. Vauterin, P. Vandamme, K. Kersters, P. De Vos. 1996. Applicability of combined amplified ribosomal D NA restriction analysis (ARD RA) patterns patterns in phylogeny and taxonomy. J. Method. Microbiol. 26:247-259. Rainey, F. ., R.-U. Ehlers, and E. Stackebrandt. 1995. Inability of the polyphasic approach to systematics to determine the relatedness of the genera Xenorhabdus and Photorhabdus. Int. J. Syst. Bacteriol. 45:379-381. Wiesburg, G. W., S. M. Barns, D. A. Pelletier, and D. J. Lane. 1991. 16S ribosomal DNA amplification for phylogenetic study. J. Bacteriol. 173:697-703.

63

APPLICATION OF PCR-RFLP OF 16S rDNA TO STUDY THE SYMBIONTS OF TROPICAL EPNs

by Marion Fischer-Le Saux1, H. Maulon2, . Brunei3 and N. Boemare1
Institut National de la Recherche Agronomique, Laboratoire de Pathologie compare, CP. 101, Universit Montpellier II, F-34095 Montpellier Cedex 5 ', Institut National de la Recherche Agronomique, Unit de Recherches en Productions Vgtales, BP 515, Domaine Duelos, F-97165 Pointe--Pitre Cedex , and Ecole Nationale Suprieure Agronomique de Montpellier, Institut National de la Recherche Agronomique, Laboratoire de Recherche sur les Symbiotes des Racines and Unit de Recherches et de Formation des Sciences du Sol, Place P. Viala, F-34 060 Montpellier Cedex 1 3, France Summary To leam more about the diversity in the Caribbean basin of Xenorhabdus and Photorhabdus in relation with their nematode host and their environment, an extensive survey was conducted in this area. A total of 77 Caribbean isolates, recovered from 72 Heterorhabditis and 5 Steinernema , were studied. In order to type faster these numerous bacterial isolates, a method previously assessed with reference strains (see Brunei et al, in this book), Restriction Fragment Length Polymorphism analysis of PCR-amplified 16S rRNA genes, was used. The ribosomal genotypic patterns were compared to those of 40 reference strains. Most of the EPN isolates had been previously identified and ecological data about the samples origin were available. The five Xenorhabdus isolates exhibited a high degree of genetic polymorphism and only one was genotypically identical to a reference strain. In contrast, very little variability was evidenced within the 72 Photorhabdus isolates, as only four closely related genotypes, over the 12 already identified, were detected. When we compared these results with geographical and ecological data, no correlation could be drawn, excepted those directly associated to the environmental habitat of host nematodes. In fact, a closely relatedness between bacterial genotypes and nematode species was found. This result was particularly obvious within the Photorhabdus-Heterorhabditis couple where two bacterial genotypes were exclusively associated to H. indica and two others to H. bacteriophora. Within the Xenorhabdus-Steinernema association, new symbiont genotypes corresponded to new nematode species. The availability on one hand of a fast molecular identification of bacterial symbionts, and on the other hand the characterization with satellite D NA probes of most EPNs, allowed to compare their respective taxonomie structure on a large scale. Thereby, a congruence was evidenced which strongly supports an early co-evolution of the symbiotic partners of both couples.

64

Introduction Few data to relate Xenorhabdus and Photorhabdus diversity to ecological criteria are available. This is mainly due to previous difficulties to rapidly type numerous Xenorhabdus and Photorhabdus strains. The molecular method based on Restriction Fragment Length Polymorphism (RFLP) analysis of PCR-amplified 16S rRNA genes (rDNA), developed by Brunei et al (1997) provides now a fast and accurate tool to study the biodiversity of these genera on a large scale. An extensive survey, to isolate entomopathogenic nematodes (EPNs), was performed among 14 islands in the Caribbean basin, including the Guadeloupe islands where an exhaustive soil sampling was conducted. The purpose of this work was to evaluate the EPNs symbionts diversity and to relate it to host, geographical, and environmental (when available) origins. In order to determine if the Caribbean bacterial isolates represents new 16S rDNAs genotypes, their patterns were compared to those of 40 reference strains o Xenorhabdus and Photorhabdus worldly distributed (EPNs symbionts and clinical isolates). Materials and methods Soil sampling and isolation of EPNs In Trinidad and Tobago, Saint Vincent, Martinique, Puerto Rico, Dominican republic, Jamaica, and Cuba EPNs were randomly collected from various ecosystems (croplands, orchards, grass-lands, salt-marshes, and forests) by using the Galleria trap method or by capturing parasitized insects. In contrast, in the Guadeloupe islands, soil sampling was conducted according to a square grid following a precise protocol previously described (Constant et al, 1998). Then the Galleria trap technique was used to isolate EPNs. EPNs were identified by using morphological criteria, isozyme analysis, and satellite DNA probes (Grenier et al, 1996 ; Laumond et al, 1998). In the Guadeloupe survey, presence of nematodes was related to site location, elevation, rainfall, soil type, and vegetation (Constant et al, 1998). Bacterial isolates and reference strains Bacterial symbionts were isolated from infective juveniles by the hanging drop technique (Poinar & Thomas, 1966). To identify the 77 isolates, their phenotypic properties and their RFLP patterns were compared to those of 40 reference strains from various localities worldwide and from diverse nematode host species (also including clinical Photorhabdus strains). Phenotypic characterization and 16S rDNA PCR-RFLP analysis To verify in a preliminary step that the isolates belonged to Xenorhabdus and Photorhabdus genera, we used the conventional phenotypic tests previously described (Boemare & Akhurst, 1988) including cellular morphology and motility, dye adsorption, pigmentation, antimicrobial activity, bioluminescence, catalase, phospholipase (on yolk agar),

65

and lipase (on Tweens 20, 40, 60, 80, and 85) activities, and several other tests on API20E and API 20 strips (BioMrieux). DNA extraction was performed using the Nucleic Acid Extraction Kit Isoquick (Orca Research). 16S rDNAs were amplified and digested using the primers and the reaction conditions previously described (Brunei et al, 1997). As there were few Xenorhabdus isolates that corresponded to new nematode species, the six restriction enzymes which proved to be polymorphic, were used to type them {Cfo I, Hae III, Msp I, Alu I, Dde I, and Hinfl). In contrast, as the Photorhabdus isolates were numerous, only three endonucleases, which proved to be sufficient to generate all the Photorhabdus genotypes previously studied, were used : Cfo I, Alu I, and Hae III. In control, we tested the six enzymes on a subset of eight Photorhabdus selected among the different genotypes obtained, and no more polymorphism was evidenced. PCR-RFLP data analysis was conducted as previously described (Brunei et al, 1997), but the neighbor joining clustering algorithm was used in this study. Results and Discussion In the Guadeloupe islands, from the 538 soil samples, 31 Heterorhabditis were recovered. No Steinernema was found in this area; Twenty seven nematodes were identified as H. indica, three as H. bacteriophora, and one was still not identified. EPNs were mainly found in the coastal areas at middle and low altitudes (see map in Fischer Le Saux et al, 1998). In the other Caribbean islands 41 Photorhabdus were isolated from Heterorhabditis nematodes. Thirty five of them were identified as H. indica, 5 as H. bacteriophora, and one was still not identified (Fig. 1; Fig. 2; Fig. 3). Five Xenorhabdus were isolated from five different Steinernema, CUOI from S. cubanum (from Cuba), JM26 from 5'. bicornutum (from Jamaica), PR06-A from S. puertoricense (from Puerto Rico), and VC01 and FRM16 from two new species not yet described (from Saint Vincent and Martinique, respectively) (see Fig. 1 ; Fig. 2). Thus, the genus Steinernema is few represented in the Caribbean basin, whereas Heterorhabditis are more abundant; only two species of this genus are recovered : H. indica which is the predominant species, and H. bacteriophora. All the symbiotic bacteria isolated belonged to Xenorhabdus and Photorhabdus genera as they shared the common phenotypic properties of the corresponding reference strains. Few variability was found among the 16S rDNAs of the 72 Caribbean Photorhabdus which were divided into 4 genotypes (numbered 12, 13, 27, 28 in Fig. 4). Genotypes 12 and 27 were represented by reference strains Is5 and Dl, two symbionts of//, indica. Genotype 13 matched the reference strain HP88 from H. bacteriophora, and genotype 28 was not observed in reference strains. In contrast, a high degree of polymorphism was evidenced among the five Xenorhabdus isolates which divided into five different genotypes. Four of them were not represented by any reference strain. The Xenorhabdus isolate from S. cubanum shared the same genotype as the type strain of X. poinarii, symbiont of S. glaseri. Then, the two Steinernema species S. glaseri and S. cubanum may be associated to the same Xenorhabdus species. This result is corroborated by the close relatedness of these two EPNs species, which share morphological and ITS based similarities (Hominick et al, 1997).

66

DommcAX RXPtTBUC ruxKTo RICO

C3?4
. OUADIXOtTPI * rRH03|12| * 0527]. 05|27|: KARTDOQUr^ MjummonzS^ TvwcMrr*vcoi(20|

r _ - O A? h . . . C . V>

*V.

FIG. 1. General map of the investigated Caribbean Islands. Sites where entomopathogenic bacterium nematode complexes were isolated are indicated by symbols identifying the nematode species : ~k H. indica ; # H. bacteriophora ; % Heterorhabditis sp. ; S. bicornutum ; S. cubanum ; O S. puertoricense ; # Steinernema sp. The symbiont strain and genotype number are shown beside each symbol. For the extensive surveys see detailed maps (Fig. 2, Fig. 3 in this study, and Fig. 1 in Fischer Le Saux et al, 1998).

67 N
W
Ml0-A[19] O * * 112) PR17I12)

AfuabllU

liajafaz

FIG. 2. Sites where entomopathogenic bacterium nematode complexes were isolated in Puerto Rico. For symbols, strain designation, and genotype number see legend for Fig. 1. FIG. 3. Sites where entomopathogenic bacterium nematode complexes were isolated in the Dominican Republic. For symbols, strain designation, and genotype number see legend for Fig. 1.

6S

As new restriction patterns were identified, the number of restriction enzymes to type all Xenorhabdus and Photorhabdus strains has to be reconsidered. Five endonucleases are required to generate all the Xenorhabdus genotypes : Cfo I, Dde I, Hae III, H mil, and Msp I. To differentiate all the Photorhabdus genotypes Alu I, Cfo I, and Hae III are to be used. The new genotypes were added in the clustering analysis without substantially modify the phylogenetic tree (Fig. 2, Brunei et al in this book). Most of the phylogenetic positions obtained were corroborated by 16S ribosomal sequencing studies (Liu et al, 1997 ; Suzuki et al, 1996 ; Szlls et al, 1997). In the Xenorhabdus group, the new Caribbean genotypes were not closely related to any reference strains, except the isolate JM26 from Jamaica that clustered at the limit of the genus with the symbiont of S. riobrave. A higher degree of genetic diversity was evidenced in the Xenorhabdus genus when compared to previous descriptions (Brunei et al, 1997 ; Liu et al, 1997 ; Rainey et al, 1995 ; Suzuki et al, 1996 ; Szlls et al, 1997). This diversity is congruent with the wide diversity of the associated nematodes which originate from various localities, and from 16 different Steinernema species. For instance, S. scapterisci, S. serratum, and S. arenarium, whose symbionts (UY61, CN01, and SaV, respectively) were not previously typed, proved to harbor divergent Xenorhabdus symbionts, distantly related to described species (Fig. 4). Most of those genotypes were so divergent that they may represent different species. However more identical strains are needed to define new Xenorhabdus species. The four genotypes of the Caribbean Photorhabdus isolates clustered together in the group Il-a (in red and orange in Fig. 4). Genotypes 12 and 27 were closely related (0,2% divergence); genotype 13 and 28 were less related (0,6%). In fact, two groups of Photorhabdus (Il-a in red and orange, and fl-b in blue, Fig. 4) seem to be correlated to ecological data. The group Il-a encompasses symbionts of H. indica and H. bacteriophora, which are found in tropical regions, whereas group Il-b includes symbionts of//, megidis, H. heliothidis, and H. zealandica, that are restricted to temperate regions. When we compared the 16S rDNA genotypes of the Caribbean isolates to the nematode host species, it appeared that the genotypes 12 and 27 were exclusively associated to H indica, and the genotypes 13 and 28 to H. bacteriophora. Thus, the close host specificity which was proved between Xenorhabdus and Steinernema also exist in PhotorhabdusHeterorhabditis associations, despite only one species o Photorhabdus is defined at present time. No geographical and ecological data could be related to the distribution of the bacterial genotypes except those directly associated to the habitat preference of the host nematodes (Fischer Le Saux et al, 1998). The rapid DNA based molecular techniques used to characterize both symbiotic partners allow to study large samples, and thus, are a good way to investigate the diversity of these entomopathogenic complexes isolated during extensive and exhaustive surveys. Thereby, a close correspondence between the taxonomie structures of the bacteria and the EPNs has been evidenced which supports an early co-evolution between both partners of the symbiosis.

69

Figure 4
22 KROl i 2 A24
6 228

18 JP02 X. japon i eu s ^"

netmt0

Phus

7 Si 8 F3 -19 PR06-A - 2 0 VC01

. bovien

9 Sav 24 CNOl 23 UY61 - 4 77 . ponari ' 3 G6'r .S < 1^8' . beddinsii 26 U STX62 25 JM26 -21 FRM16 10 HbT P. luminescens 27 JM12 Photorhabdus from H. indica 12 Is5 - I l C8406 28 FRG26 Photorhabdus from H. bacteriophora 13 HP88 30 Q614 29 1216-79 Photorhabdus from clinical strains 14 XlNach .15 Meg Photorhabdus from H. megidis, 16 Cl H. heliothidis, and H. zealandica 17 NZH3 31 Proteus vulgaris 0.0041

rj

I\
FIG. 4. Cluster analysis (Neighbor joining method) of the 33 Xenorhabdus and Photorhabdus 16S rDNA genotypes obtained by PCR-RFLP. The genotype number (in bold character) and the name of the representative strain are indicated beside each branch. Within the Photorhabdus genus, the color code indicates the two groups defined in Brunei et al. (1997) (Il-a in red and orange, and Il-b in blue). Within group Il-a the different shades show different host species origins. The clinical strains form a separate branch (in green). Similarly the already described species o Xenorhabdus (group I) are indicated in different colors.

71

Acknowledgments The technical assistance of E liane Bonifassi and Anne Lanois is acknowledged. Authors are very grateful to Drs E va Arteaga, E nrique Cabanillas, gueda Carro, Wilfredo Figueroa, Luis Garrido, Nelson Simes, Grover Smart, and Patricia Stock for providing EPNs. This work was supported by the MENRT grant 95-5-10697. References Boemare, N. E. & Akhurst, R. J. (1988). Biochemical and physiological characterization of colony form variants in Xenorhabdus spp. (Enterobacteriaceae). J. Gen. Microbiol. 134, 751761. Brunei, B., Givaudan, ., L anois, ., Akhurst, R. J. & Boemare, N. (1997). Fast and accurate identification o Xenorhabdus and Photorhabdus species by restriction analysis of PCR-amplified 16S rRNA genes. Appi. Environ. Microbiol. 63, 574-580. Constant, P., Marchay, L ., Fischer-L e Saux, M., Panoma, S. & Maulon, . (1998). Natural occurrence of entomopathogenic nematodes (Rhabditida: Steinemematidae and Heterorhabditidae) in Guadeloupe islands. Fund. appi. Nematol 21, 6 : in press. Fischer L e Saux, M., Maulon, ., Constant, P., Brunei, B. & Boemare, . E. (1998). PCR-ribotyping of Xenorhabdus and Photorhabdus isolates in the Caribbean region in relation with the taxonomy and geographic distribution of their nematode hosts. Appi Environ. Microbiol. 64, 11 : in press. Grenier, E., Bonifassi, E., Abad, P. & Laumond, C. (1996). Use of species-specific satellite DNAs as diagnostic probes in the identification of Steinemematidae and Heterorhabditidae entomopathogenic nematodes. Parasitology 113, 483-489. Hominick, W. M., Briscoe, B. R., Garcia del Pino, F., Heng, J., Hunt, D. J., Kozodoy, E., Mracek, Z., Nguyen, ., Reid, A. P., Spiridonov, S., Stock, P., Sturhan, D., Waturu, C. & Yoshida, M. (1997). Biosystematics of entomopathogenic nematodes : current status, protocols and definitions. Journal of Helminthology 71, 271-298. Laumond, C , Maulon, ., Figueroa, W., Arteaga, E. & Garrido, L . (1998). Identification of Heterorhabditis and Steinernema from Caribbean islands. Nematologica in press. Liu, J., Berry, R., Poinar, G. & Moldenke, A. (1997). Phylogeny of Photorhabdus and Xenorhabdus species and strains as determined by comparison of partial 16S rRNA gene sequences. Int. J. Syst. Bacteriol 47, 948-951. Poinar, G. O., Jr. & Thomas, G. M. (1966). Significance of Achromobacter nematophilus Poinar and Thomas (Achromobacteraceae: E ubacteriales) in the development of the nematode DD-136 {Neoaplectana sp., Steinemematidae). Parasitology 56, 385-390.

72

Rainey, F. ., Ehlers, R-U. & Stackebrandt, E. (1995). Inability of the polyphasic approach to systematics to determine the relatedness of the genera Xenorhabdus and Photorhabdus. Int. J. Syst. Bacteriol. 45, 379-381. Suzuki, T ., Yabusaki, H. & Nishimura, Y. (1996). Phylogenetic relationships of entomopathogenic nematophilic bacteria: Xenorhabdus spp. and Photorhabdus sp. J .Basic Microbiol 36, 351-354. Szlls, E., Koch, C , Fodor, ., Burghardt, J., Buss, O., Szentirmai, ., Nealson, . H. & Stackebrandt, E. (1997). Phylogenetic evidence for the taxonomie heterogeneity of Photorhabdus luminescens. Int. J. Syst. Bacteriol. 47, 402-407.

73

PHYLOGENETIC ANALYSIS OF SYMBIOSIS BETWEEN ENTOMOPATHOGENIC NEMATODES AND THEIR BACTERIA

by Jie Liu, Ralph Berry, and George Poinar Department of Entomology, Oregon State University, Corvallis, OR 97331, USA

Abstract
Mutualistic symbioses between entomopathogenic nematodes and their bacteria are fundamental to reproductive biology, ecology, and systematics of the nematodes. Phylogenetic analysis can be used to describe these symbiotic associations by comparing both cladistic relationships and branch lengths. To describe the association between the nematodes and their bacteria, we developed a method to detect and determine symbiotic bacteria from the nematode Heterorhabditis marelatus and Steinernema oregonense. In our studies, only symbiotic bacteria in the genera Photorhabdus and Xenorhabdus were detected and determined from infective juveniles, males, and females of the nematodes. It is suggested that only the symbiotic bacteria in the genera Photorhabdus and Xenorhabdus are naturally associated with the nematodes. Shifts among nematode-bacteria associations occurred but may direct toward bacterial species available in the environment. We sequenced and compared partial ribosomal DNA in the nematodes and their bacteria. Phylogenetic analyses with maximum likelihood and parsimony methods provided evidence for cospeciation between the nematodes and their bacterial symbionts. A congruent and parallel branching pattern suggests an ancient association between the nematodes and their bacterial symbionts.

Introduction
Entomopathogenic nematodes (Heterorhabditidae and Steinemematidae) are widely available for use in biological control. These nematodes are characterized by their mutualistic relationship with symbiotic bacteria. The bacterial symbionts appear to be carried monoxenically throughout the whole intestine of the infective juveniles in Heterorhabditidae or in a special vesicle in infective stage juveniles in Steinemematidae. The nematodes provide protection and transportation for their bacterial symbionts. The bacterial symbionts contribute to the mutualistic relationship by killing insect hosts, establishing and maintaining suitable conditions for nematode reproduction, providing nutrients and antimicrobial substances that inhibit the growth of a wide range of microorganisms. Interaction between entomopathogenic nematodes and their bacteria may be one of the principal forces driving diversification and specialization in these organisms. Phylogenetic analysis allows us to infer the history of the interaction. Combining the topology of phylogenetic relationships for the nematodes and their bacteria provides a test on if the association was inherited from ancestral taxa (Mitter and Brooks, 1983). Recent advances in molecular systematics facilitate rigorous analysis of the history of the nematode and their bacteria association by providing mutually independent phylogenies for the

74

nematodes and the bacteria. Molecular data are also attractive because they provide a common yardstick for comparing phylogenies in taxonomically disparate groups, such as the nematodes and the bacteria. Homologous genes can be readily found in most taxa, whereas homologous morphological features can not be found in the nematodes and their bacteria. Furthermore, because molecular methods yield a large number of potentially informative characters, molecular phylogenetic studies can play a key role to understand the process of co-speciation between the nematodes and their bacteria. Symbiotically associated the entomopathogenic nematodes and their bacteria offer an abundance of tractable system for applying this approach. The nematode and bacterium symbioses are mutualisms, in which the bacteria benefit from protection sites and the nematode benefit from food supplied by the bacterium. However, the nature of the nematode-bacterium symbiosis has not been precisely defined. The bacteria in the genera Photorhabdus and Xenorhabdus are generally considered the main symbionts of the nematodes and the species primarily responsible for the death of the insect hosts and the growth of the nematodes within the insect cadaver. However, a number of other bacterial species have been isolated from the nematodes (Lysenko & Weiser 1974; Aguillera et al. 1993; Jackson et al. 1995) and the nematodes developed and reproduced on a number of other bacterial species (Aguillera et al. 1993, Jackson et al. 1995). These findings seem to be laboratory facts because the other bacterial species have not been linked to field-collected nematodes (Akhurst 1995, Boemare 1995). Association between the nematodes and their symbiotic bacteria are of fundamental importance for nematode infectivity, mass reproduction, and registration as biological control agents. This system provides opportunities to examining many questions about the co-speciation. However, little work has been done so far on phylogenetic analysis of these mutualisms. Initial phylogenetic patterns in most association suggest that the diversification have been accompanied more often by the co-speciation than host shifts (Liu et al., 1997). Our objectives in this study are to detect and determine the bacteria associations from the nematodes and to detect possible shift among the associations from natural populations as well as designed lab tests. We also compare the phylogenies of the nematodes and their bacteria to obtain evidences of co-speciation. To assess their symbiotic history, we map phylogenies of the nematodes and their bacteria. Materials and Methods Nematode isolates The isolates were obtained or collected from different geographic locations worldwide. Nematode species were confirmed by microscopic examination based on original description. Pure cultures of each isolate were maintained in larvae of the greater wax moth, Galleria mellone!la (L.). DNA extraction Total genomic DNA was extracted from single first generation females of each isolate and from other stages of the nematodes to detect the symbiotic bacteria. The female was obtained from the larvae of the greater wax moth, Galleria mellonella (L.) about 4 days after infection and washed 3 times with sterile distilled water. The female was then pulverized in 40 ! of 5% Chelex 100 resin (Bio-Rad, Richmond, CA) with a motor-driven pestle in an Eppendorf tube. The tube was incubated at 60C overnight. The tube contents were mixed vigorously,

75

centrifuged at 10,000 rpm for 5 min, and the supernatant was collected and used as DNA samples. PCR and sequencing Targeted gene sequences (18S rRNA gene and mtDNA ND4 genes for the nematodes and 16S rRNA gene for the bacteria) were amplified by the polymerase chain reaction (PCR) in a 25 reaction mixture containing 1 of each primer, 100 of each dNTP, 4 of template DNA, 1.5 of bovine serum albumin (BSA), and 1.4 U Taq DNA polymerase (Saiki et al. 1988) in PCR buffer [10 mM Tris-HCl (pH 9.2), 1.5 mM MgCl 2 , 75mM KCl]. The reaction mixtures were heated to 94C for 3 min, followed by 35 cycles of 94C for 1 min, 45C for 1 min, and 72C for 2 min, and a final extension at 72C for 8 min. Twenty-five microliters of PCR products were loaded onto a 1.2% (w/v) agarose gel containing 0.5 /ml of ethidium bromide and electrophoresed at 60 V for 1.5 hr. 100 bp DNA ladder (Life Technologies, Gaithersburg, MD) was molecular size standards. Amplified DNA bands were excised from the gel and purified with Agarose Spin Columns (SUPELCO, Bellefonte, PA). Purified DNA was sequenced directly by using the Taq DyeDeoxy Terminator Cycle sequencing kit (Applied Biosystems) according to the manufacturer's protocol. Alignments and phylogenetic analysis Clustal W (Thompson et al., 1994) was used to obtain multiple alignments based on sequence similarity. The published ND4 gene sequences of Caenorhabditis elegans and Ascaris suum (Okimoto et al., 1992) were also aligned with the region sequenced from Heterorhabditis nematodes. Tamura-Nei distances among the nematode isolates were computed by using MEGA (Kumar et al., 1993). The aligned sequence data were analyzed with different outgroup by two different methods: maximum likelihood by DNAML (in PHYLIP package Ver. 3.572) (Felsenstein, 1995) and maximum parsimony by PAUP Ver. 3.1.1. For maximum likelihood analyses, global search was used. The transition/transversionratio option was set to its default value of 2.0. For maximum parsimony analyses, all data were entered unordered. The most parsimonious trees were constructed by using branch and bound searches. Support for nodes of the trees were evaluated by bootstrap (1000 replications) analysis with branch and bound search. When several of the most parsimonious trees were obtained, a 75% consensus tree was constructed and treated as the most parsimonious tree for constructing figures. To analyze if the nematode-bacteria relationship is due to the association by descent, we used COMPONENT (Page, 1993a) to map the bacteria phylogeny to the nematode phylogeny by invoking possible shifted associates. Shifts among the nematode-bacteria Ten isolates of H. marelatus from Florence and Newport of Oregon, USA were sequenced for the ND4 gene for the nematodes and the 16S rRNA gene for the bacteria. In designed lad experiments, 10 infective juveniles of H. marelatus were mixed with 10 infective juveniles of H. megidis to infect a wax worm. All females obtained from the wax worm were also sequenced.

76

Results
Bacteria associations of the nematodes The partial 16S rRNA gene sequences were amplified from the bacteria purified from the hemocoel of the wax worms within 48 hours of infection by the nematodes and from different stages of the nematodes H. marelatus and S. oregonense. They all produced a single band of about 630 bp. The alignments of partial 16S rRNA gene sequences comprise approximately 616 bp from different stages H. marelatus and its bacterium or 609 bp from different stages of S. oregonense. The DNA sequences were almost identical from different stages of the nematodes. We found that more reliable sequencing data can be obtained from first and second-generation females or males. The symbiotic bacterium of the nematode H. marelatus had the minimum Tamura-Nei distance of 0.04 to other symbiotic bacterial species associated with the nematode Heterorhabditis species. The symbiotic bacterium of the nematode 5. oregonense was closely related to the symbiotic bacterium of the nematode S. feltiae. The Tamura-Nei distance was 0.02. The symbiotic bacteria appeared to be the only species associated with different stages of the nematodes H. marelatus and S. oregonense. No other bacteria species were detected from the nematodes. Shifts among the associations In total of 20 natural nematode and bacteria populations, we found only one shift occurred in a Newport nematode population, in which the nematode was associated with the symbiotic bacterium of Florence nematode population. In designed lab experiments, two nematode species were found from one wax worm but only one symbiotic bacteria species can be established in one wax worm. Phylogenies of the nematodes and their bacteria Molecular phylogenies indicated that Heterorhabditis and Steinernema belong to different monophylies. Their symbiotic bacteria also belong to different monophylies. Different nematode species appeared to be associated with different symbiotic bacteria species. If the nematode species are closed related, their symbiotic bacteria are also closed related. S.feltiae and S. oregonense are sister species. Their symbiotic bacteria also are sister species. In Heterorhabditis species, different populations of H. marelatus formed a monophyly. Their symbiotic bacteria also formed a monophyly. Maps between the phylogenies of the nematodes and their bacteria In Steinernema, 10 taxa have to be added to reconcile 9 nematode species and their bacteria phylogeny, which indicated possible host switching or extinction (Fig. 1). Statistics indicated that the bacteria associated with the nematode S. intermedia might have switched host. When its bacteria species removed from our analysis, only three taxa needed to be added to obtain a reconciled phylogeny (Fig. 2). Similar results were also obtained from the reconciled phylogeny o Heterorhabditis. Three taxa have to be added to reconcile 6 nematode species (Fig. 3). We found the symbiotic bacteria of the nematode H. indicus may have switched its host. When this bacterium removed from our analysis, no taxa was needed to reconcile the phylogeny (Fig. 4).

77

Discussion
Mutualistic symbioses between distantly related organisms have generated major innovations in the evolution of organic complexity (Chapela et al, 1994). The reconstruction of symbiotic history requires explicit phylogenies for both symbionts, a necessity that has limited the number of such studies to date. Here, we reported the results of a phylogenetic analysis of the entomopathogenic nematode-bacteria symbioses, an ancient and successful association that has lead the nematodes as excellent biological control agents. In our study, the symbiotic bacteria appeared to be the only species associated with different stages of the nematodes H. marelatus and S. oregonense. No other bacteria species were detected from the nematodes. The symbiotic bacteria associated with the nematode H. marelatus and S. oregonense are possible new species because of their unique DNA sequences from other described bacterial species. Shifts among these associations can occur but its frequency can be very low partially because the bacteria have not been found from natural habitats other than from the nematodes as well as the insect hosts that killed by the nematodes.

S.intlB

S.fellB

Reconciled tree statistics Duplication = 2 Total leaves = 19 Leaves added = 10 Losses = 7

S.orelB S.fel2B

S.riolB S.glalB S.analB

S.car2B S.carIB

Fig. 1 Reconciled Tree of Steinernema and Their Bacteria

7S

S.intIN S.riolN S.glaIN S.riolB S.glalB

S.anaIN S.analB

Reconciled tree statistics: Duplication = 1 Total leaves = 11 Leaves added = 3 Losses = 3

S.fellN S.fel2N

S.feHB

S.oreIN S.orelB S.feMN S.tel2N S.oreIN S.car2N S.car2B S.carIN S.carIB S.fel2B

Fig. 2 Reconciled Tree of Steinernema and Their Bacteria after Possible Shifted Bacteria Removed
H.mar4N H.mar4B H.mar5N H.mar5B H.indIN

Reconciled tree statistics: Duplication = 1 Total leaves = 13 Leaves added = 3 Losses = 3

H.meg1NH.meg1B H.mar4N H.mar5N H.indIN H.indlB H.megIN H.bac5N H.bac5B H.badN H.badB

Fig. 3 Reconciled Tree of Heterorhabditis

and Their Bacteria

79

H.mar4N H.mar4B

H.mar5N H.mar5B

Reconciled tree statistics: Duplication = 0 Total leaves = 5 Leaves added = 0 Losses = 0

H.bac5N H.bac5B

^ H.megIN H.megIB

H.badN H.badB

Fig. 4 Reconciled Tree of Heterorhabditis and Their Bacteria after Possible Shifted Bacteria Removed

Comparison of the topology of the symbiotic bacteria phylogeny with the phylogeny of the nematodes revealed a congruent branching pattern, which suggests a long history of symbioses. However, incongruence between the phylogenies also found. The shifts among the association have been detected, which is one of the causes for the incongruence. The incongruence can also arise from a number of other causes, such as the presence of multiple lineages of parasites coupled with the bacteria extinction, or failure of the symbiotic bacteria to colonize the nematodes, or collection failure (Page, 1993b). The co-speciation between the nematodes and their bacteria is an excellent example of how symbionts, often unseen or overlooked microorganisms, can underlie the preservation of more obvious and much studies macrobiota. In this example, it is unlikely that the entomopathogenic nematodes would survive if their bacteria partners were exterminated by some environmental factors.

so

Acknowledgment
J. Liu is grateful to Dr. N. Boemare for his invitation and valuable discussions during a COST 819 meeting. We would like to thank B. J. Adams, M. E. Barbercheck, S. P. Stock, D. Strong, L. Wennemann, and P. S. Wilhelm for providing some nematode cultures. This study was supported in part by an USDA grant as well as the Research Council and the Department of Entomology, Oregon State University. Sequence data were analyzed with a high-speed computer system donated by an Intel Higher Education Grant.

References
Aguillera, M. M. 1993. Development, reproduction, and pathogenicity of Steinernema scapterisci in monoxenic culture with different species of bacteria. J. Inverte. Pathol. 62: 289-294. Akhurst, R. J. 1995. From then to now - A brief review of entomopathogenic nematodes and their symbiotic bacteria. Abstracts of Second International Symposium on Entomopathogenic Nematodes and Their Symbiotic Bacteria. Pp. 3-8. Boemare, N. E. 1995. Controversial interpretations on the symbiotic relationships in Steinernema and Heterorhabditis nematode bacterium complexes. Abstracts of Second International Symposium on Entomopathogenic Nematodes and Their Symbiotic Bacteria. Pp 14-22. Brunei, B. et al. 1997. Fast and accurate Identification o Xenorhabdus and Photorhabdus species by restriction analysis of PCR-amplified 16S rRNA genes. App. Environ. Microbiol. 63:574-580. Chapela, I. H. et al. 1994. Evolutionary history of the symbiosis between fungus-growing ants and their fungi. Science 266: 1691-1694. Jackson, T. J. et al. 1995. Isolation of insect pathogenic bacteria, Providencia reiteri, from Heterorhabditis spp. J. App. Bacteriol. 78: 237-244. Kumar, S. et al. 1993. MEGA: Molecular evolutionary genetic analysis, version 1.01. The Pennsylvania State University, University Park, PA 16802. Liu, J. et al. 1997. Phylogeny o Photorhabdus and Xenorhabdus species and strains as determined by comparison of partial 16S rRNA gene sequences. Int. J. Syst. Bacteriol. 47:948-951. Lysenko, O. et al. 1974. Bacteria associated with the nematode Neoaplectana carpocapsae and the pathogenicity of this complex for Galleria mellonella larvae. J. Inverte. Pathol. 24: 332-336. Mitter, C. T. et al. 1983. Phylogenetics aspects of coevolution. In Coevolution (D. J. Futuma and M. Siatkin, eds.), pp. 65-98, Sinauer, Sounderland, MA.

81

Page, R. D. M. 1993a. COMPONENT, tree comparison software for Microsoft Windows, Ver 2.0, The Natural History Museum, London. Page, R. D.M. 1993b. Parasites, phylogeny and cospecieation. Int. J. Parasitol. 23: 499-506. Swofford, D. L. 1993. PAUP: phylogenetic analysis using parsimony, version 3.1.1 Illinois Natural History Survey, Champaign. Szallas, E. et al. 1997. Phylogenetic evidence for the taxonomie heterogeneity of Photorhabdus luminescens. Int. J. Syst. Bacteriol. 47: 402-407. Thompson, J. D. et al. 1994. CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specification gap penalties and weight matrix choice. Nucleic Acids Res. 22: 352-360.

IDENTIFICATION OF PHOTORHABDUS LUMINESCENS SUBSPECIES

by Ralf-Udo Ehlers
Institute for Phytopathology, Department for Biotechnology and Biological Control, Christian-Albrechts-Universitt Kiel, Klausdorfer Str. 28-36, 24223 Raisdorf Germany

Sequence variation within the variable region of the 16S rRNA allowed the synthesis of specific PCR primers for the identification of five groups within the species Photorhabdus luminescens, symbionts of entomopathogenic nematodes of the genus Heterorhabditis. For the second primer the highly conserved region was used. The P. luminescens type strain specific primer could not recognize any other P. luminescens isolates. The primer complementary to strain HSH2 (isolated from North West European H. megidis) identified all P. luminescens isolated from H. megidis from North West Europe and two isolated from closely related nematode strains from Ireland. For this group the subspecies temperatus is proposed. The primer based on strain LN2 (isolated from nematode type strain H. indicus) identified P. luminescens of tropical origin isolated from H. indicus. For this group the subspecies tropicus is proposed. Symbionts of //. bacteriophora cannot yet be separated into well described groups. A comparison of sequence data indicates that the non-symbiotic P. luminescens isolates are distinct in the variable region. Details of this paper will be published in Systematic and Applied Microbiology : Ehlers, R.-U., and I. Nielman. 1998. Molecular Identification of Photorhabdus luminescens strains by amplification of specific fragments of the 16S ribosomal DNA. Syst. Appi Microbiol. 21 : in press

85

MOLECULAR PHYLOG ENETICS AND TAXONOMIC HETEROGENEITY OF PHOTORHABDUS LUMINESCENS : OLD AND NEW DATA
by E. Szallas, Andrs Fodor, G. Kovacs, C. Koch, J. Burghart, O. Buss 1, K. Nealson 2, and E. Stackeb randt3
'Department of Genetics, Etvs University, H-1088 Budapest, Muzeum krt. 4'A, Hungary ; Center for Great Lake Studies, University of Wisconsin-Milwaukee, Milwaukee, WI53204 USA ; DSMZ-Deutche Sammlung Von Mikroorganismen und Zellkulturen GmbH, D-38124 Braunschweig, Germany. The partial (469) and/or almost complete (1,436) sequences of 16S rRNA gene of 76 strains of bacterial symbionts isolated from entomopathogenic nematodes Heterorhabditis spp. and those of 14 bacterial symbionts of the entomopathogenic nematodes Steinernema spp. were determined and compared to each other and to the sequences of several reference strains of Enterobacteriaceae. References included type strains o Photorhabdus luminescens, Xenorhabdus nematophilus, X. bovieni, X. beddingi, X. poinarii, X. japonicus. The data confirmed the separate status of the two genera of symbionts of entomopathogenic rhabditid nematodes, first suggested by Boemare. The symbionts of Heterorhabditis spp. clustered with the type strain of P. luminescens, while the symbionts of Steinernema spp. grouped with the Xenorhabdus species. X. japonicus clustered with other Xenorhabdus species. (Szlls et al., 1997). So did our recently isolated symbionts of S. anomalii, S. intermedia, S. scapteriscii, S. glaserii, S. lucskai and several strains of S. feltiae. Phylogenetic analysis of 16 almost complete 16S rDNA sequences of Heterorhabditis symbionts and 40 (++) partial sequences of P. luminescens strains indicated the presence of several subclusters within the species. The results of DNA-DNA hybridisation studies performed with a few selected strains of each of the 16S rDNA sublusters supports the existence of several genospecies within P. luminescens. Our studies involved P. luminescens strains originated from Heterorhabditis strains of (i) North American and North Western E uropean H.megidis; (ii) E uropean, Australian and North American H. bacteriophora; (iii) unknown host in Wisconsin and Australia; (iv) human wounds into several subclusters (Szlls et al, 1997) and were then extended to other H. bacteriophora symbionts from (i) Azores; (ii) Hungary; (iii) Guadeloupe; (iv) Nebraska as well as to symbionts of type strains of other Heterorhabditis species. Phenotypic differences that correlate with the genomic subclusters were not apparent, especially, because of the variability of the strains. Data on geographical correlation are rather astonishing. Subclusier III involves all but one mid-western North American (Wisconsin and Ohio) strains, but this one (WXI3) is totally different in every respects (RFLP and RAPD polymorphism; antibiotics production/resistance profile, morphology, nutritional requirements and growth rate, physiology, biochemist-try, Fodor et al., in preparation). The Australian (DSM3369) and American (DSM3368) type strain belongs to different sublusters ( and I, respectively. There are a rather sharp border between the territories pre-dominated by H. megidis (North) and H. bacteriophora (South) in E urope. Subclusier II includes all but one known

symbionts of North Western European H. megidis, but this one, {PJun), phylo-genetically takes a unique position, at the branching point of the separation of Xenorhabdus and Photorhabdus genera. Isolates from Azores did not form a single subcluster. Neither those from Hungary. H.b. symbiont Az38 proved identical with a Subcluster 11 strain {PEGB) while others {AZ36, AZ37 and AZ39) are identical to V6-1, belonging Subcluster IV, which includes two other Australian {0614, Brecon), one American (Utah) as well as a Hungarian (SZZS1) symbionts of H. bacteriophora, while AZ29, proved practically identical with Al from Hungary and has a sequence very similar to that of P. Mo! (from Moldavia) forming another Subcluster. The non-symbiotic strains of P. luminescens form a highly homogenous subcluster by themselves {Subcluster VI). 1S5 and WX13 are completely different from the rest of the strains and from each other, (temporarily put into Subcluster V). Future investigations, which will extended to the bacterial symbionts of Heterorhabditis species other than bacteriophora and megidis, should include those molecules that may decide whether or not to describe the subclusters as novel (sub)species as revealed by D NA hybridisation. In order to determine whether the sub-clusters that emerged from the analysis of a single gene could be recovered when genomic similarities were determined, D NA/D NA reassociation experiments were performed with selected strains from rD NA Subclusters I to V. These experiments are necessary as (i) D NA-D NA similarity values are needed to delimit (sub)species in the polyphasic approach to classification and (ii) D NA similarities cannot be extrapolated from the 16S rDNA similarities be-cause of the non-linear correlation between these two taxonomie parameters. P. luminescens type strain DSM3368 and its relatives (Subclusier 1) exhibited moderately high levels of reassociation with members of Subcluster IV (levels of similarity, 56 to 74%). Similar values were obtained for members o Subclusters II and /// (levels of similarity, 56 to 76%). 1S5, a member of the temporary Subcluster V, exhibited moderately high levels of reassociation with members Subclusters I, HI and IV (levels of similarity, 59 to 71%). All other intersubcluster values were lower, ranging from 40 to 66% DNA similarity. The DNA similarity values determined for selected strains of P. luminescens and E. coli were less then 20%.

87

IDENTIFICATION AND CHARACTERISATION OF INSECTICIDAL TOXIN GENES AND BACTERIA ASSOCIATED WITH ENTOMOPATHOGENIC NEMATODES
by Morgan, J.A.W.1, P. Jarrett2, D.J.Ellis2, M.A.Ousley1

Department of Plant Pathology and Microbiology, ' Department of Entomological Sciences, Horticulture Research International, Wellesbourne, Warwickshire UK CV35 9EF. alun.morgan@hri.ac.uk. One of the most successfully exploited microorganisms used to control insect pest species is the bacterium Bacillus thuringiensis. The delta endotoxins produced during sporulation have variable activity against mainly Dipteran, Lepidopteran and Coleopteran insects. With this large sample of toxins it has been disappointing that a greater diversity of genes capable of controlling many more insects have not been discovered. This may reflect the fact that no toxins significantly different from those already described and capable of controlling new groups of insects, are present in this species. In addition the recent use of B. thuringiensis on such a large scale, combined with the development and use of transgenic plants with delta endotoxin genes, has led to the development of insect populations resistant to some major B. thuringiensis toxins. Therefore, this has led to a new drive to identify novel insecticidal toxins, particularly protein toxins that are capable of controlling insects. In the search for new insecticidal genes a number of microbial proteins have already been identified. These include the vegetative insecticidal protein (VIP protein) from B. cereus and B. thuringiensis, cholesterol oxidases from Streptomyces species, and various proteinase inhibitors and enzymes including a range of chitinases from bacteria, fungi and viruses. In an attempt to find new insecticidal toxins we have identified a number of insect active bacterial strains, originally isolated from insect parasitic nematodes (IPNs), belonging to the two genera Photorhabdus luminescens and Xenorhabdus. X. nematophilus was shown to have activity against some pest species of Lepidoptera and mosquitoes. The insecticidal activity was shown to have a high molecular weight, to be heat sensitive and digested by certain proteases, therefore the toxin was believed to have a proteinaceous structure. To enable the entomocidal activity to be transferred to other micro-organisms and plants for the control of insects, the structure of the protein toxin and an understanding of its expression is needed. A 40 Kb region of DNA from X. nematophilus PMFI296 that encodes insecticidal activity has been cloned and sequenced. A preliminary identification of sequences important for toxicity has been made by transposon mutagenesis. A targeted search programme has also isolated 200 strains of Xenorhabdus sp., which have all been tested against pest species belonging to three orders of insects. Strains have been identified with differing host ranges and activities to all insects tested. Selected strains have also been characterised and shown to be diverse by PCR amplification and RFLP analysis of the 16S rRNA and flagellin genes. Insect activity was also found in P. luminescens strains, and the toxins appear to be more diverse at the sequence level. However, they seem to have much lower potency but may be active towards a broader range of insects. In addition, sequence analysis of our original toxin shows the remnants of bacteriophage elements and insertion sequences. These results may

8S

indicate that this group of toxins are mobile and are transferred between Photorhabdus and Xenorhabdus strains.

89

NUTRITIONAL SIGNIFICANCE OF PHOTORHABDUS BACTERIA FOR THE GROWTH AND REPRODUCTION OF HETERORHABDITIS NEMATODES
by Richou Han and Ralf-Udo Ehlers
Institute for Phytopathology, Department for Biotechnology and Biological Control, Christ ian-Albrechts-University Kiel, Klausdorf er Str. 28-36, D-24223 Raisdorf Germany Introduction Unlike the genus Xenorhabdus, which contains five species {X. nematophilus, X. bovienii, X. poinarii, X. beddingii, and X. japnica) (Akhurst and Boemare, 1988), the genus Photorhabdus only has one species: P. luminescens. Although P. luminescens is considered to represent the taxonomie heterogeneity of strains, based on phenotypic characteristics (Akhurst, 1983; Akhurst and Boemare, 1988; Boemare and Akhurst, 1988; Hurlbert et al, 1989), the isozyme analysis by electrophoresis (Hotchkin and Kaya, 1984), whole cell fatty acid analysis (Janse and Smits, 1990), DNA-DNA hybridization evidence (Grimont et al, 1984; Fanner et al, 1989; Boemare et al, 1993; Akhurst et al, 1996), 16S rRNA gene (Ehlers et al, 1988; Putz et al, 1990; Smits and Ehlers, 1991; Rainey et al, 1995; Brunei et al, 1997; Liu et al, 1997; Szallas et al, 1997), bacterium-nematode recombination results (Han et al, 1990, 1991; Gerritsen and Smits, 1993), the separation into different species is yet unsuccessful. Recently, two subspecies temperatus and tropicus have been proposed under the species P. luminescens, by amplification of specific fragments of the 16S Ribosomal DNA (Ehlers, this meeting). Although classical and modem techniques exist for the taxonomie and phylogenetic analysis o Photorhabdus symbionts, following difficulties still exist: only a few strains with limited geographical and nematode origin have been studied; phase variation markedly influences the results of phenotypic tests (Akhurst et al, 1996); variation in the methods or conditions is used for classical phenotypic characteristics; little information is available on the adaptive relationships between nematodes and bacteria and species description in Heterorhabditis lacks. Food-related adaptation in Heterorhabditis I Photorhabdus complexes have been developed through a long term co-evolution, on two levels (Akhurst and Boemare, 1990): nutrient provision and bacterial retention. Nematodes properly maintain and carry the bacteria in their intestine (undigested) to new food sources; the bacteria provide nutrients for reproduction of new vectors (nematodes). These two steps are necessary for a nematodebacterium combination to make full use of the food source successfully. As cross-breeding (Dix et al, 1992) and molecular methods (Joyce et al, 1994) provide a more accurate definition of Heterorhabditis species, the artificial monoxenic recombination of different Heterorhabditis strains and Photorhabdus isolates is a useful method to elucidate the taxonomie position and phylogenetic relationships o Photorhabdus.

90

Nutritional associations between Heterorhabditis nematodes and different cultures Axenic cultures (without bacteria) Attempts to establish axenic cultures of Heterorhabditis were tried by combination of axenized nematode eggs with different growth factors (Lunau et al, 1993), but none of these methods initiated axenic growth of the tested heterorhabditid nematodes. Heterorhabditis spp. cannot yet be cultured without the metabolic activity or presence of P. luminescens (Akhurst and Boemare, 1990; Ehlers et al, 1990; Han et al, 1990; Lunau et al, 1993; Grerritsen and Smits, 1993). /Vort-Photorhabdus bacteria culture It was demonstrated that the bacteria other than Xenorhabdus spp. can provide nutrients for the in vitro reproduction o Steinernema (Boemare et al, 1983; Ehlers et al, 1990; Aguillera et al, 1993), although nematodes reproduce most effectively when their natural symbiont is dominant. Little comparable data are available for Heterorhabditis spp. Jackson et al (1995) reported that Providencia rettgeri, the non-symbiotic strain isolated from Irish Heterorhabditis strains did not provide suitable nutritional requirements for nematode growth. Food signal test described by Strauch & Ehlers (1998, in press) showed that the Xenorhabdus spp. do not produce acceptable food signals or provide nutritional factors essential for nematodes o Heterorhabditis (Han & Ehlers, 1998, in press). This indicates the strict association between Heterorhabditis spp. and Photorhabdus isolates. Non-host P. luminescens bacteria culture Specificity studies have shown significant differences in the ability of various isolates of P. luminescens to support in vitro monoxenic cultures of non-host Heterorhabditis spp. and in the ability of the nematodes to retain the bacterial cells in the intestines of infective juveniles (Akhurst et al, 1990; Han et al, 1990, 1991; Gerritsen et al, 1993, 1997), although the food signals produced by non-symbiotic bacterial isolates are not species-specific for the recovery of non-host nematodes (Han &. Ehlers, 1998, in press). Several axenic Heterorhabditis spp. strains (Table 1) were combined with each other's original P. luminescens symbiont. Most strains, to a degree, developed successfully on four or five of the P. luminescens isolates to which they had been exposed. None of the strains developed on all isolates tested. Four groups were clearly delineated. Group I contains the symbiont of H. megidis HNA. It seems to be the most universally acceptable medium for nematode development. All strains tested were able to reproduce on it. Group II contains the symbiont of//, bacteriophora H06 and supports all nematode strains except H. megidis HNA. Group III consists of symbionts that support each other' nematode strain and includes the bacteria of Heterorhabditis sp. G12, H3 and probably LN2 and Group IV contains the symbionts from Heterorhabditis zealandica NZH, Heterorhabditis sp. Q6 and Heterorhabditis sp. V16 that support all nematode strains except those of Group III (Table 2) (modified from Han et al, 1990, 1991). Gerritsen and Smits (1993) reported that a Dutch H. megidis HF strain does not use the symbiont of H. bacteriophora Brecon, isolated from Australia. Akhurst et al (1996) showed that the Dl strain, isolated from a //. indica population from Australia, did not support any other Heterorhabditis nematodes tested,

91

except its host population. These data provide a strong character for identification of subspecies in P. luminescens, from H. megidis and H. indica. It should be pointed out that HNA and NZH are two isolates from two described nematode species {H. megidis and H. zealandica), and H06 (=C8406) isolated from the nematode strain, which was identified as H. bacteriophora, based on an isozyme analysis and morphological characters (Akhurst, 1987). Brunei et al (1997) presented results from PCRbased RFLP analysis of 16S rRNA, showing that the symbionts from H06 and //. bacteriophora Hm are clustered in the same group. According to the biology, preliminary morphological characters and feeding experiments, G12 is similar with the type strain of//. indica LN2. If the taxonomie position of G12 is certain, the four groups o Photorhabdus isolates exactly reflect four described nematode species. There are still contradicting results on some undescribed nematode strains. The results showed that Heterorhabditis sp. VI6 and Q6 were delineated into the same Group with H. zealandica NZH, but Joyce et al. (1994) suggested that strain VI6 belongs to the same subgroup of the H. bacteriophora species complex like the strain H. bacteriophora CI (=NC1). It is worth noting that G12, a non-pigment-producing isolate, and H3, a nonluminous isolate are both compatible with the same nematode strains and equally retained by dauer juveniles. This indicates that luminescence and pigment production are not the necessary traits required for successful symbiosis with Heterorhabditis species. The nonluminous isolates of Q614 from Akhurst and Boemare (1986) and the Chinese H3 were not compared, but differences in the compatibility with H. bacteriophora Hb and H. zealandica NZH are found. H. bacteriophora Hb cannot grow on the bacteria of Q614 (Akhurst et al., 1996), which parallels the result obtained with H. bacteriophora H06 and Heterorhabditis sp. H3 (Han et al, 1990), However, H. zealandica NZH can use Q614 bacteria but H3 can not. Q614 and H3 are probably not the same bacterial species. Bacterial transmission between Heterorhabditis nematodes and different Photorhabdus isolates It was demonstrated that the transport of Xenorhabdus and Photorhabdus bacteria in mature dauer juveniles is determined at a higher level of specificity (Akhurst, 1983; Dunphy et al, 1985; Han el al, 1990; Grerritsen and Smits, 1993). From the data of Han et al (1990), it can be seen that for each nematode population, the percentage of dauer juveniles containing bacteria varied with the Photorhabdus group. The dauer juveniles are much better able to carry bacterial isolates from the group to which their own symbiont belongs than those from other bacterial groups. Although the mechanisms determining the transmission of symbiotic bacteria to dauer juveniles are not well established, bacterial retention rates in different Heterorhabditis!Photorhabdus recombinations may reflect the co-evolution between the two associates. Low bacterial retention percentage may be indicative of less highly evolved nematode/bacterium relationships.

92

Table 1. Sources of nematodes and bacteria (Han et al., 1990, modified) Nematode host species H. megidis H. bacteriophora Heterorhabditis sp. Heterorhabditis sp. H. indica Heterorhabditis sp. Heterorhabditis sp. H. zealandica Origin United States China China China India China Australia New Zealand Reference(s) Poinar?/ al (1987) Akhurst (1987)

Bacterium HNA H06 (=C8406) G12 H3 LN2 Q6 V16 NZH

Poinar et al (1992) Akhurst (1987) Wouts ( 1979); Akhurst ( 1987)

a. An isozyme analysis of the nematode strains H. bacteriophora Hb and C8406 revealed that they are identical for 10 discriminating enzymes (Akhurst 1987). b. Morphology, biology and partial feeding experiments showed that G12 is similar to H. indica. They were isolated from tropical areas. Table 2. In vitro production of Heterorhabditis spp. and different Photorhabdus strains in monoxenic lipid agar cultures (Han et al., 1990) Nematode Bacterium HNA H06 + + + + + + + +

H. megidis HNA Heterorhabditis + sp. H06 + sp. G12 + sp.H3 H. indica LN2 + //. zealandica + NZH Heterorhabditis + sp. Q6 sp. V16 +

G12 + + n -

H3 + + n -

LN2 n n n + n n

HNZ + + n + + +

Q6 + + + + +

V16 + + n + + +

None -

+, positive; -, negative; n, not tested. Discussion Use of the compatibility test for the identification of Photorhabdus bacteria Until now, there is no report that one Heterorhabditis strain can maintain more than one specific Photorhabdus bacterial isolate. The incompatibility of some Heterorhabditis and Photorhabdus strains indicates that the nematodes only feed on the bacteria they like. This strict association between nematodes and bacteria, which significantly reflects the specific

93

physiological properties and nutritional requirements of both associates, provides the background for use of compatibility test as an indicator character for Photorhabdus grouping. Accurate definition of nematode species The specificity data apparently indicate a close correspondence between the taxonomie grouping of the bacteria and the nematode species of their partners. Although the bacterial grouping by compatibility tests is independent of the definition of nematode species, correct interpretation of the relationships between bacterial groups and the host nematodes requires an accurate taxonomie identification of nematode species. Nevertheless, at present, even with the cross-breeding and molecular methods, results are still variable. For example, the results of cross-breeding and satellite DNA analysis indicate that Heterorhabditis sp. strain HP88 is not a H. bacteriophora, H. indica, or H. megidis strain (Dix et al, 1992; Grenier et al., 1996), but PCR results suggested that this strain belongs to the same subgroup of the //. bacteriophora species complex as strains NCI and VI6 (Joyce et al, 1994). To establish consistent relationships between Photorhabdus bacteria and their Heterorhabditis hosts, reliable characterization for accurate definition of nematode species is still needed. Conclusion For the symbiosis of Heterorhabditis I Photorhabdus complexes, nothing is more important than their food-related adaptation. Food signals, essential and acceptable nutrients produced by bacteria, and bacterial transmission by dauer juveniles, are involved in this adaptation. The highly-specific nutritional relation between Heterorhabditis spp. /Photorhabdus isolates provides support for subdivision o P. luminescens. Acknowledgments The Fellowship to Richou Han, Guangdong Entomological Institute, Guangzhou, China, by Alexander von Humboldt Foundation, Germany, is gratefully appreciated. References Aguillera, M. M. & Smart, G. C. JR. (1993). Development, reproduction, and pathogenicity of Steinernema scapterisci in monoxenic culture with different species of bacteria. J. Invertebr. Pathol, 62, 289-294. Akhurst, R. J. (1983). Neoaplectana species: specificity of association with bacteria of the genus Xenorhabdus. Exp. Parasito!. 55, 258-263. Akhurst, R. J. (1987). Use of starch gel electrophoresis in the taxonomy of the genus Heterorhabditis (Nematoda: Heterorhabditidae). Nemaiologica 33, 1-9. Akhurst, R. J. and Boemare, N. E. (1986). A non-luminescent strain of Xenorhabdus luminescens (Enterobacteriaceae). J. Gen. Microbiol. 132, 1917-1922. Akhurst, R. J. and Boemare, N. E. (1988). A numerical taxonomie study of the genus Xenorhabdus (Enterobacteriaceae) and proposed elevation of the subspecies of X. nematophilus to species. J. Gen. Microbiol 134, 1835-1845.

94

Akhurst, R. J. and Boemare, N. E . (1990). Biology and taxonomy of Xenorhabdus. In {(Entomopathogenic Nematodes in Biological Control, (E ds. R. Gaugler and H. K. Kaya). pp.75-90. CRC press, Boca Raton, FL. Akhurst, R. J., Mourant, R. G., Baud, L. and Boemare, N. E . (1996). Phenotypic and DNA relatedness between nematode symbionts and clinical strains of the genus Photorhabdus (Enterobacteriaceae). Int. J. Syst. Bacteriol. 46, 1034-1041. Boemare, N. E. (1983). Recherches sur les complexes nemato-bacteriens entomopathogenes: etude bactriologique, gnotoxenique et physiopathologique du mode d'action parasitaire de Steinernema carpocapsae Weiser (Rhabditida: Steinemematidae). these doctorat d'etat, Universite de Montpellier II. Boemare, N. E. and Akhurst, R. J. (1988). Biochemical and physiological characterization of colony form variants in Xenorhabdus spp. (E nterobacteriaceae). J. Gen. Microbiol. 134, 751-761. Boemare, N. E ., Akhurst, R. J. and Mourant, R. G. (1993). DNA relatedness between Xenorhabdus spp. (E nterobacteriaceae), symbiotic bacteria of entomopathogenic nematodes, and a proposal to transfer Xenorhabdus luminescens to a new genus, Photorhabdus gen. nov. Int. J. Syst. Bacteriol. 43, 249-255. Brunei, B., Givaudan, ., Lanois, ., Akhurst, R. J. and Boemare, N. (1997). Fast and accurate identification o Xenorhabdus and Photorhabdus species by restriction analysis of PCR-amplified 16S rRNA genes. Appi. Environ. Microbiol. 63, 574-580. Dix, H., Bunnell, A. M., Griffin, C. T., Joyce, S. ., Nugent, M. J. and Downes, M. J. (1992). The identification of biological species in the genus Heterorhabditis (Nematoda: Heterorhabditidae) by crossbreeding second-generation amphimictic adults. Parasitology 104,509-518. Dunphy, G. B., Rutherford, T. A. and Webster, J. M. (1985). Growth and virulence of Steinernema glaseri influenced by different subspecies o Xenorhabdus nematophilus. J. Nematol 17,476-482. Ehlers, R-U., Stoessel, S. and Wyss, U. (1990). The influence of phasevariants of Xenorhabdus spp. and Escherichia coli (E nterobacteriaceae) on the propagation of entomopathogenic nematodes of the genera Steinernema and Heterorhabditis. Rev. Nematol. 13,417-424. Ehlers, R.-U., Wyss, U. and Stackebrandt, E . (1988). 16S rRNA cataloguing and the phylogenetic position of the genus Xenorhabdus. Syst. Appi. Microbiol. 10, 121-125. Farmer J. J., Jorgensen, J.H., Grimont, A. D., Akhurst, R. J., Poinar, G. O. JR., Ageron, E ., Pierce, G. V., Smith, J. ., Carter, G. P., Wilson, K. L. and Hickman-Bremner, F. W. (1989). Xenorhabdus luminescens (DNA hybridization group 5) from human clinical specimen. J. Clin. Microbiol. 27, 1594-1600. Gerritsen, L. J. M. and Smits, P. H. (1993). Variation in pathogenicity of recombinations of Heterorhabditis and Xenorhabdus luminescens strains. Fundam. Appi Nenalo!. 16, 367373.

95

Gerritsen, L. J. M. and Smits, P. H. (1997). The influence o Photorhabdus luminescens strains and form variants on the reproduction and bacterial retention of Heterorhabditis megidis. Fundam, app. Nematol. 20, 317-322. Grenier, E., Bonifassi, E., Abad, P. and Laumond, C. (1996). Use of species-specific satellite DNAs as diagnostic probes in the identification of Steinemematidae and Heterorhabditidae entomopathogenic nematodes. Parasitology 113, 483-489. Grimont, P. A. D ., Steigerwalt, A. G., Boemare, N. E, Hickman-Brenner, F. W., D eval, C , Grimont, F. and Brenner, D . J. (1984). D eoxyribonucleic acid relatedness and phenotypic study of the genus Xenorhabdus. Int. J. Syst. Bacteriol. 34, 378-388. Han, R. C. and Ehlers, R.-U. (1998). Cutivation of axenic Heterorhabditis spp. dauer juveniles and their response to non-specific Photorhabdus luminescens food signals. Nematologica. (in press) Han, R. C , Wouts, W. M. and Li, L.Y. (1990). D evelopment o Heterorhabditis spp. strains as characteristic of possible Xenorhabdus luminescens subspecies. Rev. Nematol. 13, 411415. Han, R. C , Wouts, W. M. and Li, L. Y. (1991). D evelopment and virulence of Heterorhabditis spp. strains associated with different Xenorhabdus luminescens isolates. J.Invertebr. Pathol. 58,27-32. Hotchkin, P. G. and Kaya, H. K. (1984). Electrophoresis of soluble proteins from two species of Xenorhabdus, bacteria mutualistically associated with the nematodes Steinernema spp. and Heterorhabditis spp. J. Gen. Microbiol. 130, 2725-2731. Hurlbert, R. E., Xu, J. and Small, C. L. (1989). Colonica! and cellular polymorphism in Xenorhabdus luminescens. Appi Environ. Microbiol. 55, 1136-1143. Jackson, T. Wang, H , Nugent, M., Griffin, C , Bumell, A. and D owds, B. (1995). Isolation of insect pathogenic bacteria, Providencia rettgeri, from Heterorhabditis spp. J . Appi Bacteriol 78, 237-244. Janes, J. D . and Smits, P.H. (1990). Whole cell fatty acid patterns o Xenorhabdus species. Letters Appi Mierob io!. 10,131-135. Joyce, S. ., Bumell, . M. and Powers, . O. (1994). Characterization o Heterorhabditis isolates by PCR amplification of segments of mtD NA and rRNA genes. J . Nematol. 26, 260-270. Liu, J., Berry, R., Poinar, G., and Moldenke, A. (1997). Phylogeny o Photorhabdus and Xenorhabdus species and strains as determined by comparison of partial 16S rRNA gene sequences. Int. J. Syst. Bacteriol. 47, 948-951. Lunau, S., Stoessel, S., Schmidt-Peisker, A. J. and Ehlers, R. -U. (1993). Establishment of monoxenic inocula for scaling up in vitro cultures of the entomopathogenic nematodes Steinernema spp. and Heterorhabditis spp. Nematologica 39, 385-399.

90

Poinar, G. O. Jr., Karunakar, G. K. & D avid, H. (1992). Heterorhabditis indicus n. sp. (Rhabditida, Nematoda) from india: separation of Heterorhabditis spp. by infective juveniles. Funddam. appi Nematol. 15, 467-472. Putz, J. Meinert, F., Wyss, U., Ehlers, R.-U. & Stackebrandt, E. (1990). D evelopment and application of oligonucleotide probes for molecular identification o Xenorhabdus species. Appi Environ. Microbiol. 56, 181-186. Rainey, F. ., Ehlers, R.-U. and Stackebrandt, E. (1995). Inability of the polyphastic approach to systematics to determine the relatedness of the genera Xenorhabdus and Photorhabdus. Int. J. Syst. Bacteriol. 45, 379-381. Smits, P. H. & Ehlers, R.-U. (1991). Identification o Heterorhabditis spp. by morphometric characters and RFLP and of their symbiotic bacteria Xenorhabdus spp. by species-specific DNA probes. IOBOWPRS Bull. 14, 195-201. Strauch, O. & Ehlers, R.-U. (1998). Food signal production o Photorhabdus luminescens inducing the recovery of entomopathogenic nematode Heterorhabditis spp. in liquid culture. Appi Microbiol. Biotechnol. (in press) Szallas, E., Koch, C , Fodor, ., Burghardt, J. Buss, O., Szentirmai, ., Nealson, . H. and Stackebrandt, E. (1997). Phylogenetic evidence for the taxonomie heterogeneity of Photorhabdus luminescens. Int. J. Syst. Bacteriol. 47, 402-407.

97

HETERORHABDITIS SPP. AND PHOTORHABDUS SPP. : COEVOLUTION AND GNOTOBIOLOGY

by O. Oravecz, E. Szlls, H. Panjav, Andrs Fodor1, B. Adams and E. Stackebrandt3
2

Department of Genetics, Etvs University, H-l 088 Budapest, Muzeum krt. 4A, Hungary, Department ofPlant Pathology, University of Nebraska, Lincoln, NE 68583-0722. USA, ' DSMZ-Deutche Sammlung Von Mikroorganismen und Zellkulturen GmbH, D-38124 Braunschweig, Germany

The understanding of the mechanisms of the stringently strain-specific (regulated developmentally by both partners) symbiosis between heterorhabditid nematodes and their bioluminescent bacteria {Photorhabdus spp). is one of the most challenging research field in the EPN-related science. As for the bacterial symbionts, phylogenetic data, based upon molecular studies, have recently been published (Boemare at.al., 1997, Szlls et al, 1997, Liu et al., at this meeting, 1998 Ehlers, at this meeting, 1998). Our polyphasic approach to classification included 16S rRNA gene sequence and DNA/DNA reassociation data could delimit subclusters but not species within the one-species genus Photorhabdus (see abstracts Szlls et al), because of the non-linear correlation between these two taxonomie parameters. On the other hand, the symbiotic hosts are separated into different species. The evolutionary history of a group of bacteria starts when this group is isolated from the rest of its previous group-mates. We try to determine how the separation the nematode species influenced the divergence and evolution of the symbionts. Nine Heterorhabditis species have been described so far: H. hepialus, H. marelatus; H. megidis; Irish K122; H. zaelandica; H. argentiensis; H. bacteriophora; H. indicus, H. hawaiensis. H herpialus and H. marelatus are likely identical. (Stock (1997) has proposed synonymy for marelatus + hepialius, the new name will be H. marelata). So H. indicus and H hawaiensis are. It is also likely (and has been convincingly shown that) 722 is a separate species, it has yet to be described as such. Surprisingly HIT, a Hungarian isolate, does belong to this species (C. Griffin, I. Dix and A. Bumell, personal communication). The phylogenetic relations of these species have been determined (Adams, 1997). The data of analyses of the ITS sequences o Heterorhabditis strains by a new technique (Phast System Electrophoresis) presented here supports his findings. There are taxonomically unidentified isolates, such as "//. guadeloupe", WX2107 (Fodor, unpublished. 1S5 was isolated as H. bacteriophora, but evidences have been accumulated, that IS5 maybe H. indicus. Data presented here show which nematodes are capable of growing on and constructing of symbiotic complexes with symbionts of other nematodes. This is called gnotobiological analysis of the symbiosis (Boemare, 1997). In our studies axenic Jl nematodes are put on different Photorhabdus strains in agar media.

9S

It was determined, whether the nematodes (i) grow and develop or die; (ii) grow to fertile adults; (iii) grow to fertile adults; (iv) produce hyamine-resistant (dauerlarva) progeny, (v)whether the dauers were ineffective and pathogenic to Galleria mellone!la and /or Mamestra brass icae caterpillars and (vi) were capable of releasing bacteria into the insect haemocoel; and if so, (vii) the reisolated bacteria were and identical to that we used as new symbiont (byribosomalDNA analyses). While majority of the Steinernema species could be grown on A', bovienii, the Heterorhabditis strains showed a very restricted host range, even within species. Data of symbiotic host range and those concerning molecular phylogeny have been discussed. We do not know whether the "gnotobiological analysis" can either confirm or refute hypotheses of coevolution. Hypotheses of coevolution are supported by congruent histories. Our data try are discussedfromthis aspect.

99

TOWARDS THE PHYSICAL MAP OF THE GENOPHOR (CHROMOSOME) OF PHOTORHABDUS LUMINESCENS

by Zsuzsanna Kiss1, Cs. P. Ortutay1, A. Fodor1, M. Rott2, and K.H. Nealson3
Department of Genetics, Etvs University (ELTE) in Budapest, Muzeum krt. 4 A, Hungary, H-1088 ; University of Wisconsin in La Cross, USA ; ' University of Wisconsin in Milwaukee USA. Photorhabdus luminescens (Enterobacteriaceae) is a Gram negative, bioluminescent bacterium, which is symbiotically associated with entomopathogenic nematodes {Heterorhabditis spp., Heterorhabditidae). There has not been any genetic or physical map for any of the bacterial symbiont of entomopathogenic nematodes established so far. Little is known about the genomic organization of P. luminescens. We adopted a low resolution physical mapping technique, based upon Southern hybridization analysis of the complementarity of large genomic fragments obtained by different restriction endonucleases and separated by pulsed field electrophoresis. Total DNA of P. luminescens (strain Hm, which is, in fact identical with the DSM reference strain no. 3368) was digested by restriction endonucleases of Sfll, Fesl, 1-Ceul, respectively, which resulted in a limited number of well separable fragments, which were resolved by countour-clamped homogenous electric field electrophoresis. Finally, a circular map of the genophore were constructed. The estimated genome size was determined as 3,700+/-50 Kb by adding the fragment lengths in each of the Pulse Field Electorphoresis-separated digests produced by different restriction endonucleases. Several cloned genes are now located on the physical map. Each was were used as a probe in Southern hybridization analysis of of digested DNA fragments separated by oulsed field gel electrophoresis. We are going to use the P. luminescens genomic library which had been constructed by one of us (Zs. Kiss) in MPI Cologne, Germany, during her STM (supported by COST 819) to put as many cloned markers onto the the map as possible.

101

THE CONCEPT OF MONOXENIC ASSOCIATIONS IN EPNs: LIMITS OF THEIR NATURAL OCCURRENCE AND OF GNOTOBIOLOGICAL INTERPRETATIONS
by Nol Boemare, E. Bonifassi, M. Fischer Le Saux 1, and G. Smart Jr 2
1

INRA Montpellier-Antibes, France 2 University of Florida, Gainesville, USA

The occurrence of specific bacterial species established in the infective juveniles of the entomopathogenic nematodes, is worldly accepted. Many researchers, who undertook such isolations, confirmed that all Steinernema spp. carried in their gut Xenorhabdus spp. (Akhurst, 1983; Akhurst and Boemare, 1988; Akhurst and Boemare, 1990; Boemare et al, 1993; Forst et al, 1997; Thomas and Poinar, 1979) and all Heterorhabditis spp. carried Photorhabdus luminescens, formerly Xenorhabdus luminescens (Boemare et al, 1993). The first approach to characterize the symbionts of the entomopathogenic nematodes, which is an analytical way, consists in isolations of intestinal contents of freshly harvested infective juveniles from entomopathogenic nematode populations occurring naturally in soil. When results are reproduced several times, i.e. when identical microbial species are isolated at different locations and periods from the same nematode species, the residential intestinal microflora of one species of nematode can be assessed. The complementary way of the previous bacteriological analyses is a synthetic way that uses methods of the gnotobiology. To undertake such studies, the main tool is the use of germ free animals. Such axenic Steinernema were obtained from disinfection of nematode eggs (Boemare, 1983; Boemare et al, 1982). It is worth noting that axenization from the eggs is the only method which can assume that the progeny will not possess any micro-organisms. It is possible to combine in a second step several bacteria, mainly those previously isolated from the intestinal microflora of the host nematodes, in order to identify the most suitable strains for the nematode reproduction and to estimate the entomopathogenic value of the combination. For Heterorhabditis, axenized eggs have to be re-combined immediately with the bacterial strain to be tested (Lunau et al, 1993). Steinernema scapterisci have been transferred from South America and many times sub-cultured in Florida. Aguillera et al (1993) isolated from this nematode several species of bacteria, as Ochrobactrum anthropi, Paracoccus denitrificans, Xanthomonas maltophilia, and a third group that they supposed to belong to Xenorhabdus genus. Grewal et al (1997) showed that the development of non-infective stages occurred with the bacterial symbiont of S. carpocapsae, Steinernema riobrave, and its own symbiotic Xenorhabdus, but that the development of infective juveniles to the 4th stage ( dauer recovery) was significantly delayed and reduced with heteroxenic Xenorhabdus spp. However, the three later monoxenic combinations were virulent to Galleria mellonella larvae, as also found Aguillera and Smart for the associations of A', nematophilus - S. scapterisci (1993). In fact, 5. scapterisci infective juveniles containing A', nematophilus or Xenorhabdus from 5. riobrave were more virulent to G. mellonella than those containing their natural symbiont. On one hand these results evidenced the beneficial impact of the natural symbiont for the nematode development and, on the other hand, an improvement of the bacterial-helminthic complex pathogenicity with certain heteroxenic associations.

102

This report establishes for the first time that occurrence of microorganisms between the two cuticles of L3 juveniles plays an antagonistic role for the multiplication of the S. scapterisci in certain insect species, as Galleria mellonella. When these microorganisms are discarded, its Xenorhabdus symbiont acts similarly as other nematode entomopathogenic complexes. However, axenic nematodes don't kill G mellonella while its symbiont is entomopathogenic. So this is the first example where the pathogenicity of the complex is supported only by the symbiont. In the case of A', poinarii and S. glaseri, each partner by itself is not pathogenic, and only the complex is pathogenic (Akhurst, 1986). In addition, when the L3 of S. scapterisci are well surface-disinfected a production is obtained with the Bedding method. The control with the monoxenic association with Xenorhabdus probes that this is the functional status of the symbiosis establishing that the Bedding method is a convenient method for producing functional monoxenic nematodes because all the contaminants of the tegument are discarded. The occurrence of several candidates previously reported as possible symbionts of S. scapterisci are mostly coming from the contaminants between both cuticles of L3. Sodium hypochoride eliminates these surface contaminants, but in addition is also a very useful tool to induce the ex-sheathing of L3 that evade from the old cuticle. When the surface axenization is followed by a carefully control of the ex-sheathing from L2 cuticle, only one type of micro-organism is isolated. This study demonstrated that the symbiont recovered in these conditions is functional at two levels : in vitro production and pathogenicity of the complex towards G. mellonella. Our results on this respect are in good agreement with those of Grewal et al (1997) who isolated a Xenorhabdus corresponding to the third group of Aguillera et al (1993). The concept of the occurrence of a natural monoxeny in EPNs would be understood if we consider the following features. In natural conditions, if any bacterial co-associate is carried in the L3, may be the association could kill insects, but the probability of a multiplication inside these insects for giving a viable progeny is lower. Consequently, chances for an ecologist to catch in the soil such parasited insects are much more unpredictable due to the poverty of these occurrences and due to the short period allowing to catch them before any putrefaction which would destroy any traces. So in fact in the nature, ecologists harvest always nematodes coming from successful parasitism and have a very low probability to harvest the errors . For this point of view the Galleria baiting method (Bedding and Akhurst, 1975) would be inappropriate if we want to catch all the microflora associated with entomopathogenic nematodes, because insect trap selects by itself the good infective nematodes. Retention of bacteria other than natural symbionts does not occur within the vesicle of Steinernema spp. The natural occurrence of many cuticular contaminants in S. scapterisci evidences an exception which is a natural error detrimental for killing insects, except the mole cricket which is very susceptible to this symbiosis. Steinemematidae and Heterorhabditidae living in the soil, but specialized as insect parasites, belong to a large group of soil saprophytic Rhabditida, including many other nematodes feeding micro-organisms. They have probably kept a part of this ancestral microbivorous behavior. It was demonstrated in vitro, that all the occasional foreign bacteria isolated in these nematodes resist to the antimicrobial end-products of the symbionts explaining their occasional occurrence together with the symbionts.

103

Aguillera, M. M., Hodge, N. C , Stall, R. E., and Smart, G. C , Jr. (1993). Bacterial symbionts o Steinernema scapterisci. J . Invertebr. Pathol. 62, 68-72. Aguillera, M. M., and Smart, G. C , Jr. (1993). Development, reproduction, and pathogenicity o Steinernema scapterisci in monoxenic culture with different species of bacteria. J . Invertebr. Pathol 62, 289-294. Akhurst, R. J. (1983). Taxonomie study o Xenorhabdus, a genus of bacteria symbiotically associated with insect pathogenic nematodes. Int. J . Syst. Bacteriol. 33, 38-45. Akhurst, R. J. (1986). Xenorhabdus nematophilus subsp. poinarii: Its interaction with insect pathogenic nematodes. Syst. Appi Microbiol. 8, 142-147. Akhurst, R. J., and Boemare, N. E. (1988). A numerical taxonomie study of the genus Xenorhabdus (Enterobacteriaceae) and proposed elevation of the subspecies of X. nematophilus to species. J. Gen. Microbiol. 134, 1835-1845. Akhurst, R. J., and Boemare, N. E. (1990). Biology and taxonomy o Xenorhabdus. In "Entomopathogenic nematodes in biological control" (R. Gaugler and H. K. Kaya, Eds.), pp. 75-90. CRC Press, Boca Raton, Fl, USA. Bedding, R. ., and Akhurst, R. J. (1975). A simple technique for the detection of insect parasitic rhabditid nematodes in soil. Nematologica 21, 109-110. Boemare, N. E. (1983). "Recherches sur les complexes nmato-baetriens entomopathognes: Etude bactriologique, gnotobiologique et physiopathologique du mode d'action parasitaire de Steinernema carpocapsae Weiser (Rhabitida: Steinemematidae)." Universit Montpellier II, Montpellier, France. Boemare, N. E., Akhurst, R. J., and Mourant, R. G. (1993). D NA relatedness between Xenorhabdus spp. (Enterobacteriaceae), symbiotic bacteria of entomopathogenic nematodes, and a proposal to transfer Xenorhabdus luminescens to a new genus, Photorhabdus gen. nov. Int. J. Syst. Bacteriol 43, 249-255. Boemare, N. E., Laumond, C , and Luciani, J. (1982). Mise en vidence d'une toxicognse provoque par le nematode entomophage Neoaplectana carpocapsae Weiser chez l'insecte Galleria mellonella L. C. R. Acad. Sel, Paris, Sr. III. 295, 543-546. Forst, S., D owds, B., Boemare, N., and Stackebrandt, E. (1997). Xenorhabdus spp. and Photorhabdus spp.: bugs that kill bugs. Annual Review of Microbiology 51, 41'-72. Grewal, P. S., Matsuura, M., and Converse, V. (1997). Mechanisms of specificity of association between the nematode Steinernema scapterisci and its symbiotic bacterium. Parasitology 114, 483-488. Lunau, S., Stoessel, S., Schmidt-Peisker, A. J., and Ehlers, R.-U. (1993). Establishment of monoxenic inocula for scaling-up in vitro cultures of the entomopathogenic nematodes Steinernema spp. and Heterorhabditis spp. Nematologica 39, 385-399. Thomas, G. M., and Poinar, G. O., Jr. (1979). Xenorhabdus gen. nov., a genus of entomopathogenic nematophilic bacteria of the family Enterobacteriaceae. Int. J. Sysl. Bacteriol. 29, 352-360.

105

OBSERVATIONS ON THE BACTERIAL SYMBIONT ASSOCIATED WITH THE NEMATODE, STEINERNEMA ABBASI (STEINERNEMATIDAE: NEMATODA) Sami Elawad1, Rob Robson2 and Nigel Hague1 department of Agriculture, The University of Reading, Earley Gate, PO Box 236, Reading, RG6 6AT, Berkshire, UK. 2School of Animal and Microbial Sciences, The University of Reading, Whiteknights, PO Box 228, Reading, Berkshire, RG6 6AJ, UK. SUMMARY Galleria mellonella larvae inoculated with the bacterium from S. abbasi turned dark brown to black, a colour not hitherto observed in steinernematid infections. The bacterium isolated from S. abbasi was a pseudomonad with a single polar flagellum. Using partial sequencing of the 16SrRNA gene, the bacterium was identified as Flavimonas oryzihabitans. This bacterium has been isolated from soil in rice paddy in Japan and from clinical specimens. It is not known how this bacterium became symbiotically associated with S. abbasi. INTRODUCTION Entomopathogenic nematodes of the genus Steinernema are soil-inhabiting insect parasites which are potentially useful biological control agents (Kaya and Gaugler, 1993). Akhurst and Boemare (1990) reported that the bacteria symbiotically associated with steinernematids belong to the genus Xenorhabdus but recently some steinernematids have been shown to contain other bacteria (Aguillera et al, 1993; Gouge e al, 1997). Two steinernematid nematodes have been isolated from subtropical semi-arid areas, S. riobrave from pre-pupae and pupae of Helicoverpa zea and Spodoptera frugiperda in Texas, USA (Raulston et al, 1992), and S. abbasi from soil in an alfalfa field in the Sultanate of Oman (Elawad et al., 1997). Elawad (1998) noted that the cadavers of Galleria mellonella larvae infected with infective juveniles (Us) from S. abbasi turned almost black. This observation suggested that the bacterial symbiont in S. abbasi might be new to entomopathogenic nematology. This paper reports on the isolation and identification of the bacterial symbiont in S. abbasi. MATERIALS AND METHODS Steinernema abbasi was isolated from soil in an alfalfa field in the Sultanate of Oman using the standard baiting technique (Bedding and Akhurst, 1975) and S. riobrave was originally obtained from Dr R. Georgis at Biosys, Palo Alto, CA. Both species of nematode were cultured in Galleria mellonella using techniques described byWoodring and Kaya (1988).

lO

Isolation of bacteria Dead infected larvae of G. mellonella were washed in distilled water and surface sterilized in 70% alcohol and left to dry in a laminar flow cabinet. Larvae were opened with sterile scissors and needles, care being taken not to damage the gut epithelium. A small drop from the oozing haemolymph was streaked onto nutrient broth agar [NBTA: 37 g nutrient agar: 25 mg Bromothymol Blue powder: 4 ml of 1 % 2,3,5 Triphenyl-tetrazolium chloride: 1000 ml distilled water]. The plates were sealed and incubated at 28C in the dark for 24 h. Single colonies were selected and sub-cultured onto NBTA. This process was continued until uniform colonies were obtained from each species of nematode. Effect of bacterial symbionts on G. mellonella For each species of nematode the purified bacteria were put into a solution and injected into last instar larvae of G. mellonella at a dosage of 1600 cells per ml in sterile saline buffer: untreated larvae were injected with 1 ml of sterile saline buffer. Larvae were incubated at 30C and mortality was assessed after 24 h. Replication was 40-fold and the colour changes in the G. mellonella were noted and photographed. Electron micrographs of the bacteria A single colony of each bacterium from the pure cultures was put into 50 ml of liquid broth (Sigma) held in a baffled flask stoppered with sterile cotton wool and incubated at 28C in a shaking incubator at 150 r.p.m. Equal quantities of the bacterial solution and a solution of 2% (w/v) aqueous phosphotungostic acid (pH 6.8) was mixed in an Eppendorf tube. One drop of the solution (50-100 ) was placed on the plastic side of the grid on a copper palladium transmission electron microscope (Hitachi H 800). The excess liquid was removed with filter paper and allowed to dry for 5 min and the bacteria examined with the microscope operating at 75 kilovolts. Identification of the bacterial symbiont To isolate genomic DNA of the bacterium, one colony from the plate containing uniform colonies of the bacterium was selected using a sterile tooth pick and placed in 20 of distilled water in an Eppendorf tube, boiled for 5 min and then placed in ice: the heating and rapid freezing disrupts the cell membranes and facilitates the release of the DNA. This solution was then used in the reaction mixture for the Polymerase Chain Reaction (PCR). The 16S r RNA gene was amplified by PCR using universal primer FD1 (Genosys) as a forward primer (at end 5') and 1492 RU (Genosys) as a reverse primer (at end 3') in the following reaction mixture: [2.2 of MgCI2; 5 of Bio-x-act buffer; 5 of dNTP; 1 of FD1 primer; 1 of 1492 RU primer]. The reaction mixture was held in an Eppendorf tube and incubated in a thermocycler adjusted to a programme in which the mixture was pre-heated to 95C for 1 min and this was followed by 20 cycles at 94C for 1 min for denaturation, 55C for 1 min for annealing and 68C for 2 min for extension: a final extension was done at 70C for 8 min. Ten of the PCR product was loaded after addition of 2 of

107

loading buffer on a 8% TAE agarose gel containing 0.5% ethidium bromide and electrophoresed at 80 volts for 45 min with kb ladder (Bethesda Research Laboratories) as size marker. The size of the 16SrRNA amplified was found to be equal to 1.6 kb. The DNA was then excised from the gel and purified using a Qiagen Kit. To clone the DNA n E. coli, it was ligated into pTZ-19 vector (Bernard, 1988). A ligation was performed after digesting the vector as well as the PCR product with ECoR1 and BamH1. The transformation of DNA into E. coli was done by electroporation of the vector and E. coli at 1200 V. Transformant colonies of E. coli were picked off agar, plated and cleaned using the Qiagen Kit to obtain crude plasmid DNA (mini-preps). To check the size insert Plasmids were cut with EC0RI and BamH1, and the product run on agarose gel. The sample with the expected size insert was submitted to be sequenced using ALF Express automated DNA sequencer. Data obtained from the partial sequencing of the DNA was submitted through the internet to the data base of the Gene Bank using the BLAST computer programme. RESULTS Cadavers of G. mellonella injected with pure colonies of the bacterial symbiont from S. abbasi turned very dark brown to black compared to the relatively minor colour change for S. riobrave bacteria injected into G. mellonella (Fig. 1). G. mellonella larvae died within 24 h when injected with the bacterial symbionts from both species of nematode and and both nematode species reproduced well in the cadavers of G. mellonella. Fig 1. The colour changes In cadavers of G. mellonella infected with Steinernema abbasi and S. riobrave.

IOS

The bacterial cells from S. abbasi were typically of pseudomonad form with a single polar flagellum (Fig 2a) and the cell in S. riobrave was Xenorhabdid with numerous peritrichous flagella (Fig 2b).

Fig 2. The electron micrographs of bacterial cells isolated from Steinernema abbasi (a) and S. riobrave (b).

.*'

(a)

(b)

The result of the partial sequencing of the gene submitted through the internet to the BLAST programme to match the Data Base of the Gene Bank + EMBL + DDBJ + POB is shown in Fig 3. The matching gave a high score for the length of the gene (1527) submitted with a 98% positive identity for the species Flavimonas oryzihabitans (Holmes et al., 1987).

109

Fig 3. The partial nucleotide sequence of the 16SrRNA gene isolated from the bacterial symbiont of the nematode S. abbasi.

ACTCACTATAGGGAAaGCTTGCATGCCTGCAGGTCGACTCTAGAGGATCCTACCTTGTTACGACTTCACCCCAGTCATGAA TCACACCGTGGTAACCGTCCTCCCGAAGGTTAGACTACKrrACTTCTGGTGCAACCCACTCCCATCKTGTGACCGGCGGTG TGTACAAGGCCCGGGAACGTATTCACCGCGACATTCTGATTCGCGATTACTAGCGATTCCGACTTCACGCAGTCGAGTTG CAGACTGCGATCCGGACTACGATCGGTTTTGTGGGATTAGCTCCACCTCGCGGCrrGGCAACCCTCTGTACCGACCATTG TAGCACGTGTGTAGCCCAGGCCGTAAGGGCCATGATGACTTGACGTCATCCCCACCTTCCTCCGGTTTGTCACCGGCAGT CTCCTTAGAGTGCCCACCATAACGTGCTGGTAACTAAGGACAAGGGTTGCGCTCGTTACGGGACrrAACCCAACATCT CGACACGAGCTGACGACAGCCATGCAGCACCTGTGTCAGAGTTCCCGAAGGCACCAATCCATCTCTGGAAAGTTCTCTGC ATGTCAAGGCCTGGTAAGGTTCTTCGCGTTGCTTCGAATTAAACCACATGCTCCACCGCTTGTGCGGGCCCCCGTCAATT CCTTTGAGTTTCAGTCTTGCGACCGTACTCCCCAGGCGGAGTGCTTAATGCGTTASCTGgCAGCACTAAGGGGCGGAAACC CCCYAACACTTAGCACTCATCGTTrACGGCGTGGACTAACAGGGTATCYAATCCYGTTTfJCYCCCCAMGTTTTtSBSsSCTC AGTGTCAtTACAGACCGAAAGTYGCCTTCSCCACTGGGGTCCcTCAAATATCTACGCWTTCACGGTACAYTGGATCACTGg CGGYGTTTtACACGTCGTGWCTGGGAAACCCGGSGTACCAAYTwATCGCITGNAGVAATCCCTTTSGCAGBTGGGTAaTAS CGAGAGGCCSCMCCGTTGCCTTCCACAGTtGSCCCTGATGGSRATGGAaWTKARSTAAATTTG

DISCUSSION The bacterial symbiont from S. abbasi turned the cadaver of G. mellonella larvae very dark brown to black, a colour change not previously reported from steinernematids, suggesting that the bacterial symbiont might be a species new to entomopathogenic nematology. The electron micrographs confirmed that the symbiont was a pseudomonad completely unlike the Xenorhabdids normally associated with steinernematids. Using partial sequencing of the 16SrRNA gene the bacterial symbiont was identified as Flavimonas oryzihabitans. A genus closely related to that of Pseudomonas, F. oryzihabitans, described by Holmes et al. (1987), was originally described as Pseudomonas oryzihabitans by Kodarna ef al. (1985). F. oryzihabitans is widely distributed having been isolated in the UK, USA, Switzerland, Germany, Malaysia and Japan (Holmes et al., 1987). Originally isolated from rice paddy (Kodarna et al, 1985), it is also found in clinical specimens (Holmes et al., 1987; Kodarna tal, 1985). This bacterium is probably widespread in soil, which may account for its appearance in soil in the Sultanate of Oman. It is not known how it became associated with S. abbasi. Other species of bacteria, including species of Pseudomonas, have been found associated with steinernematids (Weiser, 1963; Aguillera eia/., 1993; Gouge et al, 1997) but these non-Xenorhabdid bacteria which include Pseudomonas inflorescens (Gouge et al., 1997) are not necessarily satisfactory symbionts for the reproduction

110

of nematodes. The bacterium, F. oryzihabitans found in S. abbasi, is in every way a satisfactory symbiont for infectivity, establishment and reproduction (Elawad et a/., 1996). S. abbasi will reproduce very effectively in several species of Lepldoptera, including Spodoptera Httoralis and Helicoverpa armigera, which were thought to be the likely hosts for S. abbasi in Oman (Elawad, 1998). F. oryzihabitans differs markedly from the Xenorhabdus species normally associated with steinernematids because it is aerobic and has one polar flagellum in contrast to species o Xenorhabdus which are facultatively anaerobic and have numerous peritrichous flagella (Bergey, 1994). F oryzihabitans has been found to remain viable in water at room temperature for several months (Elawad, 1998) in marked contrast to Xenorhabdus species which are unstable in the external environment (Poinar, 1990). In the course of the present investigation the bacteria were cultured in broth solutions and it was observed that all species of bacteria investigated were motile. Applications of these bacteria, in the absence of the nematode vector, are effective against larvae of both Lepidoptera and Coleptera (Elawad, 1998). Although various species of entomopathogenic nematodes are used worldwide for insect control and are exempt from normal registration procedures for biopesticides, bacteria on their own will be more difficult to develop, particularly a species such as F. oryzihabitans which has been found in clinical specimens (Holmes et al, 1987). REFERENCES Aguillera, M.M., Hodge, N.C., Stall, R.E. & Smart, G.C. (1993). Bacterial symbionts of Steinernema scapterisci. J. Invertebr. pathol. 62: 68-72. Akhurst, R.J. & Boemare, H.E. (1990). Biology and taxonomy of Xenorhabdus. In: Entomopathogenic Nematodes in Biological Control (R. Gaugler and H.K. Kaya, Eds), pp. 75-92. CRC Press, Boca Raton, Florida. Bedding, R.A. & Akhurst, R.J. (1975). A simple technique for the detection of insect pathogenic rhabditid nematodes in soil. Nematologica 21 : 109-110. Bergey, D.H. (1994). Bergey's Manual of Determinative Bacteriology (9th Edition) (Eds J.G. Holt eia/.), pp.787. Bernard, P. (1988). A practical guide to molecular cloning (2nd Edition). Wiley Interscience Publications, pp. 201-296. Elawad, S.A. (1998). Studies on the taxonomy and the biology of a newly isolated species of Steinernema (Steinemematidae: Nematoda) from the Tropics and associated bacteria. PhD Thesis, The University of Reading, Reading, UK. Elawad, S.A., Abbas, M.S. & Hague, N.G.M. (1996). The establishment and pathogenicity of a new species of Steinernema from the Sultanate of Oman in Galleria mellonella. Afro-Asian J. Nematol. 6: 40-45.

Ill

Elawad, S. Ahmad, W. & Reid, A.P. (1997). Steinernema abbasi sp. n. (Nematoda: Steinemematidae) from the Sultanate of Oman. Fundamental and Applied Nematology 20: 435-442. Gouge, D.H., Van Berkum, J.R., Lee, L.L. & Henneberry, T.J. (1997). Temporal association of entomopathogenic nematodes. J. Nematol. (Abs) 29: 579. Holmes, B., Steigerwalt, A.G., Weaver, R.E. & Brenner, D.T. (1987). Chryseomonas luteola comb. nov. and Flavimonas oryzihabitans gen. nov., comb, nov., Pseudomonas-Wke species from human clinical specimens and formerly known, respectively, as Groups Ve-1 and Ve-2. International Journal of Systematic Bacteriology 37: 245-250. Kaya, H.K. & Gaugier, R. (1993). Entomopathogenic Nematodes. Annual Review of Entomology 38: 181-206. Kodarna, K. Kimura, N. & Komagata, K. (1985). Two new species of Pseudomonas: P. oryzihabitans isolated from rice paddy and clinical specimens and P. lutela isolated from clinical specimens. International Journal of Systematic Bacteriology 35: 467-474. Poinar, G.O. (1990). Biology and taxonomy of Steinemematidae and Heterorhabditidae. In Entomopathogenic Nematodes in Biological Control (R. Gauglerand H.K. Kaya, Eds), pp. 23-62. CRC Press, Boca Raton, Florida. Weiser, J. (1963). Diseases of insects of medical importance in Europe. Bull. World Health Org. 28: 121-127. White, G.F. (1927). A method for obtaining infective juveniles from cultures. Science 66: 302-303. Woodring, J.L. & Kaya, H.K. (1988). "Steinernematid and Heterorhabditid Nematodes: A Handbook of Biology and Techniques". Southern Cooperative Series Bulletin 331, Arkansas Experimental Station, Fayetteville, AR.

European Commission EUR 18832 COST 819 Entomopathogenic nematodes Taxonomy, phylogeny and gnotobiological studies of entomopathogenic nematode bacterium complexes Edited by Nol Boemare, Paul Richardson and Franoise Coudert Luxembourg: Office for Official Publications of the European Communities 1999 111 pp. 17.6 25 cm ISBN 9282858235

Venta Salg Verkauf Sales Vente Vendita Verkoop Venda Myynti Frsljning
B ELG I QU E/BE LG I Jean De Lannoy Avenue du Roi 202/Koningslaan 202 B1190 Bruxelles/Brussel Tl. (322) 538 43 08 Fax (322) 538 08 41 Email: lean de.lannoy t infoboard.be; URL: http//www.jeandelannoy.be La librairie europenne/ De Europese Boekhandel Rue de la Loi 214 Wetstraat 244 B1040 Bruxelles/Brussel Tl. (322) 295 26 39 Fax (322) 735 08 60 Email: mail libeurop.be URL: http//www. Iibeurap.be Moniteur beige/Belgisch Staatsblad Rue de Louvain 4042/Leuvenseweg 4042 B1000 Bruxelles/Brussel Tl. (322) 552 22 11 Fax (322) 511 01 84 DANMARK J. H. Schultz Information A/S Herstedvang 1012 DK2620 Albertslund Tlf. (45) 43 63 23 00 Fax (45) 43 63 19 69 Email: Schultz schultz.dk URL: http I/Aw, schultz.dk DEUTSCHLAND Bundesanzeiger Verlag GmbH Vertriebsabteilung Amsterdamer Strae 192 050735 Kln Tel. (49221) 97 66 80 Fax (49221) 97 66 82 78 EMail: vertrieb bundesanieiger de URL: http//www.bundesanzeiger.de /GRE EC E G. C. Elettheroudakls SA International Bookstore Paneptstimiou 17 GR10564Athina Tel. (301) 331 41 8071/2/3/4/5 Fax (301) 323 98 21 Email: elebooks@netor.gr ESPAA Boletn Oficial del Estado Tratalgar. 27 E28071 Madrid Tel. (34)915 38 21 11 (Libras), 913 84 17 15(Suscrip.) Fax (34) 915 38 21 21 (Libras), 913 84 17 14(Suscrip.) Email: clientes4icom.boe.es URL: http//www. boe.es Mundi Prensa Libras, SA Castello, 37 E28001 Madrid Tel. (34) 914 36 37 00 Fax (34) 915 75 39 98 Email: libreria@mundiprensa.es URL: http://www.mundiprensa.com FRANCE Journal officiel Service des publications des CE 26, rue Desaix F75727 Paris Cedex 15 Tl. (33) 140 58 77 31 Fax (33) 140 58 77 00 URL: http/Avww.journalofficiel.gouv.fr IRELAND Government Supplies Agency Publications Section 45 Harcourt Road Dublin 2 Tel. (3531) 661 31 11 Fax (3531) 475 27 60 ITALIA Licosa SpA Via Duca di Calabria, 1/1 Casella postale 552 150125 Firenze Tel. (39) 055 64 83 1 Fax (39) 055 64 12 57 Email: licosaeftbcc.it URL: http7/www.ftbcc.it/licosa LUXEMBOURG Messageries du livre SARL 5, rue Raitfeisen L2411 Luxembourg Tl. (352)40 10 20 Fax (352)49 06 61 Email: mail@mdl.lu URL: http://www.mdl.lu NEDERLAND SDU Servicecentrum Uitgevers Christoffel Plantijnstraat 2 Postbus 20014 2500 EA Den Haag Tel. (3170) 378 98 80 Fax (3170) 378 97 83 Email: sdu@sdu.nl URL: http //www. sdu.nl STERREICH Manz'sche Verlags und Universittsbuchhandlung GmbH Kohlmarkt 16 A1014Wien Tel. (431) 53 16 11 00 Fax (431) 53 16 11 67 EMail: bestellen manz.co.at URL: http//www.manz.at/index.htm PORTUGAL Distribuidora de Livros Bertrand Ld.' Grupo Bertrand, SA Rua das Terras dos Vales, 4A Apartado 60037 P2700 Amadora Tel. (3511) 495 90 50 Fax (3511) 496 02 55 Imprensa Nacional-Casa da Moeda, EP Rua Marqus S da Bandeira, 16A P1050 Lisboa Codex Tel. (3511) 353 03 99 Fax (3511) 353 02 94 Email: del.incmmail tolepac pt URL: http//www.incm.pt SUOMI/FINLAND Akateeminen Kirjakauppa/ Akademiska Bokhandeln Keskuskatu 1/Centralgatan 1 PL/PB 128 FIN00101 Helsinki/Helsingfors P/tfn (3589) 121 44 18 E/fax (3589) 121 44 35 Shkposti: akatilausOakateeminen.com URL: http //www. akateeminen.com SVERIGE BTJAB Traktorvgen 11 S221 82 Lund Tfn (4646) 18 00 00 Fax (4616) 30 79 47 Epost: btjeupub@btj.se URL: http://www.btj.se UNITED KINGDOM The Stationery Office Ltd International Sales Agency 51 Nine Elms Lane London SW8 5DR Tel. (44171)873 90 90 Fax (44171) 873 84 63 Email: ipa.enquries@theso.co.uk URL: http/Avww.thestationeryorfce.co.uk ISLAND Bokabud Larusar Blonda) Sklavrdustig. 2 IS101 Reykjavik Tel. (354) 551 56 50 Fax (354) 552 55 60 NORGE Swets Norge AS stenjoveien 18 Boks 6512 Etterstad N0606 Oslo Tel. (4722) 97 45 00 Fax (4722) 97 45 45 SCHWEIZ/SUISSE/SVIZZERA Euro Info Center Schweiz c/o OS EC Stampfenbachstrae 85 PF 492 CH8035 Zrich Tel. (411) 365 53 15 Fax (411) 365 54 11 Email: eicsosec.ch URL: http/Avww.osec.ch/eics BALGARIJA Europress Euromedia Ltd 59. blvdVitosha BG1000 Sofia Tel. (359-2) 980 37 66 Fax (3592) 980 42 30 Email: Milena mbox.cit.bg C E S K A REPUBUKA USIS NISprodejna Havelkova 22 CZ130 00Praha3 Tel. (4202) 24 23 14 86 Fax (4202) 24 23 11 14 Email: nkpospdec.nis.cz URL: http://usiscr.cz CYPRUS Cyprus Chamber of Commerce and Industry PO Box 1455 CY1509 Nicosia Tel. (357-2) 66 95 00 Fax (3572) 66 10 44 Email: demetrapceci.org.cy EESTI Eesti Kaubandus-Tostuskoda (Estonian Chamber of Commerce and Industry) ToomKooli 17 EE0001 Tallinn Tel. (372) 646 02 44 Fax (372) 646 02 45 Email: einfo@koda.ee URL: http//www. koda.ee HRVATSKA Medlatrade Ltd Pavia Hatza 1 HR10000 Zagreb Tel. (3851) 481 94 11 Fax (3851) 481 94 11 MAGYARORSZAG Euro Info Service Europa Haz Margitsziget PO Box 475 H1396 Budapest 62 Tel. (361) 350 80 25 Fax (361) 350 90 32 Email: euroinfomail.matav.hu URL: http//www.euroinfo.hu/index.htm MALTA Miller Distributors Ltd Malta International Airport PO Box 25 Luqa LQA 05 Tel. (356) 66 44 88 Fax(356) 67 67 99 Email: gwirthusa.net POLSKA Ars Polona Krakowskie Przedmiescie 7 Skr. pocztowa 1001 PL00950 Warszawa Tel. (4822) 826 12 01 Fax (4822) 826 62 40 Email: ars_polbevy.hsn.com.p] ROMANIA Euromedia Str. Gral Berthelot Nr 41 RO70749 Bucuresti Tel. (401) 315 44 03 Fax (401) 314 22 86 ROSSIYA CCEC 60letiya Oktyabrya Av. 9 117312 Moscow Tel.(7095)135 52 27 Fax (7095) 135 52 27 SLOVAKIA Centrum VT1 SR Nm.Slobody. 19 SK81223 Bratislava Tel. (4217) 54 41 83 64 Fax (4217) 54 41 83 64 Email: europtbb1 .sllk.stuba.sk URL: http7/www.sltk.stuba.sk SLOVENIJA Gospodarskl Vestnik Dunajska cesta 5 SLO1000 Ljubljana Tel.(386)613 09 16 40 Fax (386) 613 09 16 45 Email: europgvestnik.si URL: http//www. gvestnik.si TRKIVE Du nya In to tel AS 100, Yil Mahallessi 34440 TR80050 Bagcilarlstanbul Tel. (90212) 629 46 69 Fax (90212) 629 46 27 Email: infotel dunyagazete.com.tr AUSTRALIA Hunter Publications PO Box 404 3067 Abbotsford, Victoria Tel. (613) 94 17 53 61 Fax (613) 94 19 71 54 Email: jpdaviesozemail.com.au CANADA Les ditions La Libert inc. 3020, chemin SainteFoy G1X 3V SainteFoy, Quebec Tel. (1418) 658 37 63 Fax (1800) 567 54 49 Email: liberte@mediom.qc.ca Renouf Publishing Co. Ltd 5369 Chemin Canotek Road Unit 1 K U 9J3 Ottawa, Ontario Tel. (1-613) 745 26 65 Fax (1613) 745 76 60 Email: order.dept renoufbooks.com URL: http://www.renoufbooks.com EGYPT The Middle East Observer 41 Sherif Street Cairo Tel. (202) 392 69 19 Fax (202) 393 97 32 Email: matoudameobserver.com.eg URL: http://www.meobserver.com.eg INDIA EBIC India 3rd Floor, Y. B. Chavan Centre Gen. J. Bhosale Marg. 400 021 Mumbai Tel. (9122) 282 60 64 Fax (9122) 285 45 64 Email: ebicgiasbm01 vsnl not in URL: http//www. ebicindia.com ISRAEL ROY International 41. Mishmar Hayarden Street PO Box 13056 61130 Tel Aviv Tel. (9723) 649 94 69 Fax (9723) 648 60 39 Email: royilnetvision.net.il URL: http//www. royint.co.il Subagent for the Palestinian Authority; Index Information Services PO Box 19502 Jerusalem Tel. (972-2) 627 16 34 Fax (9722) 627 12 19 JAPAN PSI-Japan Asahi Sanbancho Plaza #206 71 Sanbancho, Chiyodaku Tokyo 102 Tel. (813) 32 34 69 21 Fax (813) 32 34 69 15 Email: bookspsijapan.co.jp URL: http//www.psijapan.com

EBIC Malaysia Level 7, Wisma Hong Leong 18 Jalan Perak 50450 Kuala Lumpur Tel. (60-3) 262 62 98 Fax (603) 262 61 98 Email: ebicklmol.net.my

Mundi Prensa Mexico, SA de CV Rio Panuco No 141 Colonia Cuauhtemoc MX06500 Mexico, DF Tel. (525) 533 56 58 Fax (525) 514 67 99 Email: 101545.2361 compuserve.com

EBIC Philippines 19th Floor, PS Bank Tower Sen. Gil J. Puyat Ave. cor. Tindalo St. Makati City Metro Manilla Tel. (63-2) 759 66 80 Fax (632) 759 66 90 Email: eccpcom@globe.com.ph URL: httpZ/www.eccp.com

EBIC Sri Lanka Trans Asia Hotel 115 Sir chittampalam A. Gardiner Mawatha Colombo 2 Tel. (941) 074 71 50 78 Fax (941) 44 87 79 Email: ebicslitmin.com

EBIC Thailand 29 Vanissa Building, 8th Floor Soi Chidlom Ploenchit 10330 Bangkok Tel. (66-2) 655 06 27 Fax (662) 655 06 28 Email: ebicbkk@ksc15.th.com URL: http:/www.ebicbkk.org UNITED STATES OF AMERICA Beman Associates 4611 F Assembly Drive Lanham MD20706 Tel. (1800) 274 44 47 (toll tree telephone) Fax (1800) 865 34 50 (toll free tax) Email: querybernan.com URL: http//www. beman.com ANDERE LANDER/OTHER COUNTRIES/ AUTRES PAYS Bitte wenden Sie sich an ein Bro Ihrer Wahl/ Please contact the sales office of your choice/Veuillez vous adresser au bureau de vente de votre choix Office for Officiai Publications of the European Communities 2, rue Mercier L2985 Luxembourg Tel. (352) 29 2942455 Fax (352)29 2942758 Email: nfo.info@opoce.cec.be URL: http//europ.eu.int

NOTICE TO THE READER Information on European Commission pub lications in the areas of research and innovation can b e ob tained from: RTD info The European Commission's free magazine on European research, focusing on project results and policy issues, published every three months in English, French and G erman. For more information and subscriptions, contact: DG XII Communication Unit, rue de la Loi/Wetstraat 200, B1049 Brussels Fax (322) 2958220; email: info@dg12.cec.be

This volume describes the proceedings of the workshop held at Horticulture Research International, Wellesbourne, Warwick, United Kingdom, on 22 and 23 April 1998, by COST Action 819 and which was devoted to the bacterial symbionts of entomopathogenic nema todes (EPNs). It was divided into three sessions. During the first, the taxonomy and phy logeny of symbioses other than those involving EPNs were described in order to consider some valid comparisons in other research fields. In addition, as an introduction to the debate on the taxonomy of the bacteria, current knowledge on the taxonomy of EPNs was updated. The second session, on the taxonomy and phylogeny of EPN symbioses, gave the recent advances in the phylogeny and taxonomy of Xenorhab dus and Photorhab dus. The debate pointed out that at least three major taxons can be recognised by molecular techniques in the Photorhab dus genus with perhaps two others (or some subdivisions in the previous ones). Discussions showed difficulties in using phenotypic data for discrimi nating between these taxons and a next meeting was programmed in order to propose to the scientific community special regulations for classifying Photorhab dus strains. The third session, on gnotobiological studies on EPN symbioses, showed that these experiments are very accurate in some cases in verifying the specificity of each symbiont species for a nematode species, but not reliable in others. Therefore, gnotobiological studies have to be taken into consideration with other methodologies. Otherwise, from a physiological point of view, these techniques are very fruitful in analysing the symbiosis and in seeing the rela tionships between bacteria and nematodes, especially with respect to the nematode devel opment.

* * *
"k

op

EUR

OFFICE FOR OFFICIAL PUBLICATIONS OF THE EUROPEAN COMMUNITIES L2985 Luxembourg

ISBN

535

789282 858233