Human OTULIN haploinsufficiency impairs cell-intrinsic immunity to staphylococcal α-toxin

Abstract

The molecular basis of interindividual clinical variability upon infection with Staphylococcus aureus is unclear. We describe patients with haploinsufficiency for the linear deubiquitinase OTULIN, encoded by a gene on chromosome 5p. Patients suffer from episodes of life-threatening necrosis, typically triggered by S. aureus infection. The disorder is phenocopied in patients with the 5p- (Cri-du-Chat) chromosomal deletion syndrome. OTULIN haploinsufficiency causes an accumulation of linear ubiquitin in dermal fibroblasts, but tumor necrosis factor receptor-mediated nuclear factor κB signaling remains intact. Blood leukocyte subsets are unaffected. The OTULIN-dependent accumulation of caveolin-1 in dermal fibroblasts, but not leukocytes, facilitates the cytotoxic damage inflicted by the staphylococcal virulence factor α-toxin. Naturally elicited antibodies against α-toxin contribute to incomplete clinical penetrance. Human OTULIN haploinsufficiency underlies life-threatening staphylococcal disease by disrupting cell-intrinsic immunity to α-toxin in nonleukocytic cells.

Figure 1.. OTULIN haploinsufficiency and its molecular characterization.

(A) Genome-wide enrichment in rare, predicted deleterious variants in the cohort of N=105 patients with severe staphylococcal disease, relative to N=1,274 patients with mycobacterial disease. (B) Pedigrees of the kindreds presenting severe necrotizing staphylococcal disease and carrying heterozygous mutations of OTULIN. (C) Pedigrees of the kindreds carrying heterozygous mutations of OTULIN and presenting necrosis triggered by other etiologies. (D) Pedigrees of the kindreds of ORAS probands carrying biallelic mutations of OTULIN. (E) Functional population genetics based on minor allele frequencies (MAFs) reported in the gnomAD database and the NF-κB inhibitory capacity of the OTULIN variants, as assessed with a luciferase reporter system in transiently transfected HEK293T cells stimulated with TNF (datapoints represent the mean of N=4–5 per variant). (F) Pedigrees of the kindreds of the 5p- syndrome probands carrying deletions with a breakpoint centromeric to OTULIN. The patients are shown in ascending age order (left to right, top to bottom). (G) Pedigree of the 5p- syndrome kindred with probands carrying a deletion with a breakpoint telomeric to OTULIN. WT: wild-type allele; MT: mutant allele; LOF: loss of function. See also Figure S1 and Table S1, S2.

Figure 2.. OTULIN haploinsufficiency does not impair the hematopoietic immune system.

(A) Aggregate uniform manifold approximation and projection (UMAP) plots of peripheral blood mononuclear cells (PBMCs) from patients with autosomal dominant OTULIN deficiency and healthy controls, as assessed by cytometry by time-of-flight (CyTOF). (B) Principal component (PC) analysis plot of the transcriptional profile of PBMCs at baseline and after incubation with various stimuli, in comparison with healthy controls. Arrows have been added for visual support of the direction of variance after stimulation. NS: not stimulated; HKSA: heat-killed S. aureus. See also Figure S2.

Figure 3.. OTULIN gene dosage-dependent accumulation of M1-Ub in fibroblasts.

(A) Expression of OTULIN, HOIP, HOIL-1, and SHARPIN in primary dermal fibroblasts (PDFs). (B) Relative abundance of OTULIN peptides in PDFs, as determined by mass spectrometry analysis (median). (C) Accumulation of aggregates containing M1-Ub in immortalized fibroblasts, as assessed by immunohistochemistry. Representative images. (D) Quantification of the aggregate accumulations seen in (D) (median). (E) Quantification of aggregates containing M1-Ub in immortalized fibroblasts after rescue with wild-type OTULIN, as compared with cells rescued with an empty virus (mean ±SD). (F) Abundance of M1-Ub, as determined by AQUA-MS/MS analysis of immortalized fibroblasts (median). Statistical significance was calculated by ANOVA with Dunnett post hoc correction for multiple comparisons. See also Figure S3.

Figure 4.. OTULIN-dependent accumulation of caveolin-1 in fibroblasts.

(A) Transcriptional profile of unstimulated primary dermal fibroblasts (PDFs), as assessed by RNA sequencing, expressed as log₂-fold changes (FC) relative to the mean value for controls. Genes satisfying the threshold for statistical significance in comparisons of ORAS patients to controls are shown. (B) Proteomic profiles of unstimulated PDFs, as assessed by mass spectrometry on whole-cell lysates, expressed as Z-scores. Genes satisfying the threshold for statistical significance in comparisons of ORAS patients to controls are shown. (C) STRING analysis of the cluster of proteins with differential abundances in various genotypes identified in (B). (D) Accumulation of SDS-resistant high-MW caveolin-1-containing complexes in PDF-whole cell lysates (WCLs). (E) High-MW caveolin-1-containing complex intensities relative to β-actin intensities in PDF-WCLs, normalized against the mean value of healthy controls (datapoints indicate the mean of N=2 replicates per patient, median per group). (F) High-MW caveolin-1-containing complex intensities relative to β-actin intensities in PDF-WCLs treated with an sgRNA pool targeting OTULIN (sgOTU), normalized against the mean value in those treated with a non-targeting control sgRNA (sgNTC) (median, N=4). The statistical significance of differences was assessed in Student’s t-tests. See also Figure S4.

Figure 5.. Polyubiquitination of caveolin-1.

(A) Analysis of K63-Ub on caveolin-1 by LC-MS/MS on purified caveolin-1 from primary dermal fibroblast (PDF) whole cell lysates (WCLs). Caveolin-1 was analyzed in solution, in the high-molecular weight (MW) complex fraction, and in the monomeric (Mono) fraction. (B) High-MW K63-Ub complex intensities relative to caveolin-1 intensities in recombinant deubiquitinase-treated purified caveolin-1 from PDFs. (C) High-MW K63-Ub complex intensities relative to caveolin-1 intensities in purified caveolin-1 from PDFs after rescue with CYLD and/or OTULIN. (D) Intensities of CYLD and HOIP bound to M1-Ub, as detected in TUBE pull-downs from immortalized fibroblasts, relative to intensities of total CYLD and HOIP, respectively. Statistical significance was calculated by ANOVA with Dunnett post hoc correction for multiple comparisons. See also Figure S5.

Figure 6.. OTULIN haploinsufficiency impairs intrinsic immunity to α-toxin in fibroblasts.

(A) Colocalization of α-toxin with ADAM10 and caveolin-1 in primary dermal fibroblasts (PDFs), visualized in high-resolution images acquired by stochastic optical reconstruction microscopy, after 30 minutes of incubation in the presence or absence of α-toxin. (B) Binding of α-toxin in PDFs, as detected by flow cytometry. (C) Viability of PDFs following 2.5 h of incubation with culture supernatant from S. aureus (datapoints indicate the mean of N=3 replicates per patient, median per group). (D) Viability of PDFs following 24 h of incubation with recombinant microbial toxins (N=2–4 per group, mean ±SD per group). (E) Viability of PDFs from a 5p- syndrome patient with a breakpoint telomeric to OTULIN following 24 h of incubation with recombinant α-toxin, superimposed on that for healthy controls and 5p- patients with a breakpoint centromeric to OTULIN from panel (F) (N=3, mean ±SD). (F) Cell-surface metalloprotease activity in PDFs induced by wild type α-toxin (WT) or a toxoid mutant (H35L) following the treatment of cells with an ADAM10 inhibitor or carrier (datapoints indicate the mean of N=3 replicates per patient, median per group). (G) Viability of OTULIN-deficient PDFs treated with an sgRNA pool targeting CAV1 (sgCAV1), CAV2 (sgCAV2), or a non-targeting control sgRNA (sgNTC) following exposure to α-toxin (0.12 μg/mL; N=3, median). The statistical significance of differences was assessed in Student’s t-tests (C), or by ANOVA with Bonferroni (F) or Dunnett (G) post hoc corrections for multiple comparisons. See also Figure S6.

Figure 7.. α-toxin-neutralizing antibodies rescue OTULIN haploinsufficiency.

(A) Capacity of plasma at a dilution of 1:300 to neutralize the hemolytic activity of α-toxin (20 ng/mL), pneumolysin (PLY, 0.67 ng/mL), and streptolysin O (SLO, 42 ng/mL) in rabbit erythrocytes. (B) Anti-α-toxin immunoglobulin levels in plasma at a dilution of 1:750. We analyzed plasma from adult patients, relatives, and controls. The statistical significance of differences was determined by ANOVA with Dunnett post hoc correction for multiple comparisons. See also Figure S7.