- Email: [email protected]
Reduced immunohistochemical expression of adhesion molecules in vitiligo skin biopsies
Accepted Manuscript Title: REDUCED IMMUNOHISTOCHEMICAL EXPRESSION OF ADHESION MOLECULES IN VITILIGO SKIN BIOPSIES Author: Adriane Reichert Faria Juliana Elizabeth Jung Caio C´esar Silva de Castro Lucia de Noronha PII: DOI: Reference:
S0344-0338(16)30455-1 http://dx.doi.org/doi:10.1016/j.prp.2016.12.019 PRP 51706
To appear in: Received date:
18-9-2016
Please cite this article as: Adriane Reichert Faria, Juliana Elizabeth Jung, Caio C´esar Silva de Castro, Lucia de Noronha, REDUCED IMMUNOHISTOCHEMICAL EXPRESSION OF ADHESION MOLECULES IN VITILIGO SKIN BIOPSIES, Pathology - Research and Practice http://dx.doi.org/10.1016/j.prp.2016.12.019 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Title Page
REDUCED IMMUNOHISTOCHEMICAL EXPRESSION OF ADHESION MOLECULES IN VITILIGO SKIN BIOPSIES
Adriane Reichert Faria, a,b Juliana Elizabeth Jung, c Caio César Silva de Castro, a Lucia de Noronha b
Institutions: a
Santa Casa de Misericórdia de Curitiba Hospital. Praça Rui Barbosa, 694 –
Centro, Curitiba, PR, Brazil. CEP 80010-030 b
Experimental Pathology Laboratory, School of Medicine, Pontifical Catholic
University of Paraná. Rua Imaculada Conceição, 1155 - Prado Velho, Curitiba, PR, Brazil. CEP 80215-901 c
Erasto Gaertner Hospital, Curitiba, PR, Brazil
Abstract
Because defects in adhesion impairment seem to be involved in the etiopathogenesis of vitiligo, this study aimed to compare the immunohistochemical expression of several adhesion molecules in the epidermis of vitiligo and non lesional vitiligo skin. Sixty-six specimens of lesional and non lesional skin from 33 volunteers with vitiligo were evaluated by immunohistochemistry using anti-beta-catenin, anti-Ecadherin, anti-laminin, anti-beta1 integrin, anti-collagen IV, anti-ICAM-1 and antiVCAM-1 antibodies. Biopsies of vitiligo skin demonstrated a significant reduction in the
expression
of
laminin
and
integrin.
The
average
value
of
the
immunohistochemically positive reaction area of the vitiligo specimens was 3053.2 µm2, compared with the observed value of 3431.8 µm2 in non vitiligo skin (p=0.003) for laminin. The immuno-positive area was 7174.6 µm2 (vitiligo) and 8966.7 µm2 (non lesional skin) for integrin (p=0.042). A reduction in ICAM-1 and VCAM-1 expression in the basal layer of the epidermis in vitiligo samples was also observed (p=0.001 and p<0.001, respectively). However, no significant differences were observed with respect to the expression of beta-catenin, E-cadherin, and collagen IV between vitiligo and non lesional skin. Our results suggest that an impairment in adhesion exists in vitiligo skin, which is supported by the diminished immunohistochemical expression of laminin, beta1 integrin, ICAM-1 and VCAM-1.
Key words: vitiligo, immunohistochemistry, adhesion, skin biopsy, paraffinembedded.
Text
Introduction
Vitiligo is an acquired depigmenting disorder that is characterized by the loss of functional epidermal melanocytes. Multifactorial and overlapping pathogenic mechanisms seem to be involved in its etiology. [1] A recent theory termed melanocytorrhagy has suggested that a primary epidermal defective cellular adhesion function may be involved in the loss of melanocytes in vitiligo. [2,3] Cell adhesion, which is critical for the regulation of tissue development and maintenance of tissue architecture, is regulated by various epidermal adhesion molecules (proteins) on the cell surface. [4,5] Interactions between melanocytes and keratinocytes are mediated by cadherins, [6] which connect basal and suprabasal cells, and by catenins, which, in association with cadherins, form cell-cell adherence junctions. [5,7] Interactions between melanocytes and the basement membrane are mediated by integrins, which are adhesive proteins that are constitutively found on basal cells. Integrins mediate cell-cell and cell-matrix communication through a connection with collagen and laminin, among others. [5,6,8,9] Although they are present in smaller amounts, the expression of adhesion molecules from the immunoglobulin superfamily, such as intercellular adhesion molecule-1 (ICAM-1) and vascular adhesion molecule-1 (VCAM-1), is also observed in the epidermis. [5] Both normal and vitiligo melanocytes have shown similarly low levels of constitutive expression of ICAM-1.[10] However, it is unclear whether and how skin injuries can induce the inappropriate expression of ICAM-1 and other proinflammatory genes in melanocytes, as ICAM-1 expression has also been detected in melanocytes around active vitiligo patches as well as in surgically transplanted melanocytes.[4,11] Al-Badri et al. and van denWijngaard et al. also demonstrated the presence of ICAM-1 expression in perilesional melanocytes around active vitiligo patches. The expression of ICAM-1 by melanocytes may contribute to the abnormal immune response in vitiligo.[12,13,14] In normal skin, the expression of VCAM-1 has been observed on perivascular dendritic cells and some follicular keratinocytes.[5,15] In view of this, we decided to study the immunohistochemical expression of several adhesion molecules (beta-catenin, E-cadherin, lamninin, beta1 integrin, collagen IV, ICAM-1 and VCAM-1) found in the skin and to compare vitiligo and non lesional skin specimens.
Materials and Methods
This study was approved by the Research Ethics Committee of Pontifical Catholic University of Paraná and was conducted according the principles of the Declaration of Helsinki. Thirty-three patients who were evaluated by a single dermatologist (CCSC) were enrolled in this study. The main inclusion criteria were as follows: absence of topical or systemic treatment for vitiligo in the last 30 days, area for a biopsy of non lesional vitiligo skin with a 15-cm radius from any vitiliginous macules, as skin biopsies up to 15-cm that had already shown alterations that were not found in non lesional skin.[11] After signing an informed consent, two skin specimens (vitiligo and non lesional skin) were obtained with a 3 mm punch. The vitiligo specimen was obtained from a well-defined achromic area, without any signs of clinical inflammation (erythema). Formalin-fixed paraffin-embedded skin samples were prepared using the tissue microarray (TMA) technique, mounted from the original paraffin blocks containing vitiligo and non lesional skin. The TMA blocks were sectioned to originate multisamples slides that were analyzed by using immunohistochemistry. The immunoperoxidase procedure was used in for immunohistochemistry, as described by Chong.[16] Each immunostaining reaction included positive controls (skin from adults without vitiligo) and negative controls (without incubation with the primary antibody). To observe the immunohistochemical expression of beta-catenin, E-cadherin, ICAM-1, VCAM-1, laminin, beta1 integrin and collagen IV in vitiligo and non lesional skin using immunohistochemistry, the following primary monoclonal antibodies at their respective dilutions were used: from Novocastra® (New Castle upon Tyne, United Kingdom): anti-beta-catenin (1:800), anti-Ecadherin (1:100), anti-ICAM-1 (1:100), and anti-VCAM-1 (1:200). Polyclonal antilaminin (1:800) from DakoCytomation® (Glostrup, Denmark) anti-beta1 integrin (CD29) (1:400) from Epitomics (Burlingame, California), and anti-collagen IV (1:100) from Santa Cruz Biotechnology, Inc (Europe). All TMA staining procedures included both a negative control (which was missing primary antibody) and a positive control (normal skin, from a non vitiligo patient). An immunoperoxidase assay with modifications was part of the immunohistochemistry, as reported by Chong and colleagues.16Antigen retrieval was performed using a BioSBTMImmunoRetriever. Tissue samples were incubated with the primary antibodies in a moist chamber at room temperature for one hour. Incubations
with
the
secondary
antibody
(Dako
AdvanceTMHRP
System,DakoCytomation, Inc., CA, USA) were carried out for 30 min. Incubations
with 3,3-diaminobenzidine and hydrogen peroxide substrate (DakoCytomation, Inc., CA, USA) were performed for 3min to visualize positive staining. All specimens underwent optical microscopic analysis using a BX50 Olympus® (Tokyo,Japan), coupled to a Dinoeye video camera enhanced by image analysis software Image Pro Plus™ (Maryland, USA). For each sample, three fields of each skin specimen (200x) underwent the analysis. The observer (ARF) was unaware of the skin type and was only informed of the antibody that was used. Samples that demonstrated an inappropriate reaction were excluded. One high-quality section of immunohistochemical staining (control and skin control) for anti-beta-catenin, anti-E-cadherin, anti-laminin, anti-beta1 integrin and anti-collagen IV was chosen to serve as a “mask”, which contained adequate levels of positive tissue immunoexpression signal. The mask was then superimposed to the
samples
photomicrographs.
Based
on
the
ideal
positive
tissue
immunoexpression signal obtained from the mask the image analysis software Image Pro Plus™ identified the positive areas in the samples and is able to transform these results into a positive tissue immunoexpression area per square micron (µm2). For each case, an average of positive area was determined in three images. In this field, the observer manually selected the amount of dark brown color to be considered positive antibody expression, establishing a pattern to be followed in the morphometric analysis, performed by the Image Pro Plus™ software to minimize analysis errors. This pattern was followed using three high-power fields (HPF) of each sample of beta-catenin, E-cadherin, laminin, beta1 integrin and collagen IV staining were selected (Olympus BX50 – HPF = 400X) and evaluated using Image Pro Plus® for morphometric analysis. The “mask” was used to compare the expression of such molecules between vitiligo and non lesional skin through the total area/HPF of the positive immunohistochemical reaction obtained (the results are presented as µm2). For ICAM-1 and VCAM-1 staining (Olympus BX40 – HPF = 400X), a qualitative evaluation was performed, and the epidermal localization of positive immunohistochemical reactions was specified (the presence or absence of basal cell predominance). The resultant data were analyzed using the IBM SPSS Statistica v.20.0 software, and paired Student’s t-test (vitiligo and non lesional skin of the same patient) was used for quantitative variables. The variables normality condition was evaluated by the Kolmogorov-Smirnov test. A p value <0.05 indicated statistical significance.
Results
The median age of the individuals studied was 42.8 years, and the most common type of vitiligo was vitiligo vulgaris (75.8%). The clinical characteristics of all included patients and the biopsy location sites are described in Table I. The numeric results for the immunochemistry of adhesion molecules and the comparison of the thickness between vitiligo and non lesional skin are described in Tables II and III. The averages of the immunohistochemical positive areas for laminin (p=0.003), beta1 integrin (p=0.042), ICAM-1 (p=0.001) and VCAM-1 (p<0.001) in vitiligo samples were lower than those obtained for non lesional skin (p<0.001) (Figure 1). There was no pathological sign of inflammation in all samples (vitiligo and non lesional skin). The slides immunostained with anti-beta-1 integrin and anti-laminin antibodies revealed immunopositivity between epidermal keratinocytes. This immunopositivity was marked in the basal portions of non lesional epidermis with a decrease in the intermediate portions and it was virtually absent in the upper layers (Figure 1, B and D). This aspect presented attenuated in vitiligo skin as shown in Figure 1 A and C. Immunostained with anti-beta catenin and anti-E cadherin antibodies revealed immunopositivity in the full extent of the epidermis with a chicken wire pattern (Figure 1 F and H), the same pattern seen in vitiligo skin, but with slight attenuation as shown in Figure 1 E and G. The samples immunostained with anti-ICAM-1 and anti-VCAM-1 antibodies show light immunopositivity only in basal and parabasal layers of the epidermis, and this was observed only in 2 and 3 samples of vitiligo, respectively. The biomarkers of adhesion molecules were tested statistically versus Koebner phenomena presence and abscence, but there was no statistically significant difference between groups. Discussion
Vitiligo is an acquired pigmentation disorder whose etiology is not completely understood and that is clinically characterized by the development of white macules related to the selective loss of melanocytes. What causes damage to melanocytes and their subsequent disappearance in affected skin remains unclear. [12] Several
pathophysiologic
theories
have
been
proposed.
The
melanocytorrhagic hypothesis considers vitiligo a disease caused by the chronic detachment and transepidermal loss of melanocytes. This theory also proposes that
the primary defective adhesion in the epidermis is the major and possible primary predisposing factor for vitiligo. [2,4,12] Adhesion-mediated signaling influences several critical cellular processes including gene expression, the cell cycle and programmed cell death. [7] The basement membrane – where melanocytes reside – is a highly dynamic and complex structure, and it has a substantial role in the regulation of cell adhesion, differentiation, and motility. [17] Recently, genetic variants of a cell surface receptor named DDR1 (discoidin domain receptor 1) that is involved in cell adhesion have been associated with vitiligo,[18,19] as reduced immunohistochemical expression was found in vitiligo skin. [20] In this study, we have investigated the following adhesion molecules that are expressed in normal skin: beta1 integrin, laminin, collagen IV, E-cadherin, betacatenin, ICAM-1 and VCAM-1. This study demonstrated a reduced immunohistochemical expression of beta1 integrin in vitiligo skin compared to non lesional skin (p=0.042). Previously, it has been shown that integrin expression was not affected in vitiligo: no marked differences were found in the overall level of expression of integrins/cadherins among control, non lesional or lesional non segmental vitiligo skin. [21,22] In a comparison of the positive immunoreactivity for the polyclonal antilaminin antibody, this study noted a significant reduction of this adhesion molecule in vitiligo samples (p=0.003). Until now, no other report has examined the variation of laminin expression in vitiligo skin. The present analysis also demonstrated a diminished immunohistochemical expression of both ICAM-1 and VCAM-1 (p=0.001 and p<0.001, respectively) within the epidermal basal layer in vitiligo samples, compared with non lesional skin. In normal skin, VCAM-1 has been observed on perivascular dendritic cells and some follicular keratinocytes in addition to endothelium,[5,15] and ICAM-1 has shown similarly low constitutive expression in both normal and vitiligo melanocytes. [10] It is unclear whether and how skin injuries can induce the inappropriate expression of ICAM-1 and other proinflammatory genes in melanocytes, as ICAM-1 expression has also been detected in melanocytes around active vitiligo patches and in surgically transplanted melanocytes.[4,11] Al-Badri et al. and van denWijngaard et al. demonstrated the presence of ICAM-1 expression in perilesional melanocytes around active vitiligo patches.[12,13,14] In another immunohistochemical study, ICAM-1 was expressed in four of five sections of epidermis in the marginal skin of actively spreading lesions, but not in stable vitiligo.
[23] The expression of ICAM-1 in melanocytes may contribute to the abnormal immune response observed in vitiligo. [4] No remarkable difference was observed with regard to immunohistochemical staining with the beta-catenin or collagen IV antibodies. Because defective Ca2+ transport was demonstrated in keratinocytes and melanocytes from vitiligo skin by Schallreuter and Pittelkow in 1988, cadherins might also be candidates for defective adhesion. By immunohistochemistry, the normal expression of E-cadherin has been observed in non lesional and lesional vitiligo skin and in skin reconstructs composed of normal and vitiligo cells; these findings are enhanced by the present analysis. [2] Tarlé et al. tested for association between vitiligo and polymorphisms of CDH1 gene that codify protein E-cadherin (gene belonging to the DDR1 adhesion pathway). The results reveal that alleles of marker rs10431924 of the CDH1 gene are associated with vitiligo, especially in the presence of autoimmune comorbidities and E-cadherin has recently been shown to be absent from or discontinuously distributed across melanocyte membranes of vitiligo patients long before clinical lesions appear. [24,25]
Conclusion All of the distinguishable differences in the stained markers between vitiligo and non lesional skin, as revealed by the present study, relate to adhesion molecules located outside or as constituents of the basement membrane, where melanocytes reside. These present findings illustrate more evidence that impairment in cell adhesion might be involved in the pathogenesis of vitiligo.
References
[1]. Picardo M, Dell´Anna ML. A review and a new hypothesis for nonimmunological pathogenetic mechanisms in vitiligo. Pigment Cell Res.19 (2006) 406-11.
[2]. Cario-André M, Pain C, Gauthier Y, et al. The melanocytorrhagic hypothesis of vitiligo tested on pigmented, stressed, reconstructed epidermis. Pigm Cell Res. 20 (2007) 385-93.
[3]. Kumar R, Parsad D, Kanwar AJ. Role of apoptosis and melanocytorrhagy: a comparative study of melanocyte adhesion in stable and unstable vitiligo. Br J Dermatol. 164 (2011)187-91.
[4]. Gauthier Y, Cario-Andre M, Taïeb A. A critical appraisal of vitiligo etiologic theories. Is melanocyte loss a melanocytorrhagy? Pigment Cell Res. 16 (2003) 322–32.
[5]. Čabrijan L, Lipozenčić J. Adhesion molecules in keratinocytes. Clin Dermatol. 29 (2011) 427-31.
[6]. Wang C, Yeh Y, Tang M. DDR1/E-caderin complex regulates the activation of DDR1 and cell spreading. Am J Physiol Cell Physiol. (2009) C419-29.
[7]. Aplin AE, Howe A, Alahari SK, et al. Signal transduction and signal modulation by cell adhesion receptors: the role of integrins, cadherins, immunoglobulin-cell adhesion molecules and selectins. Pharmacol Rev. 50 (1998) 197-263.
[8]. Margadant C, Chafeddine RA, Sonnenberg A. Unique and redundant functions of integrins in the epidermis. FASEB J. 24 (2010) 4133-52.
[9]. Lowell CA, Mayadas TN. Overview: studying integrins in vivo. Methods Mol Biol. 757 (2012) 369-97.
[10]. Hedley SJ, Metcalfe R, Gawkrodger DJ, et al. Vitiligo melanocytes in long-term culture show normal constitutive and cytokine-induced expression of intercellular adhesion molecule-1 and major histocompatibility complex class I and class II molecules. Br J Dermatol. 139 (1998) 965-73.
[11]. Zhang S, Liu S, Yu N, et al. RNA released from necrotic keratinocytes upregulates intercellular adhesion molecule-1 expression in melanocytes. Arch Dermatol Res. 303 (2011) 771-6.
[12]. al Badri AM, Foulis AK, Todd PM, et al. Abnormal expression of MHC class II and ICAM-1 by melanocytes in vitiligo. J Pathol. 169 (1993) 203-6.
[13]. Van den Wijngaard R, Wankowicz-Kalinska A, Le Poole C, et al. Local immune response in skin of generalized vitiligo patients. Destruction of melanocytes is associated with the prominent presence of CLA+ T cells at the perilesional site. Lab Invest. 80 (2000):1299-309.
[14].
Abdallah
M,
Abdel-Naser
MB,
Moussa
MH,
et
al.
Sequential
immunohistochemical study of depigmenting and repigmenting minigrafts in vitiligo. Eur J Dermatol. 13 (2003) 548-52.
[15]. Groves RW, Ross EL, Barker JN, et al. Vascular cell adhesion molecule-1: expression in normal and diseased skin and regulation in vivo by interferon gamma. J Am Acad Dermatol. 29 (1993) 67-72.
[16]. Chong DC, Raboni SM, Abujamra KB, et al. Respiratory viruses in pediatric necropsies: an immunohistochemical study. Ped Develop Pathol. 12 (2009) 211216.
[17]. Hashmi S, Marinkovich MP. Molecular organization of the basement membrane zone. Clin Dermatol. 29 (2011) 398-411.
[18]. Kim H, Uhm YK, Yun JY et al. Associations between polymorphisms of discoidin domain receptor tyrosine kinase 1 (DDR1) and non-segmental vitiligo in the Korean population. Eur J Dermatol 29 (2010) 231–232.
[19]. De Castro CC, Do Nascimento LM, Walker G, et al.. Genetic variants of the DDR1 gene are associated with vitiligo in two independent brazilian population samples. J Invest Dermatol. 130 (2010) 1813-8.
[20]. Reichert-Faria A, Jung JE, Moreschi Neto V, Silva de Castro CC, Mira MT, Noronha L. Reduced immunohistochemical expression of Discoidin Domain Receptor 1 (DDR1) in vitiligo skin. J Eur Acad Dermatol Venereol. 27 (2013) 10579.
[21]. Gauthier Y, Cario-Andre M, Lepreux S, Pain C, Taieb A. Melanocyte detachment after skin friction in non lesional skin of patients with generalized vitiligo. Br J Dermatol. 148 (2003) 95–101.
[22]. Le Poole IC, van den Wijngaard RM, Westerhof W, Das PK. Tenascin is overexpressed in vitiligo lesional skin and inhibits melanocyte adhesion. Br J Dermatol. 137 (1997) 171-8.
[23]. Ahn SK, Choi EH, Lee SH, Won JH, Hann SK, Park YK. Immunohistochemical studies from vitiligo comparison between active and inactive lesions. Yonsei Med J. 35 (1994) 404-10.
[24]. Tarlé RG, Silva de Castro CC, do Nascimento LM, Mira MT. Polymorphism of the E-cadherin gene CDH1 is associated with susceptibility to vitiligo. Exp Dermatol. 24 (2015) 300-2.
[25]. Wagner RY, Luciani F, Cario-André M, Rubod A, Petit V, Benzekri L, Ezzedine K, et al. Altered E-Cadherin Levels and Distribution in Melanocytes Precede Clinical Manifestations of Vitiligo. J Invest Dermatol. 135 (2015) 1810-9.
Figure Legends
Figure 1 – Immunohistochemical staining (samples were from the same patient for each antibody tested). K – keratinocytes, Ep – epidermis, BM – basal membrane, arrows point immunopositivity area. 200x A. Positive immunoreactivity of anti-beta1 integrin in vitiligo skin. B. Positive immunoreactivity of anti-beta1 integrin in non lesional skin. C. Positive immunoreactivity of anti-laminin in vitiligo skin. D. Positive immunoreactivity of anti-laminin in non lesional skin. E. Positive immunoreactivity of anti-E-cadherin in vitiligo skin. F. Positive immunoreactivity of anti-E-cadherin in non lesional skin. G. Positive immunoreactivity of anti-beta-catenin in vitiligo skin. H. Positive immunoreactivity of anti-beta-catenin in non lesional skin.
Tables Table I. Demographic and clinical data of the population sample with vitiligo included in the study.
Gender
Age and evolution (years)
n
%
Female
22
66.7
Male
11
33.3
Total
33
100
Average + standard deviation
Median
Age
42.8 ± 15.1
41.0
Age at vitiligo
26.2 ± 16.1
25.0
16.6 ± 15.0
12.0
n
%
Vulgaris
25
75.8
Acrofacial
4
12.1
Segmental
2
6.1
Focal
2
6.1
Total
33
100.0
n
%
With
11
35.5
Without
20
64.5
Total
31
100.0
n
%
With
13
40.6
Without
19
59.4
Total
32
100.0
n
%
With
15
46.9
Without
17
53.1
Total
32
100.0
Non lesional skin
Vitiligo
onset Evolution period
Clinical Type
Autoimmune-comorbidity
Family history of vitiligo
Köebner
n
%
n
%
4
13.0
13
42.0
9
29.0
11
36.0
Chest
1
3.0
2
6.0
Abdomen
8
26.0
1
3.0
Back
9
29.0
4
13.0
Total
31
100.0
31
100.0
Upper Biopsy location
extremities Lower extremities
n= number of individuals with available information
Table II. Immunohistochemical analysis performed on paraffin sections of vitiligo lesional and non lesional skin.
Antibody
Skin sample
n
Area average +
Area median
p
standard deviation 2524 ± 801
2466
1927 ± 980
1795
Difference (vitiligo – non lesional)
-597 ± 904,5
- 448
Non lesional skin
8967 ± 5764
7926
7175 ± 3780
5785
Difference (vitiligo – non lesional)
-1792 ± 4610
-503
Non lesional skin
3581 ± 2236
2832
3296 ± 2012
3469
Difference (vitiligo – non lesional)
-285 ± 3314
-368
Non lesional skin
3478 ± 1751
3356
2900 ± 1246
2587
Difference (vitiligo – non lesional)
-578 ± 1714
-434
Non lesional skin
3277 ± 2273
2492
3201 ± 2528
2612
-76 ± 1917
-74
Non lesional skin Laminin
Integrin
Collagen type IV
β-catenin
E-cadherin
Vitiligo
Vitiligo
Vitiligo
Vitiligo
Vitiligo Difference (vitiligo – non lesional)
25
30
30
29
30
Legend: n= number of samples analyzed by microscopic analysis. Area average and area median are presented in square microns.
0.003
0.042
0.641
0.080
0.830
Table III. Number of samples with a positive immunohistochemical reaction within the epidermal basal layer.
Non lesional skin
VCAM-1
Vitiligo
p
n
%
n
%
Positive samples
26
78.8
3
10.3
Negative samples
7
21.2
26
89.7
Total
33
100.0
29
100.0
Positive samples
17
51.5
2
6.9
Negative samples
16
48.5
27
93.1
Total
33
100.0
29
100.0
<0.001
ICAM-1
0,001
S0344-0338(16)30455-1 http://dx.doi.org/doi:10.1016/j.prp.2016.12.019 PRP 51706
To appear in: Received date:
18-9-2016
Please cite this article as: Adriane Reichert Faria, Juliana Elizabeth Jung, Caio C´esar Silva de Castro, Lucia de Noronha, REDUCED IMMUNOHISTOCHEMICAL EXPRESSION OF ADHESION MOLECULES IN VITILIGO SKIN BIOPSIES, Pathology - Research and Practice http://dx.doi.org/10.1016/j.prp.2016.12.019 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Title Page
REDUCED IMMUNOHISTOCHEMICAL EXPRESSION OF ADHESION MOLECULES IN VITILIGO SKIN BIOPSIES
Adriane Reichert Faria, a,b Juliana Elizabeth Jung, c Caio César Silva de Castro, a Lucia de Noronha b
Institutions: a
Santa Casa de Misericórdia de Curitiba Hospital. Praça Rui Barbosa, 694 –
Centro, Curitiba, PR, Brazil. CEP 80010-030 b
Experimental Pathology Laboratory, School of Medicine, Pontifical Catholic
University of Paraná. Rua Imaculada Conceição, 1155 - Prado Velho, Curitiba, PR, Brazil. CEP 80215-901 c
Erasto Gaertner Hospital, Curitiba, PR, Brazil
Abstract
Because defects in adhesion impairment seem to be involved in the etiopathogenesis of vitiligo, this study aimed to compare the immunohistochemical expression of several adhesion molecules in the epidermis of vitiligo and non lesional vitiligo skin. Sixty-six specimens of lesional and non lesional skin from 33 volunteers with vitiligo were evaluated by immunohistochemistry using anti-beta-catenin, anti-Ecadherin, anti-laminin, anti-beta1 integrin, anti-collagen IV, anti-ICAM-1 and antiVCAM-1 antibodies. Biopsies of vitiligo skin demonstrated a significant reduction in the
expression
of
laminin
and
integrin.
The
average
value
of
the
immunohistochemically positive reaction area of the vitiligo specimens was 3053.2 µm2, compared with the observed value of 3431.8 µm2 in non vitiligo skin (p=0.003) for laminin. The immuno-positive area was 7174.6 µm2 (vitiligo) and 8966.7 µm2 (non lesional skin) for integrin (p=0.042). A reduction in ICAM-1 and VCAM-1 expression in the basal layer of the epidermis in vitiligo samples was also observed (p=0.001 and p<0.001, respectively). However, no significant differences were observed with respect to the expression of beta-catenin, E-cadherin, and collagen IV between vitiligo and non lesional skin. Our results suggest that an impairment in adhesion exists in vitiligo skin, which is supported by the diminished immunohistochemical expression of laminin, beta1 integrin, ICAM-1 and VCAM-1.
Key words: vitiligo, immunohistochemistry, adhesion, skin biopsy, paraffinembedded.
Text
Introduction
Vitiligo is an acquired depigmenting disorder that is characterized by the loss of functional epidermal melanocytes. Multifactorial and overlapping pathogenic mechanisms seem to be involved in its etiology. [1] A recent theory termed melanocytorrhagy has suggested that a primary epidermal defective cellular adhesion function may be involved in the loss of melanocytes in vitiligo. [2,3] Cell adhesion, which is critical for the regulation of tissue development and maintenance of tissue architecture, is regulated by various epidermal adhesion molecules (proteins) on the cell surface. [4,5] Interactions between melanocytes and keratinocytes are mediated by cadherins, [6] which connect basal and suprabasal cells, and by catenins, which, in association with cadherins, form cell-cell adherence junctions. [5,7] Interactions between melanocytes and the basement membrane are mediated by integrins, which are adhesive proteins that are constitutively found on basal cells. Integrins mediate cell-cell and cell-matrix communication through a connection with collagen and laminin, among others. [5,6,8,9] Although they are present in smaller amounts, the expression of adhesion molecules from the immunoglobulin superfamily, such as intercellular adhesion molecule-1 (ICAM-1) and vascular adhesion molecule-1 (VCAM-1), is also observed in the epidermis. [5] Both normal and vitiligo melanocytes have shown similarly low levels of constitutive expression of ICAM-1.[10] However, it is unclear whether and how skin injuries can induce the inappropriate expression of ICAM-1 and other proinflammatory genes in melanocytes, as ICAM-1 expression has also been detected in melanocytes around active vitiligo patches as well as in surgically transplanted melanocytes.[4,11] Al-Badri et al. and van denWijngaard et al. also demonstrated the presence of ICAM-1 expression in perilesional melanocytes around active vitiligo patches. The expression of ICAM-1 by melanocytes may contribute to the abnormal immune response in vitiligo.[12,13,14] In normal skin, the expression of VCAM-1 has been observed on perivascular dendritic cells and some follicular keratinocytes.[5,15] In view of this, we decided to study the immunohistochemical expression of several adhesion molecules (beta-catenin, E-cadherin, lamninin, beta1 integrin, collagen IV, ICAM-1 and VCAM-1) found in the skin and to compare vitiligo and non lesional skin specimens.
Materials and Methods
This study was approved by the Research Ethics Committee of Pontifical Catholic University of Paraná and was conducted according the principles of the Declaration of Helsinki. Thirty-three patients who were evaluated by a single dermatologist (CCSC) were enrolled in this study. The main inclusion criteria were as follows: absence of topical or systemic treatment for vitiligo in the last 30 days, area for a biopsy of non lesional vitiligo skin with a 15-cm radius from any vitiliginous macules, as skin biopsies up to 15-cm that had already shown alterations that were not found in non lesional skin.[11] After signing an informed consent, two skin specimens (vitiligo and non lesional skin) were obtained with a 3 mm punch. The vitiligo specimen was obtained from a well-defined achromic area, without any signs of clinical inflammation (erythema). Formalin-fixed paraffin-embedded skin samples were prepared using the tissue microarray (TMA) technique, mounted from the original paraffin blocks containing vitiligo and non lesional skin. The TMA blocks were sectioned to originate multisamples slides that were analyzed by using immunohistochemistry. The immunoperoxidase procedure was used in for immunohistochemistry, as described by Chong.[16] Each immunostaining reaction included positive controls (skin from adults without vitiligo) and negative controls (without incubation with the primary antibody). To observe the immunohistochemical expression of beta-catenin, E-cadherin, ICAM-1, VCAM-1, laminin, beta1 integrin and collagen IV in vitiligo and non lesional skin using immunohistochemistry, the following primary monoclonal antibodies at their respective dilutions were used: from Novocastra® (New Castle upon Tyne, United Kingdom): anti-beta-catenin (1:800), anti-Ecadherin (1:100), anti-ICAM-1 (1:100), and anti-VCAM-1 (1:200). Polyclonal antilaminin (1:800) from DakoCytomation® (Glostrup, Denmark) anti-beta1 integrin (CD29) (1:400) from Epitomics (Burlingame, California), and anti-collagen IV (1:100) from Santa Cruz Biotechnology, Inc (Europe). All TMA staining procedures included both a negative control (which was missing primary antibody) and a positive control (normal skin, from a non vitiligo patient). An immunoperoxidase assay with modifications was part of the immunohistochemistry, as reported by Chong and colleagues.16Antigen retrieval was performed using a BioSBTMImmunoRetriever. Tissue samples were incubated with the primary antibodies in a moist chamber at room temperature for one hour. Incubations
with
the
secondary
antibody
(Dako
AdvanceTMHRP
System,DakoCytomation, Inc., CA, USA) were carried out for 30 min. Incubations
with 3,3-diaminobenzidine and hydrogen peroxide substrate (DakoCytomation, Inc., CA, USA) were performed for 3min to visualize positive staining. All specimens underwent optical microscopic analysis using a BX50 Olympus® (Tokyo,Japan), coupled to a Dinoeye video camera enhanced by image analysis software Image Pro Plus™ (Maryland, USA). For each sample, three fields of each skin specimen (200x) underwent the analysis. The observer (ARF) was unaware of the skin type and was only informed of the antibody that was used. Samples that demonstrated an inappropriate reaction were excluded. One high-quality section of immunohistochemical staining (control and skin control) for anti-beta-catenin, anti-E-cadherin, anti-laminin, anti-beta1 integrin and anti-collagen IV was chosen to serve as a “mask”, which contained adequate levels of positive tissue immunoexpression signal. The mask was then superimposed to the
samples
photomicrographs.
Based
on
the
ideal
positive
tissue
immunoexpression signal obtained from the mask the image analysis software Image Pro Plus™ identified the positive areas in the samples and is able to transform these results into a positive tissue immunoexpression area per square micron (µm2). For each case, an average of positive area was determined in three images. In this field, the observer manually selected the amount of dark brown color to be considered positive antibody expression, establishing a pattern to be followed in the morphometric analysis, performed by the Image Pro Plus™ software to minimize analysis errors. This pattern was followed using three high-power fields (HPF) of each sample of beta-catenin, E-cadherin, laminin, beta1 integrin and collagen IV staining were selected (Olympus BX50 – HPF = 400X) and evaluated using Image Pro Plus® for morphometric analysis. The “mask” was used to compare the expression of such molecules between vitiligo and non lesional skin through the total area/HPF of the positive immunohistochemical reaction obtained (the results are presented as µm2). For ICAM-1 and VCAM-1 staining (Olympus BX40 – HPF = 400X), a qualitative evaluation was performed, and the epidermal localization of positive immunohistochemical reactions was specified (the presence or absence of basal cell predominance). The resultant data were analyzed using the IBM SPSS Statistica v.20.0 software, and paired Student’s t-test (vitiligo and non lesional skin of the same patient) was used for quantitative variables. The variables normality condition was evaluated by the Kolmogorov-Smirnov test. A p value <0.05 indicated statistical significance.
Results
The median age of the individuals studied was 42.8 years, and the most common type of vitiligo was vitiligo vulgaris (75.8%). The clinical characteristics of all included patients and the biopsy location sites are described in Table I. The numeric results for the immunochemistry of adhesion molecules and the comparison of the thickness between vitiligo and non lesional skin are described in Tables II and III. The averages of the immunohistochemical positive areas for laminin (p=0.003), beta1 integrin (p=0.042), ICAM-1 (p=0.001) and VCAM-1 (p<0.001) in vitiligo samples were lower than those obtained for non lesional skin (p<0.001) (Figure 1). There was no pathological sign of inflammation in all samples (vitiligo and non lesional skin). The slides immunostained with anti-beta-1 integrin and anti-laminin antibodies revealed immunopositivity between epidermal keratinocytes. This immunopositivity was marked in the basal portions of non lesional epidermis with a decrease in the intermediate portions and it was virtually absent in the upper layers (Figure 1, B and D). This aspect presented attenuated in vitiligo skin as shown in Figure 1 A and C. Immunostained with anti-beta catenin and anti-E cadherin antibodies revealed immunopositivity in the full extent of the epidermis with a chicken wire pattern (Figure 1 F and H), the same pattern seen in vitiligo skin, but with slight attenuation as shown in Figure 1 E and G. The samples immunostained with anti-ICAM-1 and anti-VCAM-1 antibodies show light immunopositivity only in basal and parabasal layers of the epidermis, and this was observed only in 2 and 3 samples of vitiligo, respectively. The biomarkers of adhesion molecules were tested statistically versus Koebner phenomena presence and abscence, but there was no statistically significant difference between groups. Discussion
Vitiligo is an acquired pigmentation disorder whose etiology is not completely understood and that is clinically characterized by the development of white macules related to the selective loss of melanocytes. What causes damage to melanocytes and their subsequent disappearance in affected skin remains unclear. [12] Several
pathophysiologic
theories
have
been
proposed.
The
melanocytorrhagic hypothesis considers vitiligo a disease caused by the chronic detachment and transepidermal loss of melanocytes. This theory also proposes that
the primary defective adhesion in the epidermis is the major and possible primary predisposing factor for vitiligo. [2,4,12] Adhesion-mediated signaling influences several critical cellular processes including gene expression, the cell cycle and programmed cell death. [7] The basement membrane – where melanocytes reside – is a highly dynamic and complex structure, and it has a substantial role in the regulation of cell adhesion, differentiation, and motility. [17] Recently, genetic variants of a cell surface receptor named DDR1 (discoidin domain receptor 1) that is involved in cell adhesion have been associated with vitiligo,[18,19] as reduced immunohistochemical expression was found in vitiligo skin. [20] In this study, we have investigated the following adhesion molecules that are expressed in normal skin: beta1 integrin, laminin, collagen IV, E-cadherin, betacatenin, ICAM-1 and VCAM-1. This study demonstrated a reduced immunohistochemical expression of beta1 integrin in vitiligo skin compared to non lesional skin (p=0.042). Previously, it has been shown that integrin expression was not affected in vitiligo: no marked differences were found in the overall level of expression of integrins/cadherins among control, non lesional or lesional non segmental vitiligo skin. [21,22] In a comparison of the positive immunoreactivity for the polyclonal antilaminin antibody, this study noted a significant reduction of this adhesion molecule in vitiligo samples (p=0.003). Until now, no other report has examined the variation of laminin expression in vitiligo skin. The present analysis also demonstrated a diminished immunohistochemical expression of both ICAM-1 and VCAM-1 (p=0.001 and p<0.001, respectively) within the epidermal basal layer in vitiligo samples, compared with non lesional skin. In normal skin, VCAM-1 has been observed on perivascular dendritic cells and some follicular keratinocytes in addition to endothelium,[5,15] and ICAM-1 has shown similarly low constitutive expression in both normal and vitiligo melanocytes. [10] It is unclear whether and how skin injuries can induce the inappropriate expression of ICAM-1 and other proinflammatory genes in melanocytes, as ICAM-1 expression has also been detected in melanocytes around active vitiligo patches and in surgically transplanted melanocytes.[4,11] Al-Badri et al. and van denWijngaard et al. demonstrated the presence of ICAM-1 expression in perilesional melanocytes around active vitiligo patches.[12,13,14] In another immunohistochemical study, ICAM-1 was expressed in four of five sections of epidermis in the marginal skin of actively spreading lesions, but not in stable vitiligo.
[23] The expression of ICAM-1 in melanocytes may contribute to the abnormal immune response observed in vitiligo. [4] No remarkable difference was observed with regard to immunohistochemical staining with the beta-catenin or collagen IV antibodies. Because defective Ca2+ transport was demonstrated in keratinocytes and melanocytes from vitiligo skin by Schallreuter and Pittelkow in 1988, cadherins might also be candidates for defective adhesion. By immunohistochemistry, the normal expression of E-cadherin has been observed in non lesional and lesional vitiligo skin and in skin reconstructs composed of normal and vitiligo cells; these findings are enhanced by the present analysis. [2] Tarlé et al. tested for association between vitiligo and polymorphisms of CDH1 gene that codify protein E-cadherin (gene belonging to the DDR1 adhesion pathway). The results reveal that alleles of marker rs10431924 of the CDH1 gene are associated with vitiligo, especially in the presence of autoimmune comorbidities and E-cadherin has recently been shown to be absent from or discontinuously distributed across melanocyte membranes of vitiligo patients long before clinical lesions appear. [24,25]
Conclusion All of the distinguishable differences in the stained markers between vitiligo and non lesional skin, as revealed by the present study, relate to adhesion molecules located outside or as constituents of the basement membrane, where melanocytes reside. These present findings illustrate more evidence that impairment in cell adhesion might be involved in the pathogenesis of vitiligo.
References
[1]. Picardo M, Dell´Anna ML. A review and a new hypothesis for nonimmunological pathogenetic mechanisms in vitiligo. Pigment Cell Res.19 (2006) 406-11.
[2]. Cario-André M, Pain C, Gauthier Y, et al. The melanocytorrhagic hypothesis of vitiligo tested on pigmented, stressed, reconstructed epidermis. Pigm Cell Res. 20 (2007) 385-93.
[3]. Kumar R, Parsad D, Kanwar AJ. Role of apoptosis and melanocytorrhagy: a comparative study of melanocyte adhesion in stable and unstable vitiligo. Br J Dermatol. 164 (2011)187-91.
[4]. Gauthier Y, Cario-Andre M, Taïeb A. A critical appraisal of vitiligo etiologic theories. Is melanocyte loss a melanocytorrhagy? Pigment Cell Res. 16 (2003) 322–32.
[5]. Čabrijan L, Lipozenčić J. Adhesion molecules in keratinocytes. Clin Dermatol. 29 (2011) 427-31.
[6]. Wang C, Yeh Y, Tang M. DDR1/E-caderin complex regulates the activation of DDR1 and cell spreading. Am J Physiol Cell Physiol. (2009) C419-29.
[7]. Aplin AE, Howe A, Alahari SK, et al. Signal transduction and signal modulation by cell adhesion receptors: the role of integrins, cadherins, immunoglobulin-cell adhesion molecules and selectins. Pharmacol Rev. 50 (1998) 197-263.
[8]. Margadant C, Chafeddine RA, Sonnenberg A. Unique and redundant functions of integrins in the epidermis. FASEB J. 24 (2010) 4133-52.
[9]. Lowell CA, Mayadas TN. Overview: studying integrins in vivo. Methods Mol Biol. 757 (2012) 369-97.
[10]. Hedley SJ, Metcalfe R, Gawkrodger DJ, et al. Vitiligo melanocytes in long-term culture show normal constitutive and cytokine-induced expression of intercellular adhesion molecule-1 and major histocompatibility complex class I and class II molecules. Br J Dermatol. 139 (1998) 965-73.
[11]. Zhang S, Liu S, Yu N, et al. RNA released from necrotic keratinocytes upregulates intercellular adhesion molecule-1 expression in melanocytes. Arch Dermatol Res. 303 (2011) 771-6.
[12]. al Badri AM, Foulis AK, Todd PM, et al. Abnormal expression of MHC class II and ICAM-1 by melanocytes in vitiligo. J Pathol. 169 (1993) 203-6.
[13]. Van den Wijngaard R, Wankowicz-Kalinska A, Le Poole C, et al. Local immune response in skin of generalized vitiligo patients. Destruction of melanocytes is associated with the prominent presence of CLA+ T cells at the perilesional site. Lab Invest. 80 (2000):1299-309.
[14].
Abdallah
M,
Abdel-Naser
MB,
Moussa
MH,
et
al.
Sequential
immunohistochemical study of depigmenting and repigmenting minigrafts in vitiligo. Eur J Dermatol. 13 (2003) 548-52.
[15]. Groves RW, Ross EL, Barker JN, et al. Vascular cell adhesion molecule-1: expression in normal and diseased skin and regulation in vivo by interferon gamma. J Am Acad Dermatol. 29 (1993) 67-72.
[16]. Chong DC, Raboni SM, Abujamra KB, et al. Respiratory viruses in pediatric necropsies: an immunohistochemical study. Ped Develop Pathol. 12 (2009) 211216.
[17]. Hashmi S, Marinkovich MP. Molecular organization of the basement membrane zone. Clin Dermatol. 29 (2011) 398-411.
[18]. Kim H, Uhm YK, Yun JY et al. Associations between polymorphisms of discoidin domain receptor tyrosine kinase 1 (DDR1) and non-segmental vitiligo in the Korean population. Eur J Dermatol 29 (2010) 231–232.
[19]. De Castro CC, Do Nascimento LM, Walker G, et al.. Genetic variants of the DDR1 gene are associated with vitiligo in two independent brazilian population samples. J Invest Dermatol. 130 (2010) 1813-8.
[20]. Reichert-Faria A, Jung JE, Moreschi Neto V, Silva de Castro CC, Mira MT, Noronha L. Reduced immunohistochemical expression of Discoidin Domain Receptor 1 (DDR1) in vitiligo skin. J Eur Acad Dermatol Venereol. 27 (2013) 10579.
[21]. Gauthier Y, Cario-Andre M, Lepreux S, Pain C, Taieb A. Melanocyte detachment after skin friction in non lesional skin of patients with generalized vitiligo. Br J Dermatol. 148 (2003) 95–101.
[22]. Le Poole IC, van den Wijngaard RM, Westerhof W, Das PK. Tenascin is overexpressed in vitiligo lesional skin and inhibits melanocyte adhesion. Br J Dermatol. 137 (1997) 171-8.
[23]. Ahn SK, Choi EH, Lee SH, Won JH, Hann SK, Park YK. Immunohistochemical studies from vitiligo comparison between active and inactive lesions. Yonsei Med J. 35 (1994) 404-10.
[24]. Tarlé RG, Silva de Castro CC, do Nascimento LM, Mira MT. Polymorphism of the E-cadherin gene CDH1 is associated with susceptibility to vitiligo. Exp Dermatol. 24 (2015) 300-2.
[25]. Wagner RY, Luciani F, Cario-André M, Rubod A, Petit V, Benzekri L, Ezzedine K, et al. Altered E-Cadherin Levels and Distribution in Melanocytes Precede Clinical Manifestations of Vitiligo. J Invest Dermatol. 135 (2015) 1810-9.
Figure Legends
Figure 1 – Immunohistochemical staining (samples were from the same patient for each antibody tested). K – keratinocytes, Ep – epidermis, BM – basal membrane, arrows point immunopositivity area. 200x A. Positive immunoreactivity of anti-beta1 integrin in vitiligo skin. B. Positive immunoreactivity of anti-beta1 integrin in non lesional skin. C. Positive immunoreactivity of anti-laminin in vitiligo skin. D. Positive immunoreactivity of anti-laminin in non lesional skin. E. Positive immunoreactivity of anti-E-cadherin in vitiligo skin. F. Positive immunoreactivity of anti-E-cadherin in non lesional skin. G. Positive immunoreactivity of anti-beta-catenin in vitiligo skin. H. Positive immunoreactivity of anti-beta-catenin in non lesional skin.
Tables Table I. Demographic and clinical data of the population sample with vitiligo included in the study.
Gender
Age and evolution (years)
n
%
Female
22
66.7
Male
11
33.3
Total
33
100
Average + standard deviation
Median
Age
42.8 ± 15.1
41.0
Age at vitiligo
26.2 ± 16.1
25.0
16.6 ± 15.0
12.0
n
%
Vulgaris
25
75.8
Acrofacial
4
12.1
Segmental
2
6.1
Focal
2
6.1
Total
33
100.0
n
%
With
11
35.5
Without
20
64.5
Total
31
100.0
n
%
With
13
40.6
Without
19
59.4
Total
32
100.0
n
%
With
15
46.9
Without
17
53.1
Total
32
100.0
Non lesional skin
Vitiligo
onset Evolution period
Clinical Type
Autoimmune-comorbidity
Family history of vitiligo
Köebner
n
%
n
%
4
13.0
13
42.0
9
29.0
11
36.0
Chest
1
3.0
2
6.0
Abdomen
8
26.0
1
3.0
Back
9
29.0
4
13.0
Total
31
100.0
31
100.0
Upper Biopsy location
extremities Lower extremities
n= number of individuals with available information
Table II. Immunohistochemical analysis performed on paraffin sections of vitiligo lesional and non lesional skin.
Antibody
Skin sample
n
Area average +
Area median
p
standard deviation 2524 ± 801
2466
1927 ± 980
1795
Difference (vitiligo – non lesional)
-597 ± 904,5
- 448
Non lesional skin
8967 ± 5764
7926
7175 ± 3780
5785
Difference (vitiligo – non lesional)
-1792 ± 4610
-503
Non lesional skin
3581 ± 2236
2832
3296 ± 2012
3469
Difference (vitiligo – non lesional)
-285 ± 3314
-368
Non lesional skin
3478 ± 1751
3356
2900 ± 1246
2587
Difference (vitiligo – non lesional)
-578 ± 1714
-434
Non lesional skin
3277 ± 2273
2492
3201 ± 2528
2612
-76 ± 1917
-74
Non lesional skin Laminin
Integrin
Collagen type IV
β-catenin
E-cadherin
Vitiligo
Vitiligo
Vitiligo
Vitiligo
Vitiligo Difference (vitiligo – non lesional)
25
30
30
29
30
Legend: n= number of samples analyzed by microscopic analysis. Area average and area median are presented in square microns.
0.003
0.042
0.641
0.080
0.830
Table III. Number of samples with a positive immunohistochemical reaction within the epidermal basal layer.
Non lesional skin
VCAM-1
Vitiligo
p
n
%
n
%
Positive samples
26
78.8
3
10.3
Negative samples
7
21.2
26
89.7
Total
33
100.0
29
100.0
Positive samples
17
51.5
2
6.9
Negative samples
16
48.5
27
93.1
Total
33
100.0
29
100.0
<0.001
ICAM-1
0,001