Prior studies identified T cells, B cells, and macrophages in the inflammatory infiltrate and up-regulation of their protein products in discoid lupus erythematosus (DLE) skin; however, they lacked rigorous analyses to define their specific locations in skin. Thus, we compared expressions of selected T cell, B cell, and macrophage markers in five areas of DLE, psoriasis, and normal skin. Immunostainings for CD3, CD4, CD8, CD20, CD68, CXCR3, CXCL10, and TIA-1 were performed in biopsies of 23 DLE lesional skin, 11 psoriasis lesional skin, and 5 normal skin. Three independent observers used a graded scale to rate each marker’s presence in the epidermis, dermatoepidermal junction (DEJ), perivascular area, periadnexal area, and deep dermis. DLE lesional skin contained an increased abundance of CD3+, CD8+, and CD68+ cells at the DEJ, and CD20+ and CD68+ cells in the periadnexal area versus psoriasis and normal skin. CXCR3, CXCL10, and TIA-1 were elevated in periadnexal sites of DLE lesional skin versus psoriasis lesional skin. The aggregation of T cells, B cells, macrophages, and their protein products (CXCR3, CXCL10, and TIA-1) in the DEJ and periadnexal area of DLE lesional skin may contribute to the pathology of DLE through a coordinated, sophisticated process.
As the most common manifestation of cutaneous lupus erythematosus [
Histopathologically, inflammatory infiltrates consisting mainly of T cells, B cells, and macrophages [
However, the literature has not been as consistent on the expression of B cells in DLE lesional skin [
While describing where these cells are located in DLE lesional skin, these previous studies lacked site-specific comparisons with control skin that would determine which cells predominate at individual areas in DLE lesional skin. Identifying site predilection of these cells will provide greater insight into understanding the pathophysiology of DLE. Thus, we compared expression of T cell, B cell, and macrophage markers and their associated proteins, TIA-1, CXCR3, and CXCL10 in DLE lesional skin, psoriasis lesional skin, and normal skin. Psoriasis was chosen as a disease control because T cells and macrophages figure prominently in the pathology [
Biopsies of DLE lesional skin lesions were performed in the outpatient clinics of Parkland Memorial Hospital and University of Texas Southwestern Medical Center at Dallas, TX, from May 2004 to May 2011. The diagnosis of DLE was made by clinicopathological correlation. DLE patients were also evaluated for systemic lupus erythematosus (SLE) by determining the numbers of American College of Rheumatology SLE criteria satisfied [
Formalin-fixed, paraffin-embedded tissue sections from each DLE, psoriasis, and normal skin biopsy specimen were cut into 4-micron thick sections and mounted on glass slides. Initial sections were stained with hematoxylin and eosin. Primary anti-human antibodies against CD3 (Cell Marque Corporation, Hot Springs, AR), CD4 (Biocare Medical, Concord, CA), CD8 (Cell Marque), CD20 (Ventana Medical Systems, Tucson, AZ), CD68 (Ventana), CXCL10 (R&D Systems, Minneapolis, MN), CXCR3 (BD Pharmingen, San Jose, CA), and TIA-1 (Biocare) were applied. Immunostaining using avidin-biotin-peroxidase complex and 3,3′-diaminobenzidine as substrate on an automated Ventana Benchmark instrument (Ventana) was performed using previously published protocols [
Under basic light microscopy, three reviewers (AG, JS, BFC) examined the staining of each marker in DLE lesional skin, psoriasis lesional skin, and normal skin in five areas: epidermis, DEJ, perivascular area, periadnexal area, and deep dermis. Each area received a rating of 0 (none), 1 (mild), 2 (moderate), or 3 (high), which was based on a subjective assessment of the percentage of positively stained cells and the staining intensity.
Since this was a pilot study, sample size was not calculated. Mean rating scores were compared amongst the three skin groups using the nonparametric Kruskal-Wallis Test and Dunn’s Multiple Comparison Test using statistical software (GraphPad Prism 6 version 6.0c). Similar preplanned analyses were done between DLE subgroups (e.g., those with and without SLE). To reduce the number of false positives, given the numerous comparisons performed,
Twenty-three DLE patients were included in the study. The average age was 46.5 years. 91.3% and 60.9% of DLE patients were female and African Americans, respectively. DLE patients with SLE (DLE+/SLE+) were more likely to be on medications, such as hydroxychloroquine, mycophenolate mofetil, prednisone, and/or quinacrine, than DLE patients without SLE (DLE+/SLE−) (Table
DLE patient characteristics.
Patient no. | SLE (Y/N) | Age | Gender | Ethnicity | No. of SLE Criteria | Treatment at biopsy | Biopsy site |
---|---|---|---|---|---|---|---|
1 | N | 55 | F | Caucasian | 3 | — | Shoulder |
2 | N | 52 | F | Caucasian | 3 | HCQ | Scalp |
3 | N | 46 | F | Caucasian | 3 | — | Scalp |
4 | N | 41 | F | AA | 1 | — | Scalp |
5 | N | 50 | F | AA | 2 | — | Chin |
6 | N | 42 | F | Caucasian | 1 | — | Cheek |
7 | N | 47 | M | AA | 2 | — | Scalp |
8 | N | 31 | F | AA | 1 | — | Scalp |
9 | N | 36 | F | AA | 3 | — | Scalp |
10 | N | 62 | F | Caucasian | 3 | — | Cheek |
11 | N | 45 | F | Caucasian | 2 | — | Nasal Ala |
12 | N | 41 | F | AA | 3 | — | Scalp |
13 | N | 50 | M | AA | 2 | — | Scalp |
14 | Y | 51 | F | Asian | 5 | HCQ, MM, prednisone | Scalp |
15 | Y | 42 | F | Hispanic | 6 | Prednisone | Arm |
16 | Y | 31 | F | AA | 9 | MM, prednisone, quinacrine | Upper thigh |
17 | Y | 39 | F | AA | 5 | HCQ, prednisone | Scalp |
18 | Y | 73 | F | Caucasian | 7 | Prednisone | Cheek |
19 | Y | 54 | F | AA | 5 | HCQ | Shoulder |
20 | Y | 54 | F | AA | 4 | — | Scalp |
21 | Y | 36 | F | AA | 5 | Prednisone | Scalp |
22 | Y | 42 | F | AA | 6 | — | Scalp |
23 | Y | 50 | F | AA | 5 | — | Shoulder |
Abbreviations: AA: African American, DLE: discoid lupus erythematosus, HCQ: hydroxychloroquine, MM: mycophenolate mofetil, SLE: systemic lupus erythematosus.
In the epidermis, psoriasis lesional skin (
Comparisons of staining of T cell, B cell, and macrophage markers (CD3, CD4, CD8, CD20, CD68, CXCR3, CXCL10, and TIA-1) in the epidermis and dermoepidermal junction (DEJ) of DLE lesional skin (
Comparisons of staining of T cell, B cell, and macrophage markers (CD3, CD4, CD8, CD20, CD68, CXCR3, CXCL10, and TIA-1) in the perivascular area of DLE lesional skin (
Comparisons of staining of T cell, B cell, and macrophage markers (CD3, CD4, CD8, CD20, CD68, CXCR3, CXCL10, and TIA-1) in the periadnexal area of DLE lesional skin (
Comparisons of staining of T cell, B cell, and macrophage markers (CD3, CD4, CD8, CD20, CD68, CXCR3, CXCL10, and TIA-1) in the deep dermis of DLE lesional skin (
Serial sections revealed that CXCR3 and CXCL10 appeared to have similar staining patterns as the CD3+ T cell and CD68+ macrophage populations, but not the CD20+ B cell populations (Figure
Serial sections of DLE lesional skin showed that CXCL10 (a) and CXCR3 (b) appeared to mirror the staining patterns of CD3+ T cells (c) and CD68+ (d) macrophages, but not CD20+ (e) B cells. Serial sections of DLE lesional skin were stained for CD3, CD20, CD68, CXCL10, and CXCR3. Magnification: 200x.
We analyzed the T cell, B cell, and macrophage populations and selected associated proteins (CXCR3, CXCL10, and TIA-1) in five different areas of DLE lesional skin. To determine where these markers were differentially expressed in DLE lesional skin; site-specific comparisons were made with psoriasis lesional skin and normal skin. We stratified our results into major or minor categories based on whether DLE was significantly discrepant from psoriasis and normal skin (major) or from only one of the two control groups (minor).
A major finding of T cells in DLE lesional skin was that CD3+ T cells and CD8+ cytotoxic T cells were significantly higher in DLE lesional skin versus both psoriasis lesional skin and normal skin at the DEJ. Minor findings included increased numbers of CD3+ T cells and CD8+ T cells in the periadnexal area and in the deep dermis in DLE versus psoriasis lesional skin and CD4+ helper T cells at the DEJ and in periadnexal areas in DLE lesional skin versus normal and psoriasis lesional skin, respectively. Although Tebbe et al. reported that most T cells were of the helper subtype [
B cells have largely been implicated in SLE due to their production of autoantibodies. In SLE, autoantibodies can form immune complexes that mediate tissue damage through antibody-dependent cell-mediated cytotoxicity (ADCC) [
Similar to CD8+ T cells, CD68+ macrophages were also significantly increased at the DEJ in DLE lesional skin versus both psoriasis lesional skin and normal skin. Additionally, significantly higher levels of CD68+ macrophages were found in the periadnexal area of DLE lesional skin versus both control groups. Of all the cell types, the CD68+ macrophage was the only cell type that was consistently significantly different when compared with normal skin in 4 of 5 locations studied (DEJ, perivascular area, periadnexal area, and deep dermis). Despite the enhanced presence of macrophages in DLE lesional skin, little is known regarding how macrophages contribute to the pathology of DLE. Previous studies have found that after CD3+ T cells, CD68+ cells are the next most common infiltrate in DLE [
CXCL10 and CXCR3, which were both up-regulated in the periadnexal areas of DLE versus psoriasis lesional skin, appeared to mimic the expression patterns of effector T cells and macrophages. As CXCR3 is predominantly found on
We also investigated if there were differences in the inflammatory infiltrates in DLE+/SLE− and DLE+/SLE+ skin because it is unclear whether DLE in the setting of systemic disease is different than DLE alone. While no significant differences in any marker between DLE+/SLE− and DLE+/SLE+ skin were found, there were trends towards stronger immunostaining of CD3, CD4, CD8, CD20, CD68, CXCR3, CXCL10, and TIA-1 in most areas of DLE+/SLE− skin compared with DLE+/SLE+ skin. We hypothesize that a dilutional effect may be present in DLE patients with systemic disease in which inflammatory cells and their protein products are dispersed in other areas of the body. However, medication effect may also explain this difference, as more DLE+/SLE+ patients were on potent immunosuppressants than DLE+/SLE− patients.
The study is limited by the reviewers not being blinded during the immunohistochemical analyses. While reviewers were instructed to focus their attention on the staining intensities of the various markers and not on the underlying pathology, distinct pathological findings in DLE, psoriasis, and normal skin could be viewed in the background. Additionally, clinical data was not available for the control groups. While no significant differences in the markers studied were seen between DLE+/SLE+ and DLE+/SLE− skin, we are pursuing further investigations in these DLE skin subtypes that broaden the number of genes and proteins studied via gene microarrays and confirmatory protein expression studies.
In conclusion, we performed site-specific analyses of the inflammatory infiltrate and their selected proteins in DLE lesional skin compared with psoriasis lesional skin and normal skin. We found that CD3+ T cells, CD8+ T cells, and CD68+ macrophages are significantly higher in the DEJ of DLE lesional skin versus normal and psoriasis lesional skin. CD20+ B cells and CD68+ macrophages were also elevated in the periadnexal areas of DLE lesional skin compared with normal and psoriasis lesional skin. These periadnexal areas also contained higher expression of TIA-1, CXCR3, and CXCL10 in DLE versus psoriasis lesional skin. Aggregation of these cells in these specific areas of DLE lesional skin likely contributes to the pathology of DLE through an intricate cross-talk and coordination amongst these cells.
Antibody-dependent cell-mediated cytotoxicity
Dermoepidermal junction
Discoid lupus erythematosus
Systemic lupus erythematosus
T-cell restricted intracellular antigen 1.
Benjamin F. Chong is an investigator for Amgen Incorporated and Daavlin Corporation.
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors are indebted to Mary Debram-Hart and the histopathology staff at Cockerell Dermatopathology for their technical assistance and expertise in carrying out the immunohistochemistry protocols. The research in this publication was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health under Award no. K23AR061441 (BFC). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.