• Azospirillum;
  • rhizosphere;
  • PGPR;
  • 1-aminocyclopropane-1-carboxylate;
  • ACC deaminase;
  • ethylene


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Deamination of the ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC) is a key plant-beneficial trait found in plant growth-promoting rhizobacteria (PGPR) and phytosymbiotic bacteria, but the diversity of the corresponding gene (acdS) is poorly documented. Here, acdS sequences were obtained by screening putative ACC deaminase sequences listed in databases, based on phylogenetic properties and key residues. In addition, acdS was sought in 71 proteobacterial strains by PCR amplification and/or hybridization using colony dot blots. The presence of acdS was confirmed in established AcdS+ bacteria and evidenced noticeably in Azospirillum (previously reported as AcdS), in 10 species of Burkholderia and six Burkholderia cepacia genomovars (which included PGPR, phytopathogens and opportunistic human pathogens), and in five Agrobacterium genomovars. The occurrence of acdS in true and opportunistic pathogens raises new questions concerning their ecology in plant-associated habitats. Many (but not all) acdS+ bacteria displayed ACC deaminase activity in vitro, including two Burkholderia clinical isolates. Phylogenetic analysis of partial acdS and deduced AcdS sequences evidenced three main phylogenetic clusters, each gathering pathogens and plant-beneficial strains of contrasting geographic and habitat origins. The acdS phylogenetic tree was only partly congruent with the rrs tree. Two clusters gathered both Betaprotobacteria and Gammaproteobacteria, suggesting extensive horizontal transfers of acdS, noticeably between plant-associated Proteobacteria.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In the associative symbiosis (cooperation) between plant growth-promoting rhizobacteria (PGPR) and the plant, the PGPR benefit the latter through a variety of indirect (e.g. competition or antagonism towards phytopathogens) and/or direct (by enhancing nutrient availability for roots, inducing disease resistance in the plant, and/or modifying plant hormonal balance) effects (Haas & Keel, 2003; Moënne-Loccoz & Défago, 2004; Bally & Elmerich, 2005). It is well documented that some of the direct phytobeneficial effects displayed by PGPR, for example nitrogen fixation, can also be implemented by bacteria such as rhizobia, in a context of obligate/mutualistic symbiosis with the plant (Sullivan et al., 2002; Ma et al., 2003a). Recently, the diminution of ethylene concentration via microbial deamination of the ethylene precursor 1-aminocyclopropane-1-carboxylate (ACC) (Glick et al., 1998; Holguin & Glick, 2003) was identified as another phytobeneficial mode of action exhibited by both PGPR and nitrogen-fixing rhizobia (Ma et al., 2003a, b).

The ACC deaminase [EC] is a pyridoxal 5′-phosphate (PLP)-dependent enzyme (Yao et al., 2000) that catalyses the conversion of ACC to ammonia and α-ketobutyrate. In PGPR closely associated with roots, this enzyme is thought to lead to a diminution of the amount of plant ACC available for conversion into ethylene (Glick et al., 1998). Because less ethylene will be synthesized by the plant, the inhibitory effects of this phytohormone on root elongation will be reduced. In parallel, the PGPR may use reaction products of ACC hydrolysis as nutrients for growth. Experimental data indicate that inoculation of plants with ACC deaminase-positive PGPR did reduce ethylene levels in the root zone while resulting in enhanced root proliferation (Penrose & Glick, 2003). Inactivation of acdS (encoding ACC deaminase) in the PGPR Pseudomonas putida GR12-2 and Enterobacter cloacae UW4 abolished the ability of the bacteria to promote the elongation of canola roots (Glick et al., 1994; Li et al., 2000). In the bacterial symbiont Rhizobium leguminosarum, inactivation of acdS reduced the extent of nodulation (Ma et al., 2003a). ACC deaminase activity is also documented in the yeast Hansenula saturnus (Minami et al., 1998) and the fungus Penicillium citrinum (Jia et al., 2000), but its ecological significance in eukaryotic microorganisms is unknown.

As ACC deaminase activity in bacteria can benefit the plant, it would be of interest to improve our knowledge of the range of bacteria displaying this activity. On one hand, ACC deaminase activity (and sometimes acdS) has mostly been documented in PGPR belonging to the genera Pseudomonas, Enterobacter and Kluyvera, and in the symbiont R. leguminosarum (Glick et al., 1995; Burd et al., 1998; Wang et al., 2001; Penrose & Glick, 2003; Ma et al., 2003b). ACC deamination ability has been identified by phenotypic means so far (Penrose & Glick, 2003). PCR amplification of acdS has been attempted before (Babalola et al., 2003) but without subsequent verification by sequencing, and there is a need to develop genetic tools to identify acdS in bacteria. Therefore, a genetic approach based on PCR and Southern hybridization was developed in this work to search for acdS in a range of Proteobacteria. On the other hand, putative ACC deaminase sequences derived from the annotation of whole-genome sequences are also available, but these sequences have not been validated by phylogenetic means. Therefore, phylogenetic methods were used to study ACC deaminase-related sequences, with the objectives of confirming their identity (based also on key residue analysis) and gaining insight into the evolutive history of acdS.

The experimental work was carried out using a collection of Proteobacteria encompassing subdivisions Alpha, Beta and Gamma, with an emphasis on the genera Azospirillum, Agrobacterium, Burkholderia and Pseudomonas. Azospirillum and Pseudomonas were selected as key PGPR genera, in which ACC deaminase activity is thought to be respectively absent (Holguin & Glick, 2001, 2003) and prevalent (Wang et al., 2001). Therefore, the current work included an investigation of AcdS activity, especially in the case of Azospirillum. With the recent recognition that certain emblematic virulence properties, for example the type-three protein secretion system (Ochman & Moran, 2001; Mazurier et al., 2004; Rezzonico et al., 2004), are widespread also in plant-beneficial bacteria, it is conceivable that key phytobeneficial traits may in fact be present in pathogens as well. To assess this hypothesis in the case of acdS, Agrobacterium and Burkholderia were included in the study because many strains from these genera have been extensively characterized as true or opportunistic pathogens affecting plants or animals.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Bacterial strains and culture conditions

The bacterial strains used in this study are listed in Table 1. The majority of strains were taken from culture collections, and the Azospirillum strains isolated in this work were obtained from the roots of field-grown plants in Pakistan, using N-free NFB medium (Nelson & Knowles, 1978). Azospirillum was grown first overnight with shaking (200 r.p.m.) in Tryptone Yeast extract (TY) (Beringer, 1974) medium, and then overnight on TY agar. Rhizobium and Agrobacterium were grown for 48 h on TY agar supplemented with 6 mM CaCl2 and on MG agar (Keane et al., 1970), respectively. Burkholderia was grown for 72 h on Pseudomonas capacia azelaic acid trytamine (PCAT) medium (Burbage & Sasser, 1982). Pseudomonas and Kluyvera were cultured overnight on King's medium B (King et al., 1954), and Enterobacter strains were cultured in Tryptic Soybean Broth (TSB; Serlabo, Bonneuil sur Marne, France). All strains were grown at 28°C.

Table 1.   Occurrence of acdS in the 71 Proteobacteria used in the study
SpeciesGeographic originSourceStrainsReferenceacdS hybridization*acdS PCR analysisAcdS activity
PCR productacdS sequence
  • *

    Hybridization results on colonies with an acdS PCR probe (from Pseudomonas fluorescens PITR2, Burkholderiacaledonica LMG19076 and/or Azospirillum lipoferum CN1) are indicated with either ++, strong signal with at least one probe; +, moderate signal with at least one probe; 0, no hybridization signal with none of the probes or ND, not done.

  • PCR results are indicated with +, successful amplification; −, no amplification or ND, not done.

  • Sequencing results are indicated with +, acdS confirmed by BLASTN; −, no homology; F, sequencing failed or ND, not done.

  • §

    This acdS sequence was obtained after hybridizing a DNA library (see Materials and methods).

  • Presence or absence of ACC deaminase activity already documented, or acdS already sequenced.

  • For certain taxa, indications are given on the pathogenic (P) or plant growth-promoting (PGPR) potential of the strains when available.

  • **

    ** The roman numerals in parentheses indicate genomovars.

Azospirillum (α)
 Azospirillum lipoferumBangladeshRiceMRB16Rhaman (1987)0 ND
FranceCornCRT1Fages & Mulard (1988)+++
FranceRice4BBally et al. (1983)++++§+
JapanRiceB52Elbeltagy et al. (2001)0+ND
JapanRiceB510Elbeltagy et al. (2001)++ND ND
JapanRiceB518Elbeltagy et al. (2001)+++ND
PakistanCottonCN1, N4This work++++
PakistanRiceRSWT1This work0 
VietnamRiceTVV3Tran Van et al. (1997)+++++
 Azospirillum brasilenseBrazilDigitariaSp7Tarrand et al. (1978)+++
BrazilWheatSp245Penot et al. (1992)+++ND
FranceRicePH1Rinaudo (1982)+++ND
FranceSorghumL4Kabir et al. (1996)+++
JapanRiceB506Elbeltagy et al. (2001)+ND ND
PakistanCornZN1This work+ 
PakistanWheatWb1, Wb3, WN1, WS1This work0 
USACynodon dactilonCdEskew et al. (1977)++ 
 Azospirillum irakenseIraqRiceKBC1Khammas et al. (1989)0+ND
PakistanRiceRSB1This work0 
 Azospirillum sp.CubaRice soilR5(6)Lavire (1998)+++ND
CubaRice soilR5(15)Lavire (1998)0+ND
MaliSoilNC9Kabir et al. (1996)0+ND
Rhizobium (α)
 Rhizobium leguminosarumUnknownPea128C53KMa et al. (2003a)0+++
Agrobacterium (α)
 Genomic group 1USACrown gallS56Mougel et al. (2002)++ ND
 Genomic group 2FranceHuman urineCIP 43-76Kiredjian (1979)++NDND
 Genomic group 3FranceHuman cephalo-rachidian liquidCIP 111-78Kiredjian (1979)0 ND
 Genomic group 4 (Agrobacterium tumefaciens)USAApple seedlingATCC 23308Yanagi & Yamasato (1993)++ ND
 Genomic group 6IsraelDahliaZutra F/1Mougel et al. (2002)++ ND
 Genomic group 7UKFlacourtia ramontchiNCPPB 1641Mougel et al. (2002)+ ND
 Genomic group 8USACherry tree gallC58Mougel et al. (2002)+++ND
 Genomic group 9South AustraliaPotting soilLMG26Mougel et al. (2002)+ ND
 Genomic group 10 (A. rhizogenes)USAAppleCFBP 2408Mougel et al. (2002)0 ND
 Genomic group 11 (A. rubi)USARubus ursinusLMG17935Mougel et al. (2002)0 ND
Achromobacter (β)
 Achromobacter xylosoxidansRussiaPea/leaf mustard stand (PGPR)Cm4Belimov et al. (2001)+++++
Burkholderia (β)
 Burkholderia caledonicaUKRhizosphere (PGPR)LMG19076Coenye et al. (2001a)++++
 Burkholderia caribiensisMartiniqueVertisolLMG18531Achouak et al. (1999)0 ND
 Burkholderia caryophylliUSADianthus caryophyllus (P)LMG2155Yabuuchi et al. (1992)+ ND
 Burkholderia cepacia (I)**USAAllium cepa (P)LMG1222Yabuuchi et al. (1992)+++++
 Burkholderia cepacia (II)BelgiumClinical isolateLMG13010Vandamme et al. (1997)++ ND
 Burkholderia cepacia (III)UKClinical isolate (P)LMG16656Vandamme et al. (1997)+++
 Burkholderia cepacia (IV)BelgiumClinical isolateLMG14294Vandamme et al. (2000)++++
 Burkholderia cepacia (V)VietnamRice (PGPR)LMG10929Gillis et al. (1995)++ ND
 Burkholderia cepacia (VI)USAClinical isolateLMG18941Coenye et al. (2001b)+++++
 Burkholderia gladioliUSAGladiolus sp. (P)LMG2216Yabuuchi et al. (1992)+++FND
 Burkholderia glatheiGermanySoilLMG14190Viallard et al. (1998)0 ND
 Burkholderia glumaeJapanRice (P)LMG2196Urakami et al. (1994)0+ND
 Burkholderia graminisFranceCornLMG18924Viallard et al. (1998)++++
 Burkholderia kururiensisJapanAquiferLMG19447Zhang et al. (2000)+ ND
 Burkholderia phenaziniumUnknownSoilLMG2247Viallard et al. (1998)0+++
 Burkholderia plantariiJapanRice (P)LMG9035Urakami et al. (1994)ND ND
 Burkholderia pyrrociniaUnknownSoilLMG14191Viallard et al. (1998)+++FND
 Burkholderia tropicalisMexicoCornBM16Estrada et al. (2002)++ND
MexicoCorn (PGPR)BM273Estrada et al. (2002)+++++
Ralstonia (β)
 Ralstonia solanacearumGuyaneTomatoGMI1000Boucher et al. (1985)ND+++
Pseudomonas (γ)
 Pseudomonas fluorescensBhutanTomatoP97.26Wang et al. (2001)++ +
Czech RepublicCucumberK94.31Wang et al. (2001)+++++
Czech RepublicWheatP97.30Wang et al. (2001)+++++
IrelandSugar beetF113Fenton et al. (1992)+++++
ItalyTomatoPILH1Keel et al. (1996)0+F+
ItalyWheatPITR2Keel et al. (1996)+++++
SwitzerlandCucumberP97.38Wang et al. (2001)ND+F+
SwitzerlandCucumberCM1'A2Fuchs & Défago (1991)+++++
SwitzerlandTobaccoCHA0Stutz et al. (1986)0 
SwitzerlandTomatoTM1A3Fuchs & Défago (1991)+++F+
Enterobacter (γ)
 Enterobacter cloacaeCanadaSoil (PGPR)UW4Glick et al. (1995)++++
USASoil (PGPR)CAL2Glick et al. (1995)+++++
Kluyvera (γ)
 Kluyvera ascorbataRussiaSoil (PGPR)SUD 165Burd et al. (1998)0ND +

PCR amplification and sequencing

PCR amplifications were performed in 100 μL volumes containing 1 × PCR buffer, 1.5 mM MgCl2, 0.5 μM of each primer, 10 μM of each dNTP, 0.5 U Taq DNA polymerase (Gibco BRL/Life technologies, Cergy Pontoise, France), and cells from one colony as DNA source. Primers for amplification of acdS were defined according to nucleotide alignment of sequences available in Genbank (Pseudomonas sp. 65G, ACP and 17, E. cloacae UW4 and CAL2), and they are listed in Table 2. PCR cycles were as follow: (i) 95°C for 5 min, (ii) 95°C for 30 s, 50°C for 30 s, 72°C for 30 s (35 cycles), and (iii) 72°C for 7 min. Universal eubacterial primers PA and PH were used to amplify 1.5 kb of the 16S rRNA gene rrs (Bruce et al., 1992). PCR cycles for rrs were as follows: (i) 95°C for 5 min, (ii) 95°C for 30 s, 65°C for 30 s, 72°C for 45 s (35 cycles), and (iii) 72°C for 7 min.

Table 2.   Primers used in this study
Target genes and primersSequenceacdS nucleotidic position in Pseudomonas fluorescens ACP*Corresponding AcdS domainReference
  • *

    Accession number M73488.

  • acdS amplicons were 792 (F1936/F1938), 558 (F1936/F1939), 750 (F1937/F1938) and 516 bp long (F1937/F1939) based on the sequence in P. fluorescens ACP.

 F1936 (forward)5′ GH GAM GAC TGC AAY WSY GGC 3′38–44EDCNSGThis work
 F1937 (forward)5′MGV AAG CTC GAA TAY MTB RT 3′52–59RKLEYLIPThis work
 F1938 (reverse)5′AT CAT VCC VTG CAT BGA YTT 3′296–302KSMHGMThis work
 F1939 (reverse)5′ GA RGC RTC GAY VCC RAT CAC 3′218–224VIGIDAThis work
 PA5′ AGA GTT TGA TCC TGG CTC AG 3′  Bruce et al. (1992)
 PH5′ AAG GAG GTG ATC CAG CCG CA 3′  Bruce et al. (1992)

DNA bands were extracted from electrophoretic gels using the QIAquick Gel Extraction Kit (Qiagen, Courtaboeuf, France) and cloned into pGEM-T vector (Promega, Charbonnière-les-Bains, France) according to supplier's instructions. The presence of an insert (acdS and rrs amplicons) was checked by restriction and PCR using vector-annealing primers SP6 and T7. DNA sequencing (Genome Express, Grenoble, France) was performed on both strands on an ABI377 sequencer FS (Perkin-Elmer, Courtaboeuf, France) using the ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction kit with AmpliTaq DNA polymerase.

Analysis of acdS and deduced AcdS sequences

BLASTN and BLASTP searches (Devereux et al., 1984; Altschul et al., 1990) were performed to detect similar DNA and deduced protein sequences (obtained using OrfFinder; NCBI web site), respectively. DNA sequences and deduced protein sequences were aligned using the multiple alignment software CLUSTAL X and CLUSTAL W version 1.74, respectively (Thompson et al., 1994). AcdS comparisons were computed in SeqVU (J. Gardner, Garvan Institute of Medical Research, Sydney, Australia). The deduced AcdS sequences from a selection of three strains (E. cloacae UW4, Pseudomonas sp. ACP and R. leguminosarum 128C53 K) were analysed by comparison with the well-studied AcdS from H. saturnus (Yao et al., 2000).

Phylogenetic analyses of acdS and deduced AcdS sequences were conducted using MEGA version 2.1 (Kumar et al., 2001). Distances between sequence pairs, the deduced phylogenetic tree, and bootstrap values were all computed using the global gap removal option. For acdS sequences, the distance-based tree-making methods used included the neighbour-joining method (NJ) (Saitou & Nei, 1987), using both the Kimura 2 parameter (K2P) and Jukes-Cantor (JC; with gamma parameter), and the maximum parsimony method (MP) (Berry & Gascuel, 1996). For deduced AcdS sequences, trees were obtained with the NJ method, using both Poisson correction and observed divergence (Galtier et al., 1996), and with the MP method. Nodal robustness of the trees was assessed using 1000 bootstrap replicates.

Analysis of acdS by hybridization

Bacterial colonies were transferred onto Gene Screen Plus nylon membranes, hybridized and washed (twice with saline-sodium citrate (SSC) 2 × for 5 min, and twice with SSC 2 × containing sodium dodecyl sulphate (SDS) 10% for 30 min) at 65°C according to the manufacturer's recommendations (NEN life science products, Le Blanc Mesnil, France). Autoradiography was performed according to Amersham-Pharmacia protocols (Orsay, France). To obtain probes, primers F1936 and F1938 were used to amplify acdS in Pseudomonas fluorescens PITR2, Burkholderia caledonica LMG19076 and Azospirillum lipoferum CN1. PCR fragments were radioactively labelled with 32P-dCTP according to the Random Primed DNA Labelling Kit (Boehringer Mannheim, Mannheim, Germany) and used as probes in DNA blot analyses. Hybridization results were confirmed when the procedure was repeated.

ACC deaminase activity in selected Proteobacteria

ACC deaminase activity in 24 representative proteobacterial strains was assessed by quantifying the reaction product α-ketobutyrate, using the procedure described by Penrose and Glick (2003). Briefly, a tube containing 7.5 mL of TY medium was inoculated with A. lipoferum (strain 4B, RSWT1, CRT1, TVV3, N4 or CN1), A. brasilense (strain L4, Sp7, Cd, WN1, WB1 or WB3), Azospirillum irakense (strain RSB1), B. caledonica (strain LMG19076), Burkholderia tropicalis (strain BM273), Burkholderia cepacia genomovar I (strain LMG1222), III (strain LMG16656), IV (strain LMG14294) or VI (strain LMG18941), Burkholderia graminis (strain LMG18924), Burkholderia phenazinium (strain LMG2247), Ralstonia solanacearum (strain GMI1000) or P. fluorescens (strains PITR2 and CHA0 as AcdS-positive and AcdS-negative controls, respectively). The tubes were incubated overnight at 28°C with shaking. Bacterial biomass was harvested by centrifugation (8000 g) for 10 min at 4°C. The cells were washed twice in 5 mL of DF salt minimal medium (Penrose & Glick, 2003) and resuspended in 7.5 mL of DF salt minimal medium supplemented with 3 mM ACC to induce acdS transcription. The bacterial cells were incubated for 24 h with shaking at 28°C and harvested by centrifugation, as described above. The pellets were washed twice in 5 mL 0.1 M Tris HCl, pH 7.6. The cells were lysed using 30 μL of toluene, and the procedure of Penrose & Glick (2003) was completed by vigourous shaking of the tubes for 1 min in a MM200 Retsch mixer mill (Bioblock, Vaulx Milieu, France) to ensure complete cell lysis. The amount of α-ketobutyrate produced was determined with a colorimetric method (absorbance at 540 nm), by comparison with a standard curve, according to Penrose & Glick (2003). Total protein content in the assay was measured using the BCA Protein Assay Kit (Pierce, Rockford, IL). ACC deaminase activity was assessed at least twice for each strain. Data were log-transformed before means and standard deviations were computed, and analysis of variance was performed followed by Fisher's LSD tests (P<0.05) (S-PLUS 6 for Windows; Insightful Corporation, Seattle, WA).

Accession numbers

acdS and AcdS sequence data already available and included in this work were obtained from the NCBI database for E. cloacae UW4 (GenBank accession no. AF047710) and CAL2 (AF047840), Enterobacter aerogenes Cal3 (AY604544.1), Caulobacter crescentus CB15 (AE005876), Salmonella typhimurium LT2 (NC_003197), Salmonella enterica CT18 (NC_003198), Erwinia carotovora SCRI1043 (NC_004547), Yersinia pestis KIM (NP_669768) and CO92 (CAC90662), Yersinia pseudotuberculosis IP 32953 (BX936398), Bordetella parapertussis 12822 (BX640435), Desulfovibrio vulgaris Hildenborough (NC_002937), Pseudomonas sp. 6G5 (M80882), 17 (U37103) and ACP (M73488), Pseudomonas syringae DC3000 (AE016853), GR12-2 (AY604545.1) and B728a (CP000075), Pseudomonas marginalis (AY604542.1), Pseudomonas brassicacearum Am3 (AY604528.1), P. putida Bm3 (AY604533.1), P. fluorescens PfO-1(NZ_AAAT03000010), Ralstonia eutropha JMP134 (NZ_AADY01000003), R. solanacearum GMI1000 (AL646080), Serratia proteamaculans SUD (AY604543.1), Achromobacter sp. CM1 (AY604541.1), Achromobacter xylosoxidans A551 (AY604539.1) and BM1 (AY604540.1), Variovorax paradoxus 5C2 (AY604531.1), 2C1 (AY604530.1) and 3C3 (AY604532.1), Burkholderia xenovorans LB400 (NZ_AAAJ02000016), Burkholderia pseudomallei K96243 (NC_006350), Burkholderia mallei ATCC 23344 (CP000011), B. cepacia R18194 (NZ_AAEI01000001) and R1808 (NZ_AAEH01000001), Rhizobium sullae HCNT-1 (AY604534.1), R. leguminosarum 128C53 K (AF421376 and AY172673) and 99A1 (AY604535.1), Bradyrhizobium japonicum USDA110 (NC_004463), and Agrobacterium tumefaciens d3 (AF315580). Nucleotide and deduced protein sequence data obtained in this work are available at the NCBI database for A. lipoferum 4B (GenBank accession no. DQ125242), CN1 (DQ125253), N4 (DQ125255), and TVV3 (DQ125257), B. cepacia LMG16656 (DQ125243), LMG1222 (DQ125251), LMG14294 (DQ125259), and LMG18941 (DQ125258), B. caledonica LMG19076 (DQ125247), B. graminis LMG18924 (DQ125249), B. phenazinium LMG2247 (DQ125252), B. tropicalis BM273 (DQ125254), P. fluorescens PITR2 (DQ125244), TM1A3 (DQ125245), CM1′A2 (DQ125246), P97.30 (DQ125248), K94.31 (DQ125250), and F113 (DQ125256), and A. xylosoxidans Cm4 (DQ133504).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Analysis of acdS PCR amplicons in Proteobacteria

Two forward and two reverse degenerative primers for acdS amplification were designed based on three Pseudomonas and two Enterobacter cloacae sequences available in Genbank (Table 2). The four primer combinations were tested on nine AcdS-positive pseudomonads (strains CM1′A2, TM1A3, P97.38, F113, K94.31, P97.30, P97.26, PITR2, PILH1). One PCR band, which was of the expected size, was obtained in all strains, except that P. fluorescens P97.26 did not yield any PCR amplicon. Primers F1936 and F1938 were selected for further use because (i) they were the only ones that worked with Azospirillum (a key taxon targetted in this study), and (ii) they yielded the longest amplicon (about 800 bp).

Using primers F1936 and F1938, a PCR product of the expected size (along with other PCR bands in certain strains) was obtained with 42 of the 68 Proteobacteria tested, including three strains in which acdS had already been sequenced (Table 1). Sequencing was performed on the 39 other 800-bp PCR products. Sequencing failed in five cases (despite several attempts). BLASTN analysis confirmed the presence of acdS for 19 bacteria, whereas no significant hit with GenBank sequences was obtained for the other 15.

The acdS gene was found in all three proteobacterial subdivisions tested (Alpha, Beta and Gamma), including taxa for which (to our knowledge) acdS or ACC deaminase activity has not previously been published, for example Azospirillum. Indeed, we obtained an acdS sequence for A. lipoferum (in three of ten strains). In the genus Burkholderia, an acdS sequence was obtained for eight strains (including three clinical isolates), which encompassed the species caledonica, graminis, phenazinium, tropicalis and genomovars I, III, IV and VI within the cepacia complex (Table 1).

Phylogenetic validation of partial acdS sequences obtained from databases

In addition to sequences obtained in this work, all putative ACC deaminase sequences available in databases were considered. Among the acdS-related sequences available in GenBank, BLAST alignment and phylogenetic analysis of the deduced proteic sequence for the Alphaproteobacteria C. crescentus CB15 (AE005876), Betaproteobacteria Achromobacter sp. CM1 (AY604541.1), A. xylosoxidans BM1 (AY604540.1), B. parapertussis 12822 (BX640435), R. eutropha JMP134 (NZ_AADY01000003), B. pseudomallei K96243 (NC_006350), Deltaproteobacteria D. vulgaris Hildenborough (NC_002937) and the Gammaproteobacteria S. typhimurium LT2 (NC_003197), S. enterica CT18 (NC_003198), E. carotovora SCRI1043 (NC_004547), Y. pestis KIM (NP_669768) and CO92 (CAC90662), Y. pseudotuberculosis IP 32953 (BX936398), P. syringae GR12-2 (AY604545.1) and B728a (CP000075), P. fluorescens PfO-1(NZ_AAAT03000010), P. marginalis (AY604542.1), E. aerogenes Cal3 (AY604544.1) and S. proteamaculans SUD (AY604543.1) showed that these sequences compared poorly with established AcdS sequences, whereas significant homology was found with sequences of d-cystein desulfhydrases (which, like AcdS, are PLP-dependent enzymes) (Yao et al., 2000; Soutourina et al., 2001). This means that these sequences cannot be considered as acdS sequences.

In contrast, relevant acdS sequences in terms of phylogenic properties were recovered by screening databases in the case of R. sullae HCNT-1 (Hedysarum coronarium isolate from Italy), R. leguminosarum 99A1, B. japonicum USDA110 (soybean nodule isolate from the USA), A. tumefaciens d3 (i.e. genomic group 4; soil isolate from the USA), A. xylosoxidans A551, B. xenovorans LB400 (soil isolate from the USA), B. cepacia R18194 (genomovar unknown) and R1808 (genomovar V; soil isolate from the USA), B. mallei ATCC 23344 (human isolate from Burma), V. paradoxus 2C1, 3C3 and 5C2, P. brassicacearum Am3 (pea or Indian mustard isolate from Russia), Pseudomonas chlororaphis 6G5, P. putida Bm3 (pea or Indian mustard isolate from Russia), P. syringae DC3000 (tomato isolate from the UK), and Pseudomonas sp. strains 17 (soil isolate from South Africa) and ACP.

Three proteobacterial partial sequences of acdS found in GenBank (P. brassicacearum Am3, P. putida Bm3 and V. paradoxus 3C3) were markedly different from all the others based on sequence alignment and phylogenetic analyses (data not shown), for example at distances of at least 1.73 base substitutions per site with the NJ method using K2P (data not shown). Therefore, they were removed from the dataset for acdS phylogenetic analysis (but were included in AcdS studies).

Phylogeny of acdS sequences

Significant polymorphism was found when comparing the acdS sequences obtained in this work or recovered from GenBank, as 227 of 415 sites were informative. Phylogenetic analyses of partial acdS sequences using the NJ method with K2P (Fig. 1a), the NJ method with Poisson correction or JC, or the MP method gave the same overall topology (data not shown). Three main acdS groups were arbitrarily defined for clusters that were highly robust and that contained a significant number of sequences (Fig. 1a). Variovorax paradoxus and R. solanacearum were not included within group I because their position varied somewhat according to the method, and R. sullae and B. japonicum were not included within group III because the distance was too long. The three acdS groups gathered bacteria displaying acdS pairwise distances up to 0.253 (group I), 0.257 (group II) and 0.253 (group III) base substitutions per site within each group.


Figure 1.  Phylogenetic relationship between Proteobacteria based on partial acdS (a) and deduced AcdS (b) sequences. The AcdS tree includes (indicated with a star) Pseudomonas brassicacearum Am3, Pseudomonas putida Bm3 and Variovorax paradoxus 3C3 in addition to sequences used in the acdS tree. The unrooted neighbour-joining phylogenetic trees were inferred using the Kimura 2 parameter (acdS) and Poisson correction (AcdS). Bootstrap values (1000 replicates) are shown when higher than 70%. The scale bar represents the percentage of substitutions per site. Numbers I to III indicate well-resolved clusters arbitrarily defined based on the topology of the tree in (a). AcdS+ bacteria are indicated with a solid square (when the AcdS+ status is a result from the current work) or circle (when a published result), and AcdS bacteria with an empty square.

Download figure to PowerPoint

Groups I and II both gathered sequences originating from the Beta- and Gammaproteobacterial subdivisions studied, whereas the seven sequences in group III corresponded to Alphaproteobacteria (three Rhizobiaceae and four A. lipoferum). The two E. cloacae strains clustered in acdS group I, whereas Burkholderia and Pseudomonas species clustered in groups I and II. Therefore, the phylogenetic relationship derived from acdS analysis was only partly congruent with the one based on 16S rRNA gene sequence analysis (data not shown). In addition, the GC content (%) of the acdS fragment for acdS group I (58.7±1.4; n=11) was statistically different (Kruskal–Wallis and Wilcoxon tests; P<0.005) from the one for acdS group II (65.3±1.8; n=13) or III (64.1±3.0; n=7).

Geographic distribution and habitat origin of acdS+Proteobacteria in relation to acdS groups

Each of the three acdS groups comprised strains originating (i) from several countries in different continents, and (ii) from different microbial habitats (Table 3). For instance, acdS group I included isolates from Europe (Ireland, UK, Switzerland, Italy, the Czech Republic), North America (Canada, USA) and South Africa, recovered from monocots, dicots or soil. Moreover, several geographic locations and host plants studied yielded isolates from more than one acdS group. However, all acdS+ clinical isolates analysed belonged to acdS group II.

Table 3.   Geographic (A) and habitat (B) distributions across acdS phylogenetic groups I, II and III of proteobacterial strains for which an acdS sequence is known
AGeographic origin*
Irl and UK (n=5)W Europe (n=15)C Europe (n=3)Russia (n=3)SE Asia (n=5)Pakistan (n=9)Africa (n=2)N America (n=17)C and S America (n=8)Unknown (n=10)
acdS I3311   2 1
acdS II121 1  413
acdS III 1  1211 2
Others 1 2   114
BHost plantBulk soil (n=15)Clinical isolates (n=6)
Onion (n=1)Corn (n=5)Rice (n=12)Wheat (n=7)Cotton (n=2)Cucumber (n=3)Pea (n=2)Sugarbeet (n=1)Tomato (n=5)Unknown (n=10)
  • *

    Irl, Ireland; W Europe, western Europe (France, Belgium, Switzerland and Italy); C Europe, central Europe (Germany and Czech Republic); SE Asia, south-eastern Asia (Burma, Vietnam, Bangladesh and Bhutan), N America, northern America (Canada and USA); C and S America, central and southern America (Mexico, Cuba, Martinique, French Guyana, Brazil). Strains from Israel (n=1), Iraq (n=1), Japan (n=5), Australia (n=1) were tested but none yielded an acdS sequence.

  • Number of strains studied by acdS PCR or for which an acdS sequence was already available.

  • Strains from sorghum (n=1), Digitaria (n=1), tobacco (n=1), Cynodon (n=1), apple tree (n=2), dahlia (n=1), soybean (n=1), Hedysarum (n=1), Flacourtia (n=1), cherry tree (n=1), Rubus (n=1), Dianthus (n=1), Gladiolus (n=1) were tested but none yielded an acdS sequence.

acdS I   1 2 1215 
acdS II12 1     234
acdS III  2 2 1  11 
Others      1 14  

Phylogenetic analysis of deduced AcdS sequences

The phylogenetic tree derived from NJ analysis (using the number of differences) of deduced AcdS sequences is shown in Fig. 1b (61 of 137 sites were informative). Only minor topological differences were found when comparing with trees obtained using the NJ (with Poisson correction) or MP methods (data not shown). acdS and AcdS phylogenies were highly congruent when considering the Proteobacteria used in Fig. 1a, regardless of the method, and it appeared that the acdS-based groups I, II and III were also relevant with AcdS sequences. The three strains whose acdS sequence was very different from all the others (and therefore were not included in the acdS trees) clustered with acdS group I strains (i.e. for P. brassicacearum Am3), with the two V. paradoxus (i.e. for V. paradoxus 3C3), or with the two A. xylosoxidans (i.e. for P. putida Bm3) in the AcdS trees (see e.g. Fig. 1b).

Analysis of AcdS domains and residues based on deduced proteic sequences

Alignment of complete AcdS sequences (338 or 339 amino acids) for three strains encompassing acdS phylogenetic groups I (E. cloacae UW4), II (Pseudomonas sp. ACP) and III (R. leguminosarum 128C53K) showed a high level of conservation (Fig. 2). Identity levels ranged between 64 and 81%, and similarity levels between 79 and 87%. Structural analysis of the dimeric AcdS protein in H. saturnus (Yao et al., 2000) evidenced a PLP-binding domain (domain I) and an internal, smaller domain (domain II), which were readily identified in the bacterial sequences studied (Fig. 2).


Figure 2.  Comparison of complete AcdS sequences deduced for Enterobacter cloacae UW4 (acdS phylogenetic group I), Pseudomonas sp. ACP (group II) and Rhizobium leguminosarum 128C53K (group III). Conserved residues are boxed and similar residues indicated in grey. The α-helices and β-strands described in AcdS from the yeast Hansenula saturnus are indicated. Residues are numbered as in Pseudomonas sp. ACP. The lysine residue serving as pyridoxal 5′-phosphate binding site (i.e. K51) is indicated with a star. The sequence recovered by PCR is between positions 50 and 314.

Download figure to PowerPoint

Many residues considered to play a key role in protein conformation and functioning were highly conserved, such as the lysine residue K51. K51 is the PLP-binding site in H. saturnus (Yao et al., 2000) and is required for enzymatic activity in P. fluorescens ACP (Murakami et al., 1997), probably because it functions as the base for proton abstraction (Yao et al., 2000). Other residues (numbered as in Pseudomonas sp. ACP) important for conformation and functioning of AcdS in the yeast H. saturnus (Yao et al., 2000) and found in the three sequences included (i) four residue pairs implicated in monomer−monomer contact via hydrogen bonds, i.e. R23–A89, R38–G44, A89–E286 and R112–S332; (ii) four of 11 pairs of van der Waals contacts, i.e. F13–P17, L45–E286, A89–E286 and I121–F335; (iii) four aromatic (W102, Y110, Y268 and Y294) and three hydrophobic (V103, I137, and V293) residues forming two cavities involved in substrate/product transportation; (iv) E295 (specific of AcdS enzymes), which is within hydrogen bonding distance of the N1 nitrogen (pyridinium N) of the cofactor pyridine ring and stabilizes the PLP; (v) five residues (K54, V198, T199, G200 and T202) forming a pocket to which the phosphate group of PLP is tightly fixed; (vi) the tyrosine residue Y294, thought to contribute to hydrogen binding with the substrate (Yao et al., 2000); (vii) the serine residue S78, thought to act as nucleophile in the enzymatic attack (Yao et al., 2000); and (viii) T81, which is hydrogen-bonded to S78. The residue in position 162, which is important for enzymatic activity in Pseudomonas sp. ACP (Honma, 1985), is occupied by cysteine or alanine in PLP-dependent enzymes (e.g. C162 in AcdS from H. saturnus), which was the case for the three bacterial sequences studied. Therefore, it appears that functional traits of AcdS were highly conserved across the three acdS groups.

Analysis of acdS by colony hybridization in Proteobacteria

A total of 69 Proteobacteria were tested by acdS hybridization, performed using three different probes, for a more extensive assessment of the presence of the gene in the collection studied. Different hybridization results were obtained depending on the probe used. Briefly, when considering strains for which an acdS sequence is available, it appears that (i) the probes from P. fluorescens PITR2 and B. caledonica LMG19076 (acdS group I) hybridized with strains from acdS groups I and II; and (ii) the probe from A. lipoferum CN1 (acdS group III) hybridized with strains from acdS groups II and III.

Overall, a positive hybridization signal with at least one of the probes was obtained with 22 of 37 Alphaproteobacteria (i.e. 59%), 15 of 19 Betaproteobacteria (i.e. 79%) and 9 of 12 Gammaproteobacteria strains (i.e. 75%) (Table 1). Hybridization failed to identify acdS in a few strains in which an acdS sequence was obtained (i.e. R. leguminosarum 128C53 K and B. phenazinium LMG2247). More importantly, however, hybridization (especially with the P. fluorescens probe) evidenced acdS in additional strains (A. lipoferum CRT1 and B518, and A. brasilense L4, PH1, Sp245, Sp7, Cd and ZN1), genomovars (Agrobacterium genomic groups 1, 6, 7, 8 and 9, and B. cepacia genomovars II and V) and species (the Azospirillum species brasilense, and the Burkholderia species caryophylli, kururiensis, gladioli and pyrrocinia) when compared with acdS sequence results.

Analysis of ACC deaminase activity

In this work, 24 representative strains belonging to all proteobacterial taxa studied were selected to quantify ACC deaminase activity. Results indicated that (i) 11 of the 24 bacteria showed ACC deaminase activity in vitro; and (ii) levels of specific activity could differ statistically from one AcdS+ strain to the next, including within a particular species (Fig. 3). As expected, the positive control P. fluorescens PITR2 displayed ACC deaminase activity and the negative control P. fluorescens CHA0 did not. In A. lipoferum, five of six strains tested showed ACC deaminase activity, at levels ranging from 95.1±1.3 to 101.1±0.6 nmol α-ketobutyrate (mg protein)−1 h−1 (Fig. 3). Several acdS+A. brasilense were studied, but ACC deamination in vitro was not found. ACC deaminase activity was also evidenced in Burkholderia species and R. solanacearum. The prevalence of strains showing ACC deaminase activity in vitro was 80% in acdS phylogenic group I, 50% in group II and 100% in group III.


Figure 3.  Specific 1-aminocyclopropane-1-carboxylic acid deaminase activity of selected strains of Pseudomonas fluorescens (PITR2 and CHA0), Azospirillum lipoferum (4B, CN1, N4, CRT1 and TVV3), Burkholderia tropicalis (BM273), Burkholderia cepacia genomovar I (LMG1222) and VI (LMG18941), Burkholderia phenazinium (LMG2247) and Ralstonia solanacearum (GMI1000). One unit corresponds to 1 nmol of α-ketobutyrate formed per milligram protein per hour (Honma & Shimomura, 1978). Bars represent standard deviations. Letters a to i are used to show the statistical relationship between treatments based on analysis of variance and Fisher's LSD tests (P<0.05).

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Understanding the genetic basis differentiating bacteria of contrasting ecology (e.g. pathogenic vs. beneficial bacteria) is a key issue. In recent years, features such as type-three protein secretion genes or genetic islands, thought to be specific to pathogens, have also been found in non-pathogens, including symbiotic bacteria (Ochman & Moran, 2001; Mazurier et al., 2004; Rezzonico et al., 2004). Similarly, certain genes often associated with beneficial interactions (e.g. nifU and nifS) may be important also for pathogens (Olson et al., 2000), but this possibility has received less attention in comparison.

In this context, we focused on ACC deaminase activity, a plant-beneficial property extensively surveyed by phenotypic means in several (but not all) key PGPR and phytosymbiotic bacteria (Glick et al., 1995; Wang et al., 2001; Penrose & Glick, 2003; Ma et al., 2003a, b), and whose assessment in other types of bacteria remained to be carried out. Our results indicate that ACC deaminase activity was not detected in vitro for several acdS+Proteobacteria, which means that these bacteria could not have been identified with the phenotype-based traditional methods used for screening strain collections (e.g. Wang et al., 2001). Therefore, the DNA approaches developed in the current work were useful to extend our knowledge on the distribution of acdS in Proteobacteria. PCR amplification of acdS was not as efficient as hybridization to detect acdS+Proteobacteria (Table 1). This is not surprising, and can be explained by variations in the short DNA sequences of the priming sites (data not shown). In addition, in a few cases, the PCR approach yielded nonspecific bands, or bands of the expected size but for which all sequencing attempts failed. Here, presence of the gene was established (i) in a range of pathogenic bacteria (i.e. certain Burkholderia species including several cepacia genomovars, and five Agrobacterium genomovars), and (ii) in a key PGPR genus believed so far to be devoid of ACC deaminase activity (i.e. Azospirillum) (Table 1).

First, acdS is present in several pathogenic Proteobacteria, and therefore the gene is not restricted to plant-beneficial taxa. acdS was evidenced in phytopathogens such as the Gladiolus pathogen B. gladioli, the onion pathogen B. cepacia genomovar I, the carnation pathogen B. caryophylli, and Agrobacterium strains (all of them harboring a pTi virulence plasmid) (Gelvin et al., 2000) belonging to genomovars 1, 4, and 6−9. In addition, the gene is also present in opportunistic human pathogens isolated from cystic fibrosis patients and belonging to B. cepacia genomovars II, III, IV and VI, such as the genomovar III strain LMG16656 implicated in the death of several patients (Govan et al., 1993), as well as in the true pathogen B. mallei (Nierman et al., 2004). ACC deaminase activity being a plant-beneficial trait, the occurrence of acdS (and sometimes ACC deaminase activity) in true or opportunistic pathogens is striking and raises the question of its contribution to pathogenesis, which may be especially relevant when dealing with phytopathogens. In addition, it questions the potential role of this trait for the survival of true and opportunistic human pathogens in terrestrial and/or non-terrestrial environments, because ACC deaminase activity may be important for nitrogen nutrition in microbial habitats containing ACC (Glick et al., 1998). For instance, strains belonging to B. cepacia genomovar III are found in the rhizosphere of corn (Balandreau et al., 2001), but it is not known whether the same strain can colonize both the rhizosphere and cystic fibrosis patients. The significance of acdS and/or ACC deaminase activity for survival in non-terrestrial environments is not obvious, because to our knowledge the occurrence of ACC is not documented in animal or human hosts.

Second, in the PGPR taxa Azospirillum, acdS was found in two of three species tested, and several acdS+Azospirillum strains displayed ACC deaminase activity in vitro (Table 1). Holguin & Glick (2001) concluded that ACC deaminase activity was absent from A. brasilense Cd and Sp245, which is in accordance with the current phenotypic results obtained in vitro, but nevertheless acdS was discovered here in these and other A. brasilense strains (some of the latter actually displaying enzymatic activity). So far, the phytobeneficial effets of Azospirillum PGPR strains have been attributed to nitrogen fixation (Bally & Elmerich, 2005) and phytohormone production (Dobbelaere et al., 2001), and in certain cases biocontrol of phytopathogenic bacteria (Bashan & de-Bashan, 2002) or phytoparasitic plants (Michéet al., 2000). In light of the current results, this will need to be reassessed by including the contribution of ACC deaminase activity.

Phylogenetic analysis evidenced three main acdS phylogenetic groups. BLASTP comparison of deduced AcdS sequences within each of the groups failed to distinguish conserved domains from variable regions (data not shown). Indeed, polymorphism was observed throughout the 264-amino acid sequence, even though key residues, for example K51, S78, T81, C162/A162, Y294 and E295 (see Fig. 2), were conserved. Therefore, information is not available to suggest that AcdS sequence polymorphism has the potential to modulate enzymatic functioning in a group-dependent manner.

Membership of proteobacterial strains to a particular group did not correlate with a particular geographic or habitat origin (Table 3). For instance, strains from a given acdS group originated from different plant hosts, similarly to findings derived from the study of other genes implicated in plant−microbe interactions (Wang et al., 2001; Ramette et al., 2003; Rezzonico et al., 2004). All opportunistic human pathogens were found in a single Burkholderia cluster within acdS group II. This observation will need confirmation using a larger number of sequences, as here an acdS sequence was available for only three opportunistic human pathogens. When analysing the complete dataset, however, it appeared that there was no obvious relation with the ecological strategy of the bacteria, as each acdS group included both pathogenic and plant-beneficial bacteria. This means that acdS alleles will not be useful as molecular markers to distinguish pathogens from nonpathogenic bacteria.

Each of the acdS phylogenetic groups I and II gathered strains belonging to the Beta-and Gammaproteobacterial subdivisions, whereas group III was composed of Alphaproteobacteria (Fig. 1). Therefore, although the strain collection studied did not cover the entire proteobacterial division, it is clear that the relationship with the rrs phylogeny was not extensive in comparison with congruence levels often found when studying bacterial genes important for interaction with a eukaryotic host (Ramette et al., 2003; Lerat & Moran, 2004; Mazurier et al., 2004; Rezzonico et al., 2004). This is probably a result of the past occurrence of extensive horizontal gene transfers. This hypothesis is supported by the observations that acdS is plasmid-borne in R. leguminosarum (Ma et al., 2003a) and is located within a symbiotic island in R. loti (Sullivan et al., 2002), genetic islands being prone to such transfers.

Insight into the consequence of a horizontal gene transfer of acdS can be gained from experiments in which acdS from E. cloacae UW4 was intentionally introduced into A. brasilense Cd and Sp245. Indeed, the introduced acdS gene was readily expressed in the transconjugants and resulted in enhanced phytobeneficial effects (Holguin & Glick, 2001, 2003). Similarly, introduction of the gene into the acdS indoleacetic acid-producing strain P. fluorescens CHA0 increased the ability of the latter to protect potato tubers from E. carotovora-mediated soft rot (in certain but not all experimental conditions) and cucumber from Pythium damping-off, but not tomato from Fusarium crown and root rot (Wang et al., 2000). In addition, it conferred direct plant-growth promotion ability on canola cultured in artificial soil. These experiments illustrate how transferred acdS can benefit the plant, and perhaps this may in turn benefit acdS+ root-colonizing bacteria via increased root exudation.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This work was supported in part by the Ministère de la Recherche, Paris, France. A postdoctoral fellowship awarded to M.S.M. by the Higher Education Commission, Pakistan is gratefully acknowledged. We are grateful to A. Menard, X. Nesme and P. Portier (UMR CNRS 5557 Ecologie Microbienne, Lyon 1) for Burkholderia and Agrobacterium strains; B. Glick (University of Waterloo, Waterloo, Ontario) for E. cloacae UW4 and CAL2, K. ascorbata SUD 65 and R. leguminosarum 128C53K; A. Belimov (University of Saint Petersburg, Saint Petersburg, Russia) for A. xylosoxidans Cm4; and G. Défago (ETH, Zürich, Switzerland) for Pseudomonas strains CHA0, TM1A3, CM1′A2, PITR2, P97.30, K94.31, F113 and P97.38. We thank V. Segealet, D. Mourier des Gayets, J. Morel and O. Crassard for technical assistance, L. Jocteur Monrozier and A. Dechesne (UMR CNRS 5557 Ecologie Microbienne, Lyon 1) for advice on statistical analysis, and J. Balandreau, X. Nesme and R. Bally (UMR CNRS 5557 Ecologie Microbienne, Lyon 1) and A. Ramette (Michigan State University, East Lansing, Michigan) for useful discussions.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Achouak W, Christen R, Barakat M, Martel MH & Heulin T (1999) Burkholderia caribensis sp. nov., an exopolysaccharide-producing bacterium isolated from vertisol microaggregates in Martinique. Int J Syst Bacteriol 49: 787794.
  • Altschul SF, Gish W, Miller W, Myers EW & Lipman DJ (1990) Basic local alignment search tool. J Mol Biol 215: 403410.
  • Babalola OO, Osir EO, Sanni AI, Odhiambo GD & Bulimo WD (2003) Amplification of 1-amino-cyclopropane-1-carboxylic (ACC) deaminase from plant growth promoting rhizobacteria in Striga-infested soil. Afr J Biotechnol 2: 157160.
  • Balandreau J, Viallard V, Cournoyer B, Coenye T, Laevens S & Vandamme P (2001) Burkholderia cepacia genomovar III is a common plant-associated bacterium. Appl Environ Microbiol 67: 982985.
  • Bally R & Elmerich C (2005) Biocontrol of plant diseases by associative and endophytic nitrogen-fixing bacteria. Associative and Endophytic Nitrogen-Fixing Bacteria and Cyanobacterial Associations, (ElmerichC & NewtonWE, eds), pp. 171190. Kluwer Academic Publishers, Dordrecht (in press).
  • Bally R, Thomas-Bauzon D, Heulin T, Balandreau J, Richard C & De Ley J (1983) Determination of the most frequent N2-fixing bacteria in the rice rhizosphere. Can J Microbiol 29: 881887.
  • Bashan Y & De-Bashan LE (2002) Protection of tomato seedlings against infection by Pseudomonas syringae pv. tomato by using the plant growth-promoting bacterium Azospirillum brasilense. Appl Environ Microbiol 68: 26372643.
  • Belimov AA, Safronova VI, Sergeyeva TA, et al. (2001) Characterization of plant growth promoting rhizobacteria isolated from polluted soils and containing 1-aminocyclopropane-1-carboxylate deaminase. Can J Microbiol 47: 642652.
  • Beringer JE (1974) R-factor transfer in Rhizobium leguminosarum. J Gen Microbiol 84: 188198.
  • Berry V & Gascuel O (1996) Interpretation of bootstrap trees: threshold of clade selection and induced gain. Mol Biol Evol 13: 9991011.
  • Boucher CA, Barberis PA, Trigalet AP & Demery DA (1985) Transposon mutagenesis of Pseudomonas solanacearum: isolation of Tn5-induced avirulent mutants. J Gen Microbiol 13: 24492457.
  • Bruce KD, Hiorns WD, Hobman JL, Osborn AM, Strike P & Ritchie DA (1992) Amplification of DNA from native populations of soil bacteria by using the polymerase chain reaction. Appl Environ Microbiol 58: 34133416.
  • Burbage DA & Sasser M (1982) A medium selective for Pseudomonas cepacia. Phytopathology 72: 706.
  • Burd GI, Dixon DG & Glick BR (1998) A plant growth-promoting bacterium that decreases nickel toxicity in seedlings. Appl Environ Microbiol 64: 36633668.
  • Coenye T, Laevens S, Willems A, Ohlen M, Hannant W, Govan JR, Gillis M, Falsen E & Vandamme P (2001a) Burkholderia fungorum sp. nov. and Burkholderia caledonica sp. nov., two new species isolated from the environment, animals and human clinical samples. Int J Syst Evol Microbiol 51: 10991107.
  • Coenye T, LiPuma JJ, Henry D, Hoste B, Vandemeulebroecke K, Gillis M, Speert DP & Vandamme P (2001b) Burkholderia cepacia genomovar VI, a new member of the Burkholderia cepacia complex isolated from cystic fibrosis patients. Int J Syst Evol Microbiol 51: 271279.
  • Devereux J, Haeberli P & Smithies O (1984) A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 12: 387395.
  • Dobbelaere S, Croonenborghs A, Thys A, et al. (2001) Responses of agronomically important crops to inoculation with Azospirillum. Austr J Plant Physiol 28: 871879.
  • Elbeltagy A, Nishioka K, Sato T, Suzuki H, Ye B, Hamada T, Isawa T, Mitsui H & Minamisawa K (2001) Endophytic colonization and in planta nitrogen fixation by a Herbaspirillum sp. isolated from wild rice species. Appl Environ Microbiol 67: 52855293.
  • Eskew DL, Focht DD & Ting P (1977) Nitrogen fixation, denitrification and pleomorphic growth in a highly pigmented Spirillum lipoferum. Appl Environ Microb 34: 582585.
  • Estrada P, Mavingui P, Cournoyer B, Fontaine F, Balandreau J & Caballero-Mellado J (2002) A N2-fixing endophytic Burkholderia sp. associated with maize plants cultivated in Mexico. Can J Microbiol 48: 285294.
  • Fages J & Mulard D (1988) Isolement de bactéries rhizosphériques et effet de leur inoculation en pots chez Zea mays. Agronomie 8: 309314.
  • Fenton AM, Stephens PM, Crowley J, O'Callaghan M & O'Gara F (1992) Exploitation of gene(s) involved in 2,4-diacetylphloroglucinol biosynthesis to confer a new biocontrol capability to a Pseudomonas strain. Appl Environ Microbiol 58: 38733878.
  • Fuchs J & Défago G (1991) Protection of cucumber plants against black root rot caused by Phomopsis sclerotioides with rhizobacteria. Plant Growth-Promoting Rhizobacteria – Progress and Prospects (IOBC/WPRS Bull XIV/8), (KeelC, KollerB & DéfagoG, eds), pp. 5762. IOBC/WPRS, Interlaken, Switzerland.
  • Galtier N, Gouy M & Gautier C (1996) SEAVIEW and PHYLO_WIN: two graphic tools for sequence alignment and molecular phylogeny. Comput Appl Biosci 12: 543548.
  • Gelvin SB (2000) Agrobacterium and plant genes involved in T-DNA transfer and integration. Annu Rev Plant Physiol Plant Mol Biol 51: 223256.
  • Gillis M, Van Tran V, Bardin R, Goor M, Hebbar P, Willems A, Segers P, Kersters K, Heulin T & Fernandez MP (1995) Polyphasic taxonomy in the genus Burkholderia leading to an amended description of the genus and proposition of Burkholderia vietnamiensis sp. nov. for N2-fixing isolates from rice in Vietnam. Int J Syst Bacteriol 45: 274289.
  • Glick BR, Jacobson CB, Schwarze MMK & Pasternak JJ (1994) 1-aminocyclopropane-1-carboxylic acid deaminase mutants of the plant growth promoting rhizobacterium Pseudomonas putida GR12-2 do not stimulate canola root elongation. Can J Microbiol 40: 911915.
  • Glick BR, Karaturovíc DM & Newell PC (1995) A novel procedure for rapid isolation of plant growth-promoting pseudomonads. Can J Microbiol 41: 533536.
  • Glick BR, Penrose DM & Li J (1998) A model for the lowering of plant ethylene concentrations by plant growth-promoting bacteria. J Theor Biol 190: 6368.
  • Govan JRW, Brown PH, Maddison J, Doherty CJ, Nelson JW, Dodd M, Greening AP & Webb AK (1993) Evidence for transmission of Pseudomonas cepacia by social contact in cystic fibrosis. Lancet 342: 1519.
  • Haas D & Keel C (2003) Regulation of antibiotic production in root-colonizing Pseudomonas spp. and relevance for biological control of plant disease. Annu Rev Phytopathol 41: 117153.
  • Holguin G & Glick BR (2001) Expression of the ACC deaminase gene from Enterobacter cloacae UW4 in Azospirillum brasilense. Microb Ecol 41: 281288.
  • Holguin G & Glick BR (2003) Transformation of Azospirillum brasilense Cd with an ACC deaminase gene from Enterobacter cloacae UW4 fused to the Tetr gene promoter improves its fitness and plant growth promoting ability. Microb Ecol 46: 122133.
  • Honma M (1985) Chemically reactive sulfhydryl groups of 1-aminocyclopropane-1-carboxylate deaminase. Agric Biol Chem 49: 567571.
  • Honma M & Shimomura T (1978) Metabolism of l-aminocyclopropane-1-carboxylic acid. Agric Biol Chem 42: 18251831.
  • Jia YJ, Ito H, Matsui H & Honma M (2000) 1-aminocyclopropane-1-carboxylate (ACC) deaminase induced by ACC synthesized and accumulated in Penicillium citrinum intracellular spaces. Biosci Biotechnol Biochem 64: 299305.
  • Kabir M, Faure D, Heulin T, Achouak W & Bally R (1996) Azospirillum populations in soils infested by a parasitic weed (Striga) under Sorghum cultivation in Mali, West Africa. Eur J Soil Biol 32: 157163.
  • Keane PJ, Kerr A & New PB (1970) Crown gall of stone fruit. II. Identification and nomenclature of Agrobacterium isolates. Aust J Biol Sci 23: 585595.
  • Keel C, Weller DM, Natsch A, Défago G, Cook RJ & Thomashow LS (1996) Conservation of the 2,4-diacetylphloroglucinol biosynthesis locus among fluorescent Pseudomonas strains from diverse geographic locations. Appl Environ Microbiol 62: 552563.
  • Khammas KM, Ageron E, Grimont PA & Kaiser P (1989) Azospirillum irakense sp. nov., a nitrogen-fixing bacterium associated with rice roots and rhizosphere soil. Res Microbiol 140: 679693.
  • King EO, Ward MK & Raney DE (1954) Two simple media for the demonstration of pyocyanin and fluorescin. J Lab Clin Med 44: 301307.
  • Kiredjian M (1979) Le genre Agrobacterium peut-il être pathogène pour l'homme? Med Malad Infect 9: 223235.
  • Kumar S, Tamura K, Jakobsen IB & Nei M (2001) MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17: 12441245.
  • Lavire L (1998) Variation de phase chez Azospirillum: un mécanisme d'adaptation à la rhizosphère? DEA thesis, Université Claude Bernard Lyon1, Villeurbanne, France.
  • Lerat E & Moran NA (2004) The evolutionary history of quorum-sensing systems in bacteria. Mol Biol Evol 21: 903913.
  • Li J, Ovakim DH, Charles TC & Glick BR (2000) An ACC deaminase minus mutant of Enterobacter cloacae UW4 no longer promotes root elongation. Curr Microbiol 41: 101105.
  • Ma W, Guinel FC & Glick BR (2003a) Rhizobium leguminosarum biovar viciae 1-aminocyclopropane-1-carboxylate deaminase promotes nodulation of pea plants. Appl Environ Microbiol 69: 43964402.
  • Ma W, Sebestianova SB, Sebestian J, Burd GI, Guinel FC & Glick BR (2003b) Prevalence of 1-aminocyclopropane-1-carboxylate deaminase in Rhizobium spp. Antonie van Leeuwenhoek 83: 285291.
  • Mazurier S, Lemunier M, Siblot S, Mougel C & Lemanceau P (2004) Distribution and diversity of type III secretion system-like genes in saprophytic and phytopathogenic fluorescent pseudomonads. FEMS Microbiol Ecol 49: 455467.
  • Miché L, Bouillant ML, Rohr R & Bally R (2000) Physiological and cytological studies of the inhibitory effect of soil bacteria of the genus Azospirillum on striga seeds germination. Eur J Plant Pathol 106: 347351.
  • Minami R, Uchiyama K, Murakami T, Kawai J, Mikami K, Yamada T, Yokoi D, Ito H, Matsui H & Honma M (1998) Properties, sequence, and synthesis in Escherichia coli of 1-aminocyclopropane-1-carboxylate deaminase from Hansenula saturnus. J Biochem (Tokyo) 123: 11121118.
  • Moënne-Loccoz Y & Défago G (2004) Life as a biocontrol pseudomonad, Pseudomonas. Genomics, Life Style and Molecular Architecture, Vol. 1, (RamosJL, ed.), pp. 457476. Kluwer Academic/Plenum Publishers, New York.
  • Mougel C, Thioulouse J, Perrière G & Nesme X (2002) A mathematical method for determining genome divergence and species delineation using AFLP. Int J Syst Evol Microbiol 52: 573586.
  • Murakami T, Kiuchi M, Ito H, Matsui H & Honma M (1997) Substitutions of alanine for cysteine at a reactive thiol site and for lysine at a pyridoxal phosphate binding site of 1-aminocyclopropane-1-carboxylate deaminase. Biosci Biotechnol Biochem 61: 506509.
  • Nelson LM & Knowles R (1978) Effect of oxygen and nitrate on nitrogen fixation and denitrification by Azospirillumbrasilense grown in continuous culture. Can J Microbiol 24: 13951403.
  • Nierman WC, DeShazer D, Kim HS, et al. (2004) Structural flexibility in the Burkholderia mallei genome. Proc Natl Acad Sci USA 101: 1424614251.
  • Ochman H & Moran NA (2001) Genes lost and genes found: evolution of bacterial pathogenesis and symbiosis. Science 292: 10961098.
  • Olson JW, Agar JN, Johnson MK & Maier RJ (2000) Characterization of the NifU and NifS Fe-S cluster formation proteins essential for viability in Helicobacter pylori. Biochemistry 39: 1621316219.
  • Penot I, Bergès N, Guinguené C & Fages J (1992) Characterization of Azospirillum associated with maize (Zea mays) in France using biochemical tests and plasmid profiles. Can J Microbiol 38: 798803.
  • Penrose DM & Glick BR (2003) Methods for isolating and characterizing ACC deaminase-containing plant growth-promoting rhizobacteria. Physiologia Plantarum 118: 1015.
  • Ramette A, Frapolli M, Défago G & Moënne-Loccoz Y (2003) Phylogeny of HCN synthase-encoding hcnBC genes in biocontrol fluorescent pseudomonads and its relationship with host plant species and HCN synthesis ability. Mol Plant–Microbe Interact 16: 525535.
  • Rezzonico F, Défago G & Moënne-Loccoz Y. (2004) Comparison of ATPase-encoding type III secretion system hrcN genes in biocontrol fluorescent pseudomonads and in phytopathogenic proteobacteria. Appl Environ Microbiol 70: 51195131.
  • Rhaman M (1987) Amelioration de la fixation d'azoste dans la rhizosphère du riz cultivé sur différents sols du Bengladesh. PhD thesis, Université Claude Bernard Lyon 1, Villeurbanne, France.
  • Rinaudo G (1982) Fixation hétérotrophe de l'azote dans la rhizosphère du riz. Thèse de Doctorat d'Etat, Université Paris-Sud, Orsay, France.
  • Saitou N & Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4: 406425.
  • Soutourina J, Blanquet S & Plateau P (2001) Role of d-cysteine desulfhydrase in the adaptation of Escherichia coli to d-cysteine. J Biol Chem 276: 4086440872.
  • Stutz EW, Défago G & Kern H (1986) Naturally occurring fluorescent pseudomonads involved in suppression of black root rot of tobacco. Phytopathology 76: 181185.
  • Sullivan JT, Trzebiatowski JR, Cruickshank RW, et al. (2002) Comparative sequence analysis of the symbiosis island of Mesorhizobium loti strain R7A. J Bacteriol 184: 30863095.
  • Tarrand JJ, Krieg NR & Döbereiner J (1978) A taxonomic study of the Spirillum lipoferum group, with descriptions of a new genus, Azospirillum gen. nov. and two species, Azospirillum lipoferum (Beijerinck) comb. nov. and Azospirillum brasilense sp. nov. Can J Microbiol 24: 967980.
  • Thompson JD, Higgins DG & Gibson TJ (1994) Clustal W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 46734680.
  • Van Tran T, Ngoke S, Berge O, Faure D, Bally R, Hebbar P & Heulin T (1997) Isolation of Azospirillumlipoferum from the rhizosphere of rice by a new simple method. Can J Microbiol 43: 486490.
  • Urakami T, Ito-Yoshida C, Araki H, Kijima T, Suzuki K & Komagata K (1994) Transfer of Pseudomonas plantarii and Pseudomonas glumae to Burkholderia as Burkholderia spp. and description of Burkholderia vandii sp. nov. Int J Syst Bacteriol 44: 235245.
  • Vandamme P, Holmes B, Vancanneyt M, et al. (1997) Occurrence of multiple genomovars of Burkholderia cepacia in cystic fibrosis patients and proposal of Burkholderia multivorans sp. nov. Int J Syst Bacteriol 47: 11881200.
  • Vandamme P, Mahenthiralingam E, Holmes B, Coenye T, Hoste B, De Vos P, Henry D & Speert DP (2000) Identification and population structure of Burkholderia stabilis sp. nov. (formerly Burkholderia cepacia genomovar IV). J Clin Microbiol 38: 10421047.
  • Viallard V, Poirier I, Cournoyer B, Haurat J, Wiebkin S, Ophel-Keller K & Balandreau J (1998) Burkholderia graminis sp. nov., a rhizospheric Burkholderia species, and reassessment of [Pseudomonas] phenazinium, [Pseudomonas] pyrrocinia and [Pseudomonas] glathei as Burkholderia. Int J Syst Bacteriol 48: 549563.
  • Wang C, Knill E, Glick BR & Défago G (2000) Effect of transferring 1-aminocyclopropane-1-carboxylic acid (ACC) deaminase genes into Pseudomonas fluorescens strain CHA0 and its gacA derivative CHA96 on their growth-promoting and disease-suppressive capacities. Can J Microbiol 46: 898907.
  • Wang C, Ramette A, Punjasamarnwong P, Zala M, Natsch A, Moënne-Loccoz Y & Défago G (2001) Cosmopolitan distribution of phlD-containing dicotyledonous crop-associated biocontrol pseudomonads of worldwide origin. FEMS Microbiol Ecol 37: 105116.
  • Yabuuchi E, Kosako Y, Oyaizu H, Yano I, Hotta H, Hashimoto Y, Ezaki T & Arakawa M (1992) Proposal of Burkholderia gen. nov. and transfer of seven species of the genus Pseudomonas homology group II to the new genus, with the type species Burkholderiacepacia (Palleroni and Holmes 1981) comb. nov. Microbiol Immunol 36: 12511275.
  • 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: 115120.
  • Yao M, Ose T, Sugimoto H, Horiuchi A, Nakagawa A, Wakatsuki S, Yokoi D, Murakami T, Honma M & Tanaka I (2000) Crystal structure of 1-aminocyclopropane-1-carboxylate deaminase from Hansenula saturnus. J Biol Chem 44: 3455734565.
  • Zhang H, Hanada S, Shigematsu T, Shibuya K, Kamagata Y, Kanagawa T & Kurane R (2000) Burkholderia kururiensis sp. nov., a trichloroethylene (TCE)-degrading bacterium isolated from an aquifer polluted with TCE. Int J Syst Evol Microbiol 50: 743749.