Characterization of tropical tree rhizobia and description ...horizon.documentation.ird.fr/exl-doc/pleins_textes/pleins_textes_7/b_fdi_51-52/... · purity by repeated streaking and

< Printed in Great Britain > International Journal of Systematic Bacteriology (1 998), 48, 369-382

1 Laboratoire de Microbiologie des Sols, ORSTOM BP 1386, Dakar, Senegal, West Africa

2 Department of Applied Chemistry & Microbiology, University of Helsinki, PO Box 56, Biocentre 1, Viikinkaari 9, FIN-O0014 Helsinki, Finland

3 Laboratorium voor - Microbiologie, Universiteit Gent, K.-L. Ledeganckstraat 35, B- 9000 Ghent, Belgium

4 Microbiology Department, Reading Laboratory, Institute of Food Research, Whiteknights Road, Reading RG6 6BZ, UK

Agricultura de Lavras, Caixa postal 37, CEP 37200- 970, Lavras MG, Brazil

5 Escola Superior de

Characterization of tropical tree rhizobia and description of Mesorhizobium plurifarium sp. nov.

Philippe de/Lajudie,’t3 Anne W i I l e m ~ , ~ ~ ~ Giselle Nick,’ Fatima Moreiraf5 Flore Molouba,’ Bart Hostef3 Urbain Torckf3 Marc Neyra,’

and Monique Gillis3 Matt R ew D. Collinsf4 Kristina LindstrÖm,‘/Bernard Dreyfuslt

Author for correspondence: Monique Gillis. Tel: $32 9 264 5117. Fax: +32 9 264 5092. e-mail: [email protected]

A collection of strains isolated from root nodules of Acacia species in Senegal was analysed previously by electrophoresis of total cell protein, auxanographic tests, rRNA-DNA hybridization, 165 rRNA gene sequencing, DNA base composition and DNA-DNA hybridization [de Lajudie, P., Willems, A., Pot, B. & 7 other authors (1994). In t l Syst Bacterio/ 44,715-7331. Strains from Acacia were shown to belong to two groups, Sinorhizobium terangae, and a so-called gel electrophoretic cluster U, which also included some reference strains from Brazil. Further taxonomic characterization of this group using the same techniques plus repetitive extragenic palindromic-PCR and nodulation tests is presented in this paper. Reference strains from Sudan and a number of new rhizobia isolated from nodules of Acacia senegal, Acacia tortilis subsp. raddiana and Prosopis ju/Íf/ora in Senegal were included. As a result of this polyphasic approach, the creation of a new species, Mesorhizobium p/urifarÍum, is proposed for a genotypically and phenotypically distinct group corresponding to the former cluster U and containing strains isolated from Acacia, Leucaena, Prosopis and Chamaecrista in West Africa (Senegal), East Africa (Sudan) and South America (Brazil). The type strain of Mesorhizobium plurifarium ORS 1032 has been deposited in the LMG collection as LMG 11892.

Keywords: tropical tree rhizobia, Acacia, Mesorhizobiuin plurifarium sp. nov., polyphasic taxonomy

INTRODUCTION

In recent years numerous changes have occurred in the classification of the Rhizobium-Agrobacteriunz group which, together with the genera Ochrobactrum, Phyllo- bacteriunt, Brucella and Bartonella, constitute a separate rRNA cluster in the alpha subclass of the

I\

‘1 ;$

Proteobacteria (for reviews see 25, 48, 49). Although the close relationship between Agrobacterium and Rhizobium has been known for a long time, the complex intertwined relationship of the different species of the group is only gradually being elucidated by polyphasic taxonomic studies and 16s rRNA gene comparisons (6, 13, 14, 36, 39, 46, 47). Within the Agrobacteriunz-Rhizobium rRNA group, several sub- groups can be distinguished: a first sub-group contains Rhizobium leguininosaruin, Rhizobium tropici types A and By Rhizobium etli and Agrobacteriunz biovar 2; a second one consists of Rhizobium galegae, Agro- bacterium biovar 1 , Agrobacteriunz vitis and Agro- bacterium rubi; a third sub-group contains Sino- rhizobium, including Sinorhizobium Ji-edii, Sino- rhizobium terangae, Sinorhizobium saheli, Sinorhizo- bium medicae and Sinorhizobium rneliloti (6, 35); the

?Present address: LSTM ORSTOMKIRAD-Foret, Baillarguet, BP 5035, 34032 Montpellier Cedex 1, France. Abbreviations: ARDRA, amplified rDNA restriction analysis; REP-PCR,

- \ repetitive extragenic palindromic-PCR; UPGMA, unweighted pair-group method with averages. The EMBLaccession numbersfgrthe new sequence data areY14158for M. plurifarium LMG 11892T, Y14161 forM,p/urifariumLMG 10056.Y14159for \

- - ‘i

_ - - _-

P. de Laiudie and others

fourth sub-group is the most divergent branch from the other three sublineages and corresponds to the recently proposed genus Mesorhizobium (16, 20) including Mesorhizobium loti (15), Mesorhizobizim huakuii (4), Mesorhizobium ciceri (30), cluster U (6), Rh izob ìum t iaiishanense (5), Rhizobium mediterraneum (29) and Rhizobium sp. (Cicer) genomic groups 3 and 4 (29). Several emerging rhizobial groups can already be identified as belonging to Mesorhizobium (18, 28, 31,

In a previous polyphasic study of tropical rhizobia (6) we described a new group belonging to the Meso- rhizobium branch, the so-called cluster U, comprising strains isolated from different leguminous species in Africa, Brazil and New Zealand. Our results showed that cluster U constituted a separate protein gel electrophoretic cluster. Two representative strains (ORS 1001 and ORS 1002) had identical 16s rRNA gene sequences, and showed 98 and 99.6% sequence similarity with M. loti and M. huakzrii, respectively. The recently proposed species M. mediterraneum has 98.7% 16s rRNA gene similarity with cluster U representative strain ORS 1001 (29).

Considerable internal heterogeneity demonstrated in cluster U by protein gel electrophoretic and auxanographic results was confirmed by DNA-DNA hybridization which revealed at least two different genospecies (6).

Here we report more taxonomic data on cluster U, including nodulation tests on diverse legume plants and further genotypic characterization by repetitive extragenic palindromic-PCR (REP-PCR), 16s rRNA gene sequencing and DNA-DNA hybridization. We found additional members of cluster U by screening more isolates from Senegal. We also included representatives of the phenotypic clusters previously formed by 60 rhizobial strains from Acacia senega1 and Prosopis chilensis in the Sudan (50) and representative strains of genospecies 3 and 4 reported by Nour et al. (30) within a group of Cicer isolates. As a result of this polyphasic approach, we propose the creation of a new species, Mesorhizobium plurifarium, for this group which is clearly separate from the other Mesorhizobium species.

43).

,

METHODS

Bacterial strains. Rhizobium strains were isolated from root nodules harvested on young seedlings of species of Acacia or Prosopis grown in tubes in the presence of soil suspension as previously described (6).

All strains used are listed in Table 1. They were checked for purity by repeated streaking and by microscopical exam- ination. The identity of the nodulating strains was checked by plant infection tests on the original host plants.

We included type or representative strains of the different Rhizobium, Bradyrhizobizim, Azorhizobium, Mesorhizobium, Sinorhizobium and Agrobacterium species, clusters of rhizobia described in Brazil by Moreira et al. (27), in Sudan by

Zhang et al. (50) and genospecies 3 and 4 found among Cicer isolates by Nour et al. (30). Growth and culture conditions. All Rhizobium and Brady- rhizobium strains were maintained on yeast mannitol agar (YMA), containing (g 1-l): mannitol, 10; sodium glutamate, 0.5; K,HPO,, 0.5; MgSO,.7H,O, 0.2; NaC1, 0.05; CaCl,, 0.04; FeCl,, 0.004; yeast extract (Difco), 1 ; pH 6.8; agar, 20. Azorhizobium and Agrobacteritim strains were maintained on yeast extract peptone medium (YEB) containing (in g per litre 0.01 M phosphate buffer, pH 7.2): peptone (Oxoid), 5; yeast extract (Oxoid), 1 ; beef extract (Oxoid), 5; sucrose, 5 and MgSO, .7H,O, 0.592. All strains were stored at - 80 "C on the same medium plus 15 % (v/v) glycerol. For protein and DNA preparations we used tryptone yeast extract medium (TY) containing (in g l-l, pH 6.8-7): tryptone (Oxoid), 5; yeast extract (Oxoid), 0.75; KH,PO,, 0.454; Na,HPO,. 12H,O, 2.388; CaCl,, 1 ; agar, 20 (LabM was used for protein preparation agar). Incubation was at 30 "C for 48 h for Rhizobium, Sinorhizobium, Mesorhizobium, Azorhizobium and Agrobacterium strains, and for 3 d for Bradyrhizobium strains. Morphological tests. Cell dimensions and morphology were determined on living and Gram-stained cells by phase- contrast microscopy. PAGE of total bacterial proteins. This was performed using small modifications of the procedure of Laemmli (17), as described previously (6). The normalized densitometric traces of the protein electrophoretic patterns were grouped by numerical analysis, using the GelCompar 2.2 software package (40). The similarity between each pair of traces was expressed by the Pearson product-moment correlation coefficient (r) converted for convenience to a percentage (32,

REP-PCR. Cells for DNA isolation were grown at 28 "C in 5 ml yeast extract-mannitol broth (19) to saturation. Total DNA was isolated from 1.5 ml cultures using hexadecyl- trimethyl ammonium bromide (2). The concentration was verified by comparing the DNA samples with the known concentrations of LDNA in agarose gel electrophoresis. The oligonucleotide primers (41) were synthesized by the Institute of Biotechnology, University of Helsinki. PCR amplifications were carried out in 25 pl reaction volume containing about 50 ng chromosomal DNA, 35 pmol primer (BOXAlR or GTG5) or primers (REPlR-1 and REP2-1 or ERIClR and ERIC2), 1 mM dNTPs, 2 U Dynazyme DNA polymerase (Finnzymes) and PCR buffer (10 mM Tris/HCl, pH 8.8 at 20 OC, 7 mM MgCl,, 50 mM KCl,O.l% Triton X- 100, 10 % DMSO, 170 mg BSA ml-l). The amplifications were performed with a DNA thermal cycler (Minicycler; MJ Research). For the REP primers the following temperature profile was used: initial denaturation at 95 "C for 6 min, 30 cycles at 94 "C for 1 min, at 40 "C for 1 min and at 65 "C for 8 min; final extension at 65 "C for 16 min. For ERIC, BOX and GTG primers, the first 7 min at 95 "C was followed by 30 cycles at 94 "C for 1 min, at 52 "C for 1 min, at 65 "C for 8 min; final extension at 65 "C for 16 min. After the reaction, 8 pl of the.products were separated on 1.5 % agarose gels for 6 h at 60 V, stained with ethidium bromide and photographed using Polaroid type 55 film. The REP-PCR fingerprints were analysed by the GelCompar 2.2 program and unweighted pair group method with averages (UPGMA) dendrograms were constructed from the profiles. Plant infection tests. The seeds were scarified and surface- sterilized with concentrated sulfuric acid. The durations

33).

370 International Journal of Systematic Bacteriology 48

Tropical tree rhizobia

Table 7. Strains used .... . ......... .. .......... .. ............ . ................ . .................................................................................................................... .. .................... I

Original strain number, or as received, is given. Abbreviations: ATCC, American Type Culture Collection, Rockville, MD, USA; BR and FL, strains from the CNPBS/EMBRAPA, Centro Nacional de Pesquisa em Biologia do Solo, Seropedica 23851, Rio de Janeiro, Brazil/Emprasa Brasiliera de Pesquisa Agropequaria; CFN, Centro de Investigación sobre Fijación de Nitrógeno, Universidad Nacional Autónoma de México, Cuernavaca, Mexico ; CIAT, Rhizobiurn Collection, Centro International de Agricultura Tropical, Cali, Columbia; HAMBI, Culture Collection of the Department of Microbiology, University of Helsinki, Helsinki, Finland; INPA, National Institute of Amazonia Research, Brazil; LMG, Collection of Bacteria of the Laboratorium voor Microbiologie, B-9000 Ghent, Belgium; NCPPB, National Collection of Plant-Pathogenic Bacteria, Harpenden Laboratory, Hertfordshire, UK; NZP, Culture Collection of the Department for Scientific and Industrial Research, Biochemistry Division, Palmerston North, New Zealand; ORS, ORSTOM Collection, Institut Français de Recherche Scientifique pour le Développement en Coopération, BP 1386, Dakar, Senegal; Pan., C. Panagopoulos, Crete, Greece; UPM, Universidad Politécnica Madrid, Spain; USDA, US Department of Agriculture, Beltsville, MD, USA.

9

Strain LMG HAMBI Host plantlorigin Geographical Reference/ no. no. origin source

Mesorhizobiuin plurifariuin ORS 1001 ORS 1015 ORS 1005 ORS 1004 ORS 1014tl ORS 1010 ORS 1002 ORS 13 ORS 1088 ORS 1018 ORS 1020 ORS 1024 ORS 1026 ORS 1029 ORS 1030 ORS 1031 ORS 1032T ORS 1035 ORS 1036 ORS 1037 ORS 1038 ORS 1040 ORS 38 ORS 42 HAMBI 1487 ORS 1093 ORS 1095 ORS 1096 HAMBI 260 BR 3804 INPA 12A INPA 78B INPA 1 18A INPA 129A INPA 338A Mesorhizobium loti 3F3C1 3F6g2 NZP 2230 NZP 22 1 3T NZP 2037 NZP 2014 NZP 2234

7836 7839 7845t1 7848 7849t1 7853 7854 7921

11880 11881 11883 11884 11886 11889 11890 11891 1 1892T 11893 11894 11895 11896 11898 11931 11935 14925

15297 15298 15633 9970

10031 10056 10059 10061 10093

4269 4284 6126 6125T 6123 6124

193

206

197 207 194

1994

1995

208T 195 196

1996

1998

2025

204

205

1 12gT 1338

1148

Acacia senegal Acacia senegal Acacia sp. Acacia senegal Acacia senegal Acacia seiiegal Acacia senegal Acacia sp. Acacia seyal Acacia senegal Acacia seizegal Acacia senegal Acacia senegal Acacia senegal Acacia senegal Acacia senegal Acacia senegal Acacia senegal Acacia senegal Acacia senegal Acacia senegal Acacia senegal Prosopis juliflora Prosopis juliflora Acacia senegal Acacia senegal Acacia tor tilis subs p . raddiana Acacia tor tilis subs p. raddiana Acacia senegal Chanzaecrista ensiformis Leucaena leucocephala Leucaena diversifolia Leucaena pulvuruleiita Leucaena puhurulenta Leucaena diversifolia

Wisteria frutescens Caragana arborescens Lotus maroccanus Lotus coriiiculatiu Lotus divaricatus Lotus corniculatus

Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Senegal Sudan Senegal Senegal Senegal Sudan Brazil Brazil Brazil Brazil Brazil Brazil

Colorado, USA Morocco New Zealand New Zealand

6 6 6 6 6 6 6 6 6 6 6 6

6 6 6 6 6 6 6 6 6

This work

This work This work 49 6

This work 6

49 27 27 27 27 27 27

14 14 14 14 14 14

(continued overleaj)

International Journal of Systematic Bacteriology 48 ~

371

P. de Lajudie and others

Table 1. (conta

Strain LMG HAMBI no. no.

Mesorhizobium ciceri UPM-Ca7T 1498gT 1750T UPM-Ca7T 17150T 522 17149 Mesorhizobium mediterraneum Ca-36T 17148T UPM-Ca142 14990 Mesovhizobium sp. (Cicer avietinum) genospecies 3 Ca-7 14994 Mesovhizobium sp. (Cicer arietinum) genospecies 4 IC-60 14995 Mesorhizobium huakuii IAM 14158T 14107T Mesorhizobium tianshanense A-lBST 017A 6 Sinorhizobium fredii USDA 205T USDA 191 Sinorhizobium meliloti NZP 4009 NZP 4027T 1021 Balsac Sinorhizobium medicae m75 m102 Sinorhizobium terangae ORS 51 ORS 52 ORS 100gT ORS 1013 ORS 1016 ORS 1072 ORS 1073 Sinorhizobium saheli ORS 60gT ORS 609t2 ORS 600 ORS 611 ORS 12 Rhizobium leguminosaruin NZP 561 CB 596 Biovar trifolii ATCC 14480 Biovar viciae ATCC 10004T Rhizobium tropici Group A

CFN 299

15767T 15768 15769

62 1 7T 8317

6130 6133T

16579 16580

6464 11859 7834T 7844t1 7851t1

11925 11926

7837T 8309t2 11864 8310 7835

6122

8820

8817T

9517

1674T

1 870T

1956

2075T 1337

1463 1167

1808 1809

200 220T

212 209 210

215T

217 213 214

1125

Host plantlorigin Geographical Reference/ origin source

Cicer arietinuin L. Cicer arietinum L. Cicer arietinum L.

Cicer arietinum L. Cicer arietinzini L.

Cicer arietinum L.

Cicer arietiiium L.

Astragalus sinicus

Glycyrrhiza pallidij7ora Sophora alopecuroides Glycyrrhiza uralensis

Glycine max Soil

Medicago sativa Medicago sativa

Medicago sativa

Medicago sativa Medicago sativa

Sesbailia rostrata Sesbania rostrata Acacia laeta Acacia senegal Acacia laeta Acacia senegal Acacia senegal

Sesbania cannabina Sesbania cannabina Sesbania pachycarpa Sesbania grandiJEora Unknown

Trifolium repens

Trifolium pratense L.

Piszim sativiim

Phaseolus vulgaris L.

Spain Spain Russia

Spain Spain

India

India

Nanjing, China

Xinjiang, China Xinjiang, China China

Henan, China Shanghai, China

Australia

USA Uruguay

Senegal Senegal Senegal Senegal Senegal Senegal Senegal

Senegal Senegal I

Senegal Senegal Senegal

Australia

USA

Brazil

30 30 30

29 29

29

29

4

5 5 5

14 14

F. M. Ausubel

11 11

6 6 6 6 6 6 6

6 6 6 6

This work

B. Jarvis

26


Table 1. (contd)


Strain LMG HAMBI Host plantlorigin Geographical Reference/ no. no. origin source

Group B CIAT 899T

Rhizobium etlì ORS 645T (CFN 42T) Rhizobium galegae HAMBI 540T HAMBI 1147 Agrobacterium biovar 1 B6' B2a Agrobactevium rhizogenes ATCC 1 1325T Agrobacterium rubi ATCC 13335' Agrobacterium vitìs Pan. AG61 NCPPB 1771 Azorhizobium cauliuodans ORS 571' FY12 ORS 478 Bradyrhizobium japonìcum NZP 5533 NZP 5549T TAL 309 Bradyrhizobium elkauiì NZP 5531T NZP 5532

9503T

1 1937T

6214T 6215

1 87T 268

1 50T

1 56T

257 233

6465T 11352 11820

6136 6138' 8316

6134T 6135

1 163T Phaseolus vulgaris L. Columbia 26

1727T Phaseolus vulgaris L. Mexico 37

Galega orientalis Finland Galega orientalis Russia

Lycopersicon lycopersicon Lycopersicon lycopersicon

Sesbania rostrata Sesbania rostrata Sesbania rostrata

Glycine nzax Glycine hispida Macrotyloma afiicanuin

Glycine max Glycine inax

Senegal 10 Senegal 34 Senegal 6

USA 14 Japan 14 Zimbabwe 1

14 USA 14

(min) of treatment in H,SO, for the different plant species were: Acacia senegal, 14; Acacia seyal, 30; Acacia tortilis subsp. raddiana, 150; Sesbania rostrata, 30-60 ; Sesbania pubescens, 60 ; Sesbania grandiflora, 60 ; Neptunia oleracea, 30 ; Leucaena leucocephala, 30 ; Acacia nilotica, 120 ; Prosopis julifEora, 30; Lotus corniculatus, 20; Ononis repens, 15. After acid treatment, the seeds were washed with water until all traces of acid were removed. The seeds were germinated in sterile Petri dishes on 1 % water agar for 24-48 h and then transferred to tubes containing Jensen seedling slant agar (42) for root nodulation trials (8-10 plants were routinely tested with each strain). Root nodules appeared 1&20 d after inoculation and 3 weeks later they were fully developed. Analysis of 165 rRNA genes. A large fragment of the 16s rRNA gene [positions 21-1521, Escherichia coli numbering system (3)] was amplified by PCR and sequenced directly as described previously (45, 46). The sequences determined, together with reference sequences from the EMBL database, were aligned using the PILEUP and PRETTY programs in the Genetics Computer Group (GCG) analysis package (9). The alignment was verified and corrected manually. In total, a continuous stretch of 1398 bases (including gaps) was used for further analysis. Distances were calculated using the DNADIST program of the Phylogeny Inference Package, P ~ I P (12). The program NEIGHBOR of the same package (12) was used to produce an unrooted phylogenetic tree. The stability of the groupings was verified by a bootstrap analysis of 500 data sets using the programs DNABOOT, DNADIST,

NEIGHBOR and CONSENSE (12). Similarity values were calculated using the DISTANCES program in the GCG package.

DNA base composition. Cells were grown for 2-3 d in Roux flasks on TY medium. High-molecular-mass DNA was prepared by the method of Marmur (22). The G + C content was determined by thermal denaturation (7) and calculated by the equation of Marmur & Doty (23), as modified by De Ley (7). DNA from E. coli LMG 2093 was used as a reference.

DNA-DNA hybridization. DNA-DNA hybridizations were performed with the initial renaturation rate method as described previously (8). Renaturations using approximately 50 pg DNA ml-l were carried out at 79-8 OC, the optimal renaturation temperature in 2 x SSC (SSC is 015 M NaC1, 0.015 M sodium citrate, pH 7.0).

RESULTS

Collection of Rhizobium isolates

We studied strains isolated previously (6) , together with a number of new isolates from different ecological areas in Senegal and originating from root nodules of native Acacia species (Acacia senegal, Acacia tortilis subsp. raddiana) and of the introduced species Prosopis juliflora. We also included representatives of the

International Journal of Systematic Bacteriology 48 373


70 IIIIIIIIIIIlll

80 90 100 j 1 I I I I I I I I I I I I I I I I I I I I t i

LMG61341 LM06135 Mû6136 LMG 8316 LMG 6217' LMG 8317

YMALMGl715a' YMALMG 17149

LMG 613 LMG6133' LMG 16580 LMG 16579

NZP 5531' NZP 5532 NZP 5533 TAL 309 USOA205' USOA 191

522 NZP 4W9 NZP 4027' Hambi 1809 Hambi 1808

UPMca7'

elkanii japonicum

fredii

meliloti medicae

LMG6214' Hambi 540'

YMALMG6124 NZP201.I LMG 14925 Hambi 1487 ¡MG6126 NZP2UO

4-1- EWE' t z 1 3 '

LMG 11935 ORS42 LMG 11894 ORS 1036

YMALMG 11693 ORS 1035 LMG 11895 ORS 1037 LMG7848 ORSlW4 LMG7P53 ORS1010 LMG 7w5 11 ORS 1W5 LMG7921 ORS13 LM07836 ORSlWl LMG 7849 11 ORS 1014 LMG11880 ORS1088 LMG7B39 ORS1015 - LMG 11883 ORS 1020 LMG 11881 ORS 1018 LMG7W ORSlW2 LMG 11884 ORS 1024 LMG 11889 ORS 1029 LMG 11886 ORS 1026 LMGlW58 lNPA788 LMG 1W59 INPA ll8A - LMG 1w93 INPA 338A LMGS370 BR3804 LMG 1w61 INPA 129A LMG 11898 ORS 1040 LMG 11931 ORS38 LMG 11895 ORS 1038 LMG 11890 ORS 1030 LMG 11892' ORS 1032' LMG 11891 ORS 1031 LMG lW31 INPA 1ZA LMG 15298 ORS 1096

u106124 NZP2014 LMG6123 NZPM37 LMG14994 Ca4

YMALMG14994 C2-7

-

- -

I

-

LMG 15297 ORS 1095

M. loti

a

Cluster U

b

M. loti Mesomizobium sp.

LMG4269 3F3C1 YMALMG 15767' A-18s' YMALMG157M) 017A YMALMG15769 8

LMG 14983' UPM.Ca7' Y" 14989' uPWa7' I M. ciceri

YMAVKj 15633 Hambi 260

YMALMG 14107' IAM 14158' LMG 14107~ IAM 14158' ] M. huakuii

YMALMG 11884 ORS 1024 LUG 14595

ATCC ORS 571' 11325'

Mesorhizobium sp.

AZ. caulinodans A. rhizoaenes

I YMALMG9970 Er3604

~~ K;; fa;y7711 A. vitis YMALMG 17148' ca36' M. meditemneum

I if :i ] Agmbacterium bv. 7 LMG 14990 UPM Ca142

.................................................... .............................................................................. ......................................................................................... Fig. 1. Dendrogram showing the relationships between the electrophoretic protein patterns from Senegalese, Sudanese, Brazilian and reference strains of Mesorhizobium, Rhizobium, Bradyrhizobium, Azorhizobium, Sinorhizobium and Agrobacterium species. The mean correlation coefficient (r) was represented as a dendrogram, constructed by UPGMA. Positions 10-320 of the 400 point traces were used for calculation of similarities between individual pairs of traces. The scale represents the r value converted to a percentage. Strains grown on YMA are indicated as such.

Y

i i

different phenotypic clusters identified by Zhang et al. (50) among strains isolated from nodules of Acacia senegal and Prosopis chilensis in Sudan (East Africa,

374

same climatic zone as Senegal). Here we consider only the strains (Table 1) which grouped in cluster U (see below).

International Journal of Systematic Bacteriology 48


HAMEl204 LMG 10056 - HAMEIZO5 LMG 10093 r HAMBl207 ORS 1016 HAME1194 ORS 1020 HAMBl 1996 ORS 1037 HAMEl 193 ORS 1001 HAMBl 195 ORS 1035 HAME11994 ORS 1024 HAMEl 196 ORS 1036 - HAME1206 ORS 1002 HAMEl 197 ORS 1088 HAMEI 1487 LMG 14925 HAMEl260 LMG 15633 HAME11938 ORS 1040 HAMEI 2025 ORS 1095 HAMEl 1995 ORS 1030 - HAM81208 ORS 1032

- + i - I

I, I

ClusterU

HAMBl 209 ORS 1072 S. f " y e

HAMEIZO0 ORS 52

HAMEI 210 ORS 1073 1 HAMEI 220T ORS 10ogT

I HAMBIZ12 ORS1016 J HAMBI 1337 USOA 191 HAMBl2075T USDA 205T 1 " "ed" HAMBl 1870T A-IEST M. HAME11956 6 1 HAMBl 1463 1021 HAMBl 1167 Balsac

1 " melilo'i

fig. 2. Numerical analysis of REP-PCR patterns of 17 cluster U strains (M. plurifarium) and of strains from M. loti, M. huakuii, M. ciceri, M. tianshanense, R. galegae, R. leguminosarum, R. tropici, R. etli, Sinorhizobium meliloti, Sinorhizobium fredii, Sinorhizobium terangae and Sinorhizobium saheli. The dendrogram was constructed using UPGMA.

SDSPAGE of total bacterial protein

A total of 700 protein profiles, including all strains previously found in cluster U (6), new isolates from Senegal, representatives from the phenotypic clusters from Sudan (50) and from the different Rhizobium, Sinorhizobium, Mesorlzizobiurn, Azorhizobium and Agrobacterium species and groups were compared. In the resulting dendrogram, cluster U was found to be clearly distinct from all the previously described species of Rhizobium, Sinorhizobium, Mesorhizobium, Azo- rhizobium and Agrobacterium Since we will focus on cluster U, a limited dendrogram (90 protein patterns) containing this group, its closest relatives and some representatives of the other named species is shown in Fig. 1. Cluster U (6) has been enlarged by five Senegalese strains: ORS 1026 from Acacia senegal, ORS 1095 and ORS 1096 from Acacia tortilis subsp. raddiana, and ORS 42 and ORS 38 from Prosopis jul$ora. At or above a mean correlation coefficient ( r ) of 82 %, cluster U now contains 32 strains.

The four previously distinguished (6) subclusters Ul-U4 are slightly modified. Subcluster U1 and U2 plus strains ORS 1002 and ORS 1088 are now found intermixed in our new analysis and form a single but rather heterogeneous group, a (Fig. 1). The 11 strains (five from Senegal, six from Brazil) previously found in subcluster U3 plus three additional Senegalese strains (ORS 1095, ORS 1096 and ORS 38) now group in one subcluster, b. The former subcluster U4, which consisted of M. loti strains LMG 6123 and LMG 6124 (6) now groups outside cluster U and appears to be more similar to Mesorhizobium sp. (Cicer).

With the exception of cluster U, M. loti and M. huakuii, representatives of the species of the genus Mesorhizobium [M. ciceri, Mesorlzizobium sp. (Cicer), M. rnediterraneum and M. tiaizshanense] did not grow significantly on TY medium, which is used for all the rhizobia in our database. Therefore strains of the Mesorhizobium species were grown on YMA and the resulting protein profiles were compared to those of representatives of the different subclusters of cluster U, also grown on YMA. All profiles obtained on YMA are marked as such in Fig. 1. We found that the growth medium used had no effect on the protein pattern of Mesorhizobium sp. (Cicer) LMG 14994 and 14995, M. ciceri LMG 14989 and M. huakuii LMG 14107 since profiles obtained on the two media grouped together (Fig. 1). By contrast, some members of cluster U (LMG 9970 and ORS 1024) grouped outside cluster U when grown on YMA. Our results show that M. meditersaneuni, M. ciceri, M. huakuii, M. tiarzshanense and Mesorhizobium sp. (Cicer) strains from genomic species 3 and 4 (29) do not cluster in cluster U. Two particular Sudanese isolates from Acacia senegal, HAMBI 260 and HAMBI 1487, belonging respectively to phena 4 and 13 (50) and having an identical 16S- PCR-RFLP genotype to cluster U (28), were also included in the study. HAMBI 260 was only grown on YMA and groups with strain ORS 1024 when grown on YMA, indicating that HAMBI 260 is probably a genuine member of cluster U; HAMBI 1487 groups with LMG 6124 when grown on YMA.

Amplification of the genomic DNA by REP-PCR

The amplifications resulted in specific fingerprints with



. Y

Fig. 3. Levels of DNA-DNA hybridization at 79.8 O C between DNAs from Mesorhizobium and cluster U (M. plurifarium) strains.

fragment sizes ranging from approximately 0.3-6.5 kb. A UPGMA dendrogram was constructed from the combined fingerprint profiles obtained with primers REP, ERIC, BOX and GTG (Fig. 2). In the dendrogram the following species were represented by a single or several strains: M. ciceri, M. huakuii, M . loti, M. tianshanense, R. etli, R. galegae, R. leguminosarum and R. tropici, Sinorhizobium fredii, Sinorhizobium meliloti, Sinorhizobium saheli and Sino- rhizobium terangae. Except for M. loti, all species represented by several strains, formed tight clusters. M. loti strains HAMBI 1338 (syn. LMG 6123, NZP 2037) and HAMBI 1148 clustered together but separately from M. loti strain HAMBI 112gT. The 17 cluster U strains used in this analysis formed a separate cluster not related to any of.the recognized species; however, cluster U was heterogeneous and formed several subclusters.

G + C content

The G + C content of strains of cluster U was 62.c64.4 mol% (Fig. 3).

LMG 6123 and LMG 6124. Two clear DNA groups and several strains with borderline values of DNA hybridization were distinguished in cluster U. Members of electrophoretic subcluster a had 77-88 % DNA hybridization and constitute DNA group I. Strain HAMBI 1487 from Sudan also belonged to this group with 91 % DNA hybridization with strain ORS 1024. DNA group II consisted of members of electrophoretic subcluster b among which 72-100% DNA hybridization was measured. DNA groups I and II exhibited a mean DNA hybridization value of 39 %. An analogous mean DNA hybridization value (38 %) was found between strains LMG 6123 and LMG 6124. No significant DNA hybridization was found between these strains and members of DNA groups I and II. In addition, three strains from electrophoretic subcluster b had debatable borderline values of DNA hybridization. Two of them, LMG 9970 and LMG 10056, had 57% DNA hybridization between them and 28-47 % with representatives from DNA groups I and II. The third strain, ORS 1096 (LMG 15298), exhibited DNA hybridization of 40 % with DNA group II and 55 % with DNA group I.

i'

L

DNA-DNA hybridizations

The genomic heterogeneity reported previously among the members of cluster U (6) was investigated further by extensive DNA-DNA hybridizations. More representative strains from each protein electrophoretic subcluster were included. The results are presented in Fig. 3. None of the species or groups of the Mesorhizobium branch gade significant levels of DNA hybridization (mean 14.6%) with members of cluster U or M. loti

16s rRNA gene sequence analysis

In addition to the previously reported 16s rDNA sequence of strains ORS 1001 and ORS 1002 (SDS- PAGE subcluster a) (6), we determined the 16s rDNA sequences of strains LMG 10056 and LMG 11892 (SDS-PAGE subcluster b), and of R. loti strains LMG 6123 and LMG 6124, which were previously con- sidered to be members of cluster U (6). These sequences were compared with all the available sequences for other members of Mesorhizobium and representatives


Tropical tree rhizobia 4

Bartonella elizabefhae F9251 Bartonella henselae Houston-I

vinsonii ATCC VR-152 Bartonella quintana ATCC VR-358

100 Sinorhizobium saheli LMG 7837

Zooglea ramigera ATCC 19623

Phyllobacterium mbiaceamm IAM 13587 rPhyi1obacterium myrsinacearum IAM 13584

Mesohizobium loti R8CS

Sinorhizobium meliloti LMG 6133

i - Mycoplana dimorpha IAM 13154

-

e4 - Mesorhizobium loti LMG 6125' Mesohizobium ciceri UPM-Ca7T Mesohizobium loti ATCC 3366gT - Mesorhizobium meditemneum UPM-Ca36'

Mesorhizobium loti ICMP 3153

Mesohizobium loti IAM 1 3588T Mesorhizobium loti LMG 6123

Mesorhizobium plurifarium ORS 1032' Ochrobactrum antropi IAM 14119

Brucella abortus 11 -1 9

100

84 Rhizobium etli Tal 182 Rhizobium galegae LMG 6214

Agmbacterium vitis LMG 8750 Agmbacterium tumefaciens 66

Agmbacterium Nbi LMG 156 Azorhizobium caulinodans LMG 6465

-

................. ............ .............................................. ...................... .. , , Fig. 4. 165 rRNA gene sequence-based dendrogram obtained by neighbour-joining, showing the phylogenetic position of M. plurifarium ORS 1032T (= LMG 118927, M. loti LMG 6123 and M. loti LMG 6124 among other Mesorhizobium species within the alpha subclass of the froteobacteria. Significant bootstrap probabilities are indicated at the branching points.

of other related taxa. A distance matrix analysis was performed and the dendrogram obtained by the neighbour-joining method is shown in Fig. 4. The Mesorlzizobiurn strains form a separate cluster with sequence similarities of at least 97-6 %. The strains of the different subclusters in cluster U were found to have identical 16s rDNA sequences. They occupy a separate position in the dendrogram with other Meso- rhizobium strains showing 97.6-99.4 '%O sequence similarity with cluster U. M. loti LMG 6123 and LMG 6124 showed 99.7% sequence similarity (4 base differences) with each other and 99.4% (9 base differences) with strains of cluster U. Strain LMG 6124

that of M. loti IAM 13588T (IAM, Institute of Applied Microbiology, The University of Tokyo, Japan). Two other sequences (GenBank accession nos X67229 and D14514) are available for the type strain of M. loti. These sequences were obtained from strains LMG 6125T and ATCC 33669T, respectively. They show 99.3 YO similarity with each other (8 base differences), which is not particularly high for strains that represent the same original isolate. In fact, the sequence of LMG 6125T has greater similarity to that of M. ciceri UPM- Ca7T (99.7 % similarity) than to that of ATCC 33669T. From our results (Fig. 4), it is nevertheless clear that LMG 6125T and ATCC 33669T group relatively closely together and group separately from IAM 1358gT. Their relatively low levels of 16s rDNA sequence

c was found to have a 16s rDNA sequence identical to

similarity (97.6-97.7 %) with strain IAM 1358gT sug- gest that the identity of the IAM type strain is questionable.

M. loti strains group in two separate clusters: five strains, LMG 6125T, ATCC 33669T, LMG4284, R88B and R8CS group together with M. ciceri UPM-Ca7T and Mesorhizobiuin sp. strains CJ2 and CJ5, two representatives of a group of non-symbiotic field isolates from the rhizosphere of Lotus corniculatus (38), at or above a similarity level of 99 YO (Fig. 4). A bootstrap value of 100% shows this grouping to be statistically significant. Four other strains, IAM 13588T, LMG 6124, ICMP 3153 and LMG 6123, group together at or above a similarity level of 99.6 %. They are closely related to M. lzuakuii IAM 1415gT (99.7-99.8 % similarity), cluster U (99.3-99.4 YO) and M. mediterraneum UPM-Ca36T (99.1 YO), whereas their level of sequence similarity with the first group of M. loti strains is 97-8-9806 %.

Host specificity

The plant host origins of strains belonging to cluster U are diverse (Table 1). Lortet et al. (21) reported that 18 strains from cluster U genospecies I and II exhibited similar host specificity towards Sesbania (Sesbania rostrata, Sesbania grand@ora, Sesbania pubesoens), Acacia (Acacia senegal, Acacia tortilis subsp. raddiana,



Table 2. Carbon source utilization as discriminative features in Mesorhizobium

+ , All strains are positive; - , all strains are negative; + / -, some strains are positive. The values are the percentage of positive strains.

Substrates M. pluvifavium" M. huakuiit M. lotis M. nieditevratieum§ M. &evi§ M. tianshattetise II

Erythritol 2-Ketoglutarate D-Tryptophan, DL-kynurenine, Zaminobenzoate, 3-aminobenzoate, 4-aminobenzoate

L-Xylose D-Fructose, D-mannose,

D- and L-arabinose, sorbitol, sucrose, pyruvate

Inositol Lactose D-Melibiose Xylitol D-Lyxose D- and L-Fucose Fumarate, L-malate Glycolate D-Malate L-Ornithine L-Arginine L-Proline DL-4-Aminobutyrate Citrate DL-3-Hydroxybutyrate Acetate L-a-Alanine, L-leucine D-Tagatose DL-Glycerate Sarcosine Histamine Dulcitol ß-Gentiobiose L-Ly sine L-Isoleucine Trigonelline ß-Alanine DL-5-Amino-valerate

Methyl-D-glucoside L-Valine L-Threonine L-Sorbose L-Phenylalanine L-Aspartate L-Arabitol, ethanolamine Butyrate, isobutyrate 2-Ketogluconate Arbutin Oxalate

D-Rafhose

- - -

+ +

+ + + + + + + + + + + + + 90 85 80 80 70 70 70 65 60 60 60 50 50 50 45 40 35 35 35 30 30 30 25 20 15 10 -

50 - -

+ +

+ + 50 50 + + + + 50 50 50 + + 50 + 50

+

-

-

- - - 50 50 50 50 50 + 50

+ 50 50

+ 50 50

-

-

- -

-

+

+ +/-

- +/- -

+ -

-

+ +/-

-

+/-

+/ -

-

* 26 strains were studied. Results from de Lajudie et al. (6). t LMG 14107T was studied. Results from de Lajudie et al. (6). 8 Strains LMG 6125T and LMG 4284 were studied. Results from de Lajudie et al. (6). 0 Results from Nour et al. (30). II Results from Chen et al. (5).



Acacia nilotica), Prosopis julifora and Leucaeria leuco- cepliala. We extended this study with four strains of cluster U which showed intermediate DNA hybridization results, as well as strains LMG 6123 and LMG 6124. We also tested nodulation on Acacia seyal, Neptunia oleracea, Lotus corniculatus and Ononis repens.

With the exception of LMG 1003 1, all strains of cluster U tested exhibited the same host specificity and were able to induce nodules on Acacia senegal, Acacia tortilis subsp. raddiana, Acacia nilotica, Acacia seyal, L. leucocephala and N . oleracea, but none could nodulate any of the Sesbania species tested, or O. repens or L. corniculatus. LMG 10031 had poor symbiotic properties compared to other cluster U strains: it could not nodulate Acacia senegal and Acacia tortilis subsp. raddiana and only 10-50 % of the plants were found nodulated in the case of Acacia seyal, N . oleracea and L. leucocephala.

Strain LMG 6123 could nodulate Acacia seyal, N . oleracea, and to some extent (10-50% of the plants nodulated) L. corniculatus and Acacia senegal. LMG 6124 could nodulate Acacia seyal and to some extent (10-50 % of the plants nodulated) Acacia tortilis subsp. raddiana but could not nodulate Acacia senegal and L. cornicula tus.

DISCUSSION

In a previous polyphasic taxonomic study of rhizobia we described a new group, designated cluster U, for a number of isolates from different leguminous plants in Africa, Brazil and New Zealand. We found this group to belong to the Mesorhizobium rRNA branch in the alpha subclass of the Proteobacteria, in the vicinity of M. huakuii (6). In the present study we extended the characterization of cluster U by including four new Senegalese isolates and additional reference strains from species and groups of the Mesorhizobium rRNA branch in the SDS-PAGE analysis. We performed more DNA-DNA hybridizations and 16s rRNA sequencing and introduced an additional technique,

On the basis of our extensive analysis of protein patterns, cluster U now comprises 32 strains (Fig. 1). It is clearly distinct from other species in related genera such as Rhizobium, Azorlzizobiurn, Sinorhizobium and Bradyrhizobiuni and also from the different Meso- rhizobium species included. This separate position of cluster U is confirmed by the phylogenetic analysis of 16s rDNA genes. All the Mesorhizobium sequences available were included, as well as representatives from related taxa. From the results it is evident that cluster U strains form a separate lineage on the larger Mesorlzizobium rRNA branch. Their nearest neighbour on this rRNA branch is M . huakuii IAM 14158T at 99.6 % sequence similarity (Fig. 4), corresponding to 6 base differences.

REP-PCR.

Because of the inclusion of additional strains in the SDS-PAGE study, the internal structure of cluster U has been modified. The four subclusters Ul-U4 evident in our previous analysis (6) have been re- arranged: in the present analysis we distinguish only two main subclusters, a and b. The first subcluster (a) consists of all strains that previously made up subclusters U1 and U2, supplemented by two new isolates and two previously separate strains; the second subcluster (b) consists of strains that previously formed subcluster U3, supplemented with three new isolates. The two strains of the former subcluster U4 (M. loti strains LMG 6123 and LMG 6124) now group outside cluster U in the SDS-PAGE analysis (Fig. 1). Although SDS-PAGE does allow the cluster U strains to be identified, some reorganization in the internal groupings occurred in our new analysis and it is clear that sub-groupings on the basis of protein patterns are not stable, indicating that this technique alone is not sufficient for the study of the internal structure of cluster U. REP-PCR experiments with representative cluster U strains distinguished clearly subcluster a, while subcluster b was further divided into two parts.

Despite the considerable internal structure of cluster U shown by these two techniques, and previous evidence of the presence of genotypic sub-groups (6), 16s rRNA gene sequences of four members representing cluster U were found to be identical. This was confirmed by the single genotype found for 18 strains by amplified rDNA restriction analysis (ARDRA) of the 16s rRNA gene (28). Therefore more extensive DNA-DNA hybridization experiments were performed. More strains representing the different electrophoretic subclusters were included as well as representative strains of the other Mesorhizobium species and Mesorhizobiunz sp. (Cicer), the unnamed genomic group 3 described by Nour et al. (29). Our results reveal the presence of at least two separate DNA groups (Fig. 3) in cluster U.

Group I consists of strains from electrophoretic subcluster a, representatives of which show at least 77% internal DNA hybridization. As suggested by Nick et al. (28) and REP-PCR results here, strain HAMBI 1487 from Sudan also belongs to this DNA group (with high DNA hybridization) although it does not group inside cluster U by SDS-PAGE. The second Sudanese strain having the same genotype as cluster U in ARDRA (28) was not included in DNA-DNA hybridizations, but is confirmed as a regular cluster U strain because of its REP-PCR profile and its similar protein profile on YMA.

Group II consists of strains of electrophoretic subcluster b. Since no significant DNA hybridization was found between the two DNA groups of cluster U and any of the other Mesorhizobium species included, the two DNA groups could be regarded as separate genospecies in the genus Mesorhizobium In accord- ance with the guidelines for the definition of species (44), it could be argued that a new Mesorhizobium species should be created for each of these genospecies



since they can be distinguished phenotypically (6). However, at present we believe it is not advisable to do so because the DNA hybridization matrix is not complete and we have identified at least three strains belonging to cluster U with intermediate DNA hybridization values with DNA groups I and II. The mean DNA hybridization between DNA groups I and II was 39 % ; it was 36 % between borderline strains of cluster U and DNA group II; it was 39.8% between borderline strains of cluster U and DNA group I. In comparison, the mean DNA hybridization between cluster U strains and strains of other species on the Mesorhizobiaim branch was much lower (14-7 %). In addition we have indications (unpublished data) that inclusion of more strains could lead to a blurring of the boundaries between the present genospecies : i.e. more intermediate strains may exist, showing that cluster U consists of a continuum of strains rather than several distinct, clearly delineated genospecies. Since cluster U is phylogenetically distinct (Figs 2 and 4) from all the other Mesorhizobium species and can also be differentiated phenotypically from these taxa (Table 2), we propose to create a single species, M. plurifariam for these strains. An analogous decision was adopted to describe R. tropici for a group of strains where two different genospecies (types A and B) sharing low DNA-DNA hybridization (36 %) and having differences in the sequence of their 16s rRNA genes could be distinguished; later, R. tropici strains intermediate between types A and B were described (24).

M. loti strains LMG 6123 and LMG 6124 grouped together inside cluster U in our previous SDS-PAGE analysis of protein patterns (6) whereas now (Fig. 1) they still group together, but outside cluster U, in the vicinity of Mesorhizobium sp. (Cicer). There was 38 % DNA hybridization between strains LMG 6123 and LMG 6124 and no significant DNA hybridization between these strains and the strains tested of cluster U or Mesorhizobium. In particular, levels of DNA hybridization with Mesorhizobium sp. (Cicer) LMG 14994, were less than 21 %, indicating that these strains may represent two additional genotypic groups in the genus Mesorhizobium This is confirmed by 16s rDNA analysis (Fig. 4) where strain LMG 6123 clearly occupies a separate position in the Mesorhizobium rRNA branch close to strain LMG 6124, which groups together with M. loti strains IAM 13588T and ICMP 3 153. These M. loti strains cluster separately from the majority of M. loti strains (LMG 6125T, ATCC 3366gT, LMG 4284, R88B and R8CS), which form a separate rRNA subcluster together with several nonsymbiotic field isolates from the rhizosphere of Lotus corniculatus (38) and the type strain of M. ciceri. Likewise, in the SDS-PAGE analysis of protein patterns M. loti strains were recovered in different places: M. loti strain LMG 4269 clusters separately from three M. loti strains including the type strain LMG 6125T. It is clear that the relationships of the M. loti strains need further study to define the different groups and establish their taxonomic status.

Description of Mesorhizobium plurifarium sp. nov.

Mesorhizobium plurifarium (plu.ri.fa'ri.um. N.L. adj. from plurifariam adv., in different places, referring to the fact that this species contains strains isolated ín several places in East Africa, West Africa and South America).

Aerobic, Gram-negative, non-spore-forming rods that are 0-5-0.7 pm wide by 1-3 pm long, except strain ORS 1005 which can be up to 4 pm long. Motile in liquid medium. Growth on yeast mannitol medium at temperatures as high as 42 "C (6). Colonies on YMA are beige, round, convex to drop-like, 0.5-2" diameter within 2-3 d at 30 OC, smooth with a creamy and gleaming aspect, mucoid but not mucilaginous. A wide range of carbohydrates, organic acids and amino acids are utilized as sole carbon sources for growth (6). Features that discriminate M. plurifarium from other Mesorhizobium species are given in Table 2. Most of the strains can nodulate Acacia senegal, Acacia tortilis subsp. raddiana, Acacia nilotica, Acacia seyal, Leu- caena leucocephala and Neptzinia oleracea, but none could nodulate Sesbania rostrata, Sesbania pubescens, Sesbania grandiflora, Ononis repens and Lotus corniculatus. Can be differentiated at the molecular level from other Mesorhizobium species and related genera by ARDRA of the 16s rRNA gene (28), SDS-PAGE of proteins, total DNA hybridization and 16s rDNA sequencing. The G + C content is 62-6-64.4 mol %. The type strain is the well-studied strain ORS 1032T (= LMG 1 1 892T). The G + C content of the type strain is 644mol %. All strains have been deposited in the Culture Collection of the Laboratorium voor Micro- biologie, University of Ghent, Belgium, and in the Culture Collection of the Laboratory of Soil Micro- biology, ORSTOM, Dakar, Senegal.

ACKNOWLEDGEMENTS

We thank O. Diagne and S. Badji for kindly providing their Rhizobium strains. We thank D. Monget and bioMérieux, Montalieu-Vercieu, France, for supplying API galleries. We thank B. Pot for helpful discussion and software assistance. This work was supported by the Commission of the European Communities (STD3 programme, contract TS2 0169-F; BRIDGE programme, contracts BIOT-CT91-0263 and BIOT-CT91-0294 ; by the French and Belgian Embassies through Programme d'Actions Intégrées franco-belge Tour- nesol 94085. M.G. is indebted to the Fund for Scientific Research, Flanders, Belgium, for research and personnel grants. A. W. is indebted to the Fund for Scientific Research for a position as postdoctoral research fellow.

REFERENCES 1. Alazard, D. (1991). La nodulation caulinaire dans le genre

Aeschynomene. PhD thesis, Université Claude Bernard, Lyon I.

2. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A. & Struhl, K. (1989). Current Protocols in Molecular Biology. Greene & Wiley.

. b'

J

d

f


’* Tropical tree rhizobia

3. Brosius, J., Palmer, M. L., Kennedy, P. J. & Noller, H. F. (1978). Complete nucleotide sequence of a 16s ribosomal RNA gene from Escherichia coli. Proc Nat1 Acad Sci USA 75, 4801-4805.

4. Chen, W. X., Li, G. S., Qi, Y. L., Wang, E. T., Yuan, H. L. & Li, J. L. (1991). Rhizobium huakuii sp. nov. isolated from the root nodules of Astragalus sinicus. bit J Syst Bacteriol41,

5. Chen, W., Wang, E., Wang, S., Li, Y., Chen, X. & Li, Y. (1995). Characteristics of Rhizobium tianshanense sp. nov., a moderately and slowly growing root nodule bacterium isolated from an arid saline environment in Xinjiang, People’s Republic of China. Int JSyst Bacteriol45,153-159.

6. de Lajudie, P., Willems, A., Pot, B. & 7 other authors (1994). Polyphasic taxonomy of rhizobia: emendation of the genus Sinorhizobium and description of Sinorhizobium ineliloti comb. nov., Sinorhizobium saheli sp. nov., and Sino- rliizobium teranga sp. nov. Int J Syst Bacteriol44,715-733.

7. De Ley, J. (1970). Reexamination of the association between melting point, buoyant density, and chemical base composition of deoxyribonucleic acid. J Bacteriol101,737-754.

8. De Ley, J., Cattoir, H. & Reynaerts, A. (1970). The quantitative measurement of DNA hybridization from renaturation rates. Ezir J Biochem 12, 133-142.

9. Devereux, J., Haeberli, P. & Smithies, O. (1984). A com- prehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 12, 387-395.

10. Dreyfus, B., Garcia, J. L. & Gillis, M. (1988). Characterization of Azorhizobiuni caulinodans gen. nov., sp. nov., a stem- nodulating nitrogen-hing bacterium isolated from Ses- bania rostrata. Int J Syst Bacteriol38, 89-98.

11. Eardly, B. D., Materon, L. A., Smith, N. H., Johnson, D. A., Rumbaugh, M. D. & Selander, R. K. (1990). Genetic structure of natural populations of the nitrogen-fixing bacterium Rhizobium nieliloti. Appl Environ Microbiol 56, 187-194.

12. Felsenstein, J. (1982). Numerical methods for inferring evolutionary trees. Q Rev Biol 57, 379-404.

13. Jarvis, B. D. W., Downer, H. L. & Young, J. P. W. (1992). Phylogeny of fast-growing soybean-nodulating rhizobia supports synonymy of Sinorhizobium and Rhizobium and assignment to Rhizobium fredii. Int J Syst Bacteiiol 42,

14. Jarvis, B. D. W., Gillis, M. & De Ley, J. (1986). Intra- and intergeneric similarities between the ribosomal ribonucleic acid cistrons of Rhizobium and Bradyrliizobiuin species and some related bacteria. Int J Syst Bacteriol36, 129-138.

15. Jarvis, B. D. W., Pankhurst, C. E. & Patel, J. J. (1982). Rhizo- bium loti, a new species of legume root nodule bacteria. Int J Syst Bacteriol32, 378-380.

16. Jarvis, B. D. W., van Berkum, P., Chen, W. X., Nour, S. M., Fernandez, M. P., Cleyet-Marel, J. C. & Gillis, M. (1997). Transfer of Rhizobium loti, Rhizobiuni huakuii, Rhizobium ciceri, Rliizobium mediteiraneum, and Rhizobium tiansha- neme to Mesorhizobium gen. nov. Int J Syst Bacteriol47,

17. Laemmli, U. K. (1970). Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 227, 680-685.

18. Limeschenko, E., Vorobyov, N. & Novikova, N. (1995). Diversity of bacteria isolated from root nodules of Ono- brychis spp. plants. In Nitrogen Fixation : Fundamentals and Applications, p. 707. Edited by I. A. Tikhonovich, N. A.

275-280.

4

4

a

93-96.

1

895-898.

Provorov, V. I. Romanov & W. E. Newton. Dordrecht, The Netherlands : Kluwer.

19. Lindström, K., Sarsa, M., Polkunen, J. & Kansanen, P. (1985). Symbiotic nitrogen fixation of Rhizobium (Galega) in acid soils and its survival in soil under acid and cold stress. Plant Soil 87, 293-302.

20. Lindström, K., van Berkum, P., Gillis, M., Martinez, E., Novikova, N. & Jarvis, B. (1995). Report from the roundtable on Rhizobium taxonomy. In Nitrogen Fixation : Funda- mentals and Applications, pp. 807-810. Edited by I. A. Tikhonovich, N. A. Provorov, V. I. Romanov & W. E. Newton. Dordrecht, The Netherlands : Kluwer.

21. Lortet, G., Mear, N., Lorquin, J., Dreyfus, B., de Lajudie, P., Rosenberg, C. & Boivin, C. (1996). Nod factor thin-layer chromatography profiling as a tool to characterize symbiotic specificity of rhizobia1 strains : application to Sino- rhizobiunz salieli, S. teranga and Rhizobium sp. strains isolated from Acacia and Sesbania. Mol Plant-Microbe Interact 9, 736-747.

22. Marmur, J. (1961). A procedure for the isolation of deoxyribonucleic acid from microorganisms. J Mol Biol 3,

23. Marmur, J. & Doty, P. (1962). Determination of the base composition of deoxyribonucleic acid from its thermal denaturation temperature. J Mol Biol 5, 109-1 18.

24. Martínez-Romero, E. (1996). Comments on Rhizobium systematics. Lessons from R. tropici and R. etli. In Biology of Plant-Microbe Interactions, pp. 503-508. Edited by G. Stacey, B. Mullin & P. M. Gresshoff. St Paul, MN: Inter- national Society for Molecular Plant-Microbe Interactions.

25. Martínez-Romero, E. & Caballero-Mellado, J. (1996). Rhi- zobium phylogenies and bacterial genetic diversity. Crit Rev Plant Sci 15, 113-140.

26. Martlnez-Romero, E., Segovia, L., Mercante, F. M., Franco, A. A., Graham, P. & Pardo, M. A. (1991). Rhizobium tropici, a novel species nodulating Phaseolus vulgaris L. beans and Leucaeiia sp. trees. Int J Syst Bacteriol41, 417-426.

27. Moreira, F. M. S., Gillis, M., Pot, B., Kersters, K. & Franco, A. A. (1993). Characterization of rhizobia isolated from different divergence groups of tropical Legunzinosae by com- parative polyacrylamide gel electrophoresis of their total proteins. Syst Appl Microbiol 16, 135-146.

28. Nick, G., Räsänen, L. A., de Lajudie, P., Gillis, M. & Lindström, K. (1998). The biodiversity among rhizobia, agrobacteria and Agrobacteriuin-like strains assessed by 16s rDNA PCR-RFLP. Appl Eizviron Microbiol (submitted).

29. Nour, 5. M., Cleyet-Marel, J.-C., Normand, P. & Fernandez, M. P. (1995). Genomic heterogeneity of strains nodulating chickpeas (Cicer arietinuin L.) and description of Rhizobium mediterraneum, sp. nov. Int J Syst Bacteriol45, 64Ck648.

30. Nour, S. M., Fernandez, M. P., Normand, P. & Cleyet-Marel, J.-C. (1994). Rhizobium ciceri sp. nov., consisting of strains that nodulate chickpeas (Cicer arietinurn L.). Int J Syst Bacteriol44, 51 1-522.

31. Novikova, N., Limeschenko, E., Bustos, P., Rodriguez, O., Davila, G. & Vorobyov, N. (1995). Taxonomy of Rhizobium strains from legumes of the temperate zone. In Nitrogen Fixation: Fundamentals and Applications, p. 716. Edited by I. A. Tikhonovich, N. A. Provorov, V. I. Romanov & W. E. Newton. Dordrecht, The Netherlands: Kluwer.

32. Pot, B., Gillis, M., Hoste, B., Van De Velde, A., Bekaert, F., Kersters, K. & De Ley, J. (1989). Intra- and intergeneric

208-218,

~ ~~~~



relationships of the genus Oceanospirillum. Int J Syst Bacteriol39, 23-34.

33. Pot, B., Vandamme, P. & Kersters, K. (1994). Analysis of electrophoretic whole organism protein fingerprints. In Modern Microbial Methods (Chemical Methods in Pro- caryotic Systematics series), pp. 493-521. Edited by M. Goodfellow & A. G. O’Donnell. Chichester: Wiley.

34. Rinaudo, G., Orenga, S., Fernandez, M. P., Meugnier, H. & Bardin, R. (1991). DNA homologies among members of the genus Azorhizobium and other stem- and root-nodulating bacteria isolated from the tropical legume Sesbania rostrata. Int J Syst Bacteriol41, 114-120.

35. Rome, S., Fernandez, M. P., Brunel, B., Normand, P. & Cleyet- Marel, J. C. (1996). Sinorhizobium medicae sp. nov., isolated from annual Medicago spp. Int JSyst Bacteriol46,972-980.

36. Sawada, H., leki, H., Oyaizu, H. & Matsumoto, S. (1993). Proposal for rejection of Agrobacteriunz tumefaciens and revised descriptions for the genus Agrobacterium and for Agrobacterizmz radiobacter and Agrobacterium rhizogenes. Int J Syst Bacteriol43, 694-702.

37. Segovia, L., Young, J. P. W. & Martinez-Romero, E. (1993). Reclassification of American Rhizobium leguminosarum biovar phaseoli type I strains as Rhizobium etli sp. nov. Int J Svst Bacteriol43, 374-311.

38. Sullivan, J. T., Eardly, B. D., van Berkum, P. & Ronson, C. W. (1 996). Four unnamed species of nonsymbiotic rhizobia isolated from the rhizosphere of Lotus corniculatus. Appl Environ Microbiol 62,2818-2825.

39. van Berkum, P., Beyene, D. & Eardly, B. D. (1996). Phylo- genetic relationships among Rhizobium species nodulating the common bean (Phaseolus vulgaris L.). bit J Syst Bacteriol46, 240-244.

40. Vauterin, L. & Vauterin, P. (1992). Computer-aided objective comparison of electrophoresis patterns for grouping and identification of microorganisms. Eur Microbiol 1, 37-41.

41. Versalovic. J., Schneider, M., de Bruijn, F. J. & Lupski, J. R. (1994). Genomic fingerprinting of bacteria using repetitive sequence-based polymerase chain reaction. Methods Mol Cell Bio1 5, 25-40.

42. Vincent, J. M. (1970). A manual for the practical study of root-nodule bacteria. International Biological Programme Handbook, no. 15, pp. 73-97. Oxford: Blackwell.

43. Wang, E. & Martinez, E. Personal communication. 44. Wayne, L. G., Brenner, D. J., Colwell, R. R. & 9 other authors

(1987). Report of the ad hoc committee on reconciliation of approaches to bacterial systematics. Int J Syst Bacteriol37, 463-464.

45. Willems, A. & Collins, M. D. (1992). Evidence for a close genealogical relationship between AJipia the causal organism of cat scratch disease, Bradyrhizobium japonicum and Blastobacter denitrijïcans. FEMS Microbiol Lett 96,

46. Willems, A. & Collins, M. D. (1993). Phylogenetic analysis of rhizobia and agrobacteria based on 16s ribosomal DNA sequences. Int J Syst Bacteriol43, 305-313.

47. Yanagi, M. & Yamasato, K. (1993). Phylogenetic analysis of the family Rhizobiaceae and related bacteria by sequencing of 16s rRNA gene using PCR and DNA sequencer. FEMS Microbiol Lett 107, 115-120.

48. Young, J. P. W. (1991). Phylogenetic classification of nitrogen-fixing organisms. In Biological Nitrogen Fixation, pp. 43-86. Edited by G. Stacey, R. H. Burris & H. J. Evans. New York: Chapman & Hall.

49. Young, J. P. W. & Haukka, K. E. (1996). Diversity and phylogeny of rhizobia. New Phytol 133, 87-94.

50. Zhang, X., Harper, R., Karsisto, M. & Lindström, K. (1991). Diversity of Rhizobium bacteria isolated from the root nodules of leguminous trees. Int J Syst Bacteriol 41, 104-113.

241-246.


Characterization of tropical tree rhizobia and description ...horizon.documentation.ird.fr/exl-doc/pleins_textes/pleins_textes_7/b_fdi_51-52/... · purity by repeated streaking and

Documents