Comparative genome analysis of 15 clinical Shigella flexneri strains regarding virulence and antibiotic resistance

Abstract

Shigellosis is the major cause of dysentery globally. It is mainly attributed to two Shigella species, Shigella sonnei and Shigella flexneri, which leads to approximately 165 million infections and 1.1 million deaths each year. Rapid increase and widening of spectrum in antibiotics resistance make Shigella hard to be adequately controlled through existing prevention and treatment measures. It has also been observed that enhanced virulence and advent of antibiotic resistance (AR) could arise almost simultaneously. However, genetic linkages between the two factors are missing or largely ignored, which hinders experimental verification of the relationship. In this study, we sequenced 15 clinically isolated S. flexneri strains. Genome assembly, annotation and comparison were performed through routine pipelines. Differential resistant profiles of all 15 S. flexneri strains to nine antibiotics were experimentally verified. Virulence factors (VFs) belonging to 4 categories and 31 functional groups from the Virulence Factor Database (VFDB) were used to screen all Shigella translated CDSs. Distribution patterns of virulence factors were analysed by correlating with the profiles of bacterial antibiotics resistance. In addition, multi-resistant S. flexneri strains were compared with antibiotic-sensitive strains by focusing on the abundance or scarcity of specific groups of VFs. By doing these, a clear view of the relationships between virulence factors and antibiotics resistance in Shigella could be achieved, which not only provides a set of genetic evidence to support the interactions between VFs and AR but could also be used as a guidance for further verification of the relationships through manipulating specific groups of virulence factors.

Keywords: HMMER; Prokka; Shigella; antibiotics resistance; comparative genomics; virulence factor.

Figure 1.. Phylogenomic tree generated via core genes of 27 Shigella strains, which divided S. sonnei and S. flexneri into two branches. Genome size (black bar) and degree of antibiotic resistance (red bar) were incorporated, accordingly. Bootstrapping values (1000 times) were visualized through red square symbols of varying size on the branches. The presence (filled squares) and absence (empty squares) of nine antibiotics that were tested in this study, which included Amoxicillin/Clavulanic acid (AMC), Ceftiophene (CFT), Cefotaxime (CTX), Gentamicin (GEN), Nalidixic acid (NAL), Norfloxacin (NOR), Tetracycline (TBT), and compound Sulfamethoxazole (SMZ), were presented. No square means intermittent level of resistance. AMC and NOR shows the most apparent resistance difference between S. flexneri and S. sonnei. ^# Most sensitive strains with MDR value of 0 and 1 (vertical line 1). *Most resistant strains with MDR value of 8 and 9 (vertical line 8).

Figure 2.. Genome comparison of 15 isolated S. flexneri strains against reference genome S. flexneri 2a str. 301 generated by BRIG 0.95. The inner cycle (black) represents the complete genome of the reference strain and the shade of each colors denote the similarities between each strain with reference strain. GC content and GC skew (+/-) were illustrated in-between.

Figure 3.. Venn diagram of core- and pan-genome of 15 S. flexneri strains. 3742 core genes were shared by all strains while varied number of genes were present in each strain as unique genes. The two sensitive strains (S13106 and S15054) have the lowest number of unique genes (476 and 464) while the two most resistant strains (S13028 and S13073) have the highest number of unique genes (908 and 993).

Figure 4.. Principal component analysis (PCA) of the relationship between antibiotic resistance and virulence factors in 15 sequenced Shigella flexneri strains. S13016 and S15054 are antibiotics-sensitive strains while the rest strains are multi-drug resistant. The two sensitive strains have comparatively less virulence factors in a majority of functional groups while resistant strains have comparatively higher number of certain groups of virulence factors. Thus, resistance and virulence could have a mutual improvement relationship, although exceptions do exist.