Staphylococcus aureus Triggers Induction of miR-15B-5P to Diminish DNA Repair and Deregulate Inflammatory Response in Diabetic Foot Ulcers

Abstract

Diabetic foot ulcers (DFUs) are a debilitating complication of diabetes in which bacterial presence, including the frequent colonizer Staphylococcus aureus, contributes to inhibition of healing. MicroRNAs (miRs) play a role in healing and host response to bacterial pathogens. However, the mechanisms by which miR response to cutaneous S. aureus contributes to DFU pathophysiology are unknown. Here, we show that S. aureus inhibits wound closure and induces miR-15b-5p in acute human and porcine wound models and in chronic DFUs. Transcriptome analyses of DFU tissue showed induction of miR-15b-5p to be critical, regulating many cellular processes, including DNA repair and inflammatory response, by suppressing downstream targets IKBKB, WEE1, FGF2, RAD50, MSH2, and KIT. Using a human wound model, we confirmed that S. aureus-triggered miR-15b-5p induction results in suppression of the inflammatory- and DNA repair-related genes IKBKB and WEE1. Inhibition of DNA repair and accumulation of DNA breaks was functionally confirmed by the presence of the pH2AX within colonized DFUs. We conclude that S. aureus induces miR-15b-5p, subsequently repressing DNA repair and inflammatory response, showing a mechanism of inhibition of healing in DFUs previously unreported, to our knowledge. This underscores a previously unknown role of DNA damage repair in the pathophysiology of DFUs colonized with S. aureus.

Figure 1.. S. aureus infection of acute wounds induces miR-15b-5p.

a. Histology shows acute wounds 4 days post-wounding and infection of acute ex-vivo human wound model. Black arrows indicate initial site of wounding, ET = epithelial tongue; red arrows indicate bacterial aggregates. Separation of epidermis from dermis was observed in S. aureus infected wounds (red dashed line). b. Immuno-localization of K6 (red) at the wound edge and epithelial tongue upon infection with S. aureus isolate from DFU in the human skin ex-vivo model demarcates epithelialization. Nuclei are visualized with DAPI (blue). White dashed lines indicate DAPI stained bacterial aggregates. Scale bars: 50μm. c. qPCR showing up-regulation of miR-15b-5p a in S. aureus infected human (hsa, Homo sapiens) ex-vivo wounds in comparison to control wounds 4 days post infection and MRSA USA300 infected pig (ssc, Sus scrofa) wounds in vivo 2 days post infection. Paired and unpaired t-test were used respectively to determine statistical significance (n=3 independent experiments, * = p-value<0.05). d. qPCR showing suppression of miR-15b-5p in human uninfected acute wound model. Data is normalized to the 0h time point at which wounds were created (n=3 independent experiments) e. Agarose gel showing the amplification of the nuc gene in DFU samples. AW= human acute ex-vivo wound (negative control), AW + S. a. = human acute ex-vivo wound infected with S. aureus (positive control). f. qPCR showing up-regulation of miR-15b-5p in DFU (n=12 per group, FS = foot skin, Mann-Whitney U-test was used to determine statistical significance, ** = p-value<0.01).

Figure 2.. Microarray analysis of differentially expressed genes in DFUs vs. FS and functional enrichment analysis.

a. Heatmap of differentially expressed genes in DFUs vs. FS. The majority of DFU regulated genes were suppressed (75%). b. Top enriched functions from IPA functional enrichment analysis performed on DFU regulated genes identifies DNA repair mechanisms as the top regulated. Z-scores (activation score) represent the predicted directionality of the enriched function (Z-scores ≥ 2 are considered as induced and ≤ −2 are considered as suppressed). c. Functional assessment of DNA repair suppression was perform by quantification of pH2AX using western blot. pH2AX confirms the presence of DNA double strand breaks in DFUs, but not control FSs. Quantification of all samples shown in bar graph (n=5 per group) d. qPCR showing IL1A, a pro-inflammatory cytokine that can be induced by DNA damage, is induced in DFUs (n=5 per group, Mann-Whitney U-test was used to determine statistical significance, * = p-value<0.05, ** = p-value < 0.01).

Figure. 3.. Network analysis and validation of DFU de-regulated genes.

a. Network analysis of the genes deregulated in DFUs was built using IPA focusing on genes with known regulatory functions (blue=suppression; red=induction). b. qPCR validation of microarray data. IKBKB, WEE1, MSH2, RAD50, FGF2, and KIT showed significant down-regulation in DFU compared to FS. IGF1 showed a trend of down-regulation but it did not reach statistical significance. c. Western blot confirming suppression of IKBKB in DFU compared to FS; graph represents quantification of all samples (n=5 per group). d. Immuno-fluorescence staining showing decreased expression of MSH2 in DFUs compared to FS. E=epidermis; D=Dermis; 100X Scale bars: 100 μm, 400X Scale bars: 50 μm, (n=10 per group, Mann-Whitney U test was used to determine statistical significance, * = p-value<0.05, ** = p-value < 0.01, *** = p-value < 0.001).

Figure 4.. miR-15b-5p, predicted to be induced in DFUs, and suppression of its targets, DNA repair and inflammatory regulatory genes.

a. IPA identified miR-15b-5p as the top enriched miR predicted to target DFU regulated genes from microarrays. b. miR-15b-5p predicted targets comprise regulators of the inflammatory response and genes involved in DNA repair. c. IKBKB and WEE1, are down-regulated in HaCaT cells transfected with mimic miR-15b-5p. MSH2 and RAD50 showed a trend of down-regulation but did not reach statistical significance. ctrl = mimic negative control #1. d. IKBKB and WEE1 are down-regulated in S. aureus infected human ex-vivo wounds in comparison to control un-infected wounds. (n=3 independent experiments, paired t-test was used to determine statistical significance, * = p-value<0.05, ** = p-value < 0.01). e. Luciferase reporter assay showing miR-15b-5p directly targets the 3’UTRs of IKBKB and WEE1. ctrl = mimic negative control #1. (n=3, unpaired t-test was used to determine statistical significance, ** = p-value ≤ 0.01, *** = p-value ≤ 0.001).

Figure. 5.. Diagram summarizing how miR-15b-5p orchestrates processes involved in pathophysiology of DFUs.

Persistent S. aureus colonization in DFUs leads to over-expression of miR-15b-5p resulting in increased DNA damage through suppression of WEE1 and chronic sub-optimal inflammation by targeting IKBKB. DNA repair mechanisms are further inhibited by down-regulation of multiple genes in DFUs including MSH2, RAD50, KIT, FGF2 and TP53. DNA damage feeds into a positive feedback loop by causing the release of IL-1A.