Molecular identification and genetic diversity among Photorhabdus and Xenorhabdus isolates

Five bacterial strains were isolated from the hemocoel of the greater wax moth larvae (Galleria mellonella) infected with the entomopathogenic nematodes: Heterorhabditis bacteriophora HP88, Heterorhabditis indicus RM1 and Heterorhabditis sp (S1), Steinernema abbasi and Steinernema sp. (S II). Strains were identified as Photorhabdus luminescens HRM1, P. luminescens HS1, P. luminescens HP88, Xenorhabdus indica and X. nematophila ATTC19061 using 16S rDNA sequence analysis. To reveal the genetic diversity among these strains, three molecular markers (RAPD, ISSR and SRAP) were employed. RAPD analysis showed 73.8 and 54.5 polymorphism percentages for the Photorhabdus and Xenorhabdus strains, respectively. ISSR analysis resulted in 70.1 and 75.2 polymorphism percentages among the Photorhabdus and Xenorhabdus strains, respectively. The SRAP analysis indicated that 75.6 and 61.2% genetic polymorphism was detected among Photorhabdus and Xenorhabdus strains, respectively. The cluster analysis grouped the three Photorhabdus strains together in one cluster and the two Xenorhabdus strains together in another cluster indicating the phylogenetic relationships among them. The genotype-specific markers detected from the three molecular markers (RAPD, ISSR and SRAP) were sufficient to distinguish between the different bacterial strains tested and can be used in the future IBM program that could be built on the use of these strains.


Introduction
Xenorhabdus spp. and Photorhabdus spp. are motile, Gram-negative bacteria belonging to the family Enterobacteriaceae. These bacteria are symbiotically associated with nematodes of the families, Steinernematidae and Heterorhabditidae, respectively (Forst et al. 1997;Sheets et al. 2011;Ferreira et al. 2014). These symbiotic complexes are highly pathogenic for a wide range of insects and are, therefore, used worldwide as biological agents in crop control (Ehlers 2003;Goodrich-Blair and Clarke 2007;Del Valle et al. 2013).
Many symbiotic bacterial species and isolates are still to be identified or described. Several identification techniques are now available that range from total protein, isozyme profiles to DNA/DNA hybridization and sequence analysis of the 16S rDNA region (Akhurst and Boemare 1990; Boemare and Akhurst 2007; Lee and Stock 2010).
The availability of molecular markers such as 16S rDNA (Christensen et al. 1998), RAPD (Williams et al. 1990;Welsh et al. 1991), ISSR (Zietkiewicz et al. 1994) facilitates the development of phylogenetic relationships, identification and characterization of bacterial species.
The sequence-related amplified polymorphism (SRAP) system was developed by Li and Quiros (2001) to target overlapping coding and non-coding regions of the genome. Depending on the amplification of Open-Reading Frames (ORFs) using the GC-rich exons and the promoter (Li and Quiros 2001), SRAP not only amplifies the interval between genes and their non-coding flanking regions, but also tightly links to actual genes, which would generate a fingerprint of the coding sequences and permit easy isolation of these bands for sequencing (Yu et al. 2008). SRAP is used in genetic map construction, genealogical classification, gene tagging and cloning, population structure, genetic diversity and genetic linkage map of plants (Li et al. 2007;Zheng et al. 2010;Jiang and Liu 2011;Lu et al. 2012;Alghamdi et al. 2012). Furthermore, it was used to study genetic diversity in parasites of human and animal health (Li et al. 2009;Song et al. 2011).
In the present work, five bacterial isolates (three Photorhabdus and two Xenorhabdus) were identified at the molecular level based on the 16S rDNA region. Furthermore, the genetic polymorphism among the five bacterial isolates was investigated at the molecular levels using RAPD, ISSR and SRAP analyses. The obtained data were used to construct the phylogenetic tree. The genotypespecific markers were also determined.

Bacterial strains and culture
Five bacterial isolates were isolated from the hemocoel of the greater wax moth larvae (Galleria mellonella) infected with different entomopathogenic nematode strains (Steinernema abbasi, Steinernema sp. (S II), Heterorhabditis bacteriophora HP88, Heterorhabditis indicus RM1 and Heterorhabditis sp. (S1)). The bacterial isolates were plated on Nutrient Bromothymol Blue Agar (NBTA) medium (Akhurst 1980). On which Phase I is distinguished from Phase II by its adsorption of bromothymol blue to produce a red core colony overlaid by dark blue and surrounded by a clear zone after 3-4 days of incubation at 25°C (Wang et al. 2008). The NBTA medium contained 20 g nutrient agar (Difco), 25 mg bromothymol blue (s.d. FiNE_CHEM ltd), and 40 mg tripheyltetrazolium (BDH, England) in 1 L distilled water. The Phase I colony was selectively transferred to 5 mL LB broth (Difco), and incubated at 25°C for 48 h with gentle agitation (100 rpm).

Data analysis
All the genotypes were scored for the presence and absence of the RAPD, ISSR and SRAP bands. And the data were entered into a binary matrix as discrete variables, 1 for presence and 0 for absence of the character and this data matrix was subjected to further analysis.
The Excel file containing the binary data was imported into NT Edit of NTSYS-pc 2.02J. The 0/1 matrix was used to calculate similarity as DICE coefficient using SIMQUAL subroutine in SIMILARITY routine. The resultant similarity matrix was employed to construct dendrogram using sequential agglomerative hierarchical nesting (SAHN) based on the unweighted pair group method with arithmetic means (UPGMA) to infer genetic relationships and phylogeny (Sneath and Sokal 1973).

Results and discussion
In the present study, three symbiotic bacterial strains isolated from Heterorhabditis family and two from Steinernema family were identified using 16S rDNA sequence analysis. The data indicated that 1530 bp fragment was amplified; these bands were eluted and sequenced (Fig. 1). The complete 16S rDNA gene sequences of all strains (1530 nucleotides) were aligned to the homologous sequences of Photorhabdus and Xenorhabdus using the BLASTN program in the NCBI (Altschul et al. 1990). The data indicate that the three symbiotic bacterial strains that have been isolated from Heterorhabditis family were highly similar to Photorhabdus. The identified bacterial strains were registered on the GeneBank under the accession number KC237382 to P. luminescens HRM1 and KC237383 for the P. luminescens HS1, while the third sequence was identical with P. luminescens HP88. The data of the two symbiotic bacteria which have been isolated from Stinernema family indicate that one of them was identical to X. indica and the second sequence was similar to X. nematophila ATTC19061 16S rDNA gene sequence with the score of 96%.
The five bacterial strains identified were characterized at the molecular levels using three molecular markers (RAPD, ISSR and SRAP) to determine the genetic polymorphism among them and also to determine the genotypespecific markers for each strain.
Eleven random RAPD primers were used to detect the genetic polymorphism among these five bacterial strains. All the primers tested resulted in clear bands (Fig. 2). The three Photorhabdus strains generated 126 bands: 93 bands out of them were polymorphic and can be considered as useful bands for the three bacterial isolates with 73.8 polymorphism percentages. The Xenorhabdus strains showed 123 total bands and 76 bands out of the total bands were polymorphic with 54.5% polymorphism ( Table 2). The RAPD genotype-specific bands as presented in Table 3 indicate that the three Photorhabdus strains showed 41 genotype-specific markers that represent 44% from the polymorphic band detected and 32.5% from the total band numbers (Tables 2, 3). The highest genotype-specific number was recorded for the bacterial strain P. luminescens HP88 (37 markers) followed by P. luminescens HS1 (2 markers) and P. luminescens RM1 (2 markers). Xenorhabdus strains showed 76 specific markers: 39 markers of them are specific to X. nematophila S2 and 37 for X. indica AB. The genotype-specific markers obtained for Xenorhabdus strains represent 61.7% of the total band number and 100% of the polymorphic bands (Tables 2, 3).
The presence of simple sequence repeats (SSR) in prokaryotes is well documented (Gur-Arie et al. 2000), and some SSRs show extensive length polymorphisms (Yang  Sreenu et al. 2006). Successful use of PCRbased SSR amplification followed by amplicon size determination to analyze the spread of microbial pathogens has been reported for Haemophilus influenzae and Candida albicans (Bretagne et al. 1997;van Belkum et al. 1997). Furthermore, the utility of inter-simple sequence repeat-PCR (ISSR-PCR) assay in the characterization and elucidation of the phylogenetic relationship between the pathogenic and nonpathogenic isolates of Vibrio cholerae was demonstrated (Kumar et al. 2007). They proposed that ISSR-PCR is an efficient tool in phylogenetic classification of prokaryotic genomes in general and diagnostic genotyping of microbial pathogens in particular. A DNA sample representing the five bacterial strains was subjected to PCR analysis using twelve ISSR primers. The primers used show stable and repeatable banding pattern for the strain tested (Fig. 3). Data presented in Table 2 indicate that 95 ISSR bands were generated among the three Photorhabdus strains; 28 bands out of total band numbers were monomorphic bands representing a common band in the genus Photorhabdus and 67 bands which are polymorphic can be considered as useful bands for the three bacterial isolates with 70.1 polymorphism percentages. The primers showed different polymorphism percentages that ranged from 33% for the primer IS3 to 100% for UBC815. The Xenorhabdus strains showed 113 total bands; 85 bands out of the total bands were polymorphic with 75.2% polymorphism ( Table 2).
The ISSR genotype-specific bands were determined. Data presented in Table 4 show that ISSR marker can distinguish between the different bacterial strains. The three Photorhabdus strains showed 34 genotype-specific ISSR markers that represent 50.7% from the polymorphic band detected and 75.2% from the total band numbers ( Table 4). The highest genotype-specific number was recorded for the bacterial strain P. luminescens HP88 (21 markers) followed by P. luminescens RM1 (9 markers) and P. luminescens HS1 (4 markers). Xenorhabdus strains showed 85 specific markers: 45 markers of them are specific to X. nematophila S2 and 40 for X. indica AB. The genotype-specific markers obtained for Xenorhabdus strains represent 75.2% of the total band numbers and 100% of the polymorphic bands (Table 4). Li and Quiros (2001) reported that the SRAP forward primer preferentially amplified exonic regions and the reverse primer preferentially amplified intronic regions. Therefore, the SRAP technique could detect polymorphisms arising from variations in the length of introns, promoters, and spacers, among both genotypes and species when different forward and reverse primers are randomly combined (Li et al. 2014).
In the present study, twenty-five SRAP combinations of five forward and five reverse primers were used for fingerprinting the five bacterial strains. All the SRAP primer combinations resulted in a scorable and reproducible bands. The data presented in Fig. 4 illustrate the SRAP banding pattern of the five bacterial strains. Data presented in Table 2 indicated that 230 bands were generated among the three Photorhabdus strains: 174 bands out of the total band numbers were polymorphic and can be considered as useful bands for the three bacterial isolates with 75.6 polymorphism percentages. The primer combinations showed different polymorphism percentages. The Xenorhabdus strains showed 191 total bands: 117 bands out of the total bands were polymorphic with 61.2% polymorphism ( Table 2).

Conclusion
Based on the data obtained from the present study, the three molecular markers (RAPD, ISSR and SRAP) in addition to the 16S rDNA could be efficiently used to identify and characterize the five bacterial strains. The genotype-specific markers were sufficient to distinguish between the different bacterial strains and can be used in the future IBM program that could be built on the use of these strains.