Advertisement

Myeloid Dendritic Cells Are Major Producers of IFN-β in Dermatomyositis and May Contribute to Hydroxychloroquine Refractoriness

  • Author Footnotes
    4 These authors contributed equally to this work.
    Kristen L. Chen
    Footnotes
    4 These authors contributed equally to this work.
    Affiliations
    Department of Dermatology, Corporal Michael J. Crescenz VAMC, Philadelphia, Pennsylvania, USA

    Department of Dermatology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
    Search for articles by this author
  • Author Footnotes
    4 These authors contributed equally to this work.
    Jay Patel
    Footnotes
    4 These authors contributed equally to this work.
    Affiliations
    Department of Dermatology, Corporal Michael J. Crescenz VAMC, Philadelphia, Pennsylvania, USA

    Department of Dermatology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
    Search for articles by this author
  • Majid Zeidi
    Affiliations
    Department of Dermatology, Corporal Michael J. Crescenz VAMC, Philadelphia, Pennsylvania, USA

    Department of Dermatology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
    Search for articles by this author
  • Maria Wysocka
    Affiliations
    Department of Dermatology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
    Search for articles by this author
  • Muhammad M. Bashir
    Affiliations
    Department of Dermatology, Corporal Michael J. Crescenz VAMC, Philadelphia, Pennsylvania, USA

    Department of Dermatology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
    Search for articles by this author
  • Basil Patel
    Affiliations
    Department of Dermatology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
    Search for articles by this author
  • Spandana Maddukuri
    Affiliations
    Department of Dermatology, Corporal Michael J. Crescenz VAMC, Philadelphia, Pennsylvania, USA

    Department of Dermatology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
    Search for articles by this author
  • Barbara White
    Affiliations
    Corbus Pharmaceuticals Holdings, Inc, Norwood, Massachusetts, USA
    Search for articles by this author
  • Victoria P. Werth
    Correspondence
    Correspondence: Victoria P. Werth, Department of Dermatology, Perelman Center for Advanced Medicine, Suite 1-330A, 3400 Civic Center Boulevard, Philadelphia, Pennsylvania 19104, USA.
    Affiliations
    Department of Dermatology, Corporal Michael J. Crescenz VAMC, Philadelphia, Pennsylvania, USA

    Department of Dermatology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
    Search for articles by this author
  • Author Footnotes
    4 These authors contributed equally to this work.
Open ArchivePublished:March 03, 2021DOI:https://doi.org/10.1016/j.jid.2020.12.032
      Dermatomyositis pathogenesis remains incompletely understood; however, recent work suggests a predominant IFN-1 response. We explored dermatomyositis pathogenesis by quantifying the inflammatory cells in the skin, comparing myeloid with plasmacytoid dendritic cell release of IFN-β, and assessing myeloid dendritic cell (mDC) contribution to hydroxychloroquine refractoriness. Immunohistochemistry was performed to assess cell-type expression in lesional skin biopsies from 12 patients with moderate-to-severe cutaneous dermatomyositis. Immunofluorescence, laser-capture microdissection, and flow cytometry were used to assess mDC release of IFN-β in lesional skin biopsies and blood of patients with dermatomyositis. Immunohistochemistry was utilized to determine whether myeloid or plasmacytoid dendritic cells were increased in hydroxychloroquine nonresponders. CD4+, CD11c+, and CD69+ cells were more populous in lesional skin of patients with dermatomyositis. mDCs colocalized with IFN-β by immunofluorescence and laser-capture microdissection revealed increased IFN-β mRNA expression by mDCs in lesional skin of patients with dermatomyositis. In blood, both mDCs and plasmacytoid dendritic cells were major producers of IFN-β in patients with dermatomyositis, whereas plasmacytoid dendritic cells predominately released IFN-β in healthy controls (P < 0.01). mDCs were significantly increased in the skin of hydroxychloroquine nonresponders compared with that in the skin of responders (P < 0.05). mDCs cells appear to play an important role in dermatomyositis pathogenesis and IFN-β production.

      Abbreviations:

      FFPE (formalin fixed and paraffin embedded), HC (healthy control), HCQ (hydroxychloroquine), HPF (high-power field), KC (keratinocyte), LCM (laser-capture microdissection), mDC (myeloid dendritic cell), pDC (plasmacytoid dendritic cell)

      Introduction

      Dermatomyositis is a systemic autoimmune condition primarily affecting the skin, muscle, and lungs, among other organs. The estimated incidence is 9.6 per 1 million persons per year (
      • Bendewald M.J.
      • Wetter D.A.
      • Li X.
      • Davis M.D.
      Incidence of dermatomyositis and clinically amyopathic dermatomyositis: a population-based study in Olmsted County, Minnesota.
      ), and it can significantly impair patients’ QOL (
      • Goreshi R.
      • Chock M.
      • Foering K.
      • Feng R.
      • Okawa J.
      • Rose M.
      • et al.
      Quality of life in dermatomyositis.
      ). However, the pathogenesis of dermatomyositis remains incompletely understood. It is thought that predisposing genes, environmental stressors, and immune- and nonimmune-mediated mechanisms induce the susceptibility and onset of dermatomyositis (
      • Mainetti C.
      • Terziroli Beretta-Piccoli B.
      • Selmi C.
      Cutaneous manifestations of dermatomyositis: a comprehensive review.
      ).
      IFNs that regulate innate and adaptive immunity are thought to be key players in dermatomyositis pathogenesis (
      • Kao L.
      • Chung L.
      • Fiorentino D.F.
      Pathogenesis of dermatomyositis: role of cytokines and interferon.
      ) because high levels of IFN-induced gene products have been found in the blood (
      • Huard C.
      • Gullà S.V.
      • Bennett D.V.
      • Coyle A.J.
      • Vleugels R.A.
      • Greenberg S.A.
      Correlation of cutaneous disease activity with type 1 interferon gene signature and interferon β in dermatomyositis.
      ) and skin tissue (
      • Wong D.
      • Kea B.
      • Pesich R.
      • Higgs B.W.
      • Zhu W.
      • Brown P.
      • et al.
      Interferon and biologic signatures in dermatomyositis skin: specificity and heterogeneity across diseases.
      ) of patients with dermatomyositis. IFNs have inherent amplification mechanisms because many of the IFN-induced downstream molecules are regulated by IFN itself (
      • Hall J.C.
      • Rosen A.
      Type I interferons: crucial participants in disease amplification in autoimmunity.
      ). With regard to autoimmunity, amplification of this response may result in increased autoantigen presentation because IFNs may increase major histocompatibility complex I and the maturation of antigen-presenting cells, contributing to increased self-targeted inflammation (
      • Hall J.C.
      • Rosen A.
      Type I interferons: crucial participants in disease amplification in autoimmunity.
      ).
      Plasmacytoid dendritic cells (pDCs) are considered major producers of IFN-1, IFN-α and IFN-β, and have been identified to infiltrate both the skin and muscle affected by dermatomyositis (
      • Greenberg S.A.
      Dermatomyositis and type 1 interferons.
      ). They are thought to produce high levels of IFN-1 after stimulation of toll-like receptor 7 and toll-like receptor 9 (
      • Greenberg S.A.
      Dermatomyositis and type 1 interferons.
      ). In addition, there appears to be an autostimulatory effect where the accumulation of pDCs amplifies the production of IFN-1 (
      • Liao A.P.
      • Salajegheh M.
      • Morehouse C.
      • Nazareno R.
      • Jubin R.G.
      • Jallal B.
      • et al.
      Human plasmacytoid dendritic cell accumulation amplifies their type 1 interferon production.
      ). Little work has been done to identify the contribution of other antigen-presenting cells, such as myeloid dendritic cells (mDCs), in the IFN-1 response.
      Our objectives were to quantify inflammatory cells in dermatomyositis skin, compare mDC release with pDC release of IFN-β in the skin and blood affected by dermatomyositis, and evaluate mDC versus pDC contribution to refractoriness to hydroxychloroquine (HCQ). A better understanding of disease pathogenesis will aid in the development of targeted therapies.

      Results

      CD4+ T cells, CD11c+ cells, and CD69+ cells are the predominant inflammatory cells in moderate-to-severe dermatomyositis skin lesions

      We evaluated the cellular infiltrates of the skin with dermatomyositis in 12 patients with moderate-to-severe cutaneous dermatomyositis disease despite standard background therapy. We found that CD4+ T cells (67.9 ± 62.5 cells per high-power field [HPF]), CD11c+ cells (34.5 ± 46.4 cells per HPF), and CD69+ cells (28.6 ± 25.6 cells per HPF) (Figures 1a–c) were more populous in lesional skin of patients with dermatomyositis than in CD8+ T cells (14.4 ± 13.8 cells per HPF), mast cells (13.6 ± 14.1 cells per HPF), and CD123+ cells (6.1 ± 13.3 cells per HPF) (Figure 1d–f). The graphical representation can be seen in Figure 1g. The isotypes are illustrated in Figure 1h and i.
      Figure thumbnail gr1
      Figure 1Inflammatory cells in lesional skin of patients with moderate-to-severe dermatomyositis (n = 12). Immunohistochemistry for (a) CD4, (b) CD11c (mDC), (c) CD69, (d) CD8, (e) Tryptase (mast cells), and (f) CD123 (pDC). (g) Cells were quntified per HPF and graphed. Immunohistochemistry (h) mouse isotype and (i) rabbit isotype. Imaged at original magnification ×20. Bar = 50 μm. Graph shows median ± interquartile range. DM, dermatomyositis; HPF, high-power field; mDC, myeloid dendritic cell; pDC, plasmacytoid dendritic cell.

      CD11c is a reliable mDC marker in lesional skin of patients with dermatomyositis

      To evaluate CD11c as a marker for mDCs, we used Image Mass Cytometry to triple label lesional skin in patients with dermatomyositis with CD11c, CD163, and HLA-DR. We used metal-conjugated antibodies CD11c (Figure 2a, red), CD163 (Figure 2b, blue), and HLA-DR (Figure 2c, green). CD11c and CD163 both overlapped with HLA-DR, identifying antigen-presenting cells, as seen in Figure 2d and e. Colabeling of CD11c and CD163 showed that there is no overlap between these two cell markers (Figure 2f). Triple labeling revealed a separation between CD11c+HLA‒DR+ cells and CD163+HLA‒DR+ cells (Figure 2g), suggesting that CD11c is a reliable marker for separating positive mDC populations and negative macrophage populations in skin of patients with dermatomyositis.
      Figure thumbnail gr2
      Figure 2Separation of mDCs from macrophages in lesional skin of patients with moderate-to-severe dermatomyositis (n = 7). Image Mass Cytometry for (a) CD11c (mDCs), (b) CD163 (macrophages), (c) HLA-DR; (d) colocalization of CD11c and HLA-DR; (e) colocalization of CD163 and HLA-DR; (f) separation of CD11c and CD163; and (g) colocalization of CD11c, CD163, and HLA-DR demonstrating the separation of mDCs (white arrows) from macrophages (yellow arrows). Nuclei are represented with Ir-DNA Intercalator (gray). Bars = 10 μm. mDC, myeloid dendritic cell.

      IFN-β is produced by mDCs in lesional skin of patients with dermatomyositis

      Given the high number of CD11c+ mDCs, we evaluated their role in IFN-β production. We used immunofluorescence to identify CD11c+ mDCs (Figure 3a, red) and IFN-β (Figure 3b, green). Immunofluorescence colabeling showed colocalization of IFN-β and CD11c+ cells (Figure 3c, yellow or orange) in the skin of patients with dermatomyositis, suggesting that mDCs produce IFN-β. To investigate IFN-β production at an mRNA level, we implemented laser-capture microdissection (LCM) to capture the CD11c+ mDC cell population and compare it with the population of keratinocytes (KCs) within each skin biopsy. Figure 4a shows the identification and circling of CD11c+ mDCs to be dissected (red) using rapid immunofluorescence staining. Subsequent laser dissection along the red outline and extraction of the CD11c+ mDC are shown in Figure 4b. RT-PCR of the acquired CD11c+ mDCs and KCs revealed an increased relative mRNA expression of IFN-β by mDCs (3.45 ± 5.64 vs. 0.37 ± 0.79, P < 0.05) (Figure 4c).
      Figure thumbnail gr3
      Figure 3Colocalization of CD11c and IFN-β in lesional skin of patients with moderate-to-severe dermatomyositis (n = 7). Immunofluorescence of (a) CD11c (mDC) and (b) IFN-β and (c) colocalization of CD11c and IFN-β (yellow). Bar = 100 μm. mDC, myeloid dendritic cell.
      Figure thumbnail gr4
      Figure 4IFN-β mRNA in CD11c+ cells in lesional skin of patients with moderate-to-severe dermatomyositis (n = 4). (a) Rapid immunofluorescence of CD11c+ mDC (outlined in red). (b) LCM successfully procures outlined CD11c+ mDCs, leaving a rim of fluorescence from laser-mediated high-intensity light absorption and reflection. (c) Graph of IFN-β mRNA expression of KCs versus that of CD11c+ cells. Bar = 10 μm. Graphs show median ± interquartile range. ∗P < 0.05. KC, keratinocyte; LCM, laser-capture microdissection; mDC, myeloid dendritic cell.

      IFN-β is produced by both mDCs and pDCs in the peripheral blood of patients with dermatomyositis

      Flow cytometry showed that peripheral blood from both healthy controls (HCs) (Figure 5a) and patients with dermatomyositis (Figure 5b) have similar percentages of mDCs (HC: 10 ± 8.5; patients with dermatomyositis: 9.9 ± 8.9) and pDCs (HC: 4.7 ± 7.0; patients with dermatomyositis: 5.6 ± 5.3). In HCs, IFN-β is almost exclusively produced by pDCs (mDC: 1.4 ± 3.0 vs. pDC: 40.56 ± 22.1) (P < 0.01) (Figure 5c), whereas in patients with dermatomyositis, IFN-β is produced by both mDCs and pDCs at high levels, and there is no significant difference between the cell sources (mDC: 58.0 ± 34.6 vs. pDC: 76.92 ± 25.2) (Figure 5d). Examples of our gating strategy for Figure 5c and d are illustrated, respectively, in Figure 5e and f.
      Figure thumbnail gr5
      Figure 5IFN-β release by CD11c+ mDCs and CD123+ pDCs in peripheral blood of HCs (n = 5) and patients with dermatomyositis (n = 5). Flow cytometry analyses quantifying the percentages of HLA-DR+Lin− cells that are CD11c+ mDCs or CD123+ pDCs in (a) HCs and (b) patients with dermatomyositis. Flow cytometry analyses quantifying IFN-β (MFI) release by mDCs and pDCs in (c) HCs and (d) patients with dermatomyositis. Example of gating strategy on FlowJo analysis in (e) HCs and (f) patients with dermatomyositis. Note: IFN-β release was stimulated with R848. Graphs show median ± interquartile range. ∗∗P < 0.01. DM, dermatomyositis; HC, healthy control; Lin, lineage; mDC, myeloid dendritic cell; MFI, mean fluorescence intensity; ns, nonsignificant; pDC, plasmacytoid dendritic cell.

      HCQ nonresponders have significantly increased numbers of mDCs in Dermatomyositis skin

      Our analysis showed significantly increased numbers of CD11c+ mDCs in HCQ nonresponders (43.9 ± 49.2) compared with those in HCQ responders (17.4 ± 32.2) (P < 0.05) (Figure 6a–c). There was no significant difference in the number of pDCs between the HCQ nonresponders (4.2 ± 20.2) and the HCQ responders (2.2 ± 10.7) (Figure 6d–f). In addition, there were significantly more CD11c+ cells than CD123+ cells in both HCQ responders (P < 0.05) and nonresponders (P < 0.01). The expression of IFN-α was found to be higher in HCQ responders than in HCQ nonresponders (P < 0.01) (Figure 6g–i), whereas there was no significant difference in IFN-β expression between the two populations (P > 0.05) (Figure 6j–l).
      Figure thumbnail gr6
      Figure 6CD11c+ mDCs and CD123+ pDCs in lesional skin of patients with moderate-to-severe dermatomyositis from HCQ responders (n = 8) and HCQ nonresponders (n = 8). Immunohistochemistry of CD11c+ cells in (a) HCQ responders and (b) HCQ nonresponders. (c) Graph quantifying CD11c+ cells per HPF in HCQ responders versus that in nonresponders. Immunohistochemistry of CD123+ cells in (d) HCQ responders and (e) HCQ nonresponders. (f) Graph quantifying CD123+ cells per HPF in HCQ responders versus that in nonresponders. Immunofluorescence staining of IFN-α in (g) HCQ responders and (h) HCQ nonresponders. (i) Graph quantifying IFN-α MFI in HCQ responders versus that in nonresponders. Immunofluoresence staining of IFN-β in (j) HCQ responders and (k) HCQ nonresponders. (l) Graph quantifying IFN-β MFI in HCQ responders versus that in nonresponders. Bars = 30 μm. Graphs show median ± interquartile range. ∗P < 0.05 and ∗∗P < 0.01. HCQ, hydroxychloroquine; HPF, high-power field; mDC, myeloid dendritic cell; MFI, mean fluorescence intensity; ns, nonsignificant; pDC, plasmacytoid dendritic cell.

      Discussion

      Dermatomyositis pathogenesis is poorly understood. To better understand this disease, we began by quantifying several cell types in moderate-to-severe dermatomyositis lesional skin.
      • Caproni M.
      • Torchia D.
      • Cardinali C.
      • Volpi W.
      • Del Bianco E.
      • D’Agata A.
      • et al.
      Infiltrating cells, related cytokines and chemokine receptors in lesional skin of patients with dermatomyositis.
      reported CD4+ T lymphocytes as predominant infiltrating cells, which we similarly found. However, we also identified CD11c+ mDCs and CD69+ cells as major cell types in skin of patients with moderate-to-severe dermatomyositis.
      Given the high number of mDCs in skin of patients with dermatomyositis, particularly juxtaposed to the scarce presence of pDCs, we hypothesized that mDCs may play a larger role in dermatomyositis pathogenesis than previously known. Recent studies have implicated IFN-1, in particular IFN-β, as a driver of the disease pathogenesis (
      • Huard C.
      • Gullà S.V.
      • Bennett D.V.
      • Coyle A.J.
      • Vleugels R.A.
      • Greenberg S.A.
      Correlation of cutaneous disease activity with type 1 interferon gene signature and interferon β in dermatomyositis.
      ;
      • Wong D.
      • Kea B.
      • Pesich R.
      • Higgs B.W.
      • Zhu W.
      • Brown P.
      • et al.
      Interferon and biologic signatures in dermatomyositis skin: specificity and heterogeneity across diseases.
      ), which is generally assumed to be derived from pDCs (
      • Fitzgerald-Bocarsly P.
      • Dai J.
      • Singh S.
      Plasmacytoid dendritic cells and type I IFN: 50 years of convergent history.
      ). Previous findings have suggested the likelihood of other cellular sources of IFN-β, including KCs and/or endothelial cells, because skin of patients with dermatomyositis has shown discordant expression of pDCs and IFN-1, with some showing impressive IFN-1 expression and rare pDCs and vice versa (
      • Magro C.M.
      • Segal J.P.
      • Crowson A.N.
      • Chadwick P.
      The phenotypic profile of dermatomyositis and lupus erythematosus: a comparative analysis.
      ). We considered that mDCs may additionally be significant producers of IFN-β because they too are capable of expressing a variety of pattern recognition receptors that may stimulate IFN-1 response (
      • Ali S.
      • Mann-Nüttel R.
      • Schulze A.
      • Richter L.
      • Alferink J.
      • Scheu S.
      Sources of type I interferons in infectious immunity: plasmacytoid dendritic cells not always in the driver's seat.
      ).
      Immunofluorescence imaging of CD11c and IFN-β showed colocalization in the lesional dermatomyositis skin, suggesting that mDCs produce IFN-β at the protein level. LCM showed increased IFN-β relative mRNA expression from mDCs compared with that from KCs. These findings suggest that mDCs play an important role in the overall IFN-1 signature in the skin. Our analysis by flow cytometry of the blood of patients with dermatomyositis compared with that of blood of HC supports this finding. We additionally demonstrated that in the peripheral blood of HCs, pDCs are almost exclusive producers of IFN-β, but in the blood of patients with dermatomyositis, IFN-β was produced by both mDCs and pDCs. The percentage of mDCs versus that of pDCs in the peripheral blood of patients with dermaotomyositis compared with that in the peripheral blood of HCs was not significantly different. This study directly compared IFN-β production from mDCs with that from pDCs in patients with dermatomyositis. This suggests that mDCs, in addition to pDCs, are important in dermatomyositis disease pathogenesis.
      Given these results, we further hypothesized that patients with dermatomyositis may not respond to HCQ owing to high numbers of mDCs. A previous report found that only approximately 25% of patients with dermatomyositis respond to HCQ, with the remainder requiring second- and third-line therapies (
      • Anyanwu C.O.
      • Chansky P.B.
      • Feng R.
      • Carr K.
      • Okawa J.
      • Werth V.P.
      The systemic management of cutaneous dermatomyositis: results of a stepwise strategy.
      ). We found significantly higher numbers of CD11c+ mDCs in the HCQ nonresponders than in the HCQ responders, similar to our previously reported findings in cutaneous lupus erythematosus (
      • Zeidi M.
      • Kim H.J.
      • Werth V.P.
      Increased myeloid dendritic cells and TNF-α expression predicts poor response to hydroxychloroquine in cutaneous lupus erythematosus.
      ). Interestingly, IFN-α was lower in HCQ nonresponders, and there was no difference in IFN-β expression between the two groups. This suggests that HCQ may have effects on high IFN-α‒expressing responders. HCQ nonresponders may not respond owing to differences in cell sources, such as mDCs that alter the IFN-1 composition. Other inflammatory stimuli may drive refractory disease as well, but it is unclear which cytokines drive the HCQ nonresponder population, and future studies will aim to elucidate this. Our finding of increased mDCs in patients with dermatomyositis who are HCQ nonresponders compared with that in those who are HCQ responders again suggests that these cells are important in dermatomyositis disease pathogenesis.
      Although mDCs have not been described previously in dermatomyositis, there are some reports of similar infiltrating cells in psoriasis, rhinitis, and inflammatory bowel disease (
      • Beitnes A.C.
      • Ráki M.
      • Brottveit M.
      • Lundin K.E.
      • Jahnsen F.L.
      • Sollid L.M.
      Rapid accumulation of CD14+CD11c+ dendritic cells in gut mucosa of celiac disease after in vivo gluten challenge.
      ;
      • Eguíluz-Gracia I.
      • Bosco A.
      • Dollner R.
      • Melum G.R.
      • Lexberg M.H.
      • Jones A.C.
      • et al.
      Rapid recruitment of CD14(+) monocytes in experimentally induced allergic rhinitis in human subjects.
      ;
      • Grimm M.C.
      • Pullman W.E.
      • Bennett G.M.
      • Sullivan P.J.
      • Pavli P.
      • Doe W.F.
      Direct evidence of monocyte recruitment to inflammatory bowel disease mucosa.
      ;
      • Jenner W.
      • Motwani M.
      • Veighey K.
      • Newson J.
      • Audzevich T.
      • Nicolaou A.
      • et al.
      Characterisation of leukocytes in a human skin blister model of acute inflammation and resolution.
      ;
      • Zaba L.C.
      • Fuentes-Duculan J.
      • Eungdamrong N.J.
      • Abello M.V.
      • Novitskaya I.
      • Pierson K.C.
      • et al.
      Psoriasis is characterized by accumulation of immunostimulatory and Th1/Th17 cell-polarizing myeloid dendritic cells.
      ). These cells have been termed monocyte-derived dendritic cells and are not normally present in HCs (
      • Collin M.
      • Bigley V.
      Human dendritic cell subsets: an update.
      ). They may play an important role in antigen presentation and amplification of the adaptive immune system (
      • Collin M.
      • Bigley V.
      Human dendritic cell subsets: an update.
      ). The IFN-1 response has pleiotropic effects, promoting the maturation, survival, and differentiation of dendritic cells, T cells, and B cells (
      • Crouse J.
      • Kalinke U.
      • Oxenius A.
      Regulation of antiviral T cell responses by type I interferons.
      ;
      • Gallucci S.
      • Lolkema M.
      • Matzinger P.
      Natural adjuvants: endogenous activators of dendritic cells.
      ;
      • Kiefer K.
      • Oropallo M.A.
      • Cancro M.P.
      • Marshak-Rothstein A.
      Role of type I interferons in the activation of autoreactive B cells.
      ). Thus, initial IFN-1 signatures may promote further inflammation and augment the IFN-1 response, leading to an amplification loop as implicated in other autoimmune diseases (
      • Crow M.K.
      • Ronnblom L.
      Type I interferons in host defence and inflammatory diseases.
      ). Further work will need to be done to identify the initiating insult for IFN production and the origin of these cells, specifically in dermatomyositis.
      There are a few limitations to the study. First, we have a relatively limited sample size. Second, although we found that both mDCs and pDCs are major sources of IFN-β, we did not survey other cell types or markers, and there may be overlapping phenotypes and other cellular sources of IFN-β in dermatomyositis. Other IFN-1 types, such as IFN-α, IFN-κ, and IFN-ω, exist and may also contribute to disease with similar or distinct expression patterns. We will explore this in future studies. Third, using LCM, we were unable to measure mRNA of HC mDCs or pDCs owing to low-cell counts in the tissues. pDCs were also too scarce in patients with dermatomyositis to extract RNA. Therefore, only IFN-β of mDCs and KCs in dermatomyositis skin biopsies could be compared. Finally, in comparing mDCs and pDCs in HCQ responders with those in nonresponders, there were more patients with systemic disease in the HCQ nonresponder group than in the responder group.
      Several studies have illustrated that IFN-β is an important driver in dermatomyositis disease pathogenesis (
      • Chen M.
      • Quan C.
      • Diao L.
      • Xue F.
      • Xue K.
      • Wang B.
      • et al.
      Measurement of cytokines and chemokines and association with clinical severity of dermatomyositis and clinically amyopathic dermatomyositis.
      ;
      • Huard C.
      • Gullà S.V.
      • Bennett D.V.
      • Coyle A.J.
      • Vleugels R.A.
      • Greenberg S.A.
      Correlation of cutaneous disease activity with type 1 interferon gene signature and interferon β in dermatomyositis.
      ;
      • Wong D.
      • Kea B.
      • Pesich R.
      • Higgs B.W.
      • Zhu W.
      • Brown P.
      • et al.
      Interferon and biologic signatures in dermatomyositis skin: specificity and heterogeneity across diseases.
      ). Our findings suggest that mDCs, in addition to pDCs, are major producers of IFN-β in patients with dermatomyositis but not in HCs. Increased numbers of mDCs were found in HCQ nonresponders. This poor response to HCQ may be a result of high IFN-β production by mDCs; however, further investigation is warranted.

      Materials and Methods

      Ethical statement

      The University of Pennsylvania Institutional Review Board approved human subject involvement in this study. All subjects in this study signed an informed consent document before donating blood or tissue.

      Patients

      All patients were diagnosed with dermatomyositis (classic or amyopathic) by VPW using either Bohan and Peter criteria (
      • Bohan A.
      • Peter J.B.
      Polymyositis and dermatomyositis (first of two parts).
      ) or Sontheimer criteria (
      • Concha J.S.S.
      • Tarazi M.
      • Kushner C.J.
      • Gaffney R.G.
      • Werth V.P.
      The diagnosis and classification of amyopathic dermatomyositis: a historical review and assessment of existing criteria.
      ;
      • Sontheimer R.D.
      Dermatomyositis: an overview of recent progress with emphasis on dermatologic aspects.
      ). In our immunohistochemistry quantification of inflammatory cells, our immunofluorescence of mDC and IFN-β, our LCM of mDCs, and our mass cytometry experiments, lesional skin biopsies were obtained from 12 patients with moderate-to-severe cutaneous dermatomyositis despite standard of care background therapy. Details on demographics are presented in Supplementary Table S1. In our flow cytometric analysis of mDC versus that of pDC production of IFN-β, we compared PBMCs from five HCs and five patients with dermatomyositis (three amyopathic, two classic) from a longitudinal dermatomyositis database. Their demographic details are given in Supplementary Table S2. Finally, in our evaluation of mDCs, pDCs, IFN-α, and IFN-β expression in HCQ responders versus that in nonresponders, lesional dermatomyositis skin was obtained from biopsies of patients subsequently characterized in terms of responsiveness to HCQ on the basis of disease activity data prospectively collected in a longitudinal database. Their demographic details in given Supplementary Table S3. Response to HCQ was defined as sufficient improvement in skin lesions so that further escalation of therapy was not needed.

      Immunohistochemistry

      Formalin fixed and paraffin embedded (FFPE) 4-mm skin biopsies were cut into 5-μm sections and placed onto glass slides. Slides were incubated at 60 °C overnight, deparaffinized in Citrisolv (Thermo Fisher Scientific, Waltham, MA), and rehydrated with serial ethanol dilutions. Antigen retrieval was achieved in EDTA (Thermo Fisher Scientific) in a 95 °C pressure cooker for 20 minutes. Peroxidase was deactivated with 0.3% hydrogen peroxide (Thermo Fisher Scientific) in deionized water for 5 minutes. Tissues were blocked in a 5% BSA protein block (Dako, Carpinteria, CA) for 4 hours and incubated with primary antibody at 4 °C overnight. Tissues were stained for CD4 T cells, CD8 T cells, mast cells (tryptase), CD69 cells, mDCs (CD11c), and pDCs (CD123). Details of the antibodies and dilutions used are given in Supplementary Table S4. Samples were probed with biotinylated anti-goat and/or anti-mouse and/or anti-rabbit (ab64257, Abcam, Cambridge, United Kingdom) for 50 minutes and streptavidin-horseradish peroxidase (Dako) for 30 minutes. Tissues were developed using NovaRED chromogen (Vector Laboratories, Burlingame, CA) for 5–10 minutes. Tissues were dehydrated using ethanol serial dilution and Citrisolv. Tissues were mounted with Permount (Thermo Fisher Scientific) and glass coverslips. Sections were analyzed with the Nikon Eclipse 80i microscope (Nikon, Tokyo, Japan). Cells in the dermis were quantified in five nonoverlapping ×40 objective magnification fields.

      Image mass cytometry triple labeling

      FFPE 4-mm biopsies were prepared, deparaffinized, and antigen retrieved as described earlier. Metal-conjugated antibodies were prepared using three Maxpar X8 Antibody Labeling Kits for metals: 173Yb, 174Yb, and 175Lu (Fluidigm, San Francisco, CA). Tissues were blocked in 3% BSA for 1 hour at room temperature and then incubated with a cocktail of metal-conjugated CD11c, HLA-DR, and CD163 antibodies. Details of the antibodies and dilutions used are given in Supplementary Table S5. Tissues were washed and incubated with Intercalator-Ir (Fluidigm) for 30 minutes at room temperature. The final wash was done in deionized water, and slides were subsequently air dried for 30 minutes. Regions of interest (2 mm × 1 mm) were ablated at a frequency of 200 Hz on the Hyperion Imaging System (Fluidigm). Images were extracted using MCDViewer (Fluidigm), and composites were created in ImageJ (National Institutes of Health, Bethesda, MD).

      Immunofluorescence

      FFPE 4-mm skin biopsies were prepared on glass slides, deparaffinized, and antigen retrieved as described earlier. Tissues were blocked in 5% BSA (Dako) for 1 hour and incubated with CD11c, IFN-α, and IFN-β primary antibodies (Supplementary Table S4) at 4 °C overnight. Tissues were washed and incubated with secondary antibodies—either goat-anti-mouse-594 or goat-anti-rabbit-488 (Thermo Fisher Scientific). Tissues were washed again and treated with TrueView (Vector Laboratories) for 2 minutes to reduce autofluorescence, after which they were washed, stained with DAPI (Thermo Fisher Scientific), and mounted using Vector antifade mounting medium (Vector Laboratories). Composite images were created using ImageJ.

      LCM

      FFPE slides were cut RNase free onto polyethylene naphthalate membrane slides and deparaffinized in xylene three times for 20 seconds each and in serial dilutions of ethanol for 30 seconds two times each. Slides were blocked with 10% normal goat serum (Dako) and incubated with CD11c (ab52632, Abcam) in 5% normal goat serum and/or 2% BSA and/or 0.05% Tween-20 and/or RNase inhibitor for 10 minutes and then incubated with secondary antibody goat anti‒rabbit-594 (Thermo Fischer Scientific) for 15 minutes. All steps were used with RNase-free agents. CD11c+ cells were identified and laser cut into adhesive caps (Zeiss, Oberkochen, Germany). KCs were captured for control comparison and identified by morphology and location. RNA was extracted from laser-captured cells using the RNeasy FFPE Kit (Qiagen, Valencia, CA) and converted into cDNA using the High-Capacity cDNA Reverse Transcriptase Kit (Applied Biosystems, Foster City, CA). cDNA was amplified by quanitative real-time reverse transcriptase-PCR using the TaqMan custom-designed array card assay, and mRNA was measured through the VIIA 7 Real-Time PCR (Applied Biosystems). mRNA levels were normalized to PPIA (Hs99999904_m1, Applied Biosystems); relative IFN-β mRNA expression was identified using the following probe: Hs01077958_s1 (IFN-β, Applied Biosystems). Relative gene expression was calculated using the comparative CT method, that is, ΔΔCt.

      Flow cytometry

      To analyze the cellular source of IFN-β, blood samples were obtained by venous puncture from five HCs and five patients with cutaneous dermatomyositis. PBMCs were isolated by density gradient over Ficoll-Hypaque by standard procedures. Isolated PBMCs at 1 × 106 cells/ml were cultured in the presence of R848 (1 μg/ml) (Invivogen, San Diego, CA) and Brefeldin A (BioLegend, San Diego, CA) for a total of 5 hours. A total of 2 × 106 cells were aliquoted into each FACS tube, pretreated with staining buffer (2% fetal calf serum), blocked with mouse IgG2b (ab18428, Abcam), and stained with surface antibodies (Lineage cocktail, HLA-DR, CD11c, and CD123) for 25 minutes. The details of the conjugated antibodies used are given in Supplementary Table S6. Cells were washed in PBS, fixed and permeabilized, stained for IFN-β for 20 minutes, and resuspended in 0.2 ml PBS. Single-cell suspensions underwent flow cytometric analysis on an LSR Fortessa B flow cytometer (BD Biosciences, San Jose, CA) and were analyzed with FlowJo software (Flowjo, Ashland, OR). A total of 150,000–200,000 events were collected for each analysis. Dendritic cells were identified by gating on leukocytes and by identifying the HLA-DR+ and Lineage negative, that is, Lin−, subpopulations. This excluded CD3+ T cells, CD14+ monocytes and macrophages, CD16+ and CD56+ NK cells, CD16+ neutrophils, and CD19+ and CD20+ B cells. Using this subpopulation, mDCs and pDCs were then identified by CD11c and CD123, respectively. To control for experimentally observed nonspecific binding of the IFN-β antibody, we established gates on the basis of IFN-β‒positive cells using mean fluorescence intensity on unstimulated cells. These gates were applied to R848 stimulated cells to identify the true IFN-β‒positive cells.

      Statistics

      Statistical analysis was performed using GraphPad Prism7 software. Mann‒Whitney U test was used to compare relative IFN-β mRNA expression from KCs and CD11c+ cells, percentages of HLA-DR+Lin− cells that expressed CD11c or CD123 measured by flow cytometry, IFN-β release from either mDCs or pDCs measured by flow cytometry, and the expression of CD11c+ and CD123+ cells in HCQ responders versus that in HCQ nonresponders. All statistics were reported as median ± interquartile range because the data were nonparametric.

      Data availability statement

      There are no datasets available for this submission.

      Conflict of Interest

      BW is an employee and stockholder of Corbus Pharmaceuticals, a company that provided the lenabasum used in this study. VPW (University of Pennsylvania, Philadelphia, PA) owns the copyright for the Cutaneous Dermatomyositis Disease Area and Severity Index. The remaining authors state no conflict of interest.

      Acknowledgments

      This project is supported by the National Institutes of Health ( R21 AAR066286 ), the Myositis Association, the Department of Veterans Affairs Veterans Health Administration, Office of Research and Development , Biomedical Laboratory Research and Development, and Core A of the Penn Skin Biology and Diseases Resource-based Center ( P30 AR069589-01 ).

      Author Contributions

      Conceptualization: KLC, JP, VPW; Formal Analysis: KLC, JP, MZ, MW, VPW; Funding Acquisition: VPW; Investigation: KLC, JP, MZ, MW, MMB, BP, SM; Methodology: KLC, JP, MZ, MW, MMB, BP, SM, VPW; Project Administration: VPW; Resources: BW, VPW; Supervision: VPW; Validation: KLC, JP, MZ, MW, MMB, BP, SM; Visualization: KLC, JP, MZ, MW, BP; Writing - Original Draft Preparation: KLC, JP; Writing - Review and Editing: MZ, MW, MMB, BP, SM, BW, VPW

      Supplementary Materials

      Supplementary Table S1Demographics for Lesional Skin of Patients with Moderate-to-Severe Dermatomyositis Despite Background Therapy
      DemographicsLesional Skin of Patients with Moderate-to-Severe Dermatomyositis (n = 12)
      Age, y, median ± IQR56 ± 9.5
      Sex, n (%)
       Male1
       Female11
      Race, n (%)
       Caucasian11
       Non-Caucasian1
      Smoking history,
      All subjects quit >20 years before.
      n (%)
       Yes6 (50)
       No6 (50)
      Disease subtype, n (%)
       Classic0 (0)
       Amyopathic12 (100)
      CDASI activity score, median ± IQR33.5 ± 9
      Background therapyAzathioprine (n = 1)
      HCQ (n = 6)
      Methotrexate (n = 4)
      Mycophenolate mofetil (n = 5)
      Quinacrine (n = 2)
      Topical corticosteroid (n = 2)
      Abbreviations: CDASI, Cutaneous Dermatomyositis Disease Area and Severity Index; HCQ, hydroxychloroquine; IQR, interquartile range.
      1 All subjects quit >20 years before.
      Supplementary Table S2Demographics for PBMC Flow Cytometry
      DemographicsHCs (n = 5)Patients with Dermatomyositis (n = 5)
      Age, y, median ± IQR41 ± 22.551 ± 35
      Sex, n (%)
       Male31
       Female24
      Race, n (%)
       Caucasian14
       Non-Caucasian40
      Smoking history, n (%)
       Yes00
       No00
      Disease subtype, n (%)
       ClassicNA2
       AmyopathicNA3
      CDASI activity score, median ± IQRNA17 ± 15.5
      Abbreviations: CDASI, Cutaneus Dermatomyositis Disease Area and Severity Index; HC, healthy control; IQR, interquartile range; NA, not applicable.
      Supplementary Table S3Demographics for mDCs and/or pDCs for HCQ Responders versus HCQ Nonresponders
      DemographicsHCQ Responders (n = 8)HCQ Nonresponders

      (n = 8)
      Age, y, median ± IQR60 ± 27.554 ± 10
      Sex, n (%)
       Male01
       Female87
      Race, n (%)
       Caucasian87
       Non-Caucasian01
      Smoking history, n (%)
       Yes04
       No84
      Disease subtype, n (%)
       Classic37
       Amyopathic51
      Abbreviations: HCQ, hydroxychloroquine; IQR, interquartile range; mDC, myeloid dendritic cell; pDC, plasmacytoid dendritic cell.
      Supplementary Table S4Primary Antibodies and Dilutions for Immunohistochemistry and Immunofluorescence
      AntibodyCompany, Catalog NumberHost SpeciesDilution
      CD11cAbcam, ab52632Rabbit1:200
      CD11cThermo Fisher Scientific, 14-9761-82Mouse1:40
      CD123Cell Marque, 6H6Mouse1:25
      CD4Abcam, ab133616Rabbit1:2,000
      CD8Abcam, ab93278Rabbit1:4,000
      CD69Abcam, ab233396Rabbit1:400
      IFN-αSanta Cruz, sc373757Mouse1:100
      IFN-βAbcam, ab140211Rabbit1:300
      TryptaseAbcam, ab2378Mouse1:25,000
      Supplementary Table S5Conjugated Antibodies and Dilutions for Mass Cytometry
      AntibodyCompany, Catalog NumberHost SpeciesDilution
      CD11cAbcam, ab52632Rabbit1:100
      CD163Bio-Rad, mca1853Mouse1:100
      HLA-DRAbcam, ab20181Mouse1:400
      Supplementary Table S6Conjugated Antibodies and Dilutions for Flow Cytometry
      AntibodyCompany, Catalog NumberHost SpeciesExperimental Dilution (Per 1 ml)
      APC anti-human lineage cocktailBioLegend, 348803Mouse20 μl
      APC/Cy7 anti-HLA-DRBioLegend, 307618Mouse3 μl
      PE/Cy7 anti-human CD11cBioLegend, 337216Mouse5 μl
      PE/Cy7 anti-human CD123 (6H6)BioLegend, 306010Mouse5 μl
      PE anti-CD123BioLegend, 306006Mouse5 μl
      FITC anti-ΙFΝ−βPBL, 21400-3Mouse0.5 μl
      Abbreviation: APC, antigen-presenting cell.

      References

        • Ali S.
        • Mann-Nüttel R.
        • Schulze A.
        • Richter L.
        • Alferink J.
        • Scheu S.
        Sources of type I interferons in infectious immunity: plasmacytoid dendritic cells not always in the driver's seat.
        Front Immunol. 2019; 10: 778
        • Anyanwu C.O.
        • Chansky P.B.
        • Feng R.
        • Carr K.
        • Okawa J.
        • Werth V.P.
        The systemic management of cutaneous dermatomyositis: results of a stepwise strategy.
        Int J Womens Dermatol. 2017; 3: 189-194
        • Beitnes A.C.
        • Ráki M.
        • Brottveit M.
        • Lundin K.E.
        • Jahnsen F.L.
        • Sollid L.M.
        Rapid accumulation of CD14+CD11c+ dendritic cells in gut mucosa of celiac disease after in vivo gluten challenge.
        PLoS One. 2012; 7: e33556
        • Bendewald M.J.
        • Wetter D.A.
        • Li X.
        • Davis M.D.
        Incidence of dermatomyositis and clinically amyopathic dermatomyositis: a population-based study in Olmsted County, Minnesota.
        Arch Dermatol. 2010; 146: 26-30
        • Bohan A.
        • Peter J.B.
        Polymyositis and dermatomyositis (first of two parts).
        N Engl J Med. 1975; 292: 344-347
        • Caproni M.
        • Torchia D.
        • Cardinali C.
        • Volpi W.
        • Del Bianco E.
        • D’Agata A.
        • et al.
        Infiltrating cells, related cytokines and chemokine receptors in lesional skin of patients with dermatomyositis.
        Br J Dermatol. 2004; 151: 784-791
        • Chen M.
        • Quan C.
        • Diao L.
        • Xue F.
        • Xue K.
        • Wang B.
        • et al.
        Measurement of cytokines and chemokines and association with clinical severity of dermatomyositis and clinically amyopathic dermatomyositis.
        Br J Dermatol. 2018; 179: 1334-1341
        • Collin M.
        • Bigley V.
        Human dendritic cell subsets: an update.
        Immunology. 2018; 154: 3-20
        • Concha J.S.S.
        • Tarazi M.
        • Kushner C.J.
        • Gaffney R.G.
        • Werth V.P.
        The diagnosis and classification of amyopathic dermatomyositis: a historical review and assessment of existing criteria.
        Br J Dermatol. 2019; 180: 1001-1008
        • Crouse J.
        • Kalinke U.
        • Oxenius A.
        Regulation of antiviral T cell responses by type I interferons.
        Nat Rev Immunol. 2015; 15: 231-242
        • Crow M.K.
        • Ronnblom L.
        Type I interferons in host defence and inflammatory diseases.
        Lupus Sci Med. 2019; 6: e000336
        • Eguíluz-Gracia I.
        • Bosco A.
        • Dollner R.
        • Melum G.R.
        • Lexberg M.H.
        • Jones A.C.
        • et al.
        Rapid recruitment of CD14(+) monocytes in experimentally induced allergic rhinitis in human subjects.
        J Allergy Clin Immunol. 2016; 137: 1872-1881.e12
        • Fitzgerald-Bocarsly P.
        • Dai J.
        • Singh S.
        Plasmacytoid dendritic cells and type I IFN: 50 years of convergent history.
        Cytokine Growth Factor Rev. 2008; 19: 3-19
        • Gallucci S.
        • Lolkema M.
        • Matzinger P.
        Natural adjuvants: endogenous activators of dendritic cells.
        Nat Med. 1999; 5: 1249-1255
        • Goreshi R.
        • Chock M.
        • Foering K.
        • Feng R.
        • Okawa J.
        • Rose M.
        • et al.
        Quality of life in dermatomyositis.
        J Am Acad Dermatol. 2011; 65: 1107-1116
        • Greenberg S.A.
        Dermatomyositis and type 1 interferons.
        Curr Rheumatol Rep. 2010; 12: 198-203
        • Grimm M.C.
        • Pullman W.E.
        • Bennett G.M.
        • Sullivan P.J.
        • Pavli P.
        • Doe W.F.
        Direct evidence of monocyte recruitment to inflammatory bowel disease mucosa.
        J Gastroenterol Hepatol. 1995; 10: 387-395
        • Hall J.C.
        • Rosen A.
        Type I interferons: crucial participants in disease amplification in autoimmunity.
        Nat Rev Rheumatol. 2010; 6: 40-49
        • Huard C.
        • Gullà S.V.
        • Bennett D.V.
        • Coyle A.J.
        • Vleugels R.A.
        • Greenberg S.A.
        Correlation of cutaneous disease activity with type 1 interferon gene signature and interferon β in dermatomyositis.
        Br J Dermatol. 2017; 176: 1224-1230
        • Jenner W.
        • Motwani M.
        • Veighey K.
        • Newson J.
        • Audzevich T.
        • Nicolaou A.
        • et al.
        Characterisation of leukocytes in a human skin blister model of acute inflammation and resolution.
        PLoS One. 2014; 9: e89375
        • Kao L.
        • Chung L.
        • Fiorentino D.F.
        Pathogenesis of dermatomyositis: role of cytokines and interferon.
        Curr Rheumatol Rep. 2011; 13: 225-232
        • Kiefer K.
        • Oropallo M.A.
        • Cancro M.P.
        • Marshak-Rothstein A.
        Role of type I interferons in the activation of autoreactive B cells.
        Immunol Cell Biol. 2012; 90: 498-504
        • Liao A.P.
        • Salajegheh M.
        • Morehouse C.
        • Nazareno R.
        • Jubin R.G.
        • Jallal B.
        • et al.
        Human plasmacytoid dendritic cell accumulation amplifies their type 1 interferon production.
        Clin Immunol. 2010; 136: 130-138
        • Magro C.M.
        • Segal J.P.
        • Crowson A.N.
        • Chadwick P.
        The phenotypic profile of dermatomyositis and lupus erythematosus: a comparative analysis.
        J Cutan Pathol. 2010; 37: 659-671
        • Mainetti C.
        • Terziroli Beretta-Piccoli B.
        • Selmi C.
        Cutaneous manifestations of dermatomyositis: a comprehensive review.
        Clin Rev Allergy Immunol. 2017; 53: 337-356
        • Sontheimer R.D.
        Dermatomyositis: an overview of recent progress with emphasis on dermatologic aspects.
        Dermatol Clin. 2002; 20: 387-408
        • Wong D.
        • Kea B.
        • Pesich R.
        • Higgs B.W.
        • Zhu W.
        • Brown P.
        • et al.
        Interferon and biologic signatures in dermatomyositis skin: specificity and heterogeneity across diseases.
        PLoS One. 2012; 7: e29161
        • Zaba L.C.
        • Fuentes-Duculan J.
        • Eungdamrong N.J.
        • Abello M.V.
        • Novitskaya I.
        • Pierson K.C.
        • et al.
        Psoriasis is characterized by accumulation of immunostimulatory and Th1/Th17 cell-polarizing myeloid dendritic cells.
        J Invest Dermatol. 2009; 129: 79-88
        • Zeidi M.
        • Kim H.J.
        • Werth V.P.
        Increased myeloid dendritic cells and TNF-α expression predicts poor response to hydroxychloroquine in cutaneous lupus erythematosus.
        J Invest Dermatol. 2019; 139: 324-332