Autoantibodies Present in Hidradenitis Suppurativa Correlate with Disease Severity and Promote the Release of Proinflammatory Cytokines in Macrophages

Abstract

Hidradenitis suppurativa (HS), also known as acne inversa, is a debilitating inflammatory skin disorder that is characterized by nodules that lead to the development of connected tunnels and scars as it progresses from Hurley stages I to III. HS has been associated with several autoimmune diseases, including inflammatory bowel disease and spondyloarthritis. We previously reported dysregulation of humoral immune responses in HS, characterized by elevated serum total IgG, B-cell activation, and antibodies recognizing citrullinated proteins. In this study, we characterized IgG autoreactivity in HS sera and lesional skin compared with those in normal healthy controls using an array-based high-throughput autoantibody screening. The Cy3-labeled anti–human assay showed the presence of autoantibodies against nuclear antigens, cytokines, cytoplasmic proteins, extracellular matrix proteins, neutrophil proteins, and citrullinated antigens. Most of these autoantibodies were significantly elevated in stages II–III in HS sera and stage III in HS skin lesions compared with those of healthy controls. Furthermore, immune complexes containing both native and citrullinated versions of antigens can activate M1 and M2 macrophages to release proinflammatory cytokines such as TNF-α, IL-8, IL-6, and IL-12. Taken together, the identification of specific IgG autoantibodies that recognize circulating and tissue antigens in HS suggests an autoimmune mechanism and uncovers putative therapeutic targets.

INTRODUCTION

Hidradenitis suppurativa (HS) is a chronic inflammatory disease characterized by the development of painful nodules with malodorous purulent drainage that progresses to interconnected tunnels and scarring in intertriginous areas (Saunte and Jemec, 2017). HS is estimated to affect around 1% of the population (range from 0.05–4.10%) and tends to affect more women of African American descent (Garg et al., 2017; Lee et al., 2017; Reeder et al., 2014; Saunte and Jemec, 2017; Vlassova et al., 2015). The etiology of HS remains unknown, but obesity, smoking, and hormonal factors have been linked to the disease (Saunte and Jemec, 2017). Genetic mutations in some members of the Notch signaling pathway have been identified in a fraction of HS cases (Ingram et al., 2013; Li et al., 2020; Liu et al., 2016; Melnik and Plewig, 2013; Wang et al., 2010).

Dysregulation of the innate and adaptive arms of the immune system has been reported in HS. Neutrophils in HS are prone to form neutrophil extracellular traps (NETs), externalizing molecules that can activate and promote pathogenic immune responses (Byrd et al., 2019). B cells and plasma cells are increased in HS skin lesions, and upregulated cytokines, such as IL-10, can promote B-cell differentiation. Given these observations, one emerging concept in the pathophysiology of HS is the identification of plasma cells, B cells, and the potential role of autoimmune responses in HS (Byrd et al., 2019; Constantinou et al., 2019; Frew et al., 2020; Gudjonsson et al., 2020; Lowe et al., 2020). Total IgG has been reported to be elevated in HS sera (Byrd et al., 2019; Frew et al., 2018; Hoffman et al., 2018; Musilova et al., 2020; van der Zee et al., 2012; van Straalen, 2020), and serum antibodies recognizing citrullinated proteins are prevalent in these patients (Byrd et al., 2019). Furthermore, IgG deposition as well as IgG-positive plasma cells have been reported in HS skin lesions (Byrd et al., 2019). Whether these IgGs can recognize other self-antigens associated with autoimmune responses in systemic autoimmune diseases and their pathogenic role in HS has not been systematically explored and has been challenged by various groups (Frew, 2020; Frew et al., 2020).

A multidisciplinary approach to the care of patients with HS has been proposed (Garg et al., 2021; Narla et al., 2020). To this end, understanding the mechanisms of immune dysregulation and their role in organ dysfunction may lead to a better understanding of HS and its related comorbidities and disease associations. In this article, we provide a comprehensive analysis of autoantibodies present in the serum and skin lesions from patients with HS with different Hurley stages. We show that various autoantibodies correlated with disease severity and that specific immune complexes (ICs) can activate myeloid cells to promote a proinflammatory environment in HS.

RESULTS

Autoantibodies are present in HS serum and correlate with disease severity

We previously reported the presence of antibodies recognizing citrullinated peptides in patients with HS (Byrd et al., 2019). Furthermore, B cells from patients with HS display an activated phenotype, suggesting that the adaptive immune system is dysregulated in these patients (Hoffman et al., 2018; Lowe et al., 2020) To explore the presence of a broader spectrum of autoantibodies in patients with HS, we screened HS sera for known autoantigens—several known to be targeted by the immune system in systemic autoimmune diseases—using a bead assay. We found that antibodies against nuclear antigens such as double-stranded DNA (dsDNA), nucleolin, and La/SSB were significantly elevated in patients with HS compared with those in healthy volunteers (Figure 1a–c). When HS serum samples were stratified by Hurley stage, we found that anti-dsDNA, anti-nucleolin, and anti-La/SSB antibodies correlate with disease severity (Figure 1f–h). Specifically, these antibodies were significantly elevated in stage II of the disease, suggesting that cell death events may be increased during this stage and may lead to dysregulation of the adaptive immune system. We corroborated the presence of antibodies against nuclear antigens utilizing the antinuclear antibody (ANA) Hep-2 assay (Figure 1k), which supported the presence of ANAs at more severe stages of the disease. Confirming our previous observations that a subset of HS antibodies recognizes citrullinated antigens, we detected antibodies against citrullinated fibrinogen to be significantly elevated in patients with HS compared with those in control (Figure 1d). Antibodies against MDA5, which are specific to patients with a specific subset of idiopathic inflammatory myopathies, were also significantly elevated in patients with HS compared with those in healthy controls (Figure 1e) and correlated with disease severity (Figure 1i and j). Unsupervised clustering of specific autoantibodies showed distinct enrichment in association with Hurley stages II and III. Specifically, antibodies recognizing chromatin, total histones, collagen III, U1, small nuclear ribonucleoproteins, PL12, protein arginine deiminases (PAD)4, and total citrullinated histones were significantly elevated in stages II and III, whereas stage I and control samples displayed minimal detection of these antibodies (Figure 1l). Taken together, patients with HS develop antibodies recognizing a spectrum of self-antigens, in association with disease severity.

Figure 1. Autoantibodies recognizing nuclear and citrullinated antigens are present in HS serum.

Ctrl (n = 9) and HS (stage I, n = 6; stage II, n = 7; stage III, n = 8) sera were tested for the presence of autoantibodies recognizing (a) dsDNA, (b) nucleolin, (c) La/SSB, (d) citrullinated fibrinogen, and (e) MDA5. Results are the mean SEM, *P < 0.05, *P < 0.01; Mann–Whitney U test analysis was used. (f–j) Autoantibodies present in HS were stratified by Hurley stage (stage I, n = 6; stage II, n = 7; stage III, n = 8). Results are the mean SEM, *P < 0.05, **P < 0.01; one-way ANOVA was used. (k) ANA assay Hep-2 assay was used to detect ANAs in HS sera. Original magnification ×20. (l) Heat map of unsupervised clustering of autoantibodies detected in HS serum stratified by Hurley stage. Ab, antibody; ANA, antinuclear antibody; Ctrl, control; dsDNA, double-stranded DNA; HS, hidradenitis suppurativa; RFU, relative fluorescence units.

Antibodies against cytokine, membrane, and nuclear components are present in HS serum

Antigens tested in this study were classified as cytokines, related to chemotaxis, bound to membrane, extracellular, nuclear, and/or cytoplasmic. Regarding anticytokine antibodies, patients with HS displayed elevated levels of serum antibodies against IL-17A and IL-1α in Hurley stage III, whereas stage II was characterized by the presence of serum antibodies recognizing TNF-α, IL-1β, IL-2, and IL-12p70 (Figure 2a). Anti-CXCL10 antibodies were elevated in stage II compared with those in other stages (Figure 2b). More broadly, stage II was also characterized by antibodies recognizing EGFRs, CTLA4, and muscarinic receptors (Figure 2c). The presence of antibodies recognizing extracellular components such as vimentin, FLG, collagen, vitronectin, and citrullinated fibrinogen was more evident in stages II and III (Figure 2d). Antibodies recognizing nuclear antigens such as histones, dsDNA, nucleosome, topoisomerase 1, and chromatin were enriched in stages II and III (Figure 2e). Conversely, the presence of antibodies recognizing neutrophil granule proteins (LL37, azurocidin, myeloperoxidase [MPO], PAD4) and metabolic enzymes (enolase, catalase) were prominent in stages II and III (Figure 2f). These results indicate a broad synthesis of autoantibodies that recognize a variety of cellular and extracellular antigens in HS, in association with specific disease severity stages.

Figure 2. Autoantibodies targeting multiple antigens detected in serum are higher in Hurley stages II and III.

Antibodies present in HS serum were classified on the basis of the role of the antigen and localization in the cells as (a) cytokine, (b) chemotaxis, (c) membrane, (d) extracellular, (e) nuclear, or (f) cytoplasmic. The average of the patients per Hurley stage was used to generate a score and perform a supervised clustering for each category. Ctrl, control; HS, hidradenitis suppurativa.

Autoantibodies are present in HS skin lesion

Because the presence of IgGs, B cells, and plasma cells have been documented in HS skin (Byrd et al., 2019; Frew et al., 2018; Gudjonsson et al., 2020; Hoffman et al., 2018; Musilova et al., 2020; van der Zee et al., 2012), we asked whether autoantibodies detected in HS sera could be identified in skin lesions. HS and healthy control skin samples were homogenized and analyzed for the presence of autoantibodies using a similar platform to that used for HS serum. Autoantibodies recognizing nuclear components such as dsDNA (Figure 3a), histone H2A (Figure 3c), citrullinated histones (Figure 3e), and histone H4 (Figure 3g) were significantly elevated in HS skin samples compared with those of the control. Significantly increased levels of anti–IL-17A (Figure 3i) and anti-LL37 (Figure 3k) antibodies were also found in HS skin samples. Stratification of the HS skin samples by Hurley stage detected antibodies recognizing dsDNA (Figure 3b), citrullinated total histones (Figure 3f), and IL-17A (Figure 3j) in stage III, and antibodies recognizing histone H2A (Figure 3d), histone H4 (Figure 3h), and LL37 (Figure 3l) had a trend to be higher in chronic stages of the disease. Unsupervised clustering analysis showed enrichment of skin autoantibodies in association with Hurley stage III. Autoantibodies recognizing FLG, IFN receptor, total citrullinated histones, collagen V, B2M, azurocidin, and IL-17A were significantly elevated in Hurley stage III shown by unsupervised stratification (Figure 3m). Taken together, these data suggest that antibodies recognizing self-antigens can be detected in HS skin samples and correlate with more severe disease.

Figure 3. Autoantibodies recognizing nuclear and citrullinated antigens are present in HS skin lesions.

Ctrl (n = 5) and HS (stage I, n = 3; stage II, n = 11; stage III, n = 11) skin tissue lysates were tested for the presence of autoantibodies against (a) dsDNA, (c) histone H2A, (e) total citrullinated histones, (g) histone H4, (i) IL-17A, and (k) LL37. Results are the mean ± SEM, *P < 0.05, **P < 0.01, ***P < 0.001; Mann–Whitney U test analysis was used. (b, d, f, h, j, l) Autoantibodies present in the lysate from HS skin lesions were stratified by Hurley stage (stage I, n = 3; stage II, n = 11; stage III, n = 11). Results are the mean± SEM, *P < 0.05; one-way ANOVA was used. (m) Heat map of unsupervised clustering of all autoantibodies detected in HS tissue stratified by Hurley stage. Ab, antibody; Ctrl, control; dsDNA, double-stranded DNA; HS, hidradenitis suppurativa; RFU, relative fluorescence units.

HS stage III skin samples are enriched in autoantibodies recognizing cytokine, membrane, and nuclear components

Autoantibodies found in HS skin lesions were classified on the basis of the cellular localization of the antigens they recognized. Most of the antibodies were higher in Hurley stage III, indicating more immune dysregulation contributing to the disease severity. Anti–IL-1α, IL-1β, IL-12 p70, TNF-α, IL-2, IL-17A, IL-8, and IFNα2 antibodies were also elevated in Hurley stage III skin lesions compared with those in stages I and II and the control skin (Figure 4a). In contrast to serum, HS skin displayed elevated levels of antibodies recognizing molecules involved in chemotaxis such as CXCL3, CXL11, CXCL2, and CXL10 in Hurley stage III but with minimal detection during early stages (Figure 4b). HS skin lesions were also characterized by elevated levels of antibodies recognizing membrane proteins such as muscarinic receptors, EGFR receptor, PD-L1, CTLA4, and ITGB4, among others (Figure 4c), compared with those in healthy controls. Antibodies recognizing extracellular antigens such as collagens II, IV, V, and VI; FLG; fibronectin (FN); and proteoglycan were also found mainly in Hurley stage III (Figure 4d). Antibodies recognizing nuclear antigens (dsDNA, histones, chromatin, high mobility group box protein 1) (Figure 4e), cytoplasmic (thyroid peroxidase, glutamate decarboxylase 1, annexin A2, PAD1, PAD2, PAD3), and neutrophil granule proteins (elastase, myeloperoxidase [MPO], proteinase 3, LL37) (Figure 4f) were also present mainly in Hurley stage III. These results indicate that in addition to serum, autoantibodies are also be found in HS skin lesions, particularly in Hurley stage III.

Figure 4. Autoantibodies detected in HS tissue are increased in Hurley stage III.

Antibodies present in HS skin samples were classified on the basis of the role of the antigen and localization in the cells as (a) cytokine, (b) chemotaxis, (c) membrane, (d) extracellular, (e) nuclear, or (f) cytoplasmic. The average of the patients per stage was used to generate a score and perform a supervised clustering for each category. Ctrl, control; HS, hidradenitis suppurativa

HS IgG ICs activate M1 and M2 macrophages to release proinflammatory cytokines

Global analysis of the antibodies found in sera and skin lesions of patients with HS showed differential prevalence depending on the source of the specimen. For example, autoantibodies recognizing lactoferrin, MDA5, PAD4, La/SSB, and complement, among others, were more prevalent in sera compared with those in skin lesions (Figure 5a). Conversely, TNF-α, histones, cytokines, LL37, and FLG antibodies, among others, were enriched in HS skin lesions compared with those in sera. In contrast, autoantibodies to citrullinated fibrinogen, MPO, nucleosome, histone H4, dsDNA, and others were equally detected in sera and HS skin lesions (Figure 5a). The presence of these autoantibodies can impact the activation status of the proteins being targeted, promote immune cell activation, and/or promote the formation of pathogenic ICs that can activate various immune cells and/or promote organ damage. Given that autoantibodies were detected in HS serum and skin and macrophages are known to play a role in the pathogenesis (Byrd et al., 2018; Hunger et al., 2008; Lowe et al., 2020; van der Zee et al., 2012), we hypothesized that antigen–antibody ICs could activate macrophages in HS. To test this hypothesis, CD14⁺ monocytes were isolated from healthy volunteers and polarized into M1-or M2-like macrophages with GM-CSF or macrophage colony–stimulating factor, respectively. Because antibodies recognizing citrullinated antigens have been found in HS specimens and previous studies indicate that ICs of specific citrullinated antigens can stimulate immune cells (Carmona-Rivera et al., 2020), we tested whether citrullinated antigen ICs can impact macrophage activation. M1- or M2-like macrophages were seeded in tissue culture plates coated with native or citrullinated versions of histone H2B, H4, or FN in the presence or absence of purified total IgG from patients with HS. Supernatants were tested for the presence of TNF-α, IL-8, IL-6, and IL-12/IL-23 (p40). Monocyte-derived macrophages incubated with cit-histone H2B or cit-histone H4 displayed enhanced release of TNF-α, IL-8, IL-6, and IL-12/IL-23 (p40) (Figures 5b–e and 6a–d), and this proinflammatory response was significantly enhanced in the presence of IgG ICs. Although incubation with citrullinated FN enhanced the release of these proinflammatory cytokines, this was not potentiated in the presence of HS IgG. These results support the notion that ICs present in patients with HS may be able to activate tissue macrophages to potentiate the release of proinflammatory cytokines that can perpetuate proinflammatory responses in skin and other tissues.

Figure 5. Native and citrullinated antigen–IgG immune complexes activate M1-like macrophages to release proinflammatory cytokines.

(a) Heat map of unsupervised clustering of all autoantibodies detected in HS serum compared with those detected in tissue. M1-like macrophages were incubated with native or citrullinated histone H2B, H4, or FN and immune complexes for 24–72 h. Supernatants were analyzed for (b) TNF-α, (c) IL-8, (d) IL-6, and (e) IL-12/IL-23p40. LPS was used as a positive control. Results are the mean SEM of four independent experiments. *P < 0.05; Mann–Whitney U test was used. FN, fibronectin; h, hour; HS, hidradenitis suppurativa; LPS, lipopolysaccharide.

Figure 6. Native and citrullinated antigen–IgG immune complexes activate M2-like macrophages to release proinflammatory cytokines.

M2-like macrophages were incubated with native or citrullinated histone H2B, H4, or FN and immune complexes for 24–72 h. Supernatants were analyzed for (a) TNF-α, (b) IL-8, (c) IL-6, and (d) IL-12/IL-23p40. LPS was used as a positive control. Results are the mean ± SEM of four independent experiments. Mann–Whitney U test was used. *P < 0.05, **P < 0.01. FN, fibronectin; h, hour; LPS, lipopolysaccharide.

DISCUSSION

Dysregulation of the immune system in patients with HS has been reported, and it is proposed to play an essential role in the progression of HS and associated comorbidities (Assan et al., 2020; Byrd et al., 2019; Frew et al., 2018; Hoffman et al., 2018; Hotz et al., 2016; Jenei et al., 2019; Kelly et al., 2015; Khandpur et al., 2013; Lowe et al., 2020; Moran et al., 2017; Navrazhina et al., 2021; Prens and Deckers, 2015; Shanmugam et al., 2019; Thomi et al., 2018). B cells, plasma cells, and Igs are present in HS chronic lesions (Byrd et al., 2019; Gudjonsson et al., 2020; Hoffman et al., 2018; Navrazhina et al., 2021; van der Zee et al., 2012). Although we previously described the enhanced levels of antibodies recognizing citrullinated antigens in HS, it was unclear whether broader autoantibody responses are present in HS and may promote exacerbation of inflammation. In this study, we show a comprehensive analysis of autoantibodies detected in HS according to their Hurley stage. Our study shows that IgGs present in the serum and skin lesions from patients with HS display broad autoantigen recognition, including various cytokines, chemokines, nuclear, cytoplasmic, and extracellular components. Furthermore, these autoantibodies correlate with disease severity, whereas specific ICs can increase the release of proinflammatory cytokines by macrophages.

We previously reported enhanced NET formation in HS lesions (Byrd et al., 2019). NETs have been described to externalize multiple antigens considered targets of the immune system in various bona fide systemic autoimmune diseases such as rheumatoid arthritis, systemic lupus erythematosus, anti-neutrophil cytoplasmic antibody–associated vasculitis, and others (Bodaño et al., 2006; Carmona-Rivera and Kaplan, 2014; Carmona-Rivera et al., 2017, 2015; Grayson and Kaplan, 2016; Hakkim et al., 2010; Kessenbrock et al., 2009; Khandpur et al., 2013; Sur Chowdhury et al., 2014). In this study, we show that patients with HS also develop autoantibodies not only to various NET components such as MPO, dsDNA, chromatin, histones, and elastase but also to a variety of targets characteristic of systemic autoimmune diseases. These observations support the hypothesis that NETs can be a source of autoantigens in HS and, more broadly, that dysregulation in cell death and/or cell death clearance may also be implicated in this disease, given the ANA response (Grayson and Kaplan, 2016; Gupta and Kaplan, 2016). This hypothesis requires future studies for corroboration. Furthermore, NETs can be internalized by antigen-presenting cells and promote pathogenic adaptive immune responses in rheumatoid arthritis (Carmona-Rivera et al., 2017). Future studies should test the role of NETs–antibody interactions in organ damage in HS, besides the skin, to better understand the multifactorial contributions and consequences of the disease.

Antibodies to nuclear antigens (i.e., ANA), cardiolipin, and other phospholipids are present in conditions such as systemic lupus erythematosus and antiphospholipid syndrome, which are characterized by high-risk cardiovascular disease and increased tendency to form blood clots (Meroni et al., 2011; Miyakis et al., 2006; Nakazawa et al., 2019). Antibodies targeting dsDNA, nucleosome, chromatin, and histones were detected in subjects with HS. Anti-SS-B/La antibodies, which are primarily found in Sjogren syndrome and systemic lupus erythematosus with both cutaneous and musculoskeletal manifestations (Novak et al., 2017), were also found in Hurley stages II and III. These antibodies have a strong association with nodal and spleen enlargement and with increased levels of gamma-globulins (Montecucco et al., 1989).

Inflammatory myopathies and dermatomyositis are characterized by the presence of antibodies to Jo1 and MDA5, respectively. MDA5 autoantibodies potentiate NET formation in neutrophils (Seto et al., 2020) and can contribute to the exacerbated NET formation found in patients with HS (Byrd et al., 2019). Anti-Jo1 is strongly associated in myositis with antisynthetase syndrome, typically presenting with interstitial lung disease, Raynaud’s phenomenon, and arthritis. MDA5 antibodies, other myositis-specific antibodies, are associated with interstitial lung disease, along with severe, treatmentresistant myositis and a very poor prognosis (Tartar et al., 2018). Future studies should address the role of anti-MDA5 and anti-Jo1 antibodies in clinical manifestations other than skin in patients with HS or whether they contribute to NET formation.

Importantly, we also note differences in the types of autoantibodies prevalent in Hurley stages II and III. For example, Hurley stage II samples appear to be enriched in autoantibodies to IL-1β, EGFR, CTL4, and muscarinic receptor, whereas stage III displays higher levels of autoantibodies to DNA, chromatin, IL-1α, chemotaxis (CXCL3, CXL11, CXCL2), FN, proteoglycan, and cytoplasmic (thyroid peroxidase, glutamate decarboxylase 1, ANXA2, PAD1, PAD2, PAD3) antigens. This could suggest that disease progression is related to increased cell death and dysregulation of the adaptive immune system and this should be further explored in future studies (Assan et al., 2020; Hoffman et al., 2018). Whether a set of antibodies present in HS can define the Hurley stage and be used as a biomarker of severity is still unknown. Several comorbidities are associated with patients with HS (increased cardiovascular disease, diabetes, arthritis, metabolic syndrome) (Garg et al., 2021). Our findings provide the impetus to expand the clinical assessment performed in patients with HS beyond dermatology, considering that a multidisciplinary approach to HS patient care is warranted.

Although there was differential detection depending on the specimen type (autoantibodies against lactoferrin, MDA5, PAD4, La/SSB, and complement, among others, were highly detected in serum from patients with HS, whereas anti-TNF-α, histones, cytokines, LL37, and FLG antibodies were more prevalent in skin lesions), future experiments are needed to better understand and delineate the autoantibodies detected in the tissue and serum. Autoantibodies to citrullinated fibrinogen, MPO, nucleosome, his-tone H4, and dsDNA, among others, were equally detected in sera and skin lesions. This may suggest a dysregulation of the adaptive immune response systemically and locally. This is consistent with reports showing B-cells infiltration in skin lesions and systemic B-cell dysregulation in HS (Byrd et al., 2019; Gudjonsson et al., 2020; Hoffman et al., 2018; van der Zee et al., 2012). Consistent with other studies, elevated levels of LL37 and other antigens suggest that the environment found in HS is conducive to the generation of antibodies against self-antigens (Bechara et al., 2012; Emelianov et al., 2012; Frew, 2019; Kelly et al., 2015; Lima et al., 2016; Navrazhina et al., 2021, 2020; Thomi et al., 2018; Vossen et al., 2019). Future studies are needed to investigate the role of NET-derived proteins ICs in the activation of innate and adaptative responses in HS (Goel and Kaplan, 2020; Kahlenberg et al., 2013). NLRP3 inflammasome has been shown to promote heart failure and atherosclerosis (Abbate et al., 2012; Bonaventura and Montecucco, 2019; Jin and Fu, 2019; Mauro et al., 2019; Zuurbier et al., 2019). NLRP3 and IL-1β have also been studied in the pathophysiology of rheumatic diseases (Dinarello, 2019).

The enrichment of autoantibodies in later Hurley stages suggests that these molecules could play an essential role in pathogenesis as the disease severity progresses. One possible mechanism through which these autoantibodies influence disease is through IC generation and macrophage activation (Byrd et al., 2019, 2018; Frew et al., 2018; Hotz et al., 2016; Witte-Händel et al., 2019; Zouboulis et al., 2020). Tissue macrophages in HS lesions release proinflammatory cytokines, promoting IL-17 synthesis by T cells (Byrd et al., 2018; Lowe et al., 2020; Moran et al., 2017; Schlapbach et al., 2011; Shah et al., 2017). We found that citrullinated his-tone H2B, H4, or FN enhanced M1 and M2 macrophage release of proinflammatory cytokines such as TNF-α, IL-8, IL-6, and IL-12/IL-23 in vitro. This was further potentiated when incubated with IgG isolated from patients with HS. Proinflammatory cytokines were not increased when native (non-citrullinated) H2B, H4, and FN were used. Citrullinated proteins, particularly citrullinated histones, have been detected in tissues where NETs are present (Khandpur et al., 2013). We have previously shown that NETs upregulate IFN-1 pathway in HS, which can promote immune dysregulation and inflammation (Byrd et al., 2019). Our current findings provide a link between IgG ICs and macrophage activation in HS pathogenesis as a putative link to tissue damage in this disease.

Our study was conducted with samples obtained mostly from African American patients, an understudied population. It will be important to know whether our results can be recapitulated in other ethnic groups because studies have shown differences in both prevalence (e.g., higher prevalence of HS in African American and biracial individuals in the United States and lower prevalence of HS in Amerindians in Brazil) and disease phenotype (e.g., men with HS in Japan are more likely to have the gluteal disease) (Garg et al., 2017; Ianhez et al., 2018; Kurokawa et al., 2015; Morss et al., 2020; Soliman et al., 2019). Another limitation of our study includes the relatively small sample size per Hurley stage. Future experiments are needed to better understand the prevalence of autoantibodies in patients with HS and whether these antibodies play an important role in some clinical manifestations of the disease.

In summary, we provided evidence of a broad autoantibody response in serum and skin samples from subjects with HS that correlate with disease activity and associate with IC generation and subsequent macrophage activation. These observations can shed light on the mechanisms involved in HS pathogenesis and potentially contribute to the development of therapeutic approaches providing better outcomes for patients living with this devastating condition. Considering the associated comorbidities and multiorgan complexities in HS (Garg et al., 2021), a more precise understanding of the mechanisms could provide insights into the multifactorial manifestations of the disease aiding in translational research efforts.

MATERIALS AND METHODS

Collection of tissue samples

The study was approved by the Johns Hopkins University Institutional Review Board (NA_00013177 and NA_00031269), and written informed consent was obtained from healthy volunteers and patients with HS from Hurley stages I to III. Normal (non-HS) sera were obtained from healthy volunteers. Normal (non-HS) skin was obtained from cosmetic abdominoplasties and mammoplasties, and lesional HS skin was obtained from surgical resections and from Johns Hopkins tissue bank obtained from patients undergoing surgical removal of HS skin. Patients with Hurley stages II and III were considered to have chronic HS, that is, sinus tracts and fibrotic tissue. Visual assessment of the patient at the time of surgery as well as clinical and histopathological diagnosis were verified. Each sample was collected and stored in liquid nitrogen for transport.

Protein isolation from HS skin lesion

Three 5-mm tissue sections were homogenized using liquid nitrogen, mortar, and pestle as previously described (Byrd et al., 2019). Homogenized tissue was resuspended in 500 μl of RIPA buffer (Sigma-Aldrich, St. Louis, MO). After 1 hour at 4 ^oC, the solution was centrifuged for 10 minutes at 14,000 r.p.m., and the supernatant was transferred to a fresh Eppendorf tube. Proteins were quantified using the BCA kit (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer’s instructions. Overall, 50 μg of total protein from healthy controls and patients with HS were analyzed from the presence of autoantibodies.

Autoantibody detection assay

Autoantigens to various autoimmune conditions, such as systemic lupus erythematosus, systemic sclerosis, Sjogren’s syndrome, idiopathic inflammatory myositis, and rheumatoid arthritis were chosen. The assay was performed as previously described (Lammert et al., 2020). Briefly, sera or skin lysates from 25 patients with HS and five healthy controls were pretreated with DNAaseI and diluted at 1:50. Samples were hybridized to a nitrocellulose membrane-coated slide containing printed antigens in duplicates (https://microarray.swmed.edu/products/category/protein-array). After washes, antibodies bound to the array were detected with Cy3-labeled antiehuman IgG (IgG; 1:2,000; Jackson ImmunoResearch Laboratories, West Grove, PA). A GenePix 4400A Microarray Scanner was used to scan the arrays and GenePix 7.0 software to analyze fluorescence intensities (both from Molecular Devices, San Jose, CA). Background signal (from PBS) was subtracted from the average signal intensity of each sample. A signal-to-noise ratio was calculated, and the antibody score was calculated as log₂(signal-to-noise ratio × net fluorescence intensity + 1) as previously described (Perez-Diez et al., 2020).

IgG isolation from serum

IgGs were purified from HS sera with a Melon Gel IgG Spin kit following the manufacturer’s instructions (Thermo Fisher Scientific).

ANA assay

ANAs were detected by immunofluorescence on HEp-2 slides following the manufacturer’s instructions (MBL International, Woburn, MA).

IgG IC–mediated macrophage activation

To obtain M1- or M2-polarized macrophages, human CD14⁺ monocytes were isolated from PBMCs using positive selection with MACs columns (Miltenyi Biotech, Bergisch Gladbach, Germany), and the cells were incubated with GM-CSF (50 ng/ml, Peprotech, Cranbury, NJ) for M1 or macrophage colony–stimulating factor (50 ng/ml, Peprotech) for M2 for 5 days. A total of 96-well plates were coated with recombinant or citrullinated recombinant human his-tone H2B, H4, or fibrinogen generated in vitro (100 ng/ml) in PBS at 4 ^oC overnight. Plates were then incubated with 100 mg of isolated HS IgG for 2 hours at room temperature. Plates were washed twice with PBS and seeded with M1 or M2 macrophages for 24 or 72 hours. Collected supernatants were analyzed by commercial ELISAs for TNF-α, IL-8, IL-6, and IL-12 (Invitrogen, Waltham, MA).

Statistical analysis

Data were analyzed using GraphPad Prism software (GraphPad software, San Diego, CA). For samples with non-Gaussian distribution, we used Mann–Whitney U test. Where indicated, one-way ANOVA Brown–Forsythe test analysis was used. Results are presented as the mean ± SEM.

Data availability statement

Datasets generated from this study have been provided in the Supplementary Materials.

Abbreviations:

ANA

antinuclear antibody

dsDNA

double-stranded DNA

FN

fibronectin

HS

hidradenitis suppurativa

IC

immune complex

MPO

myeloperoxidase

NET

neutrophil extracellular trap

PAD

protein arginine deiminases