Report
Assessment of tissue FoxP3+, CD4+ and CD8+ T-cells in active and stable nonsegmental vitiligo Marwa Abdallah1, MD, Ranya Lotfi1, MD, Wessam Othman2, MD, and Riham Galal3, MB BCH
1 Department of Dermatology and Venereology, Ain Shams University, Cairo, Egypt, 2Pathology Department, Ain Shams University, Cairo, Egypt, and 3Department of Dermatology and Venereology, Air Forces Hospital, Cairo, Egypt
Abstract Background The exact etiology of vitiligo remains obscure. Studies have indicated a role for cellular immunity in the pathogenesis of vitiligo. The aim of this study is to assess tissue FoxP3+ natural regulatory T-cells (Tregs), as well as CD4+ and CD8+ T-cells in active vs. stable nonsegmental vitiligo. Materials and methods Immunohistochemical double-staining for expression of CD4+ and
Correspondence Ranya Lotfi, MD 34 El Orouba Street, Heliopolis, Cairo Egypt E-mail: [email protected]
CD8+ T-cells with immunostaining for expression of FoxP3 in lesional, marginal, and nonlesional skin of nonsegmental vitiligo was used to evaluate the abundance of Tregs among CD4+ and CD8+ T-cells in active and stable cases of vitiligo. Results A significant increase in the number of CD4+ and CD8+ T-cells and a highly significant reduction in the number of FoxP3-expressing Tregs were detected in marginal skin in both stable and active vitiligo cases. FoxP3+ cells were decreased in tissue of
doi: 10.1111/ijd.12160
patients with vitiligo compared with healthy controls. The number of CD8+ T-cells was increased in the epidermis and dermoepidermal junction (DEJ) in comparison with the number of CD4+ T-cells. Tregs were mostly present at the DEJ. Conclusion The reduction in the number of FoxP3+ cells in the marginal skin suggests that this is the site where regulatory activity is needed to suppress the activity of helper and cytotoxic T-cells that are actively contributing to depigmentation.
Introduction Vitiligo is an idiopathic, acquired disorder of pigmentation. The exact etiology of vitiligo has not yet been established; however, the immune responses have the key role in the pathogenesis of vitiligo. Many studies have indicated a role for both cellular and humoral immunity in the pathogenesis of vitiligo.1 Melanocyte destruction in vitiligo is primarily a cell-mediated process controlled by antigen-specific T-lymphocytes. T-cells infiltrating vitiligo skin are reactive, at least in part, with the same antigens recognized by T-cells infiltrating melanoma tumors.2 Associations with MHC class I molecules specifically suggest a role of cytotoxic T-cells in disease pathogenesis.3–5 In both inflammatory and generalized vitiligo, the ratio of CD4+ to CD8+ T-cells decreases compared with unaffected control skin, supporting the involvement of a cytotoxic response to melanocytes. Indeed a prevalence of CD8+ T-cells is suggestive of a type 1 cytokine environment.6 Also supporting a cytotoxic T-cell-mediated response to melanocytes is the fact that CD8+ T-cells are frequently ª 2014 The International Society of Dermatology
found juxtaposed to the remaining melanocytes in the epidermis.7 The relevance of cellular infiltrates to progressive depigmentation was further supported by the observation that successful repigmentation of lesional skin by mini-grafting was associated with reduced expression of markers of cell-mediated immunity, particularly CD8+ expression, in graft skin specimens.8 In vitro cultures of skin-infiltrating T-cells isolated from perilesional biopsies of patients with active disease9 displayed a tendency towards type 1 cytokine profiles, and pooled clones derived from these cultures displayed cytotoxicity towards autologous melanocytes. Later studies showed that perilesional vitiligo T-cells were shown to cause apoptosis of melanocytes in nonlesional skin explants, providing direct evidence for a major etiologic role of cytotoxic T-cells in progressive vitiligo.10 As in healthy skin, part of the CD4+ population consists of natural regulatory T-cells (Tregs) in place to keep responses to self-antigens in check.11 Tregs are critical to the maintenance of immune cell homeostasis by enforcing a dominant-negative regulation of other immune cells.12 Tregs have the ability to inhibit the International Journal of Dermatology 2014
1
2
Report
FoxP3+, CD4+ and CD8+ T-cells in vitiligo
responses of CD8+ T-cells, natural killer cells, and CD4+ T-cells, so that Tregs may be important in the prevention of autoimmune diseases and in regulating inflammation and tumorigenesis.13 Markers expressed by Tregs include Forkhead box P3 (FoxP3), GITR, CTLA-4, and CD25, yet only FoxP3 expression is relatively unique to regulatory T-cells. Most of the Tregs expressing FoxP3 were also expressing CD4 and CD25 antigens.14 Tregs were found to be increased in the serum of patients with active vitiligo.15 The aim of the current study is to evaluate the number of FoxP3+ Tregs in lesional, marginal, and nonlesional skin of active vs stable vitiligo patients. Additional CD4 and CD8 staining was done in an attempt to classify FoxP3+ cells to either group. Patients and methods Patients This prospective, controlled study included 20 patients with nonsegmental vitiligo; 10 of them were suffering from active vitiligo, and the other 10 were stable cases of vitiligo with various degrees of disease extent. Ten apparently healthy age- and sex-matched controls were enrolled. The activity of the disease was detected by asking the patient about the appearance of new vitiligo lesion(s), increase in size of existing lesions and/or reappearance of any healed lesion within the last six months. Clinical assessment of vitiligo lesions was performed to determine the distribution, clinical variant, and extent. Excluded from the study were patients taking any treatment in the last four weeks prior to the study and patients with concomitant autoimmune diseases. The study was conducted in Ain Shams University Hospitals and Air Forces Hospital, Egypt, during the period from April 2010 until June 2010. Informed consent was taken from all subjects before participating in this work. The study was approved by the Ethical Committee of Ain Shams University, Cairo, Egypt. Methods Three-millimeter punch biopsies were obtained under local anesthesia from lesional, marginal, and nonlesional skin of each patient with vitiligo included in this study. Lesional (L) and marginal (M) biopsies were taken from the most recent lesion. The marginal skin biopsy was taken spanning the border between depigmented and pigmented skin. The nonlesional (NL) biopsy was taken at least 20 cm away from the edge of any existing vitiligo lesion. Sections from each skin biopsy were immunohistochemically double-stained for the expression of CD4+ and CD8+ T-cells, where CD4+ cells appeared red and CD8+ cells appeared blue. Other sections were stained for the expression of FoxP3 antigen; positively stained cells appeared brown. International Journal of Dermatology 2014
Abdallah et al.
A primary antibody cocktail (MultiVision, Fairfax, VA, USA) of CD4+–CD8+ antibody (MV-2001-R7; Labvision, Cheshire, UK) was used in conjunction with ready-to-use MultiVision polymer cocktail, composed of anti-mouse/horseradish peroxidase + antirabbit/alkaline phosphatase (Thermo Fisher Scientific-Labvision, Cheshire, UK). For the immunohistochemical detection of FoxP3 antibody, rabbit polyclonal antibody was used according to the manufacturer’s recommendations (Genway bio, San Diego, CA, USA).
Digital counting Cells were counted digitally by using Olympus soft pro software (Tokyo, Japan). Three sections from each biopsy (L, M, and NL skin) were prepared. The numbers of immunostained cells were counted in at least three randomly selected fields of each section at high-power objective magnification of 2009 in the CD4+ CD8+ double-stained slides and magnification of 4009 in the FoxP3+ slides (results were given in cells/field). The average number (mean) of positively stained cells in the nine counted fields was taken for statistical analysis. Statistical analysis Data were analyzed using the SPSS program version 15. Qualitative data were presented as frequency and percentage. Quantitative data were presented as the range and median (of the mean number of cells) as the data had extreme values. As data were not following a normal distribution, nonparametric analysis was used: Wilcoxon signed-rank tests for paired data and Mann–Whitney test for independent groups. The Spearman correlation was used to measure the correlation between the quantitative variables. A P-value of 0.05 was chosen as the level of statistical significance, and in multiple comparisons Bonferroni correction was used with a P-value of 0.016.
Results Cases and controls were age and gender matched, with a mean age for cases of 30 years (SD 14) and for controls 28 years (9; P = 0.7). Seven (35%) of the cases were males and 13 (65%) were females, while three (30%) of the controls were males and seven (70%) were females (P = 0.78). Disease duration in cases ranged from 2 to 25 years, with a median of 10 years. When all vitiligo cases were taken together, the median level of CD4+ T-cells was highest in lesional skin (L = 76 [26–101]; M = 62.5 [44–82]; NL = 27.5 [11–50]; P < 0.05); CD8+ T-cells were highest in marginal vitiligo skin (L = 29 [8–61]; M = 39 [20–54]; NL = 17.5 [4–34]; P < 0.05), making the CD4/CD8 ratio highest in lesional skin (L = 2 [1–8]; M = 2 [1–3]; NL = 1.5 [1–4]; P < 0.05). FoxP3+ cells were highest in lesional, while they were very low in marginal and nonlesional (P < 0.001) skin, with no significant difference between ª 2014 The International Society of Dermatology
FoxP3+, CD4+ and CD8+ T-cells in vitiligo
Abdallah et al.
the latter two groups (L = 18 [9–24]; M = 3 [1–7]; NL = 3 [1–8]). Similar findings were detected in active and stable vitiligo cases for CD4+ cells. CD8+ cells were higher in stable marginal vitiligo skin, while they were not different from lesional skin of active lesions. FoxP3+ cells were lowest in marginal active vitiligo, while they were lowest in nonlesional skin of stable cases compared with the rest of sites (Table 1). On comparing active with stable vitiligo, CD4+ and CD8+ T-cells were significantly higher in active cases (Fig. 1). As for FoxP3+ cells, their numbers were higher in lesional and nonlesional sections of active cases compared with stable ones, while their number was low in marginal skin of both active and stable cases without a significant difference between both groups (Fig. 2; Table 2). Furthermore, on comparing the number of FoxP3+ cells in active and stable nonlesional vitiligo skin with skin of normal healthy controls, FoxP3+ cells were significantly less in vitiligo skin than in normal healthy control skin (NL active = 5.5 [3–8]; NL stable = 2 [1–4]; healthy controls = 48 [36–54]; P < 0.001). Generally, T-cells were present perivascularly in the papillary dermis as well as close to the epidermis. T-cells were concentrated at the dermoepidermal junction (DEJ) and in the lower epidermis; sites where melanocytes are usually present. At the DEJ, CD8+ cells showed no statistically significant difference between active and stable marginal and lesional vitiligo skin (active M = 13.5 [5–18] vs stable M = 11 [0–18]; P > 0.05; active L = 5.5 [3–21] vs stable L = 6 [0–11]; P > 0.05). Notably, CD8+ cells were
Report
highest at the DEJ of marginal skin compared with lesional vitiligo skin of both active and stable cases (P < 0.05; Fig. 1). At the DEJ, FoxP3+ cells were significantly lower in marginal vs lesional vitiligo of active and stable cases (active M = 4 [1–6] vs active L = 18 [16–20]; P < 0.001; stable M = 3 [1–5] vs stable L = 14 [8–18]; P < 0.001). Active cases showed a higher number of FoxP3+ cells in marginal (P = 0.003) and lesional (P < 0.001) skin compared with stable cases (Fig. 2). Discussion Research from several groups support the involvement of autoreactive, melanocyte-specific, cytotoxic T-cells in progressive depigmentation of human skin.16–18 The fact that melanocyte-reactive CD8+ T-cells are also found in control individuals without vitiligo suggests that autoimmune reactivity is kept in check and that this checkpoint is defective in vitiligo.19 Our current study further emphasizes the immunological role in the destructive process of melanocytes in vitiligo. CD4+ and CD8+ T-cells were generally increased in active compared with stable vitiligo. CD4+ T-cells were significantly increased in lesional skin compared with marginal and nonlesional skin in both active and stable vitiligo. CD8+ T-cells were significantly increased in marginal skin compared with lesional and nonlesional skin of both active and stable disease, with a resultant significant decrease of the CD4+/CD8+ ratio in active cases in comparison to stable cases. CD8+ T-cells were more abundant at the DEJ and in the epidermis (sites where melanocytes
Table 1 Comparison between number of cells in lesional, marginal and nonlesional skin in active (left three columns) and stable (right three columns) vitiligo cases Active cases
CD4 (median, range) cells/field P-value versus lesional P-value versus marginal CD8 (median, range) cells/field P-value versus lesional P-value versus marginal CD4/CD8 (median, range) cells/field P-value versus lesional P-value versus marginal FoxP3 (median, range) cells/field P-value versus lesional P-value versus marginal
Stable cases
Lesional
Marginal
Nonlesional
Lesional
Marginal
Nonlesional
89.5 (78–101) – – 49.5 (38–61) –
70.5 (64–82) 0.005 – 48 (36–54) 0.307
38 (30–50) 0.005 0.005 25.5 (18–34) 0.005
59.5 (26–75) – – 16 (8–20) –
51.5 (44–61) 0.114 – 27 (20–37) 0.005
18.5 (11–25) 0.005 0.005 10 (4–17) 0.014
– 1.8 (1.2–2.5) – – 20 (16–24) – –
– 1.6 (1.2–1.8) 0.037 – 2.5 (1–6) 0.005 –
0.005 1.45 (1–2.2) 0.285 0.878 5.5 (3–8) 0.005 0.011
– 3.88 (1.6–7.8) – – 16 (9–22) – –
– 1.7 (1.3–3.8) 0.007 – 3 (2–5) 0.005 –
0.005 1.5 (1–3.8) 0.005 0.959 2 (1–4) 0.005 0.013
The number of cells given is cells/field. ª 2014 The International Society of Dermatology
International Journal of Dermatology 2014
3
4
Report
FoxP3+, CD4+ and CD8+ T-cells in vitiligo
Abdallah et al.
(a)
(b)
(c)
(d)
(e)
(f)
+
+
Figure 1 Skin biopsies in cases of nonsegmental vitiligo with CD4 T-cells (in red) and CD8 T-cells (in blue) in the dermis,
epidermis and DEJ with a magnification of 2009 (immunohistochemical stain). (a) L skin biopsy of an active case of nonsegmental vitiligo, showing numerous CD4+ and CD8+ T-cells in the dermis. The number of CD4+ T (95)-cells is obviously increased in the epidermis and the DEJ, and appears to be more than CD8+ T (50)-cells. (b) L skin biopsy of a stable case of nonsegmental vitiligo, in which the number of CD4+ (26) and CD8+ (20) T-cells is almost similar in the dermis, while the number of CD8+ T-cells is more in the epidermis and the DEJ. (c) PL skin biopsy of an active case of nonsegmental vitiligo. CD8+ T-cells are attacking the melanocytes at the DEJ. Perivascular cellular infiltrate composed of CD4+ (76) and CD8+ (48) T-cells. (d) PL skin biopsy of a stable case of nonsegmental vitiligo, showing CD4+ (44) and CD8+ (20) T-cells in the dermis with a decrease in their numbers in comparison with active cases. (e) NL skin biopsy of an active case of nonsegmental vitiligo, showing CD4+ (30) and CD8+ (18) T-cells in the dermis with decreased numbers in comparison with L and PL figures. Still, CD8+ T-cells are seen close to melanocytes. (f) NL skin biopsy of a stable case of nonsegmental vitiligo, showing CD4+ (23) and CD8+ (12) T-cells in the dermis less than active cases
usually reside) compared with CD4+ cells. These results are in accordance with many previous studies.6,9,17,18,20,21 Le Poole et al.22 found that melanocyte-specific T-cells can induce apoptosis of melanocytes in situ in nonlesional skin of patients with vitiligo, indicating that the effector phase of melanocyte destruction is mediated by cytotoxic T-cells. It has been reported that FoxP3+, CD4+, and CD25+ Tregs are elevated in the serum of active vitiligo patients compared with healthy controls. It was then hypothesized International Journal of Dermatology 2014
that although their number is elevated, their immune inhibitory function may be compromised.15 Recently, Klarquist et al.23 have shown a decrease of regulatory T-cells among CD3+ T-cells in lesional and marginal as well as nonlesional vitiligo skin. However, the study group did not identify differences between active and stable cases or whether Tregs were part of the CD4+ or CD8+ population. The aim of the current prospective, controlled study was to evaluate the Treg cell level among active and ª 2014 The International Society of Dermatology
FoxP3+, CD4+ and CD8+ T-cells in vitiligo
Abdallah et al.
(a)
(b)
(c)
(d)
(e)
(f)
Report
+
Figure 2 FoxP3 cells (brown cells) in active (a, c, e) and stable (b, d, f) vitiligo. Cells are seen perivascularly and at the DEJ
(arrows). FoxP3+ cells are increased in active lesional (a and b), and decreased in marginal (c and d) and nonlesional (e and f) skin. The lowest number of FoxP3+ cells is seen in marginal vitiligo sections (9400, immunohistochemical stain). (a) FoxP3+ cells in lesional skin of active vitiligo (24) showing Tregs mostly at the DEJ, which also appears more in comparison with the stable case. (b) FoxP3+ cells in lesional skin of stable vitiligo (22). (c) FoxP3+ cells in perilesional skin of active vitiligo (5), which appears more at the DEJ in comparison with the stable case, but still has the lowest number of cells in comparison with L and NL skin. (d) FoxP3+ cells in perilesional skin of stable vitiligo (which barely shows Tregs). (e) FoxP3+ cells in nonlesional skin of active vitiligo (8), which appears more in comparison to the stable case. (f) FoxP3+ cells in nonlesional skin of stable vitiligo (4), which shows Tregs mostly at the DEJ
stable cases and their probable relation to one of the two main T-cell subpopulations. The results revealed that the number of FoxP3+ T-cells was significantly reduced in the skin of vitiligo patients in comparison to normal healthy controls. FoxP3+ T-cells were lowest in marginal skin of active cases in comparison with lesional and nonlesional skin of the same group. Similarly, Tregs were lower in marginal skin of stable vitiligo compared with lesional skin of the same group. Tregs regulatory activity is needed at the marginal skin to suppress the activity of helper and cytotoxic T-cells that are actively contributing to depigmentation.23 Therefore, their deficiency plays a role in progression of the disease process. ª 2014 The International Society of Dermatology
Lesional skin showed a higher Treg number in both active and stable vitiligo compared with marginal and nonlesional skin. Because CD4+ T-cells are also increased in lesional skin, we may hypothesize that Tregs are part of the CD4+ population, as has been previously suggested.14 Klarquist et al.23 showed similar results to ours in marginal, lesional, and nonlesional vitiligo (two stable and five active cases), without differentiation between findings in active and stable cases. They further reported an increase of circulating Tregs in patients with vitiligo compared with healthy controls. In order to investigate the cause of such a discrepancy, they perInternational Journal of Dermatology 2014
5
6
Report
FoxP3+, CD4+ and CD8+ T-cells in vitiligo
Abdallah et al.
Table 2 Comparison between active and stable cases of vitiligo regarding CD4+, CD8+ T-cells, FoxP3 and CD4+/ CD8+ ratio
CD4 lesional CD4 marginal CD4 nonlesional CD8 lesional CD8 marginal CD8 nonlesional CD4/CD8 lesional CD4/CD8 marginal CD4/CD8 nonlesional FoxP3 lesional FoxP3 marginal FoxP3 nonlesional
Stable
Active
P-value
59.5 51.5 18.5 16 27 10 3.8 1.7 1.5 16 3 2
89.5 70.5 38 49.5 48 25.5 1.8 1.6 1.4 20 2.5 5.5
Assessment of tissue FoxP3+, CD4+ and CD8+ T-cells in active and stable nonsegmental vitiligo Marwa Abdallah1, MD, Ranya Lotfi1, MD, Wessam Othman2, MD, and Riham Galal3, MB BCH
1 Department of Dermatology and Venereology, Ain Shams University, Cairo, Egypt, 2Pathology Department, Ain Shams University, Cairo, Egypt, and 3Department of Dermatology and Venereology, Air Forces Hospital, Cairo, Egypt
Abstract Background The exact etiology of vitiligo remains obscure. Studies have indicated a role for cellular immunity in the pathogenesis of vitiligo. The aim of this study is to assess tissue FoxP3+ natural regulatory T-cells (Tregs), as well as CD4+ and CD8+ T-cells in active vs. stable nonsegmental vitiligo. Materials and methods Immunohistochemical double-staining for expression of CD4+ and
Correspondence Ranya Lotfi, MD 34 El Orouba Street, Heliopolis, Cairo Egypt E-mail: [email protected]
CD8+ T-cells with immunostaining for expression of FoxP3 in lesional, marginal, and nonlesional skin of nonsegmental vitiligo was used to evaluate the abundance of Tregs among CD4+ and CD8+ T-cells in active and stable cases of vitiligo. Results A significant increase in the number of CD4+ and CD8+ T-cells and a highly significant reduction in the number of FoxP3-expressing Tregs were detected in marginal skin in both stable and active vitiligo cases. FoxP3+ cells were decreased in tissue of
doi: 10.1111/ijd.12160
patients with vitiligo compared with healthy controls. The number of CD8+ T-cells was increased in the epidermis and dermoepidermal junction (DEJ) in comparison with the number of CD4+ T-cells. Tregs were mostly present at the DEJ. Conclusion The reduction in the number of FoxP3+ cells in the marginal skin suggests that this is the site where regulatory activity is needed to suppress the activity of helper and cytotoxic T-cells that are actively contributing to depigmentation.
Introduction Vitiligo is an idiopathic, acquired disorder of pigmentation. The exact etiology of vitiligo has not yet been established; however, the immune responses have the key role in the pathogenesis of vitiligo. Many studies have indicated a role for both cellular and humoral immunity in the pathogenesis of vitiligo.1 Melanocyte destruction in vitiligo is primarily a cell-mediated process controlled by antigen-specific T-lymphocytes. T-cells infiltrating vitiligo skin are reactive, at least in part, with the same antigens recognized by T-cells infiltrating melanoma tumors.2 Associations with MHC class I molecules specifically suggest a role of cytotoxic T-cells in disease pathogenesis.3–5 In both inflammatory and generalized vitiligo, the ratio of CD4+ to CD8+ T-cells decreases compared with unaffected control skin, supporting the involvement of a cytotoxic response to melanocytes. Indeed a prevalence of CD8+ T-cells is suggestive of a type 1 cytokine environment.6 Also supporting a cytotoxic T-cell-mediated response to melanocytes is the fact that CD8+ T-cells are frequently ª 2014 The International Society of Dermatology
found juxtaposed to the remaining melanocytes in the epidermis.7 The relevance of cellular infiltrates to progressive depigmentation was further supported by the observation that successful repigmentation of lesional skin by mini-grafting was associated with reduced expression of markers of cell-mediated immunity, particularly CD8+ expression, in graft skin specimens.8 In vitro cultures of skin-infiltrating T-cells isolated from perilesional biopsies of patients with active disease9 displayed a tendency towards type 1 cytokine profiles, and pooled clones derived from these cultures displayed cytotoxicity towards autologous melanocytes. Later studies showed that perilesional vitiligo T-cells were shown to cause apoptosis of melanocytes in nonlesional skin explants, providing direct evidence for a major etiologic role of cytotoxic T-cells in progressive vitiligo.10 As in healthy skin, part of the CD4+ population consists of natural regulatory T-cells (Tregs) in place to keep responses to self-antigens in check.11 Tregs are critical to the maintenance of immune cell homeostasis by enforcing a dominant-negative regulation of other immune cells.12 Tregs have the ability to inhibit the International Journal of Dermatology 2014
1
2
Report
FoxP3+, CD4+ and CD8+ T-cells in vitiligo
responses of CD8+ T-cells, natural killer cells, and CD4+ T-cells, so that Tregs may be important in the prevention of autoimmune diseases and in regulating inflammation and tumorigenesis.13 Markers expressed by Tregs include Forkhead box P3 (FoxP3), GITR, CTLA-4, and CD25, yet only FoxP3 expression is relatively unique to regulatory T-cells. Most of the Tregs expressing FoxP3 were also expressing CD4 and CD25 antigens.14 Tregs were found to be increased in the serum of patients with active vitiligo.15 The aim of the current study is to evaluate the number of FoxP3+ Tregs in lesional, marginal, and nonlesional skin of active vs stable vitiligo patients. Additional CD4 and CD8 staining was done in an attempt to classify FoxP3+ cells to either group. Patients and methods Patients This prospective, controlled study included 20 patients with nonsegmental vitiligo; 10 of them were suffering from active vitiligo, and the other 10 were stable cases of vitiligo with various degrees of disease extent. Ten apparently healthy age- and sex-matched controls were enrolled. The activity of the disease was detected by asking the patient about the appearance of new vitiligo lesion(s), increase in size of existing lesions and/or reappearance of any healed lesion within the last six months. Clinical assessment of vitiligo lesions was performed to determine the distribution, clinical variant, and extent. Excluded from the study were patients taking any treatment in the last four weeks prior to the study and patients with concomitant autoimmune diseases. The study was conducted in Ain Shams University Hospitals and Air Forces Hospital, Egypt, during the period from April 2010 until June 2010. Informed consent was taken from all subjects before participating in this work. The study was approved by the Ethical Committee of Ain Shams University, Cairo, Egypt. Methods Three-millimeter punch biopsies were obtained under local anesthesia from lesional, marginal, and nonlesional skin of each patient with vitiligo included in this study. Lesional (L) and marginal (M) biopsies were taken from the most recent lesion. The marginal skin biopsy was taken spanning the border between depigmented and pigmented skin. The nonlesional (NL) biopsy was taken at least 20 cm away from the edge of any existing vitiligo lesion. Sections from each skin biopsy were immunohistochemically double-stained for the expression of CD4+ and CD8+ T-cells, where CD4+ cells appeared red and CD8+ cells appeared blue. Other sections were stained for the expression of FoxP3 antigen; positively stained cells appeared brown. International Journal of Dermatology 2014
Abdallah et al.
A primary antibody cocktail (MultiVision, Fairfax, VA, USA) of CD4+–CD8+ antibody (MV-2001-R7; Labvision, Cheshire, UK) was used in conjunction with ready-to-use MultiVision polymer cocktail, composed of anti-mouse/horseradish peroxidase + antirabbit/alkaline phosphatase (Thermo Fisher Scientific-Labvision, Cheshire, UK). For the immunohistochemical detection of FoxP3 antibody, rabbit polyclonal antibody was used according to the manufacturer’s recommendations (Genway bio, San Diego, CA, USA).
Digital counting Cells were counted digitally by using Olympus soft pro software (Tokyo, Japan). Three sections from each biopsy (L, M, and NL skin) were prepared. The numbers of immunostained cells were counted in at least three randomly selected fields of each section at high-power objective magnification of 2009 in the CD4+ CD8+ double-stained slides and magnification of 4009 in the FoxP3+ slides (results were given in cells/field). The average number (mean) of positively stained cells in the nine counted fields was taken for statistical analysis. Statistical analysis Data were analyzed using the SPSS program version 15. Qualitative data were presented as frequency and percentage. Quantitative data were presented as the range and median (of the mean number of cells) as the data had extreme values. As data were not following a normal distribution, nonparametric analysis was used: Wilcoxon signed-rank tests for paired data and Mann–Whitney test for independent groups. The Spearman correlation was used to measure the correlation between the quantitative variables. A P-value of 0.05 was chosen as the level of statistical significance, and in multiple comparisons Bonferroni correction was used with a P-value of 0.016.
Results Cases and controls were age and gender matched, with a mean age for cases of 30 years (SD 14) and for controls 28 years (9; P = 0.7). Seven (35%) of the cases were males and 13 (65%) were females, while three (30%) of the controls were males and seven (70%) were females (P = 0.78). Disease duration in cases ranged from 2 to 25 years, with a median of 10 years. When all vitiligo cases were taken together, the median level of CD4+ T-cells was highest in lesional skin (L = 76 [26–101]; M = 62.5 [44–82]; NL = 27.5 [11–50]; P < 0.05); CD8+ T-cells were highest in marginal vitiligo skin (L = 29 [8–61]; M = 39 [20–54]; NL = 17.5 [4–34]; P < 0.05), making the CD4/CD8 ratio highest in lesional skin (L = 2 [1–8]; M = 2 [1–3]; NL = 1.5 [1–4]; P < 0.05). FoxP3+ cells were highest in lesional, while they were very low in marginal and nonlesional (P < 0.001) skin, with no significant difference between ª 2014 The International Society of Dermatology
FoxP3+, CD4+ and CD8+ T-cells in vitiligo
Abdallah et al.
the latter two groups (L = 18 [9–24]; M = 3 [1–7]; NL = 3 [1–8]). Similar findings were detected in active and stable vitiligo cases for CD4+ cells. CD8+ cells were higher in stable marginal vitiligo skin, while they were not different from lesional skin of active lesions. FoxP3+ cells were lowest in marginal active vitiligo, while they were lowest in nonlesional skin of stable cases compared with the rest of sites (Table 1). On comparing active with stable vitiligo, CD4+ and CD8+ T-cells were significantly higher in active cases (Fig. 1). As for FoxP3+ cells, their numbers were higher in lesional and nonlesional sections of active cases compared with stable ones, while their number was low in marginal skin of both active and stable cases without a significant difference between both groups (Fig. 2; Table 2). Furthermore, on comparing the number of FoxP3+ cells in active and stable nonlesional vitiligo skin with skin of normal healthy controls, FoxP3+ cells were significantly less in vitiligo skin than in normal healthy control skin (NL active = 5.5 [3–8]; NL stable = 2 [1–4]; healthy controls = 48 [36–54]; P < 0.001). Generally, T-cells were present perivascularly in the papillary dermis as well as close to the epidermis. T-cells were concentrated at the dermoepidermal junction (DEJ) and in the lower epidermis; sites where melanocytes are usually present. At the DEJ, CD8+ cells showed no statistically significant difference between active and stable marginal and lesional vitiligo skin (active M = 13.5 [5–18] vs stable M = 11 [0–18]; P > 0.05; active L = 5.5 [3–21] vs stable L = 6 [0–11]; P > 0.05). Notably, CD8+ cells were
Report
highest at the DEJ of marginal skin compared with lesional vitiligo skin of both active and stable cases (P < 0.05; Fig. 1). At the DEJ, FoxP3+ cells were significantly lower in marginal vs lesional vitiligo of active and stable cases (active M = 4 [1–6] vs active L = 18 [16–20]; P < 0.001; stable M = 3 [1–5] vs stable L = 14 [8–18]; P < 0.001). Active cases showed a higher number of FoxP3+ cells in marginal (P = 0.003) and lesional (P < 0.001) skin compared with stable cases (Fig. 2). Discussion Research from several groups support the involvement of autoreactive, melanocyte-specific, cytotoxic T-cells in progressive depigmentation of human skin.16–18 The fact that melanocyte-reactive CD8+ T-cells are also found in control individuals without vitiligo suggests that autoimmune reactivity is kept in check and that this checkpoint is defective in vitiligo.19 Our current study further emphasizes the immunological role in the destructive process of melanocytes in vitiligo. CD4+ and CD8+ T-cells were generally increased in active compared with stable vitiligo. CD4+ T-cells were significantly increased in lesional skin compared with marginal and nonlesional skin in both active and stable vitiligo. CD8+ T-cells were significantly increased in marginal skin compared with lesional and nonlesional skin of both active and stable disease, with a resultant significant decrease of the CD4+/CD8+ ratio in active cases in comparison to stable cases. CD8+ T-cells were more abundant at the DEJ and in the epidermis (sites where melanocytes
Table 1 Comparison between number of cells in lesional, marginal and nonlesional skin in active (left three columns) and stable (right three columns) vitiligo cases Active cases
CD4 (median, range) cells/field P-value versus lesional P-value versus marginal CD8 (median, range) cells/field P-value versus lesional P-value versus marginal CD4/CD8 (median, range) cells/field P-value versus lesional P-value versus marginal FoxP3 (median, range) cells/field P-value versus lesional P-value versus marginal
Stable cases
Lesional
Marginal
Nonlesional
Lesional
Marginal
Nonlesional
89.5 (78–101) – – 49.5 (38–61) –
70.5 (64–82) 0.005 – 48 (36–54) 0.307
38 (30–50) 0.005 0.005 25.5 (18–34) 0.005
59.5 (26–75) – – 16 (8–20) –
51.5 (44–61) 0.114 – 27 (20–37) 0.005
18.5 (11–25) 0.005 0.005 10 (4–17) 0.014
– 1.8 (1.2–2.5) – – 20 (16–24) – –
– 1.6 (1.2–1.8) 0.037 – 2.5 (1–6) 0.005 –
0.005 1.45 (1–2.2) 0.285 0.878 5.5 (3–8) 0.005 0.011
– 3.88 (1.6–7.8) – – 16 (9–22) – –
– 1.7 (1.3–3.8) 0.007 – 3 (2–5) 0.005 –
0.005 1.5 (1–3.8) 0.005 0.959 2 (1–4) 0.005 0.013
The number of cells given is cells/field. ª 2014 The International Society of Dermatology
International Journal of Dermatology 2014
3
4
Report
FoxP3+, CD4+ and CD8+ T-cells in vitiligo
Abdallah et al.
(a)
(b)
(c)
(d)
(e)
(f)
+
+
Figure 1 Skin biopsies in cases of nonsegmental vitiligo with CD4 T-cells (in red) and CD8 T-cells (in blue) in the dermis,
epidermis and DEJ with a magnification of 2009 (immunohistochemical stain). (a) L skin biopsy of an active case of nonsegmental vitiligo, showing numerous CD4+ and CD8+ T-cells in the dermis. The number of CD4+ T (95)-cells is obviously increased in the epidermis and the DEJ, and appears to be more than CD8+ T (50)-cells. (b) L skin biopsy of a stable case of nonsegmental vitiligo, in which the number of CD4+ (26) and CD8+ (20) T-cells is almost similar in the dermis, while the number of CD8+ T-cells is more in the epidermis and the DEJ. (c) PL skin biopsy of an active case of nonsegmental vitiligo. CD8+ T-cells are attacking the melanocytes at the DEJ. Perivascular cellular infiltrate composed of CD4+ (76) and CD8+ (48) T-cells. (d) PL skin biopsy of a stable case of nonsegmental vitiligo, showing CD4+ (44) and CD8+ (20) T-cells in the dermis with a decrease in their numbers in comparison with active cases. (e) NL skin biopsy of an active case of nonsegmental vitiligo, showing CD4+ (30) and CD8+ (18) T-cells in the dermis with decreased numbers in comparison with L and PL figures. Still, CD8+ T-cells are seen close to melanocytes. (f) NL skin biopsy of a stable case of nonsegmental vitiligo, showing CD4+ (23) and CD8+ (12) T-cells in the dermis less than active cases
usually reside) compared with CD4+ cells. These results are in accordance with many previous studies.6,9,17,18,20,21 Le Poole et al.22 found that melanocyte-specific T-cells can induce apoptosis of melanocytes in situ in nonlesional skin of patients with vitiligo, indicating that the effector phase of melanocyte destruction is mediated by cytotoxic T-cells. It has been reported that FoxP3+, CD4+, and CD25+ Tregs are elevated in the serum of active vitiligo patients compared with healthy controls. It was then hypothesized International Journal of Dermatology 2014
that although their number is elevated, their immune inhibitory function may be compromised.15 Recently, Klarquist et al.23 have shown a decrease of regulatory T-cells among CD3+ T-cells in lesional and marginal as well as nonlesional vitiligo skin. However, the study group did not identify differences between active and stable cases or whether Tregs were part of the CD4+ or CD8+ population. The aim of the current prospective, controlled study was to evaluate the Treg cell level among active and ª 2014 The International Society of Dermatology
FoxP3+, CD4+ and CD8+ T-cells in vitiligo
Abdallah et al.
(a)
(b)
(c)
(d)
(e)
(f)
Report
+
Figure 2 FoxP3 cells (brown cells) in active (a, c, e) and stable (b, d, f) vitiligo. Cells are seen perivascularly and at the DEJ
(arrows). FoxP3+ cells are increased in active lesional (a and b), and decreased in marginal (c and d) and nonlesional (e and f) skin. The lowest number of FoxP3+ cells is seen in marginal vitiligo sections (9400, immunohistochemical stain). (a) FoxP3+ cells in lesional skin of active vitiligo (24) showing Tregs mostly at the DEJ, which also appears more in comparison with the stable case. (b) FoxP3+ cells in lesional skin of stable vitiligo (22). (c) FoxP3+ cells in perilesional skin of active vitiligo (5), which appears more at the DEJ in comparison with the stable case, but still has the lowest number of cells in comparison with L and NL skin. (d) FoxP3+ cells in perilesional skin of stable vitiligo (which barely shows Tregs). (e) FoxP3+ cells in nonlesional skin of active vitiligo (8), which appears more in comparison to the stable case. (f) FoxP3+ cells in nonlesional skin of stable vitiligo (4), which shows Tregs mostly at the DEJ
stable cases and their probable relation to one of the two main T-cell subpopulations. The results revealed that the number of FoxP3+ T-cells was significantly reduced in the skin of vitiligo patients in comparison to normal healthy controls. FoxP3+ T-cells were lowest in marginal skin of active cases in comparison with lesional and nonlesional skin of the same group. Similarly, Tregs were lower in marginal skin of stable vitiligo compared with lesional skin of the same group. Tregs regulatory activity is needed at the marginal skin to suppress the activity of helper and cytotoxic T-cells that are actively contributing to depigmentation.23 Therefore, their deficiency plays a role in progression of the disease process. ª 2014 The International Society of Dermatology
Lesional skin showed a higher Treg number in both active and stable vitiligo compared with marginal and nonlesional skin. Because CD4+ T-cells are also increased in lesional skin, we may hypothesize that Tregs are part of the CD4+ population, as has been previously suggested.14 Klarquist et al.23 showed similar results to ours in marginal, lesional, and nonlesional vitiligo (two stable and five active cases), without differentiation between findings in active and stable cases. They further reported an increase of circulating Tregs in patients with vitiligo compared with healthy controls. In order to investigate the cause of such a discrepancy, they perInternational Journal of Dermatology 2014
5
6
Report
FoxP3+, CD4+ and CD8+ T-cells in vitiligo
Abdallah et al.
Table 2 Comparison between active and stable cases of vitiligo regarding CD4+, CD8+ T-cells, FoxP3 and CD4+/ CD8+ ratio
CD4 lesional CD4 marginal CD4 nonlesional CD8 lesional CD8 marginal CD8 nonlesional CD4/CD8 lesional CD4/CD8 marginal CD4/CD8 nonlesional FoxP3 lesional FoxP3 marginal FoxP3 nonlesional
Stable
Active
P-value
59.5 51.5 18.5 16 27 10 3.8 1.7 1.5 16 3 2
89.5 70.5 38 49.5 48 25.5 1.8 1.6 1.4 20 2.5 5.5
