Sulphonylurea usage in melioidosis is associated with severe disease and suppressed immune response

Abstract

Background: Melioidosis is a problem in the developing tropical regions of Southeast Asia and Northern Australia where the the Gram negative saprophytic bacillus Burkholderia pseudomallei is endemic with the risk of fulminant septicaemia. While diabetes mellitus is a well-established risk factor for melioidiosis, little is known if specific hypoglycemic agents may differentially influence the susceptibility and clinical course of infection with B. pseudomallei (Bp).

Methodology/principal findings: In this cohort study, patients with pre-existing diabetes and melioidosis were retrospectively studied.

Outcome measures: mortality, length of stay and development of complications (namely hypotension, intubation, renal failure and septicaemia) were studied in relation to prior diabetic treatment regimen. Peripheral blood mononuclear cells (PBMC) from diabetic patients and healthy PBMC primed with metformin, glyburide and insulin were stimulated with purified Bp antigens in vitro. Immune response and specific immune pathway mediators were studied to relate to the clinical findings mechanistically. Of 74 subjects, 44 (57.9%) had sulphonylurea-containing diabetic regimens. Patient receiving sulphonylureas had more severe septic complications (47.7% versus 16.7% p = 0.006), in particular, hypotension requiring intropes (p = 0.005). There was also a trend towards increased mortality in sulphonylurea-users (15.9% versus 3.3% p = 0.08). In-vitro, glyburide suppressed inflammatory cytokine production in a dose-dependent manner. An effect of the drug was the induction of IL-1R-associated kinase-M at the level of mRNA transcription.

Conclusion/significance: Sulphonylurea treatment results in suppression of host inflammatory response and may put patients at higher risk for adverse outcomes in melioidosis.

Figure 1. (A–D) Peripheral blood mononuclear cells (PBMC) from healthy donors were incubated with metformin, glyburide or insulin at the indicated doses (0.0001–0.01 mg/mL or control media) for an hour after which purified Bp antigen (1 µg/mL) was added for stimulation.

(E) PBMC from diabetic patients receiving sulphonylurea (SU)-containing regimen (n = 9) versus those not on a non-SU regimen (n = 13) were stimulated with purified Bp antigen. Cytokine responses were measured at 24 h for TNF-α and IL-1b and at 48 h for IL-10 and IFN-γ. The in-vitro data are cumulative results of at least 3 sets of experiments (n≥9) and expressed as means±standard errors of the means (SEM). * p<0.05 compared to the respective control (white bar) for (A–D) or between SU and non-SU group for (E).

Figure 2. IL-1b and TNF-α mRNA transcriptions were suppressed in glyburide-primed PBMC as compared to metformin and insulin treatment.

* p<0.05 relative to the ratio of fold increase in the mRNA levels of cells primed only in control media in absence of diabetic drugs.

Figure 3. (A–D) Western blot from cell lysates of PBMC primed with diabetic drugs for 1 hour and then stimulated with purified Bp antigen for 30 minutes.

The membranes were stripped and re-probed for MyD88, total and phosphorylated (p-) JNK, ERK and p38. Beta-actin was used as the loading control. Result is representative from 2 sets of experiment.

Figure 4. (A) IRAK-M mRNA expression is elevated by glyburide.

* p<0.05 relative to the ratio of fold increase in the mRNA levels of cells primed only in control media in absence of diabetic drugs. (B) Flow cytometry of glyburide-treated PBMC showed increased expression of IRAK-M in CD-14 positive monocytes following stimulation with purified Bp antigen. The cells were surface labeled with anti-CD14-V450 antibody, permeabilized and then labeled with primary goat anti-IRAK-M antibody. Secondary labeling with donkey anti-goat-FITC antibody was performed and cells were analyzed by flow cytometry. Result from a representative set of experiment is shown.