Characterization of Shigella sonnei in Malaysia, an increasingly prevalent etiologic agent of local shigellosis cases

Abstract

Background: Shigellosis is a major public health concern worldwide, especially in developing countries. It is an acute intestinal infection caused by bacteria of the genus Shigella, with a minimum infective dose as low as 10-100 bacterial cells. Increasing prevalence of Shigella sonnei as the etiologic agent of shigellosis in Malaysia has been reported. As there is limited information on the genetic background of S. sonnei in Malaysia, this study aimed to characterize Malaysian S. sonnei and to evaluate the prospect of using multilocus variable-number tandem-repeat (VNTR) analysis (MLVA) for subtyping of local S. sonnei.

Methods: Forty non-repeat clinical strains of S. sonnei isolated during the years 1997-2000, and 2007-2009 were studied. The strains were isolated from stools of patients in different hospitals from different regions in Malaysia. These epidemiologically unrelated strains were characterized using biotyping, antimicrobial susceptibility testing, pulsed-field gel electrophoresis (PFGE) and MLVA.

Results: The two biotypes identified in this study were biotype a (n = 29, 73%) and biotype g (n = 11, 27%). All the 40 strains were sensitive to kanamycin, ceftriaxone and ciprofloxacin. Highest resistance rate was observed for streptomycin (67.5%), followed by tetracycline (40%) and trimethoprim-sulfamethoxazole (37.5%). All the S. sonnei biotype g strains had a core resistance type of streptomycin - trimethoprim-sulfamethoxazole - tetracycline whereas the 29 biotype a strains were subtyped into eight resistotypes. All the strains were equally distinguishable by PFGE and MLVA. Overall, PFGE analysis indicated that S. sonnei biotype a strains were genetically more diverse than biotype g strains. Cluster analysis by MLVA was better in grouping the strains according to biotypes, was reflective of the epidemiological information and was equally discriminative as PFGE.

Conclusions: The S. sonnei strains circulating in Malaysia throughout the period studied were derived from different clones given their heterogeneous nature. MLVA based on seven selected VNTR loci was rapid, reproducible and highly discriminative and therefore may complement PFGE for routine subtyping of S. sonnei.

Figure 1

Dendrogram ofS. sonneiPFGE patterns generated by UPGMA clustering method using Dice coefficient. Key: all sensitive (#), ampicillin (A), chloramphenicol (C), streptomycin (S), tetracycline (T), and trimethoprim–sulfamethoxazole (Sx).

Figure 2

Dendrogram generated fromS. sonneiMLVA suptyping by the UPGMA clustering method using categorical coefficient. Key: all sensitive (#), ampicillin (A), chloramphenicol (C), streptomycin (S), tetracycline (T), and trimethoprim–sulfamethoxazole (Sx).

Figure 3

Minimum spanning tree for the 40 MalaysianShigella sonneistrains and strains from five other countries. For Malaysian strains, each circle represents one strain with a unique MLVA profile. Each circle is noted with the MLVA type of the strain. MLVA types in yellow colour are Malaysian biotype g strains whereas those in green are Malaysian biotype a strains. For the profiles of strains from five other countries, the circle size is not proportional to the number of strains belonging to the MLVA profile but it is proportional to the number of outbreaks or unrelated episodes sharing the same profile. Circles in grey colour denote global clonal group A, orange denotes group B, and blue denotes group C. Additional information on these foreign strains is provided [see Additional file 1]. The distance between two MLVA types is numbered. A distance of one locus between two closest MLVA types is marked by a thick black line, a distance of two loci is marked by a blue line, and a distance of three loci is indicated by a thin red line. A distance of four loci is indicated by a broken line and a distance of five loci is indicated by a light broken line.