Aim: This study was designed to determine whether the phenotypic antibiotic resistance pattern of B. pseudomallei is associated with the source of isolates and the genotype.
Materials and Methods: A collection of 111 B. pseudomallei isolates from veterinary cases of melioidosis and the environments (soil and water) were obtained from stock cultures of previous studies and were phylogenetically characterized by multilocus sequence typing (ST). The susceptibility to five antibiotics, namely meropenem (MEM), imipenem, ceftazidime (CAZ), cotrimoxazole (SXT), and co-amoxiclav (AMC), recommended in both acute and eradication phases of melioidosis treatment were tested using minimum inhibitory concentration antibiotics susceptibility test.
Results: Majority of isolates were susceptible to all antibiotics tested while few resistant strains to MEM, SXT, CAZ, and AMC were observed. Statistically significant association was found between resistance to MEM and the veterinary clinical isolates (p<0.05). The likelihood of resistance to MEM was significantly higher among the novel ST 1130 isolates found in veterinary cases as compared to others.
Conclusion: The resistance to MEM and SXT appeared to be higher among veterinary isolates, and the novel ST 1130 was more likely to be resistant to MEM as compared to others.
Method: Thirty-five full-length pk41 sequences from clinical isolates of Malaysia along with four laboratory lines (along with H-strain) were downloaded from public databases. For comparative analysis between species, orthologous P41 genes from P. falciparum, P. vivax, P. coatneyi and P. cynomolgi were also downloaded. Genetic diversity, polymorphism, haplotype and natural selection were determined using DnaSP 5.10 software. Phylogenetic relationships between Pk41 genes were determined using MEGA 5.0 software.
Results: Analysis of 39 full-length pk41 sequences along with the H-strain identified 36 SNPs (20 non-synonymous and 16 synonymous substitutions) resulting in 31 haplotypes. Nucleotide diversity across the full-length gene was low and was similar to its ortholog in P. vivax; pv41. Domain-wise amino acid analysis of the two s48/45 domains indicated low level of polymorphisms for both the domains, and the glutamic acid rich region had extensive size variations. In the central domain, upstream to the glutamate rich region, a unique two to six (K-E)n repeat region was identified within the clinical isolates. Overall, the pk41 genes were indicative of negative/purifying selection due to functional constraints. Domain-wise analysis of the s48/45 domains also indicated purifying selection. However, analysis of Tajima's D across the genes identified non-synonymous SNPs in the s48/45 domain II with high positive values indicating possible epitope binding regions. All the 6-cysteine residues within the s48/45 domains were conserved within the clinical isolates indicating functional conservation of these regions. Phylogenetic analysis of full-length pk41 genes indicated geographical clustering and identified three subpopulations of P. knowlesi; one originating in the laboratory lines and two originating from Sarawak, Malaysian Borneo.
Conclusion: This is the first study to report on the polymorphism and natural selection of pk41 genes from clinical isolates of Malaysia. The results reveal that there is low level of polymorphism in both s48/45 domains, indicating that this antigen could be a potential vaccine target. However, genetic and molecular immunology studies involving higher number of samples from various parts of Malaysia would be necessary to validate this antigen's candidacy as a vaccine target for P. knowlesi.