Differential expression of junctional adhesion molecules in different stages of systemic sclerosis

Authors

  • Mirko Manetti,

    Corresponding author
    1. Azienda Ospedaliero-Universitaria Careggi (AOUC), Excellence Centre for Research, Transfer and High Education on Chronic, Inflammatory, Degenerative and Neoplastic Disorders for the Development of Novel Therapies (DENOthe), University of Florence, Florence, Italy
    • Department of Anatomy, Histology, and Forensic Medicine, Section of Anatomy, University of Florence, Largo Brambilla 3, 50134 Florence, Italy
    Search for more papers by this author
    • Drs. Manetti and Guiducci contributed equally to this work.

  • Serena Guiducci,

    1. Azienda Ospedaliero-Universitaria Careggi (AOUC), Excellence Centre for Research, Transfer and High Education on Chronic, Inflammatory, Degenerative and Neoplastic Disorders for the Development of Novel Therapies (DENOthe), University of Florence, Florence, Italy
    Search for more papers by this author
    • Drs. Manetti and Guiducci contributed equally to this work.

  • Eloisa Romano,

    1. Azienda Ospedaliero-Universitaria Careggi (AOUC), Excellence Centre for Research, Transfer and High Education on Chronic, Inflammatory, Degenerative and Neoplastic Disorders for the Development of Novel Therapies (DENOthe), University of Florence, Florence, Italy
    Search for more papers by this author
  • Irene Rosa,

    1. University of Florence, Florence, Italy
    Search for more papers by this author
  • Claudia Ceccarelli,

    1. Azienda Ospedaliero-Universitaria Careggi (AOUC), Excellence Centre for Research, Transfer and High Education on Chronic, Inflammatory, Degenerative and Neoplastic Disorders for the Development of Novel Therapies (DENOthe), University of Florence, Florence, Italy
    Search for more papers by this author
  • Tommaso Mello,

    1. University of Florence, Florence, Italy
    Search for more papers by this author
  • Anna Franca Milia,

    1. Azienda Ospedaliero-Universitaria Careggi (AOUC), Excellence Centre for Research, Transfer and High Education on Chronic, Inflammatory, Degenerative and Neoplastic Disorders for the Development of Novel Therapies (DENOthe), University of Florence, Florence, Italy
    Search for more papers by this author
  • Maria Letizia Conforti,

    1. Azienda Ospedaliero-Universitaria Careggi (AOUC), Excellence Centre for Research, Transfer and High Education on Chronic, Inflammatory, Degenerative and Neoplastic Disorders for the Development of Novel Therapies (DENOthe), University of Florence, Florence, Italy
    Search for more papers by this author
  • Lidia Ibba-Manneschi,

    1. University of Florence, Florence, Italy
    Search for more papers by this author
    • Drs. Ibba-Manneschi and Matucci-Cerinic contributed equally to this work.

  • Marco Matucci-Cerinic

    1. Azienda Ospedaliero-Universitaria Careggi (AOUC), Excellence Centre for Research, Transfer and High Education on Chronic, Inflammatory, Degenerative and Neoplastic Disorders for the Development of Novel Therapies (DENOthe), University of Florence, Florence, Italy
    Search for more papers by this author
    • Drs. Ibba-Manneschi and Matucci-Cerinic contributed equally to this work.


Abstract

Objective

Systemic sclerosis (SSc) is characterized by early perivascular inflammation, microvascular endothelial cell (MVEC) activation/damage, and defective angiogenesis. Junctional adhesion molecules (JAMs) regulate leukocyte recruitment to sites of inflammation and ischemia-reperfusion injury, vascular permeability, and angiogenesis. This study was undertaken to investigate the possible role of JAMs in SSc pathogenesis.

Methods

JAM-A and JAM-C expression levels in skin biopsy samples from 25 SSc patients and 15 healthy subjects were investigated by immunohistochemistry and Western blotting. Subcellular localization of JAMs in cultured healthy dermal MVECs and SSc MVECs was assessed by confocal microscopy. Serum levels of soluble JAM-A (sJAM-A) and sJAM-C in 64 SSc patients and 32 healthy subjects were examined by enzyme-linked immunosorbent assay.

Results

In control skin, constitutive JAM-A expression was observed in MVECs and fibroblasts. In early-stage SSc skin, JAM-A expression was strongly increased in MVECs, fibroblasts, and perivascular inflammatory cells. In late-stage SSc, JAM-A expression was decreased compared with controls. JAM-C was weakly expressed in control and late-stage SSc skin, while it was strongly expressed in MVECs, fibroblasts, and inflammatory cells in early-stage SSc. Surface expression of JAM-A was higher in early-stage SSc MVECs and increased in healthy MVECs stimulated with early-stage SSc sera. JAM-C was cytoplasmic in resting healthy MVECs, while it was recruited to the cell surface upon challenge with early-stage SSc sera. Early-stage SSc MVECs exhibited constitutive surface JAM-C expression. In SSc, increased levels of sJAM-A and sJAM-C correlated with early disease and measures of vascular damage.

Conclusion

Our findings indicate that JAMs may participate in MVEC activation, inflammatory processes, and impaired angiogenesis in different stages of SSc.

Systemic sclerosis (SSc) is a multisystem connective tissue disease characterized by small-vessel vasculopathy, impaired angiogenesis, immune dysregulation, and progressive fibrosis of the skin and internal organs (1–3). Although the etiology of SSc remains unknown, complex interactions among leukocytes, endothelial cells (ECs), and fibroblasts are likely to be central to the pathogenesis of the disease (1, 4). In its early stages, SSc is characterized by EC activation/damage and perivascular mononuclear cell infiltration with an accumulation of T and B lymphocytes and monocytes in the affected tissues (2, 3, 5, 6). Furthermore, increased extracellular matrix (ECM) deposition has been suggested to be a consequence of inflammatory and immune processes clustered around small vessels in the dermis (7).

The interactions of leukocytes with ECs and fibroblasts are highly dependent on the expression and function of cell surface adhesion molecules. A number of studies have suggested that aberrant expression of adhesion molecules on leukocytes and ECs may result in the accumulation of specific subsets of activated leukocytes in the tissues of SSc patients, which may trigger the fibrotic process by the release of a large array of cytokines, chemokines, and growth factors that stimulate the transdifferentiation of resting fibroblasts into myofibroblasts and the synthesis of ECM components (8–12). Moreover, the enhanced interaction of fibroblasts with activated T cells may also contribute to abnormal ECM synthesis and deposition (13).

Adhesion molecules and their soluble forms also play important roles in angiogenesis (14). Endothelial adhesion molecules can mediate angiogenesis both directly and indirectly by promoting the influx of monocytes into tissues where they can differentiate into macrophages and secrete proangiogenic factors (14). There is also evidence that increased levels of soluble adhesion molecules released by ECs, such as soluble E-selectin, vascular cell adhesion molecule 1 (VCAM-1), and intercellular adhesion molecule 1 (ICAM-1), may reflect the ongoing EC activation state in early stages of SSc and correlate with the presence and severity of specific organ complications (15–17).

Junctional adhesion molecules (JAMs), such as JAM-A (JAM-1/F11 receptor) and JAM-C (JAM-3), are members of the immunoglobulin superfamily that regulate leukocyte transmigration across EC surfaces by their ability to undergo heterophilic binding with the leukocyte integrins lymphocyte function–associated antigen 1 (LFA-1) and Mac-1, respectively (18, 19). In addition, JAMs interact homophilically at endothelial and epithelial tight junctions and participate in the regulation of paracellular permeability. JAMs are broadly expressed on blood and lymphatic ECs, epithelial cells, fibroblasts, and circulating cells, such as monocytes, lymphocytes, and platelets (18, 19). Studies using genetically modified mice and specific blocking antibodies provided evidence of a crucial role of JAMs in the regulation of leukocyte recruitment to sites of inflammation and ischemia-reperfusion injury, vascular permeability, growth factor–mediated angiogenesis, atherogenesis, and neointima formation (18). Moreover, JAMs have been implicated in a variety of physiologic and pathologic processes in humans, including tumors, hypertension, rheumatoid arthritis, and inflammatory bowel disease (19–22).

On this basis, we hypothesized that JAMs could be involved in the complex intercellular cross-talk among ECs, inflammatory/immune cells, and fibroblasts, playing multiple roles in different stages of SSc. We therefore investigated the expression of JAM-A and JAM-C in skin and dermal microvascular ECs (MVECs), as well as circulating levels of soluble JAM-A (sJAM-A) and sJAM-C, in samples from SSc patients.

PATIENTS AND METHODS

Patients, controls, skin biopsy samples, and serum samples.

Full-thickness skin biopsy samples were obtained from the clinically involved skin of one-third of the distal forearm of 25 SSc patients (21 women and 4 men) who were recruited from the Division of Rheumatology of the University of Florence. Patients were classified as having limited cutaneous SSc (lcSSc; n = 13) or diffuse cutaneous SSc (dcSSc; n = 12) according to the classification system of LeRoy et al (23). Disease duration was calculated from the time of onset of the first clinical event (other than Raynaud's phenomenon) that was a clear manifestation of SSc. Patients were classified as having early-stage (n = 14) or late-stage (n = 11) SSc according to disease duration (<5 years for early-stage lcSSc and <3 years for early-stage dcSSc) and skin histopathology (24). Before biopsy, patients had not received any disease-modifying drugs. Skin samples from the same region of the forearm of 15 age- and sex-matched healthy donors were used as controls. Each skin biopsy sample was divided into 2 specimens. For immunohistochemistry, the specimens were fixed in 10% buffered formalin and embedded in paraffin. For protein extraction, the specimens were immersed in liquid nitrogen and stored at −80°C until used.

Serum samples were obtained from 64 SSc patients (57 women and 7 men; 41 with lcSSc and 23 with dcSSc) and 32 age- and sex-matched healthy individuals. Twenty-five patients had early-stage SSc and 39 had late-stage SSc. At the time blood was drawn, the presence of digital ulcers was recorded. All patients underwent nailfold videocapillaroscopy and were divided into 3 groups based on capillaroscopy patterns: early, active, and late (25). Patients were not taking immunosuppressive or disease-modifying drugs. Before blood was sampled, patients underwent a washout period of 10 days from oral vasodilating drugs and of 2 months from intravenous alprostadil α-cyclodextrin. Fresh venous blood samples were drawn, allowed to clot for 30 minutes before centrifugation at 1,500g for 15 minutes, and serum was collected and stored in aliquots at −80°C until used. The study was approved by the local Institutional Review Board, and written informed consent was obtained from all subjects.

Immunohistochemistry and confocal laser scanning microscopy.

Sections (5 μm thick) were deparaffinized and either stained with hematoxylin and eosin for routine histology or processed for immunohistochemistry. For antigen retrieval, skin sections were boiled for 10 minutes in citrate buffer (10 mM, pH 6.0). The sections were then washed in phosphate buffered saline (PBS) and blocked for 1 hour at room temperature with 1% bovine serum albumin (BSA) in PBS. The slides were incubated overnight at 4°C with the following anti-human antibodies: anti–JAM-A rabbit monoclonal antibody (1:50) (catalog no. ab52647; Abcam), anti–JAM-C rabbit polyclonal antibody (1:50) (catalog no. HPA003417; Atlas Antibodies), and anti-podoplanin (D2-40) mouse monoclonal antibody/lymphatic EC–specific marker (1:50) (catalog no. M3619; Dako). The immune reactions were revealed using Alexa Fluor 488–conjugated goat anti-rabbit IgG or rhodamine red X–conjugated goat anti-mouse IgG (1:200; Invitrogen). Double immunostainings with mouse and rabbit reagents were performed by mixing primary antibodies and subsequently mixing fluorochrome-conjugated reagents. Irrelevant isotype- and concentration-matched IgG (Sigma-Aldrich) were used as negative controls. Tissue sections were examined with a Leica TCS SP5 confocal laser scanning microscope (Leica Microsystems) equipped with a Leica Plan Apo 63×/1.4 numerical aperture oil immersion objective and a HeNe/Argon laser source for fluorescence measurements. Series of optical sections (1,024 × 1,024 pixels each) at intervals of 0.4 μm were obtained and superimposed to create a single composite image.

Isolation of dermal MVECs.

Dermal MVECs were isolated from biopsy samples of the involved forearm skin from 3 patients with early-stage dcSSc and from 3 healthy subjects. Patients were not taking immunosuppressive or disease-modifying drugs at the time of biopsy. Skin biopsy samples were mechanically cleaned of epidermis and adipose tissue in order to obtain a pure specimen of vascularized dermis, and were treated as previously described (26). Colonies of polygonal elements were detached with EDTA, and CD31-positive cells were subjected to immunomagnetic isolation with Dynabeads CD31 (Dynal) (27). Isolated cells were further identified as MVECs by labeling with anti–factor VIII–related antigen and anti-CD105, followed by reprobing with anti-CD31 antibodies. Healthy MVECs and SSc MVECs were maintained in MCDB medium (Sigma-Aldrich) supplemented with 30% fetal bovine serum (FBS), 20 μg/ml EC growth supplement (Calbiochem), 10 μg/ml hydrocortisone, 15 IU/ml heparin, and antibiotics. MVECs were used between the third and seventh passages in culture.

MVEC stimulation and immunocytochemistry.

MVECs were grown to confluence on glass coverslips. In some experiments, healthy MVECs were starved in MCDB with 2% FBS overnight and then stimulated with 30% serum from patients with early-stage dcSSc (n = 5), patients with late-stage dcSSc (n = 3), and healthy individuals (n = 5) for 1 and 6 hours. MVECs were fixed in 3.7% buffered paraformaldehyde and permeabilized with 0.1% Triton X-100 in PBS. Nonspecific antibody binding was blocked with 1% BSA in PBS. MVECs were incubated overnight with a mixture of anti–JAM-A rabbit monoclonal antibody (1:50; Abcam) and anti–integrin αvβ3 mouse monoclonal antibody (1:50; catalog no. MAB1976; Chemicon International), or anti–JAM-C rabbit polyclonal antibody (1:50; Atlas Antibodies) and anti–zonula occludens 1 (anti–ZO-1) mouse monoclonal antibody (1:30) (catalog no. 33-9100; Invitrogen). Alexa Fluor 488–conjugated goat anti-rabbit IgG and Alexa Fluor 635–conjugated goat anti-mouse IgG (1:200; Invitrogen) were used as secondary antibodies. Negative controls were obtained by incubation with isotype-matched normal IgG. Immunolabeled cells were examined with a Leica SP2-AOBS confocal microscope (Leica Microsystems). Series of optical sections (1,024 × 1,024 pixels each) were obtained and superimposed to create a single composite image.

Western blotting.

Proteins were extracted from skin specimens by homogenization in ice-cold lysis buffer (50 mM Tris HCl [pH 7.4], 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, and 0.25% sodium dodecyl sulfate [SDS]) supplemented with 1 mM phenylmethylsulfonyl fluoride, 1 mM sodium orthovanadate, and the protease inhibitor cocktail (Roche Diagnostics). For protein extraction from cells, confluent monolayers of MVECs were washed with PBS and then scraped in ice-cold lysis buffer. The solution was assayed for protein content using the Bradford method. Fifty micrograms of protein was subjected to electrophoresis in SDS–12% polyacrylamide gel under reducing conditions and blotted onto a nitrocellulose transfer membrane (Amersham Biosciences). The membranes were blocked in 5% nonfat dry milk with 0.05% Tween 20 in PBS, and then incubated overnight at 4°C with anti–JAM-A goat antibody (1 μg/ml) (catalog no. AF1103; R&D Systems) and anti–JAM-C goat antibody (1 μg/ml) (catalog no. AF1189; R&D Systems). After incubation with horseradish peroxidase–conjugated anti-goat IgG (Santa Cruz Biotechnology), immune complexes were detected with the enhanced chemiluminescence detection system (Amersham Biosciences). Blots were stripped using a Re-Blot Plus Western blot recycling kit (Chemicon International) and reprobed with anti–α-tubulin rabbit monoclonal antibodies (1:1,000; Cell Signaling Technology) to confirm similar loading of the gels and efficiency in electrophoretic transfer. Recombinant human JAM-A and JAM-C proteins (catalog nos. 1103JM and 1189J3, respectively; R&D Systems) were used to confirm the specificity of anti–JAM-A and anti–JAM-C antibodies. Densitometric analysis of the bands was performed using ImageJ software (National Institutes of Health; online at http://rsbweb.nih.gov/ij).

Assays for sJAM-A and sJAM-C.

Levels of sJAM-A and sJAM-C in serum samples and culture supernatants from MVECs were measured by commercial quantitative colorimetric sandwich enzyme-linked immunosorbent assay (catalog nos. E91782Hu and E80637Hu, respectively; Uscn Life Science) according to the manufacturer's protocol. The detection range was 0.312–20 ng/ml for sJAM-A and 0.156–10 ng/ml for sJAM-C. Each sample was measured in duplicate.

Statistical analysis.

Data are expressed as the mean ± SD or median and interquartile range (IQR). Student's t-test and nonparametric Mann-Whitney U test were used where appropriate for statistical evaluation of the differences between 2 independent groups. P values less than 0.05 were considered significant.

RESULTS

Histopathologic analysis of skin sections.

In skin biopsy samples from patients with early-stage SSc, the main histopathologic features were represented by perivascular inflammatory infiltrates, edema around microvessels, and a variable extent of collagen accumulation in the papillary and reticular dermis (Figure 1A). In late-stage SSc, flattening of dermal papillae, a marked reduction in microvessel density, and severe fibrotic changes with tightly packed and irregularly distributed collagen bundles were evident (Figure 1B).

Figure 1.

Histopathologic analysis of skin samples from patients with systemic sclerosis (SSc). A, Skin sample from a patient with early-stage SSc, showing dermal inflammation characterized by perivascular mononuclear cell infiltrates composed of monocytes and lymphocytes (boxed area), with edema around the microvessels and a variable extent of fibrosis in the papillary and reticular dermis. Inset, Higher-magnification view of the boxed area, showing perivascular inflammatory cells. B, Skin sample from a patient with late-stage SSc, showing prominent collagen accumulation leading to dermal thickening and deposition of dense and closely packed collagen bundles throughout the dermis, with loss of the microvasculature and dermal structures, and flattening of dermal papillae. Hematoxylin and eosin stained; original magnification × 20 in A and B; × 40 in inset.

Differential expression of JAM-A and JAM-C in skin biopsy samples from SSc patients.

Expression of JAM-A and JAM-C in skin biopsy samples was evaluated using fluorescence immunohistochemistry and confocal laser scanning microscopy. The lymphatic EC–specific marker podoplanin (D2-40) was used in double immunofluorescence analyses to differentiate blood (D2-40–negative) and lymphatic (D2-40–positive) microvessels in skin sections.

In skin samples from controls, constitutive expression of JAM-A was observed in dermal blood and lymphatic ECs, fibroblasts, and epidermal keratinocytes (Figures 2A and B). In skin samples from patients with early-stage SSc, JAM-A expression was strongly increased in blood and lymphatic microvessels and fibroblasts in both the papillary and reticular dermis (Figures 2C–F). Keratinocytes also displayed enhanced JAM-A expression (Figures 2C and E). Moreover, in skin sections from patients with early-stage SSc, perivascular inflammatory cells infiltrating the dermis showed strong JAM-A immunopositivity (Figures 2C and D). In skin samples from patients with late-stage SSc, JAM-A expression was weaker than in controls in both the epidermis and dermis, with faint or even undetectable expression in most microvessels (Figures 2G and H). In contrast, some dermal fibroblasts still exhibited strong JAM-A immunopositivity (Figure 2G). Using Western blotting, JAM-A expression levels were evaluated in protein extracts from whole skin samples. JAM-A protein expression levels were significantly increased in early-stage SSc and significantly reduced in late-stage SSc compared with control skin (P < 0.01 and P < 0.05, respectively) (Figures 2I and J).

Figure 2.

Expression of junctional adhesion molecule A (JAM-A) in skin biopsy samples. A–H, Representative photomicrographs of double immunofluorescence staining for JAM-A (green) and the lymphatic endothelial cell–specific marker D2-40/podoplanin (red) in the papillary dermis and reticular dermis in skin biopsy samples from healthy controls (A and B), patients with early-stage systemic sclerosis (SSc) (C–F), and patients with late-stage SSc (G and H). White arrows indicate D2-40–negative blood microvessels, arrowheads indicate D2-40–positive lymphatic microvessels, yellow arrows indicate dermal fibroblasts, and asterisks indicate perivascular inflammatory cell infiltrates. e = epidermal keratinocytes. Original magnification × 63. I, Western blotting of total protein extracts from the skin of healthy subjects and SSc patients. With the anti–JAM-A antibodies, a protein band with the expected molecular weight of ∼40 kd was detected. Blots were stripped and reprobed with anti–α-tubulin antibodies as a loading control for normalization. Representative immunoblots are shown. rh JAM-A = recombinant human JAM-A. J, Intensity of the bands, quantified by densitometry. Values are the mean ± SD. ∗ = P < 0.01; # = P < 0.05 versus control, by Student's t-test. OD = optical density; AU = arbitrary units.

JAM-C expression was weak in healthy skin, while it was markedly increased in blood and lymphatic ECs, fibroblasts, and keratinocytes in skin samples from patients with early-stage SSc (Figures 3A–F). Inflammatory cells found around microvessels in the papillary and reticular dermis of early-stage SSc also exhibited strong JAM-C immunopositivity (Figures 3C and D). JAM-C expression was weak and similar to that in controls in dermal microvessels in skin samples from patients with late-stage SSc, while an intense JAM-C immunostaining was still observed in some fibroblasts (Figures 3G and H). The immunohistochemical findings were confirmed by Western blot analysis. JAM-C protein expression levels were significantly higher in skin samples from patients with early-stage SSc than in those from controls (P < 0.01) (Figures 3I and J). No significant differences in JAM-C expression were detected between skin samples from patients with late-stage SSc and those from controls (Figure 3J). No differences in JAM-A and JAM-C expression were observed between skin tissue from patients with lcSSc and those with dcSSc (data not shown).

Figure 3.

Expression of junctional adhesion molecule C (JAM-C) in skin biopsy samples. A–H, Representative photomicrographs of double immunofluorescence staining for JAM-C (green) and the lymphatic endothelial cell–specific marker D2-40/podoplanin (red) in the papillary dermis and reticular dermis in skin biopsy samples from healthy controls (A and B), patients with early-stage systemic sclerosis (SSc) (C–F), and patients with late-stage SSc (G and H). White arrows indicate D2-40–negative blood microvessels, arrowheads indicate D2-40–positive lymphatic microvessels, yellow arrows indicate dermal fibroblasts, and asterisks indicate perivascular inflammatory cells. Original magnification × 63. I, Western blotting of total protein extracts from the skin of healthy subjects and SSc patients. With the anti–JAM-C antibodies, a protein band with the expected molecular weight of ∼40 kd was detected. Blots were stripped and reprobed with anti–α-tubulin antibodies as a loading control for normalization. Representative immunoblots are shown. rh JAM-C = recombinant human JAM-C. J, Intensity of the bands, quantified by densitometry. Values are the mean ± SD. ∗ = P < 0.01 versus control, by Student's t-test. OD = optical density; AU = arbitrary units.

Differential expression and subcellular localization of JAM-A and JAM-C in cultured dermal MVECs.

The expression and subcellular localization of JAMs was investigated by confocal microscopy in confluent monolayers of dermal MVECs. Cells were double immunolabeled for JAM-A and integrin αVβ3, which have been reported to form a complex on the EC surface (28), or JAM-C and the tight junction protein ZO-1 (29).

We investigated the expression and localization of JAM-A in healthy MVECs and whether they were influenced by treatment with sera from healthy individuals or patients with early-stage SSc. Consistent with previous reports (28), in resting healthy MVECs, JAM-A was mainly localized on the cell surface at intercellular junctions, where it colocalized with integrin αVβ3 (Figure 4A). A similar pattern of JAM-A/αVβ3 expression and localization was observed in healthy MVECs after treatment with healthy sera (Figure 4B). However, after treatment with sera from patients with early-stage SSc, healthy MVECs showed a strong increase in the expression of JAM-A and αVβ3 that appeared to localize not only at the junctions, but also more diffusely along the cell membrane (Figure 4C). Moreover, most JAM-A and αVβ3 were not colocalized at nonjunctional domains of the cell membrane. No significant differences were observed between 1-hour and 6-hour stimulations (data not shown). Treatment of healthy MVECs with sera from patients with late-stage SSc had no effect on JAM-A expression/localization (data not shown). In MVECs isolated from the dermis of patients with early-stage SSc, we observed a strong constitutive expression of JAM-A and αVβ3 at both junctional and nonjunctional sites of the cell membrane, with only partial colocalization (Figure 4D). Western blotting confirmed that JAM-A protein was increased in healthy MVECs upon stimulation with early-stage SSc sera and in early-stage SSc MVECs (Figure 4E).

Figure 4.

Expression and subcellular localization of junctional adhesion molecule A (JAM-A) in cultured dermal microvascular endothelial cells (MVECs). A–D, Representative photomicrographs of double immunofluorescence staining for JAM-A (green) and integrin αVβ3 (red) in basal healthy MVECs (H-MVECs) (A), healthy MVECs treated with healthy control sera (B), healthy MVECs treated with sera from patients with early-stage systemic sclerosis (SSc) (C), and MVECs isolated from the dermis of early-stage SSc patients (SSc-MVECs) (D). Right panels show merged JAM-A and integrin αVβ3 images. Arrowheads indicate localization of JAM-A at junctional sites of the cell membrane. Original magnification × 40. E, Western blotting of protein lysates from healthy MVECs and SSc MVECs, analyzed using anti–JAM-A antibodies. Blots were stripped and reprobed with anti–α-tubulin antibodies as a loading control for normalization. Representative immunoblots are shown.

Consistent with previous findings (29), in resting healthy MVECs, JAM-C was mainly intracellular and absent from interendothelial junctions (Figure 5A). Treatment with healthy sera did not influence JAM-C expression and subcellular localization in healthy MVECs (Figure 5B). Challenge with early-stage SSc sera induced an up-regulation and a rapid accumulation of JAM-C at interendothelial junctions, where it colocalized with ZO-1 (Figures 5C and D). Intense JAM-C immunofluorescence staining was also observed at nonjunctional sites of the cell surface and within cytoplasmic vesicles (Figures 5C and D), indicating that the major mechanism of JAM-C redistribution to the cell membrane involved exocytosis from intracellular compartments. Moreover, we observed that although the amount of junctional JAM-C was elevated 1 hour after stimulation with early-stage SSc sera, 6 hours after stimulation junctional JAM-C staining mostly disappeared (data not shown). Thus, JAM-C redistribution to interendothelial contacts was a rapid and transient process.

Figure 5.

Expression and subcellular localization of junctional adhesion molecule C (JAM-C) in cultured dermal microvascular endothelial cells (MVECs). A–E, Representative photomicrographs of double immunofluorescence staining for JAM-C (green) and the zonula occludens 1 (ZO-1) tight junction protein (red) in basal healthy MVECs (H-MVECs) (A), healthy MVECs treated with healthy control sera (B), healthy MVECs treated with sera from patients with early-stage systemic sclerosis (SSc) (C and D), and MVECs isolated from the dermis of early-stage SSc patients (SSc-MVECs) (E). Right panels show merged JAM-C and ZO-1 images. Arrowheads indicate localization of JAM-C at junctional sites of the cell membrane. Asterisks indicate localization of JAM-C within cytoplasmic vesicles. Original magnification × 40 in A–C and E; × 63 in D. F, Western blotting of protein lysates from healthy MVECs and SSc MVECs, analyzed using anti–JAM-C antibodies. Blots were stripped and reprobed with anti–α-tubulin antibodies as a loading control for normalization. Representative immunoblots are shown.

Treatment of healthy MVECs with sera from patients with late-stage SSc had no effect on JAM-C expression/localization (data not shown). MVECs isolated from the dermis of patients with early-stage SSc exhibited a strong constitutive expression of JAM-C at both junctional and nonjunctional sites of the cell membrane and within cytoplasmic vesicles (Figure 5E). Levels of JAM-C were increased in protein lysates from healthy MVECs treated with sera from patients with early-stage SSc and from MVECs from patients with early-stage SSc (Figure 5F).

Correlation of increased levels of sJAM-A and sJAM-C in SSc with early disease stage and measures of vascular damage.

Circulating levels of sJAM-A in the group of all SSc patients (median 0.54 ng/ml [IQR 0–1.96]) were similar to those in healthy controls (0.52 ng/ml [IQR 0–1.44]) (Figure 6A). Serum sJAM-A levels in both patients with lcSSc (median 0.42 ng/ml [IQR 0–2.15]) and patients with dcSSc (median 0.64 ng/ml [IQR 0–1.64]) were not significantly different from those in controls. In patients with early-stage SSc, circulating sJAM-A levels were significantly increased (median 1.12 ng/ml [IQR 0.64–5.27]) compared with controls (P = 0.007) and compared with patients with late-stage SSc (0 ng/ml [IQR 0–1.02]; P < 0.0001). Circulating sJAM-A levels in patients with late-stage SSc did not differ from those in controls (Figure 6A). The sJAM-A levels were significantly higher in SSc patients with an early or active capillaroscopy pattern (median 0.83 ng/ml [IQR 0–3.14]) than in those with a late capillaroscopy pattern (0 ng/ml [IQR 0–0.52]; P = 0.005) (Figure 6B). Serum sJAM-A levels were increased in SSc patients with digital ulcers (median 1.03 ng/ml [IQR 0–4.99]) compared with patients without digital ulcers (0.38 ng/ml [IQR 0–1.08]; P = 0.047) (Figure 6C).

Figure 6.

Serum levels of soluble junctional adhesion molecule A (sJAM-A) and sJAM-C, determined by quantitative colorimetric sandwich enzyme-linked immunosorbent assay. A, Serum sJAM-A levels in healthy controls, all patients with systemic sclerosis (SSc), patients with early-stage SSc, and patients with late-stage SSc. B and C, Levels of sJAM-A in patients with SSc according to nailfold capillaroscopy pattern (early/active or late) (B) and the presence or absence of active digital ulcers (DUs) (C). D, Serum sJAM-C levels in healthy controls, all patients with SSc, patients with early-stage SSc, and patients with late-stage SSc. E and F, Levels of sJAM-C in patients with SSc according to nailfold capillaroscopy pattern (early/active or late) (E), and the presence or absence of active digital ulcers (F). Data are shown as box plots. Each box represents the 25th to 75th percentiles. Lines inside the boxes represent the median. Lines outside the boxes represent the 10th and the 90th percentiles. Circles indicate outliers, and asterisks indicate the extreme values. P values were calculated by nonparametric Mann-Whitney U test.

Serum levels of sJAM-C in the group of all SSc patients (median 0.28 ng/ml [IQR 0–0.70]) and the patients with lcSSc (0.22 ng/ml [IQR 0–0.75]) were not significantly different from those in controls (0.17 ng/ml [IQR 0–0.56]). Levels of sJAM-C were slightly increased in the patients with dcSSc (0.58 ng/ml [IQR 0–0.71]) compared with controls (P = 0.034). Circulating sJAM-C levels were significantly higher in patients with early-stage SSc (0.66 ng/ml [IQR 0.24–1.15]) than in controls (P = 0.002) and patients with late-stage SSc (0.17 ng/ml [IQR 0–0.52]; P = 0.001), while they did not differ between patients with late-stage SSc and controls (Figure 6D). Serum sJAM-C levels were significantly increased in patients with an early or active capillaroscopy pattern (median 0.58 ng/ml [IQR 0–0.90]) compared with those with a late capillaroscopy pattern (0 ng/ml [IQR 0–0.23]; P = 0.001) (Figure 6E). Levels of sJAM-C were higher in SSc patients with digital ulcers (0.58 ng/ml [IQR 0.25–0.84]) than in SSc patients without digital ulcers (0 ng/ml [IQR 0–0.59]; P = 0.006) (Figure 6F).

Levels of both sJAM-A and sJAM-C were increased in culture supernatants of MVECs from patients with early-stage SSc compared with MVECs from healthy controls (for sJAM-A, mean ± SD 1.89 ± 0.34 ng/ml versus 0.63 ± 0.20 ng/ml, P = 0.005 and for sJAM-C, mean ± SD 1.07 ± 0.21 ng/ml versus 0.35 ± 0.13 ng/ml; P = 0.008).

DISCUSSION

Our present findings clearly demonstrate, for the first time, that JAM-A and JAM-C are differentially expressed in the skin of patients with early-stage SSc and those with late-stage SSc and suggest that during the course of SSc JAMs may participate in EC activation and perivascular inflammatory processes at an early stage, and in defective angiogenesis and loss of microvessels at a later stage. Increased expression of JAMs was found in dermal ECs and inflammatory infiltrates in samples from patients with early-stage SSc, while their expression was markedly reduced in ECs in skin samples from patients with late-stage SSc. Cultured MVECs from patients with early-stage SSc constitutively displayed abnormal expression and subcellular localization of JAM-A and JAM-C, and a similar pattern of JAM expression and distribution was observed in healthy MVECs after treatment with early-stage SSc sera. Moreover, circulating levels of sJAM-A and sJAM-C were selectively increased in patients with early-stage SSc and correlated with measures of microvascular damage.

Cell adhesion molecules play a critical role in the development of inflammatory and vascular diseases and have been implicated in SSc pathogenesis (8–13, 15–17). Leukocyte extravasation is mediated by their interactions with ECs in a stepwise process, which is initiated by leukocyte rolling and completed with transmigration across the endothelial barrier, which can occur via either para- or transcellular routes (30, 31). Leukocyte transendothelial cell migration involves a number of adhesion molecules, the expression of which is highly concentrated at junctions between adjacent ECs. These molecules include CD31, VCAM-1, ICAM-1, CD99, and members of the JAM family, such as JAM-A and JAM-C (30–32). Transendothelial leukocyte migration requires a temporary opening of the adhesive complex at tight junctions. Since they are strategically located at interendothelial cell tight junctions, JAMs are ideally suited to facilitate or direct the passage of leukocytes across the EC barrier. JAMs expressed on ECs may interact homophilically with JAMs expressed on leukocytes (JAM–JAM interaction) and heterophilically with integrins LFA-1 and Mac-1 expressed on leukocytes (JAM–integrin interaction) (18, 32). Moreover, JAMs may also support leukocyte recruitment and firm adhesion when redistributed to the EC luminal surface under inflammatory conditions (18, 33). Indeed, inhibition or loss of function of JAMs attenuates inflammatory responses in several in vivo experimental models, including models of skin inflammation (18, 19, 33–35). Besides their crucial role in inflammation, JAMs act as proangiogenic molecules (18).

On this basis, we hypothesized that JAMs may play multiple roles in the complex pathogenesis of SSc, which is characterized by EC activation/damage and inflammatory cell infiltration in its early stages, while lack of angiogenesis and ECM accumulation predominate in later stages. Two previous studies reported decreased expression of JAM-A and JAM-C on ECs and keratinocytes in skin samples from SSc patients, but possible differences in JAM expression according to different stages of skin involvement were not investigated (11, 36). This prompted us to evaluate the expression of JAM-A and JAM-C in a large series of skin biopsy samples from patients with early-stage SSc and patients with late-stage SSc.

The results of our immunohistochemical analysis demonstrate that dermal blood and lymphatic ECs in skin samples from patients with early-stage SSc exhibit increased expression of JAM-A and JAM-C compared with skin samples from healthy controls. Moreover, in early-stage SSc, dermal perivascular inflammatory cells and fibroblasts also strongly express both JAMs. JAM-A and JAM-C may therefore contribute to the proadhesive phenotype of dermal ECs, inflammatory/immune cells, and fibroblasts and participate in the complex intercellular cross-talk among these different cell types in early-stage SSc. Indeed, previous studies have shown that SSc ECs, peripheral blood mononuclear cells, and activated fibroblasts are hyperadhesive (8–13, 37, 38). The increased expression of JAMs on blood vascular ECs from patients with early-stage SSc may mediate the influx of inflammatory/immune cells into the dermis, where they may stimulate fibroblasts by both direct contact and the release of profibrotic mediators (1, 8). Furthermore, the strong JAM expression observed on lymphatic endothelium may allow lymphocytes and antigen-presenting cells to transmigrate into lymphatic vessels, contributing to their recirculation and trafficking from peripheral tissues to lymph nodes.

These data are in apparent contrast with the findings of Hou et al, who described a decrease in the endothelial expression of JAM-A in SSc compared with healthy skin biopsy samples (36). However, Hou et al (36) did not distinguish between early-stage and late-stage SSc patients, and differences in disease stage between the SSc skin samples used in the present study and those used by Hou et al might explain such discrepancies. Of note, in our study JAM-A and JAM-C were weakly expressed in ECs in skin samples from patients with late-stage SSc. Conversely, dermal fibroblasts from patients with late-stage SSc still displayed high expression of JAMs, which suggests their potential involvement in persistent fibroblast activation and deserves further investigation. Consistent with our findings, increased JAM-A expression on the surface of cultured SSc dermal fibroblasts was reported by Hou et al (36). As previously suggested, increased JAM-A expression on SSc fibroblasts might serve to retain myeloid cells (36). Moreover, JAMs might also play a role in keratinocyte activation, as suggested by the increased expression of JAMs observed in the epidermis in early-stage SSc. It was shown that SSc epidermis has an activated, wound-healing phenotype, and that epidermal cells activated by injury may induce and regulate local fibroblasts during wound repair (39). In addition, increased expression of adhesion molecules has previously been reported in SSc epidermis (9).

Our in vitro results provide further evidence of the important role of JAMs in early activation and the proadhesive phenotype of SSc ECs. MVECs isolated from early-stage SSc dermis display a strong up-regulation of JAM-A and JAM-C, which appear to be constitutively localized at both junctional and nonjunctional sites of the cell membrane. Consistent with the findings of previous studies, we observed that JAM-A is localized at interendothelial junctions, while JAM-C is mainly intracellular in resting healthy MVECs (28, 29). However, treatment of healthy MVECs with early-stage SSc sera induced a rapid up-regulation and subcellular redistribution of JAMs, mirroring the constitutive expression/localization pattern observed in SSc MVECs. Together with previous findings, our data suggest that JAMs may support leukocyte rolling and adhesion when redistributed to the luminal surface of activated ECs under inflammatory and ischemia-reperfusion injury conditions and may actively promote leukocyte movement through intercellular junctions (18, 28, 29, 33). Localization of JAMs at nonjunctional sites of activated/injured ECs may also facilitate the adhesion of platelets (18), which have been implicated in SSc pathogenesis via the release of bioactive molecules that regulate inflammation, angiogenesis, and ECM synthesis (3, 40).

Besides inflammation, the altered expression and subcellular localization of JAMs in SSc MVECs may also have implications for angiogenesis. JAM-A is crucial for correct EC motility, directional movement, and focal contact formation during angiogenesis (18). Indeed, overexpression of JAM-A on ECs induces EC proliferation and migration on vitronectin (18). JAM-A forms a complex with integrin αVβ3 in the absence of basic fibroblast growth factor (bFGF) and mediates bFGF-induced angiogenesis (28). After EC activation with bFGF, JAM-A redistribution is induced and the JAM-A–αVβ3 complex dissociates, which allows angiogenic signals such as the activation of MAP kinases (28). Accordingly, bFGF-induced angiogenesis is almost completely impaired in JAM-A–deficient mice (41). Our results demonstrate that JAM-A is overexpressed and only partially colocalized with integrin αVβ3 in early-stage SSc MVECs, suggesting that JAM-A may act as a proangiogenic molecule in early disease stages. However, JAM-A is strongly down-regulated on late-stage SSc dermal ECs and therefore might not be able to respond to bFGF and mediate angiogenesis, thus contributing to the loss of microvessels that characterizes SSc in its advanced stages.

JAM-C also acts as a proangiogenic molecule, and JAM-C–blocking antibodies reduce angiogenesis in models of hypoxia-induced retinal neovascularization and cancers (18, 42). The strong expression of JAM-C found in ECs in skin samples from patients with early-stage SSc, and its subsequent down-regulation in late-stage disease, suggest a differential role of JAM-C in the angiogenic process during the course of SSc.

When overexpressed on the activated endothelial layer, adhesion molecules undergo shedding and their soluble forms are detectable in serum and considered to be markers of EC activation and injury. Increasing evidence indicates that levels of soluble endothelial adhesion molecules are increased in SSc patients and that their levels may correlate with disease activity (8, 15). Recently, increased circulating levels of sJAM-A have been reported in hypertension and atherosclerosis (20, 43), while serum sJAM-C levels were found to be increased in rheumatoid arthritis (44).

Our data show that sJAM-A and sJAM-C levels are selectively increased in the serum of patients with early-stage SSc, as well as in culture supernatants from early-stage SSc MVECs. Moreover, circulating levels of sJAM-A and sJAM-C are significantly increased in SSc patients with an early or active capillaroscopy pattern compared with those with a late capillaroscopy pattern, and are associated with the presence of active digital ulcers. Taken together, these results suggest that increased serum levels of sJAM-A and sJAM-C could reflect early EC activation/injury and could be considered vascular disease markers in SSc. It has recently been demonstrated that sJAM-C is chemotactic for ECs and induces EC tube formation in vitro (44). The increased levels of proangiogenic sJAM-C found in SSc patients with an early or active capillaroscopy pattern, characterized by frequent microhemorrhages and immature and unstable giant microvessels formed during an uncontrolled angiogenic response, suggest that sJAM-C might also participate in the active derangement of the microcirculation. Further studies are needed to determine if serum levels of sJAM-A and sJAM-C may correlate with additional clinical manifestations of SSc. A prospective followup investigation of serum sJAM-A and sJAM-C is warranted.

Taken together, the results of our study suggest that JAMs are differentially expressed and may participate in EC activation, inflammatory processes, and dysregulated angiogenesis in different stages of SSc. A timely modulation of JAM expression might offer new targeted therapeutic strategies for SSc.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Manetti had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Manetti, Guiducci, Ibba-Manneschi, Matucci-Cerinic.

Acquisition of data. Manetti, Guiducci, Romano, Rosa, Ceccarelli, Mello, Conforti.

Analysis and interpretation of data. Manetti, Milia, Ibba-Manneschi, Matucci-Cerinic.

Ancillary