A vancomycin resistance-associated WalK(S221P) mutation attenuates the virulence of vancomycin-intermediate Staphylococcus aureus

Abstract

Introduction: Vancomycin-intermediate Staphylococcus aureus (VISA) is typically associated with a decline in virulence. We previously reported a WalK(S221P) mutation that plays an important role in mediating vancomycin resistance in VISA XN108. Whether this mutation is implicated in bacterial virulence remains unknown.

Objectives: This study aimed to investigate the effect of WalK(S221P) mutation on the virulence of VISA and the underlying mechanism of this effect.

Methods: The influence of WalK(S221P) mutation on VISA virulence and its underlying mechanism were explored using animal models, RNA-seq analysis, RT-qPCR, hemolytic assay, slide coagulase test, Western blot, β-galactosidase assay, and electrophoresis mobility shift assay (EMSA).

Results: Compared with XN108, WalK(S221P)-reverted strain XN108-R exacerbated cutaneous infections with increased lesion size and extensive inflammatory infiltration in mouse models. The bacterial loads of S. aureus XN108-R in murine kidney increased compared with those of XN108. RNA-seq analysis showed upregulation of a set of virulence genes in XN108-R, which exhibited greater hemolytic and stronger coagulase activities compared with XN108. Introduction of WalK(S221P) to methicillin-resistant S. aureus USA300 and methicillin-susceptible strain Newman increased the vancomycin resistance of the mutants, which exhibited reduced hemolytic activities and decreased expression levels of many virulence factors compared with their progenitors. WalK(S221P) mutation weakened agr promoter-controlled β-galactosidase activity. EMSA results showed that WalK-phosphorylated WalR could directly bind to the agr promoter region, whereas WalK(S221P)-activated WalR reduced binding to the target promoter. Inactivation of agr in S. aureus did not affect their vancomycin susceptibility but mitigated the virulence alterations caused by WalK(S221P) mutation.

Conclusion: The results of our study indicate that WalK(S221P) mutation can enhance vancomycin resistance in S. aureus of diverse genetic backgrounds. WalK(S221P)- bearing S. aureus strains exhibit reduced virulence. WalK(S221P) mutation may directly impair the activation of the agr system by WalR, thereby decreasing the expression of virulence factors in VISA.

Keywords: Agr system; Staphylococcus aureus; Vancomycin resistance; Virulence; WalK(S221P); WalKR.

Graphical abstract

Fig. 1

Reversion of WalK(S221P) enhanced VISA virulence. (A) Skin abscess areas of the S. aureus XN108- and XN108-R-infected mice. Mice were injected subcutaneously with 1 × 10⁸ CFU of bacterial cells. Abscess sizes were measured daily. The number of animals used, n = 10. Statistical significance was calculated by 2Way ANOVA. ** P < 0.01, *** P < 0.001, and ns indicates no significance. (B) Representative abscesses in mice on 2 and 7 d pi. The abscess size for each animal was indicated by a yellow box. (C) Histological observation. The skin of infected mice was obtained on 5 d pi. The histological sections were prepared, stained with HE, and observed under a microscope (Olympus BX53). The bar represented 100, 200 or 500 µm was indicated. The blue arrows indicated mouse epidemis, the red arrows showed subcutaneous tissue glands, the black arrows indicated hair follicles, the black tangles pointed to abscesses and the blue tangles showed the degenerated fat vacuoles. (D) Severity of the mouse lesions. The severity of the skin lesions in histological sections of XN108-, XN108-R-, and mock (PBS)-infected mice was numerically classified with ImageJ software, and the relative inflammatory cell infiltration area (%) was calculated and indicated. Three sections with the bars of 100 µm were used. Statistical significance was calculated by One-Way ANOVA. * P < 0.05, ** P < 0.01, and *** P < 0.001.

Fig. 2

Reversion of WalK(S221P) in VISA XN108 contributed to bacterial dissemination. (A) Dissemination of XN108 and XN108-R in organs of the infected mice. BALB/c mice (n = 7 in each group) were challenged via tail vein injection with 5 × 10⁷ CFU of the pGFP plasmid-transformed S. aureus XN108 and XN108-R, respectively. The infected mice were sacrificed on 5 d pi and their organs were isolated. The radiant efficiency of the murine organs was determined using the IVIS® Lumina LT system. The organs of the PBS-injected mice served as negative control. (B) Bacterial burdens in organs of the infected mice. After radiant efficiency determination, the kidney and liver of the infected mice were grinded to homogenate and bacteria in the tissue homogenates were counted by plate dilution method. Statistical significance was calculated by Student’s t-test. * P < 0.05 and ns represents no significance. (C) Detection of the levels of proinflammatory cytokines. Sera and organs (i.e., liver and kidney) of the XN108-pGFP-, XN108-R-pGFP-infected, and PBS-injected mice (n = 3 for each group) were respectively collected on 5 d pi, and the levels of IL-6, IL-1β, and TNF-α in the sera and organ homogenates were determined by ELISA as described in Materials and methods. Statistical significance was calculated by 2way ANOVA. ** P < 0.01, *** P < 0.001, and ns represents no significance.

Fig. 3

Transcriptomic analysis and virulence factor determination of XN108-R and XN108. (A) RNA-seq analysis showed DEGs between XN108-R and XN108. (B) Transcriptional profiles for each sample analyzed with Pairwise Euclidean distance (heat map). (C) Virulence gene set and (D) transcription regulator gene set analysis, showing upregulated genes (red) and downregulated genes in XN108-R (blue). (E) Hemolytic experiments. Sheep blood plates were inoculated with XN108 and XN108-R, cultured at 37 °C overnight, then kept at 4 °C for 24 h and photographed. (F) Western blot analysis of Hla and Hlb in S. aureus strains indicated. The total bacterial proteins served as loading control (LC). Molecular weights of the protein marker are indicated on the left. (G) Evaluation of gray values of the expressed Hla and Hlb using ImageJ software. The relative values of Hla/LC and Hlb/LC were indicated. Statistical significance was calculated by 2way ANOVA. * P < 0.05 and *** P < 0.001. (H) Slide coagulase test. More agglutinated particles were presented in XN108-R (K65) relative to those of XN108.

Fig. 4

Introduction of WalK(S221P) to S. aureus increased vancomycin resistance and attenuated bacterial virulence. E-test of vancomycin susceptibilities in (A) MRSA strains USA300 and K-USA300, and (B) MSSA strains Newman and K-Newman. RT-qPCR detection of the expression levels of virulence factors in (C) USA300 and K-USA300, and (D) Newman and K-Newman. Statistical significance was calculated by 2way ANOVA. * P < 0.05, ** P < 0.01, *** P < 0.001, and ns indicates no significance. (E) Hemolytic experiments. Sheep blood plates were inoculated with USA300 and K-USA300 or strains Newman and K-Newman, cultured at 37 °C overnight, then kept at 4 °C for 24 h and photographed. Western blot analysis of Hla and Hlb levels in (F) USA300 and K-USA300, and (H) Newman and K-Newman. Molecular weights of the protein marker are indicated on the left. LC, loading control. Evaluation of gray values of the expressed Hla and Hlb using ImageJ software. The relative values of Hla/LC and Hlb/LC in (G) USA300 and K-USA300, and (I) Newman and K-Newman were indicated. Statistical significance was calculated by 2way ANOVA. *** P < 0.001.

Fig. 5

WalK-activated WalR bound to agr promoter region to control the expression of agr in S. aureus. (A) β-galactosidase assay. The pOS1-agr^P reporter plasmid was transformed into S. aureus XN108 and XN108-R. The LacZ activity was detected and represented as mean ± SD (n = 3). Statistical significance was calculated by Student’s t-test. ***P < 0.001. (B) Predicted WalR-binding site in the promoter regions of agr P2. (C) Mutation of the predicted WalR-binding site in agr promoter regions for EMSA experiment. Interaction between agr promoter region (agr-P2) and (D) WalK-activated WalR proteins or (E) WalK(S221P)-activated WalR proteins was detected. The unlabeled DNA fragment of mecA promoter region (50 pM) was used for non-specific competition of the interaction. (F) EMSA with agr promoter mutant (agr-P2M), the sle1 promoter DNA fragment (10 pM) was used as positive control. (G) Evaluation of gray value of the bound probe in each lane using ImageJ software. The value of the free probe in the first lane (0 pM WalR protein) was used as loading control (LC), and the relative gray values of the bound probe to LC in other lanes were calculated and indicated.

Fig. 6

Agr inactivation resulted in comparable vancomycin resistance and bacterial virulence. (A) E-test of vancomycin susceptibilities in XN108/Δagr, XN108-R/Δagr, USA300/Δagr, K-USA300/Δagr, Newman/Δagr, and K-Newman/Δagr. RT-qPCR detection of the expression levels of virulence factors in (B) XN108/Δagr and XN108-R/Δagr, (C) USA300/Δagr and K-USA300/Δagr, and (D) Newman/Δagr and K-Newman/Δagr. Statistical significance was calculated by 2way ANOVA. * P < 0.05, *** P < 0.001, and ns indicates no significance. Cross streak assay for (E) XN108 and its derivatives, (F) USA300 and its derivatives, and (G) Newman and its derivatives. S. aureus RN4220 was used to represent Hlb production. The production of other hemolysins was indicated.