Phenotypic, pathogenic, molecular and phylogenetic comparisons of bacteria causing Aloe rot from three countries

Transcription

1 From the SelectedWorks of Kunal Mandal 2011 Phenotypic, pathogenic, molecular and phylogenetic comparisons of bacteria causing Aloe rot from three countries Yogeshwar Kumar Jatindra Nath Samanta Kunal Mandal Narendra A. Gajbhiye Available at:

2 Indian Phytopath. 64 (4) : (2011) RESEARCH ARTICLE Phenotypic, pathogenic, molecular and phylogenetic comparisons of bacteria causing Aloe rot from three countries YOGESHWAR KUMAR, JATINDRA NATH SAMANTA, KUNAL MANDAL* AND N.A. GAJBHIYE Directorate of Medicinal and Aromatic Plants Research, Boriavi, Anand ABSTRACT: Bacterial soft rot disease of Aloe caused by heterogeneous genus Erwinia was reported from different parts of the world. In the recent past the genus underwent major taxonomic modifications. In the present study, four Aloe pathogenic bacterial strains isolated from India, The Netherlands and Yugoslavia have been compared. Different cultural, biochemical, physiological and pathological characters and protein/lipid profiles indicated that the strains belonged to two different genera, Pectobacterium and Dickeya. Species specific amplification of pel gene sequences also supported this. Phylogenetic analysis of rrna gene (rdna) sequence placed these Dickeya strains close to D. dieffenbachiae and D. zeae. On the other hand, phenotypic tests suggested these to be either of D. dadanti, D. dieffenbachiae and D. zeae. On the basis of pathogenicity of the test strains on Dieffenbachia leaves, these were placed in D. dieffenbachiae. The Yugoslavian strain was identified as P. carotovorum. Phenotypic tests, RFLP analyses deciphered variation among the same species. However, none of the six Aloe species tested in the study showed resistance against any of the evaluated bacterial strains. Key words: Enteric bacteria, soft rot, taxonomy, variability, xerophytes Bacterial soft rot is one of the devastating diseases worldwide, especially for the herbaceous plants. It is caused by a group of pectolytic bacteria traditionally grouped under the genus Erwinia. The genus was established to include the phytopathogenic enterobacteria containing gramnegative non-spore forming rod shaped members with peritrichous flagella and facultative anaerobic metabolism (Winslow et al., 1917). These soft rot bacteria were placed at either of the species E. chrysanthemi or E. carotovora. Later, several biochemical and molecular studies showed that the group contains heterogeneous organisms and called for species delineation. A new genus, Pectobacterium was proposed to accommodate soft rot Erwinias (Waldee, 1945) however, published literature bore both the old and new names. In the late 1990s, molecular data started to sort out the taxonomic position and nomenclature for these bacteria. Phylogenetic analyses based on 16S rrna gene sequences resurrected the Pectobacterium for soft rot Erwinias (Hauben et al., 1998). On the basis of DNA-DNA hybridization, serology, phenotype and DNA sequence analyses some sub-species of E. carotovora were elevated to species level (Gardan et al., 2003). Two years later, Samson et al. (2005) proposed a new genus, Dickeya and several new species within it based on similar criteria. Due to this change in classification and taxonomic status of this group of organism in rapid succession, considerable confusion prevailed and general phytopathologists with lesser taxonomic understanding face major problems. The Aloe, belonging to family asphodelaceae, comprises of about four hundred succulent species. A. barbadensis, also known as true aloe or, medicinal aloe has been used for thousands of years for its medicinal properties. The succulent leaves contain a mucilaginous gel which is used to treat burns, wounds, skin irritations, *Corresponding author: mandal_kunal@yahoo.co.in etc. Now-a-days growing demand for aloe gel has ensured its place in major cosmetic products. It is also being marketed as health drinks. Increased demand for the raw material is met with its commercial cultivation, which is done at wide regions of the world having tropical and subtropical climates. Attempts were made for its cultivation under Indian conditions at Directorate of Medicinal and Aromatic Plants Research, Anand, Gujarat. However, after one year of plantation, a devastating soft rot disease appeared. The pathogen was identified as Pectobacterium chrysanthemi (Mandal and Maiti, 2005). Similar disease caused by E. chrysanthemi was reported from Caribbean island of Aruba (de Laat et al., 1994) and Korea (Jin et al., 1994). A soft rot disease of A. arborescens caused by E. carotovora subsp. carotovora has also been reported from Yugoslavia (Arsenijevic and Radujkov, 1987). These soft rot causing representative strains from the 3 countries of 3 continents i.e. Asia (India), Europe (The Netherlands) and North America (Yugoslavia) were compared in the present study. These bacterial strains were described and classified as per earlier system. Moreover, the organism from Yugoslavia is of uncertain identity (Dr. A. Obradovik, personal communication). Also, information on variability among these Aloe pathogens is not known. Hence, the present investigation was undertaken for proper identification and phylogenetic placement of these bacterial strains, determination of variability between the strains and identification of resistance source, if any, among different Aloe species against the bacterial strains. MATERIALS AND METHODS Bacterial strains Out of three strains, two i.e., E. chrysanthemi (PD2098 and PD2145) were from Caribbean island of Auruba, The

3 330 Indian Phytopathology 64 (4) : (2011) Netherland and third, E. carotovora subsp. carotovora (ka3a) was from Navi Sad, Yugoslavia. These bacterial strains were revived on nutrient agar (NA) plates and single purified colony was picked up for further study. Another strain, IMI was isolated from a sample collected from research farm of Directorate of Medicinal and Aromatic Plants Research, Boriavi ( N, E). These bacterial strains were maintained in aqueous suspension at 4 C (short term use) and at 70 C in 30% glycerol stock (long term preservation). Phenotypic characterization Different morphological, cultural, physiological and biochemical tests were performed following standard methods (Schaad, 1992). Carbon utilization studies were performed in minimal medium (Ayers et al., 1919) and using miniature test strips (HiMedia Laboratories, India) following manufacturer s guidelines. Unless stated otherwise, all tests were performed at 25±0.5 o C. Pathogenicity test Different vegetables (capsicum fruit, carrot root, cucumber fruit, onion bulb and tomato fruit), detached leaves of different Aloe spp. (A. barbadensis, A. pyrreyi, A. chinensis, A. rupestris, A. ciliaris and A. rauhii) and attached Dieffenbachia leaves were inoculated and incubated at 30 C. Bacterial cultures from 24 h growth in nutrient broth (NB) was adjusted to 0.01 OD with sterile water having concentration of CFU ml 1 for pathogenicity test. Observations on symptoms development were recorded up to 7 days of post inoculation. Protein and lipid profiling Overnight grown bacterial cells were adjusted to similar concentration (1 OD) and harvested by centrifugation (10000 rpm). The cell pellet was washed with buffer (10 mm Tris-Cl ph 7.1 and 30 mm NaCl) and centrifuged as earlier. The pellet cells were suspended in lysis buffer (30 mm Tris- Cl ph 7.1, 0.1 mm EDTA ph 8.0 and 20% sucrose) and sonicated (VCX 500, Sonics & Materials Inc., USA) at 30% amp for 3 10 seconds in ice. Sonicated material was centrifuged at rpm and the supernatant was collected as protein source. Protein profiling was done in 10% denaturing slab gel. In case of lipid profiling, cell pellet was sonicated, in chloroform and methanol (2:1, v/v) and debris was removed by centrifugation at rpm. The supernatant was spotted on silica gel plate 60E 254 (Merck KGaA, Germany) using CAMAG LINOMAT 5 automatic HPTLC sample applicator and air dried. The plate was developed in a twin trough chamber that has previously been equilibrated with mobile phase for 5 min. The mobile phase was chloroform, methanol and 0.2% CaCl 2 solution (55:35:8, v/v/v). Afterwards the plates were dried, briefly treated with freshly prepared ninhydrin reagent, dried and heated at 100 C for 10 minutes. The bands were visualized, photographed and analyzed using the documentation system and software. Molecular characterization Bacterial DNA was isolated from 1 ml overnight grown culture (0.3 OD) using UltraClean microbial DNA isolation kit (MoBio Laboratories Inc., USA) and used without dilution as source DNA (~100 ng template DNA in 25 ml total PCR reaction volume) for further amplification reactions. ITS and 16S rdna amplifications were performed according to Toth et al. (2001) and Shiomi et al. (1999), respectively. Pectate lyase gene (pel) amplification was achieved using species specific primers targeted to pelade (Nassar et al., 1996) and pely (Darrasse et al., 1994). All PCR reactions were performed in a thermal cycler (Mastercycler Gradient, Eppendorf AG, Germany). The amplicons (in 50 µl volume) were purified using MinElute PCR purification kit (Qiagen GmbH, Germany) and 4 µl of purified product (~100 ng DNA) separately treated with individual restriction enzymes (Fermentas International Inc., Canada) for restriction fragment length polymorphism (RFLP) of rdna and ITS. Fifteen primers (OPB2 OPB13, OPB15, OPB17 and OPB18) from Operon Biotechnologies GmbH were selected on the basis of reproducible polymorphic banding patterns for random amplified polymorphic DNA (RAPD). Products from the PCR and restriction reactions were individually loaded on 1.5% agarose gel, stained in ethedium bromide and documented in GeneGenius BioImaging System (SynGene, UK). Data analysis Qualitative data from phenotypic tests were converted to 0 ( ) and 1 (+). Similarly, nucleic acid bands from the RAPD and RFLP, protein bands from PAGE and lipid bands from TLC were scored as present (1) or absent (0) for each site. The data was then analysed using SIMQUAL operation in the NTSYSpc 2.02e software package (Exeter Software, USA). Similarity was calculated using simple matching coefficient at the default parameter for the NTSYSpc programme. The output data was then used to perform the SAHN clustering following UPGMA clustering method to produce final dendrogram. Phylogenetic anlysis PCR amplification of 16S rdna for sequencing was performed with universal primers (Shiomi et al., 1999) as per the conditions mentioned. Each 50 µl reaction mixture contained 5 µl of 10 PCR buffer, 2 µl of each primers (10 pmol), 1 µl 10 mm dntps (Fermentas), 7 µl 25 mm MgSO 4 (Fermentas), template DNA 1 µl (from isolation kit), 1.25 U proof reading Taq DNA polymerase (Fermentas) and sterile distilled water. PCR products were purified (Qiagen GmbH) and sequenced by primer walking method using BDT v3.1 Cycle sequencing kit on ABI 3730 l Genetic Analyzer through commercial service provider (Xcelris Labs Ltd., India). These sequences (GenBank accession Nos. GU GU362080) and those of the Brenneria, Dickeya and Pectobacterium type species, as mentioned in Samson et al were aligned with ClustalX2 programme and analyzed in MEGA4 (Tamura et al., 2007). Evolutionary history was inferred using the Neighbor-Joining method (Saitou and Nei, 1987) at default setting of the programme,

4 Indian Phytopathology 64 (4) : (2011) 331 while evolutionary distances were computed using the maximum composite likelihood method (Tamura et al., 2004). RESULTS AND DISCUSSION Phenotypic and pathogenic characterization Among the 54 different biochemical tests performed, 10 showed differential reactions among the tested strains (Table 1). It is interesting to note that the characteristics of three strains (IMI389157, PD2098 and PD 2145) were similar for majority of parameters but differed from the strain Ka3a. However, the Indian strain, IMI could be distinguished from the Netherland s strains with respect to lypolitic activity. None of the 6 Aloe spp. tested were resistant to either of the bacterial strains. However, the strains originally isolated from A. barbadensis (PD2098, PD2145 and IMI389157) were highly virulent with fast rotting of the leaves and bulging of leaves due to gas formation. However, the Yugoslavian strain (Ka3a) produced sunken lesion with slow progress rate. Pathogenicity on capsicum fruit and Dieffenbachia leaf also produced differential reactions for these two groups (Table 1). Protein and lipid profiling Out of 20 protein bands visible, 11 were polymorphic for at least one strain (Fig. 1A). Strains (IMI389157, PD2098, PD2145) isolated from A. barbadensis were having alomst similar protein patterns. However, two differences were observed at low molecular mass protein loci. The Indian strain was having a unique band at 48.8 kda, while the Netherland s strains possessed specific protein of 46.6 kda. The Yugoslavian strain isolated from A. arborescens showed distinct protein pattern than other three. It had at least six proteins (153.5, 112.7, 99.8, 81.0, 76.8, 44.3 kda) which were not present in others. Similar trend was observed in case of lipid profile (Fig 1B). All four strains had 8 major lipid bands. However, a band at Rf 0.20 was only present in Fig. 1. Protein (A) and lipid (B) profiles of bacterial strains, 1=IMI389157, 2=PD2098, 3=PD2145, 4=Ka3a, M=protein markers (in kd). Start and end lines for the TLC are also mentioned. 3 strains from A. barbadensis while lipid fraction having Rf 0.70 was unique for strain Ka3a. Molecular characterization Amplification of rrna gene (rdna) resulted a 1.5 kb product from all the strains. However, ITS profile distinguished the strains in two groups. Strains IMI389157, PD2098 and PD2145 produced two fragments of 0.60 and 0.44 kb length whereas, in Ka3a it was single fragment of 0.55 kb. Further, 3 restriction enzymes were used to generate RFLP of the amplicons (Fig. 2). Among these, rdna RFLP generated with EcoR1 and Hinf1 produced two distinct groups similar to ITS profile but, Sau3A failed to produce distinguishing pattern. On the other hand, ITS RFLP generated by Hinf1 and Sau3a supported the same grouping. However, EcoR1 produced altogether different RFLP. All three strains from former group had a common band of 0.46 kb but, IMI differed and had a band of 0.55 kb whereas, PD2098 was different from PD2145 due to an extra band of 0.38 kb. Strain Ka3a produced distinct RFLP pattern with a single band of 0.57 kb. Table 1. Differentiating phenotypic and pathogenic characters of the test strains Sl No Description of tests Bacterial strains IMI PD2098 PD2145 Ka3a 1 Growth at 39 o C Lipolytic activity + 3 Gas from D-glucose Indole test Lecithinase test Organic Carbon Utilization 6 D-arabinose L-arabinose Sodium malonate Pathogenicity on 9 Capsicum fruit + 10 Dieffenbachia leaf + + +

5 332 Indian Phytopathology 64 (4) : (2011) Fig. 2. RFLP profiles of rdna (A) and ITS (B) amplicons of test bacterial strains generated by different restriction enzymes (EcoR1, Hinf1 and Sau3A). M=100bp DNA ladder, 1=IMI389157, 2=PD2098, 3=PD2145, 4=Ka3a two strains had a similarity index of and 0.669, respectively with IMI The Yugoslavian strain (Ka3a) was altogether different from other 3 with similarity coefficient of <0.47. Phylogenetic analysis Fig. 3. Dendogram based on phenotypic, protein/lipid profiles, RAPD and RFLP data of four Aloe rot causing bacterial strains. A total of 138 bands were visible from 15 RAPD primers. Among these only 2 (one each from GGACTGGAGT and TGCGCCCTTC) were monomorphic, while rest were polymorphic. Strains PD2098 and PD2145 showed identical banding patterns and were distinct from IMI with similarity index of 0.457, whereas Ka3a showed least similarity. pelade gene product was detected in IMI389157, PD2098 and PD2145 while no amplification was visible from Ka3a. However, Ka3a only was positive to pely. Based on all phenotypic, protein/lipid profile and molecular data, the strains PD2098 and PD2145 were found closely related with a similarity index of (Fig. 3). These Phylogenetic analysis of 16S rdna sequences showed that three major clades (Fig. 4) where members of three different genera Brenneria, Dickeya and Pectobacterium were placed. Two test strains, isolated from A. barbadensis (IMI and PD2098), formed a group close to a clade which included D. zeae (CFBP2052) and D. dieffenbachiae (CFBP2051). Computed distance among these four varied between and However, other test strain from A. arborescence (ka3a) shared high sequence similarity (distance 0.001) with that of P. carotovorum (ATCC15713). The four Aloe pathogenic strains, isolated from three countries India, The Netherlands and Yugoslavia, showed wide variability. The strains were also found to belong to different taxa. On the basis of limited number of strains taken for the study, it is not possible to comment on the degree of variability in the population. However, it was found that within the same taxon, the strains differed in their phenotypic and molecular characters. Comparison of total (Kim et al., 2001) or membrane (Malandrin et al., 1996) protein profiles are commonly used to distinguish closely related species or pathovars. Qhobela and Claflin (1992) found protein banding patterns of Xanthomonas campestris pv. vasculorum strains formed two distinct groups according to geographical locations. Our study also distinguished major differences between the species and locations in terms of protein profiles.

6 Indian Phytopathology 64 (4) : (2011) 333 Fig. 4. Unrooted tree generated by phylogenetic analysis of 16S rrna gene sequences of type strains of Brenneria, Dickeya and Pectobacterium species and the three Aloe rot causing bacterial strains. Toth et al. (2001) proposed RFLP analysis of ITS as an sequencing independent rapid method for identification of the soft rot causing Erwinias. They tried 12 restriction enzymes and final ITS RFLP group was based on two enzymes (CfoI and RsaI). Still, variation in RFLP within the species existed, as in the present study. ITS being noncoding region, mutation in this segment will have less bearing on ecological fitment of the organism and more variation likely to occur. However, difference among the strains isolated from same location points to high variability within the species. Bacterial species have characteristic fatty acid profile signatures (O Leary, 1962) which can be used for microbial identification. In fact, the technique has been commercially exploited in MIDI system. However, it requires considerable investment in terms of infrastructure and human resource. Matsuyama (1995) developed an easy method based on whole cell TLC. Later it was modified for further ease (Khan and Matsuyama, 1998). This method could successfully differentiated Erwinia species and Pseudomonads belonging to rrna homology groups. Similarly, based on the major lipid banding patterns, present findings suggest involvement of the two Erwinia species. All the cultures were confirmed to belong to the former genus Erwinia on the basis of fermentative metabolism, absence of fluorescent pigment on King s B medium, nitrate reductase activity and absence of amylase activity (Yahhiaoui-Zaidi et al., 2003). This method has been used to differentiate the strains to species and subspecies levels in present investigation and the 3 strains (IMI389157, PD2098 and PD 2145) have been found to belong to E. chrysanthemi, and 1 (Ka3a) to E. carotovora subsp. carotovora. It was in accordance with the species specific amplification of pel gene products. Further, the categorization of the E. chrysanthemi starins was done on the basis of simplified scheme proposed by (Dickey and Victoria, 1980). It was observed that all the strains (IMI389157, PD2098 and PD 2145) belonged to subdivision IV. Also, these strains were found to belong to biovar3 as per Ngwira and Samson (1990). This biovar was reported to have wide host range (de Laat et al., 1994). As per the new speciation proposed by Samson et al. (2005) these three cultures were very similar to phenon 4 (D. dieffenbachiae) and phenon 1 (D. dadantii/ D. zeae), but the test strains did not fully match either of the phena. However, phylogenetically these were close to D.

7 334 Indian Phytopathology 64 (4) : (2011) dieffenbachiae and D. zeae. Dickey (1981) observed that only those strains of E. chrysanthemi were pathogenic on Dieffenbachia which have been originally isolated from the same host. Three E. chrysanthemi strains tested here produced positive reaction on Dieffenbachia leaf. This suggests possibility of inclusion of these strains under D. dieffenbachiae. On the basis of phenotypic and molecular characters, the Yugoslavian strain Ka3a was confirmed to be Pectobacterium carotovorum. ACKNOWLEDGEMENTS Authors are thankful to the Director, Directorate of Medicinal and Aromatic Plants Research for providing facilities for the study. We are grateful to Dr. J. D. Janse, The Netherland and Dr. A. Obradovik, Yugoslavia for supplying the bacterial cultures. REFERENCES Arsenijevic, M. and Radujkov, D. (1987). Erwinia carotovora subsp. carotovora (Jones 1901), Bergey, Harrison, Breed, Hammer and Huntoon 1923 as a parasite of the cactus Aloe arborescens Mill. Zast. Bilja. 38: Ayers, S.H., Rupp, P. and Johnson, T. (1919) A study of the alkali forming bacteria in milk. USDA Bulletin. pp Darrasse, A., Priou, S., Kotoujansky, A. and Bertheau, Y. (1994). PCR and restriction fragment length polymorphism of a pel gene as a tool to identify Erwinia carotovora in relation to potato diseases. Appl. Environ. Microbiol. 60: de Laat, P.C.A., Verhoeven, J.T.W. and Janse, J.D. (1994). Bacterial leaf rot of Aloe vera L. caused by Erwinia chrysanthemi biovar 3. Eur. J. Plant Pathol. 100: Dickey, R.S. (1981). Erwinia chrysanthemi: Reaction of eight plant species to strains from several hosts and to strains of other Erwinia species. Phytopathology 71: Dickey, R.S. and Victoria, J.I. (1980). Taxnomy and emended description of strains of Erwinia isolated from Musa paradisiaca Linnaeus. Int. J. Syst. Bacteriol. 30: Gardan, L., Gouy, C., Christen, R. and Samson, R. (2003). Elevation of three subspecies of Pectobacterium carotovorum to species level: Pectobacterium atrosepticum sp. nov., Pectobacterium betavasculorum sp. nov. and Pectobacterium wasabiae sp. nov. Int. J. Syst. Evol. Microbiol. 53: Hauben, L., Moore, E.R.B., Vauterin, L., Steenackers, M., Mergaert, J., Verdonck, L. and Swings, J. (1998). Phylogenetic position of phytopathogens within the Enterobacteriaceae. Syst. Appl. Microbiol. 21: Jin, K.S., Lee, S.W., Kim, J.J. and Ryu, H.Y. (1994). Identification of bacterial isolates obtained from diseased orchid and Aloe plants caused by Erwinia chrysanthemi. RDA J. Agri. Sci. Crop Protec. 36: Khan, A.A. and Matsuyama, N. (1998). Rapid identification of phytopathogenic bacteria by TLC. J. Fac. Agric. Kyushu Univ. 42: Kim, W.S., Hildebrand, M., Jock, S. and Geider, K. (2001). Molecular comparison of pathogenic bacteria from pear trees in Japan and the fire blight pathogen Erwinia amylovora. Microbiology 147: Malandrin, L., Huard, A. and Samson, R. (1996). Discriminant envelope protein profiles between Pseudomonas syringae pv. pisi and P. siringae pv. syringae. FEMS Microbiol. Lett. 141: Mandal, K. and Maiti, S. (2005). Bacterial soft rot of aloe caused by Pectobacterium chrysanthemi: a new report from India. Plant Pathol. 54: 573. Matsuyama, N. (1995). Trails for rapid identification of phytopathogenic bacteria by HPLC and direct colony TLC. J. Fac. Agric. Kyushu Univ. 40: Nassar, A., Darrasse, A., Lemattre, M., Kotoujansky, A., Dervin, C., Vedel, R. and Bertheau, Y. (1996). Characterization of Erwinia chrysanthemi by pectinolytic isozyme polymorphism and restriction fragment length polymorphism analysis of PCR-amplified fragment of pel genes. Appl. Environ. Microbiol. 62: Ngwira, N. and Samson, R. (1990). Erwinia chrysanthemi: description of two new biovars (bv 8 and bv 9) isolated from kalanchoe and maize host plants. Agronomie 10: O Leary, W.M. (1962). The fatty acids of bacteria. Bacteriol. Rev. 26: Qhobela, M. and Claflin, L.E. (1992). Eastern and southern African strains of Xanthomonas campestris pv. vasculovum are distinguishable by restriction fragment length polymorphism of DNA and polyacrylamide gel electrophoresis of membrane proteins. Plant Pathol. 41: Saitou, N. and Nei, M. (1987). The neighbor-joining method: A new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4: Samson, R., Legendre, J.B., Christen, R., Saux, M.F., Achouak, W. and Gardan, L. (2005). Transfer of Pectobacterium chrysanthemi (Burkholder et al. 1953) Brenner et al and Brenneria paradisiaca to the genus Dickeya gen. nov. as Dickeya chrysanthemi comb. nov. and Dickeya paradisiaca comb. nov. and delineation of four novel species, Dickeya dadantii sp. nov., Dickeya dianthicola sp. nov., Dickeya dieffenbachiae sp. nov. and Dickeya zeae sp. nov. Int. J. Syst. Evol. Microbiol. 55: Schaad, N.W. (1992). Laboratory Guide for Identification of Plant Pathogenic Bacteria, 2nd Edition. International Book Distribution Co, Lucknow. pp Shiomi, Y., Nishiyama, M., Onizuka, T. and Marumoto, T. (1999). Comparison of bacterial community structures in the rhizoplane of tomato plants grown in soils suppressive and conducive towards bacterial wilt. Appl. Environ. Microbiol. 65: Tamura, K., Nei, M. and Kumar, S. (2004). Prospects for inferring very large phylogenies by using the neighbor-joining method. Proc. Natl. Acad. Sci. USA 101: Tamura, K., Dudley, J., Nei, M. and Kumar, S. (2007). MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24: Toth, I.K., Avrova, A.O. and Hyman, L.J. (2001). Rapid identification and differentiation of the soft rot Erwinias by 16S-23S intergenic transcribed spacer-pcr and restriction fragment length polymorphism analyses. Appl. Environ. Microbiol. 67: Waldee, E.L. (1945). Comparative studies of some peritrichous phytopathogenic bacteria. Iowa State J. Sci. 19: Winslow, C.E.A., Broadhurst, J., Buchanan, R.E., Krumwiede, C.Jr., Rogers, L.A. and Smith, G.H. (1917). The families and genera of the bacteria. Preliminary report of the committee of the Society of American Bacteriologists on characterization and classification of bacterial types. J. Bacteriol. 2: Yahhiaoui-Zaidi, R., Jouan, B. and Andrivon, D. (2003). Biochemical and molecular diversity among Erwinia isolates from potato in Algeria. Plant Pathol. 52: Received for publication: August 08, 2010 Accepted for publication: August 12, 2011