Research ArticleFUNGAL INFECTIONS

A systems genomics approach identifies SIGLEC15 as a susceptibility factor in recurrent vulvovaginal candidiasis

See allHide authors and affiliations

Science Translational Medicine  12 Jun 2019:
Vol. 11, Issue 496, eaar3558
DOI: 10.1126/scitranslmed.aar3558
  • Fig. 1 Schematic overview of the study.

    We carried out WES on two independent cohorts of patients with RVVC and healthy controls to identify genes and pathways associated with RVVC. We also stimulated human PBMCs in vitro to determine genetic variants that influence C. albicans–induced cytokine release. By combining both approaches, we identified genes and pathways important in RVVC pathogenesis that we validated in vitro and in vivo.

  • Fig. 2 Analysis of RVVC-related cSNPs (RcSNPs) and cSNPs in both RVVC cohorts.

    (A) Genes containing common RCCV-associated SNPs (RcSNPs) found in both the Southern European and Northern European cohorts (SEC and NEC, respectively) with (B) GO term characterization of the overlapping genes. (C) Overlap between SNPs that are associated with RVVC (RcSNPs) and SNPs that influenced cytokine production after stimulation (cQTL) in SEC and NEC. (D) GO term characterization of the overlapping genes.

  • Fig. 3 Functional assessment of SIGLEC15 in Candida immune response in vitro.

    (A) Cytokine production after stimulation with heat-killed C. albicans yeast for 24 hours or 7 days. TNFα, IL-6, IL-1β, IL-17A, IL-22, IFNγ, and IL-23 concentrations were measured in culture supernatant after 24 hours (IL-6, IL-1β, TNFα, and IL-23) or 7 days (IL-17A, IL-22, and IFNγ). PBMCs of individuals with different genotypes (TT, n = 25; CT + CC, n = 9) for the SIGLEC15 variant rs2919643 were compared in a dominant model. (B) ROS production and C. albicans killing by PBMCs with different rs2919643 genotypes. AUC, area under curve. (C) RNA sequencing of SIGLEC15 in PBMCs stimulated with either culture medium (red) or C. albicans (black) for 4 hours. (D) Gene expression of SIGLEC family members profiled in PBMCs of eight healthy donors after stimulation with C. albicans for 4 and 24 hours. (E) Cytokine production capacity in SIGLEC15 small interfering RNA (siRNA)–treated human granulocyte-macrophage colony-stimulating factor–derived macrophages after C. albicans stimulation for 24 hours. TNFα, IL-6, and prostaglandin E2 (PGE2) concentrations were measured in culture supernatants of macrophages stimulated with C. albicans yeast for 24 hours after a transfection period of 24 hours with either SIGLEC15 or control siRNA. (F) Cytokine production capacity of human PBMCs silenced for SIGLEC15 for 24 hours and then stimulated for 24 hours. (G) Silencing efficiency in PBMCs at baseline and after stimulation with Candida for 24 hours. (H) Expression of SIGLEC15 in healthy controls (n = 4) and patients (n = 3) suffering from RVVC at baseline and after stimulation with Candida (yeast and hyphae) for 7 days. (I) Expression of NLPR3 and IL1B in HeLa cells transfected with wild-type (WT) or mutated SIGLEC15 plasmids. (J) Binding of recombinant SIGLEC15 to heat-killed C. albicans assayed by mean fluorescent intensity (MFI). Ab, antibody. (K) Heat-killed C. albicans (blue) with secondary antibody [anti-mouse–phycoerythrin (PE); green] and an antibody against β-glucan (red). (L) Heat-killed C. albicans (blue) with secondary antibody (green) and recombinant SIGLEC7 (red). (M) Live C. albicans (blue), live C. albicans with secondary antibody (anti-IgG Alexa 647; green), and live C. albicans with recombinant SIGLEC15-FC and secondary antibody (red). Data are shown as means ± SEM. Mann-Whitney U test, *P < 0.05, **P < 0.01, and ***P < 0.001. ns, not significant. Y axes vary across plots.

  • Fig. 4 ROS induced by neuramindase-treated fungi.

    ROS in PBMCs induced by (A) C. albicans (1 × 106/ml) and (B) A. fumigatus (1 × 107/ml) either treated or untreated with neuraminidase. Measurements of ROS induction in relative light units (RLU) per second were taken after the start of treatment for 1 hour in intervals of 2.23 min. (C) Area under the curve (AUC) of both experiments in (A) and (B). Data are shown as means ± SEM where *P < 0.05, **P < 0.01, and ***P < 0.001. ns, not significant. (D) Cytokine induction by PBMCs after 24 hours or 7 days of stimulation with C. albicans either treated or untreated with neuraminidase.

  • Fig. 5 SIGLEC15 activity in vaginal candidiasis.

    C57BL/6 mice (n = 6) were intravaginally inoculated with 5 × 106 Candida in 20 μl of phosphate-buffered saline (PBS). (A) Immunohistochemistry staining of vaginal sections with antibodies to SIGLEC15. (B) Siglec15 expression by real-time reverse transcription polymerase chain reaction (RT-PCR) relative to uninfected mice. (C to F) Infected mice were intravaginally treated with unmodified siRNA against murine Siglec15 (siSiglec15; 10 μg/kg) or equivalent doses of nonspecific duplex scrambled control siRNA. Intravaginal siRNA was given once the day before infection and every 2 days thereafter until sacrifice at 3, 7, or 14 dpi. One-way analysis of variance (ANOVA) with Bonferroni correction is shown. (C to E) Mice were assessed for (C) vaginal fungal burden [log10 colony-forming units (CFUs)/100 μl vaginal fluid (VF) ± SD], (D) PMN recruitment in vaginal fluid (×100 magnification), and (E) vaginal pathology as shown by periodic acid–Schiff staining of sections. (F) IL-1β/IL-1Ra concentrations (pg/mg, cytokine/total proteins for each sample) in vaginal fluid. (G) Nlrp3, Nlrc4 gene expression in ex vivo murine vaginas by real-time RT-PCR. (H) IL-6, IL-17A, IL-22, and IL-10 concentrations (pg/mg, cytokine/total proteins for each sample) in vaginal fluid. Scale bars, 200 μm. Data are presented as means ± SD. Two-way ANOVA with Bonferroni post hoc test was used unless otherwise indicated. *P < 0.05, **P < 0.001, ***P < 0.001, and ****P < 0.0001.

  • Table 1 Genes with FRV only present in RVVC cases.

    Genes with at least two FRVs only present in candidiasis cases (both NEC and SEC) but not controls were tested using a burden test. This test compares the number of FRVs between cases and controls present within a single gene.

    GeneP value (FRV
    burden test)
    FRVs in NECFRVs in SECAll FRVs
    IGFN10.002358
    ATP6AP1L0.005437
    ADAMTS200.005347
    SALL10.011426
    YTHDC20.011336
    MMEL10.011246
    MAN2A20.011246
    EVC20.011246
    PKP30.011246
    SMG80.023325
    SLC45A10.023325
    SCAF80.023325
    OR10G30.023325
    SLC5A110.023325
    HSD17B40.023325
    CCKAR0.023325
    OR4C160.023235
    PRPF40B0.023235
    NFKB10.023235
    ABHD10.023235
    RASAL20.023235
    DACH20.023235
    AREL10.048224
    LANCL10.048224
    CHAMP10.048224
    AMPH0.048224
    HACL10.048224
    FRK0.048224
    CLCN60.048224
    PPEF20.048224
    RFT10.048224
    BBS50.048224
    SYNPO20.048224
    CYP4V20.048224
    CCDC90B0.048224
    CLIP20.048224
    ABCG40.048224
    KIAA14070.048224
    DSG20.048224
  • GeneSequences
    hSIGLEC15Sense: CGCGGATCGTCAACATCTC
    αSense: GTTCGGCGGTCACTAGGTG
    hGAPDHSense: AGGGGAGATTCAGTGTGGTG
    αSense: CGACCACTTTGTCAAGCTCA

Supplementary Materials

  • stm.sciencemag.org/cgi/content/full/11/496/eaar3558/DC1

    Materials and Methods

    Fig. S1. Manhattan plot of both RVVC cohorts.

    Fig. S2. Genomic region of SIGLEC15 and surrounding genetic variants in linkage disequilibrium with rs2919643.

    Fig. S3. Validation of rs2919643 in an independent cohort.

    Fig. S4. Genotype stratification after stimulation with S. aureus.

    Fig. S5. SIGLEC15 expression in different cell types.

    Fig. S6. Binding of SIGLEC15 to different Candida strains.

    Fig. S7. Flow cytometry of mouse cells after Candida stimulation.

    Fig. S8. Silencing efficiency in mouse siRNA experiments.

    Table S1. RVVC-associated cSNPs with concordant effects in both cohorts.

    Table S2. GO characterization of genes that have cSNPs associated with RVVC in both cohorts.

    Table S3. FRVs.

    Table S4. Rare variants in both cohorts after multiple testing correction.

    Table S5. GO characterization of genes associated with both RVVC risk and QTL.

    Table S6. Proxies for rs2919643 in SIGLEC15 and their degree of linkage disequilibrium.

  • The PDF file includes:

    • Materials and Methods
    • Fig. S1. Manhattan plot of both RVVC cohorts.
    • Fig. S2. Genomic region of SIGLEC15 and surrounding genetic variants in linkage disequilibrium with rs2919643.
    • Fig. S3. Validation of rs2919643 in an independent cohort.
    • Fig. S4. Genotype stratification after stimulation with S. aureus.
    • Fig. S5. SIGLEC15 expression in different cell types.
    • Fig. S6. Binding of SIGLEC15 to different Candida strains.
    • Fig. S7. Flow cytometry of mouse cells after Candida stimulation.
    • Fig. S8. Silencing efficiency in mouse siRNA experiments.
    • Table S1. RVVC-associated cSNPs with concordant effects in both cohorts.
    • Table S2. GO characterization of genes that have cSNPs associated with RVVC in both cohorts.
    • Table S3. FRVs.
    • Legend for table S4
    • Table S5. GO characterization of genes associated with both RVVC risk and QTL.
    • Table S6. Proxies for rs2919643 in SIGLEC15 and their degree of linkage disequilibrium.

    [Download PDF]

    Other Supplementary Material for this manuscript includes the following:

    • Table S4 (Microsoft Excel format). Rare variants in both cohorts after multiple testing correction.

Stay Connected to Science Translational Medicine

Navigate This Article