Compartmentalized Cytokine Responses in Hidradenitis Suppurativa

Abstract

Background: Favorable treatment outcomes with TNF blockade led us to explore cytokine responses in hidradenitis suppurativa (HS).

Methods: Blood monocytes of 120 patients and 24 healthy volunteers were subtyped by flow cytometry. Isolated blood mononuclear cells (PBMCs) were stimulated for cytokine production; this was repeated in 13 severe patients during treatment with etanercept. Cytokines in pus were measured.

Results: CD14brightCD16dim inflammatory monocytes and patrolling monocytes were increased in Hurley III patients. Cytokine production by stimulated PBMCs was low compared to controls but the cytokine gene copies did not differ, indicating post-translational inhibition. The low production of IL-17 was restored, when cells were incubated with adalimumab. In pus, high concentrations of pro-inflammatory cytokines were detected. Based on the patterns, six different cytokine profiles were discerned, which are potentially relevant for the choice of treatment. Clinical improvement with etanercept was predicted by increased production of IL-1β and IL-17 by PBMCs at week 8.

Conclusions: Findings indicate compartmentalized cytokine expression in HS; high in pus but suppressed in PBMCs. This is modulated through blockade of TNF.

Fig 1. Monocyte subsets in HS.

Absolute counts of (A) total circulating monocytes, (B) CD14^bright/CD16^neg monocytes, (C) CD14^bright/CD16^dim monocytes, and (D) patrolling monocytes are shown for healthy volunteers (n = 14), patients with Hurley I HS (n = 37), with Hurley II HS (n = 37) and with Hurley III HS (n = 44). P values of comparisons by ANOVA and post hoc Bonferroni corrections: ^a0.004 vs healthy controls; ^b0.026 vs Hurley I; ^c0.001 vs healthy controls; ^d0.012 vs Hurley I; ^e0.008 vs Hurley II; ^f0.037 vs healthy controls.

Fig 2. Stimulation of circulating monocytes.

Production of (A) tumour necrosis factor-alpha (TNFα), (B) interleukin (IL)-1β, (C) IL-6, (D) IL-10, (E) IL-10, (F) IL-22, and (G) IL-1ra by peripheral blood mononuclear cells (PBMCs) of eight healthy controls and of 80 patients with hidradenitis suppurativa (HS). PBMCs were stimulated with LPS of Escherichia coli O55:B5; with heat-killed Candida albicans (HKCA), and with heat-killed Staphylococcus aureus (HKSA). (H) Gene transcripts of TNF and of IL-1 after stimulation with LPS. P values refer to comparisons with healthy controls by the Mann-Whitney U test; pNS: difference not-significant.

Fig 3. Stimulation of IL-10 and IL-17 in relation with disease severity

Production of IL-10 and IL-17 was stimulated in peripheral blood mononuclear cells (PBMCs) of patients with hidradenitis suppurativa (HS) by LPS of Escherichia coli O55:B5, heat-killed Candida albicans (HKCA) and heat-killed Staphylococcus aureus (HKSA). Patients are divided into those with Hurley I or Hurley II disease stage (n = 50) and into those with Hurley III disease stage (n = 40). P values refer to comparisons between disease stages by the Mann Whitney U test; pNS: difference not-significant.

Fig 4. Modulation of cytokine release by adalimumab.

Peripheral blood mononuclear cells (PBMCs) from of eight healthy controls and 40 patients with hidradenitis suppurativa were stimulated for the production of (A) interleukin (IL)-1β with LPS of Escherichia coli O55:B5; (B) IL-6 with LPS of E.coli O55:B5; (C) IL-10 with LPS of E.coli O55:B5; (D) IL-10 with heat-killed Candida albicans (HKCA); (E) IL-10 with heat-killed Staphylococcus aureus (HKSA); (F) IL-1ra with HKSA; (G) IL-17 with HKCA; (H) IL-17 with HKSA. P values refer to comparisons of cytokine production without/with adalimumab; pNS: difference not-significant.

Fig 5. Cytokine concentrations in pus of patients with HS.

(A) Concentrations of tumour necrosis factor-alpha (TNFα), interleukin (IL)-1β, IL-1α, IL-6, IL-10, IL-17 and Il-1ra were measured in the pus of the lesions of 28 patients of Hurley III disease stage. Lines represent medians. Correlations between (Β) the number of patrolling monocytes and pus TNFα; (C) the number of patrolling monocytes and pus IL-1β; and (D) IL-6 and years since disease onset. Classification patterns of patients according to pus cytokines is shown in panel E. Spearman’s coefficients of correlation and p values of significance are provided.

Fig 6. Changes of cytokine production and of C Reactive Protein with the clinical course of HS.

PBMCs were isolated and serum was sampled from patients with Hurley III disease at baseline and after 8 weeks of treatment with etanercept. Patients are divided into those experiencing worsening of symptoms (n = 6) or clinical improvement (n = 7) at the end of treatment. No change was defined when disease flare-ups persisted with heavy purulence. Improvement was defined as decrease of Sartorius score more than 30% from the baseline. PBMCs were stimulated with LPS of Escherichia coli O55:B5; with heat-killed Candida albicans (HKCA); and with heat-Staphylococcus aureus (HKSA) for the production of (A) tumour necrosis factor-alpha (TNFα), (B) interleukin (IL)-1β, (C) IL-17 and (D) C Reactive Protein (CRP). P values refer to comparisons of change of cytokine production from the baseline between improved and worsened cases by the Mann-Whitney U test.

Fig 7. A proposed scheme of pathogenesis of HS.

Patients with HS have monocytosis but cytokine responses of inflammatory monocytes and T17 lymphocytes are inhibited. This inhibition is mediated for T17 cells by tumour necrosis factor-alpha (TNFα). When cells migrate to tissues from the blood vessels through the vascular endothelium, capacity for excess production of TNFα, interleukin (IL)-1β, IL-1α, IL-6 and IL-17 is restored.