- Email: [email protected]
Enhanced cutaneous Rock2 expression as a marker of Rho Kinase pathway activation in autoimmune disease and Kohlemeier-Degos disease
Annals of Diagnostic Pathology 44 (2020) 151414
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology journal homepage: www.elsevier.com/locate/anndiagpath
Enhanced cutaneous Rock2 expression as a marker of Rho Kinase pathway activation in autoimmune disease and Kohlemeier-Degos disease
T
⁎
Cynthia Magro , Ziv Schwartz, Jad Saab, Amin Hedayat Weill Cornell Medicine, New York, United States of America
A R T I C LE I N FO
A B S T R A C T
Keywords: ROC vK2 Rho Kinase Scleroderma Lupus Dermatomyositis Degos Autoimmune
The small guanosine triphosphatase Rho and its target Rho kinase are involved in a heterogeneous spectrum of cellular activities, many of which are integral to cytoskeletal organization. Furthermore, the Rho kinases result in NF kappa beta activation and hence the induction of various pro-inflammatory cytokines including TNF-alpha, IL-1B and IL-6. ROCK2 is a downstream protein, whose expression is indicative of Rho Kinase activation. Given the diverse effects of Rho-kinase, including a potentially critical role in augmenting inflammation, ROCK2 expression was examined in biopsies of select autoimmune connective tissue diseases as compared to control diagnoses. Select cases of lupus erythematosus, dermatomyositis, autoimmune sclerodermoid disorders and Kohlmeier-Degos disease (a distinctive vasculopathy that occurs in the other aforesaid conditions but also as a forme fruste microvascular and arteriopathic syndrome) were studied. Control biopsies included normal skin and cutaneous inflammatory conditions unrelated to collagen vascular disease/autoimmune disease. We found ROCK2 expression significantly increased in biopsies of lupus erythematosus, dermatomyositis, scleroderma and Kohlmeier-Degos disease. A pattern emerged of consistent marked ROCK2 upregulation in endothelium and variable expression in inflammatory cells and epithelium. While expression was undetectable in normal skin, it was found in inflamed skin unrelated to specific autoimmune disease. The staining pattern could approach that seen in study group cases but was less pronounced and preferentially upregulated in the endothelium, with a lesser extent of staining in the epidermis and inflammatory cells. Rho kinase is a driving factor in diverse cutaneous diseases especially autoimmune disease and Kohlmeier-Degos disease. This significantly upregulated pathway defines a potential target for biologic therapy.
1. Introduction The small guanosine triphosphatase (GTP) Rho and its target Rho kinase are involved in a heterogeneous spectrum of cellular activities many of which are integral to cytoskeletal organization. The Rho GTPase, RhoA, demonstrates a specific conformational alteration, which following binding to GTP leads to intracellular protein activation. P160 ROCK is a protein serine/threonine kinase, whereby it has a kinase domain associated with Rho-binding sequences. There are two related isoforms referred to as ROCK1 and ROCK2, both being activated by Rho-kinase. When considering the varied functions of the ROCK proteins, cell migration, adhesion, cellular proliferation, apoptosis, and cellular differentiation are potentially modulated by these proteins. Both ROCK1 and ROCK2 are considered critical molecular switches integral to actin cytoskeleton organization, apoptosis, reactive oxygen species, cell migration and adhesion. They can promote a pro-inflammatory milieu with upregulation of various cytokines including
⁎
Corresponding author. E-mail address: [email protected] (C. Magro).
https://doi.org/10.1016/j.anndiagpath.2019.151414
1092-9134/ © 2019 Elsevier Inc. All rights reserved.
TNF-alpha, IL-1B and IL-6 [1]. In addition to its role in inflammation, ROCK2 has been linked with vascular remodeling. Vascular smooth muscle motility and smooth muscle contraction occurs through actin activation of myosin ATPase via ROCK2. Arterial wall smooth muscle migration, which is highly dependent on Rho kinase activation, is a critical impetus to neointimal proliferation seen in diverse settings ranging from atherosclerosis to accelerated vascular rejection post-transplantation to the fibrosing intimal vascular lesions that are observed in the setting of lupus erythematosus and Kohlmeier-Degos disease [2-4]. The Rho pathway has not been explored in the cutaneous lesions of select autoimmune conditions where inflammation and vascular disease are integral players in the clinical phenotype. Dermatomyositis, scleroderma, lupus erythematosus and Degos disease are among these disorders. Degos disease is of particular interest since it can be seen either as a forme fruste of vascular disease and or seen in the setting of select autoimmune connective disease syndromes such as dermatomyositis
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
Table 1 Lupus: ROCK 2 immunoreactivity among biopsies of lupus patients. Case
Age/sex
Diagnosis
Microscopic description
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
32/F 3 mths/F 51/F 58/F 26/F 54/M 2/F 48/F 65/F 25/F 64/M 64/F 67/M 64/F 45/F 48/M 48/M 36/F 58/F 30/F 44/F 43/M 55/F 54/F
Rowell's syndrome Neonatal lupus DLE DLE Early DLE in the setting of SLE SLE SLE SLE SLE SLE SLE/SCLE SCLE SCLE SCLE SCLE SCLE SCLE SCLE SCLE SCLE SCLE Proton pump inhibitor therapy induced SCLE Lupus profundus with overlying changes of SCLE Lupus Profundus
+1–2/3 EC; +1/3 epidermis +3/3 EC; +1/3 epidermis; +1/3 inflammatory cells +3/3 EC; +1/3 basal layer of epidermis; +1/3 inflammatory cells +2/3 EC +3/3 EC; 2/3 lower portion of epidermis; +1/3 inflammatory cells +1/3 EC +3/3 EC +3/3 EC; +2/3 epidermis; +1/3 inflammatory cells +2/3 EC +3/3 EC; +1/3 epidermis; 0/3 inflammatory cells +3/3 EC; +2/3 epidermis; +1/3 inflammatory cells +3/3 EC +3/3 EC; +2/3 epidermis; +2/3 inflammatory cells +3/3 EC +3/3 EC; +3/3 epidermis +3/3 EC; +2/3 epidermis; +2/3 inflammatory cells +3/3 EC; +2/3 epidermis; +1/3 inflammatory cells +3/3 EC; +2/3 epidermis, +2/3 inflammatory cells; +2/3 follicle +3/3 EC; +3/3 perivascular inflammatory cells +2/3 EC; +1/3 epidermis +3/3 EC +3/3 EC; +2/3 epidermis +3/3 EC; +2/3 epidermis; +3/3 inflammatory cells +3/3 EC; +2/3 inflammatory cells
Mths – Months, F – Female, M- Male, SCLE – Subacute Cutaneous Lupus Erythematosus, DLE - Discoid Lupus Erythematosus, SLE - Systemic lupus erythematosus; EC = endothelial cell staining.
[5]. We assessed ROCK2 expression in biopsies of lupus erythematosus, dermatomyositis, autoimmune sclerodermoid conditions and Kohlmeier-Degos disease. We also examined the expression of Rho kinase in normal skin, in the setting of other primary cutaneous inflammatory conditions, and in certain systemic diseases such as atypical hemolytic uremic syndrome where clinically normal skin was biopsied.
Table 3 Scleroderma: ROCK 2 immunoreactivity among biopsies of scleroderma patients.
2. Materials and methods Select cases diagnosed as lupus erythematosus, dermatomyositis, autoimmune sclerodermoid disorders were examined. Lupus erythematosus was represented by 24 cases (Table 1): 3 cases of discoid lupus erythematosus (DLE), 13 cases of subacute cutaneous lupus erythematosus (SCLE), 2 cases of lupus profundus, 7 cases of systemic lupus erythematosus (SLE), 1 case of Rowell's syndrome and 1 case of neonatal lupus erythematosus. Dermatomyositis was represented by 14 cases.(Table 2) Scleroderma was represented by 11 cases (Table 3): 9 cases of systemic scleroderma, 1 case of sclerodermatomyositis, and 1 case of morphea. Kohlmeier-Degos disease was represented by 7 cases
Case
Age/Sex
Diagnosis
Microscopic description
1 2 3 4 5 6 7 8 9 10
41/F 69/M NA NA NA NA NA NA NA 60/F
SSCL SSCL (Anti-Ro positive) SSCL SSCL SSCL SSCL SSCL SSCL SSCL Sclerodermatomyositis
11
33/F
Morphea
0/3 EC +1/3 EC +1/3 EC +1/3 EC +1/3 EC +1/3 EC +2/3 EC +1/3 EC +2/3 EC +2/3 EC; +1/3 basal layer of epidermis +3/3 EC
F – female; M – male; NA – not available; SSCL – systemic scleroderma; EC endothelial cell staining.
Table 2 Dermatomyositis: ROCK 2 immunoreactivity among biopsies of dermatomyositis patients. Case
Age/Sex
Diagnosis
Microscopic description
1 2 3 4 5 6 7 8
56/M 52/F 18/M 50/F 70/F 48/F 58/M 77/F
DM DM DM DM DM DM DM DM
9 10 11 12 13 14
76/M 74/F 26/F 48/M 57/M 46/M
DM End-Stage, Gottron's papule, DM Quiescent phase, DM Gottron's papule, DM Lymphocyte-rich DM Anti-Mi-2 Dermatomyositis
+3/3 EC; +2/3 epidermis; +2/3 inflammatory cells +3/3 EC; +1/3 inflammatory cells +2/3 EC +3/3 EC +1/3 EC; +1/3 basal layer of epidermis +2/3 EC +3/3 EC +3/3 EC; +1/3 epidermis (basilar and parabasilar keratinocytes); greatest staining superficially within the dermis; +2/3 staining of fibroblasts +2/3 EC +3/3 EC; +1/3 epidermis +2/3 EC +1–2/3 EC (most apparent superficially) +3/3 EC; +2/3 epidermis; +2/3 inflammatory cells +3/3 EC; +2/3 epidermis; +2/3 inflammatory cells
F – female; M – male; DM – dermatomyositis; EC - endothelial cell staining. 2
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
Table 4 Kohlmeier Degos: ROCK 2 immunoreactivity among biopsies of Kohlmeier Degos patients. Case
Age/Gender
Diagnosis
Microscopic findings
1 2 3 4 5 6 7
31/F 19/M 7/F NA 43/M 43/F 57/M
Degos disease like changes in the coronary vessels Multiorgan Kohlmeier Degos Disease Multiorgan Kohlmeier Degos Disease Multiorgan Kohlmeier Degos Disease Kohlmeier Degos disease Late Stage Kohlmeier Degos disease Kohlmeier Degos Disease
+3/3 EC before treatment; +1/3 EC after treatment +3/3 EC +2/3 EC +3/3 EC; weak staining of epidermis, inflammatory cells, and stromal cells 0/3 EC +2/3 EC; +0–1/3 epidermis; +0–1/3 inflammatory cells +2/3 EC; 0/3 epidermis; 0/3 inflammatory cells
F – female; M – male; NA – not available; EC - endothelial cell staining. Table 5 Controls: ROCK 2 immunoreactivity among biopsies of control subjects. Case
Age/Gender
Diagnosis
Microscopic findings
1 2 3 4 5 6 7 8 9 10
42/M 33/F 65/F 66/F 50/F 24/F 52/M 57/F 48/F 71/M
+2/EC; +0–1/3 inflammatory cells +1/3 inflammatory cells +1/3 EC; +1/3 inflammatory cells +2/3 EC; +0–1/3 inflammatory cells 0/3 EC +0–1/3 EC +3/3 EC 0/3 EC 0/3 EC +2/3 EC
11 12 13 14
43/M 46/M 91/F 59/F
Bullous eosinophilic cellulitis-like arthropod bite hypersensitivity reaction Lichenoid dermatitis Lichenoid dermatitis Lichenoid dermatitis Angiolipoma Normal skin in the setting of small fiber neuropathy Antiphospholipid antibody syndrome Normal skin biopsy in the setting of atypical Hemolytic uremic syndrome (HUS) Small fiber neuropathy, normal skin biopsy sampled Superficial fibrosing dermopathy with chronic microvascular changes and irritant epidermal changes suggestive of toxic chemotherapy effect Atypical HUS, biopsy of normal skin. Normal skin Actinic keratosis Scar
0/3 0/3 0/3 0/3
EC EC EC EC
F – female; M – male; EC - endothelial cell staining; NOS – not otherwise specified.
Fig. 1. ROCK2 expression in lupus. This biopsy shows a mild lymphocytic mediated interface dermatitis involving the acrosyringium and hair follicle compatible with early discoid lupus erythematosus in the setting of systemic lupus erythematosus (a). The Rock2 studies show significant upregulation in the epithelium, amidst inflammatory cells and in endothelium (b). In this additional case of discoid lupus erythematosus the lower power image captures the extent of staining for Rock2 in the epidermis, inflammatory cells and in endothelium (c). A higher power image is provided as well (d). In this case of subacute cutaneous lupus erythematosus, the typical cellular band like lymphocytic infiltrate is seen (e). Rock2 shows a significantly upregulated pattern of staining involving the lower portion of the epidermis, endothelium and many inflammatory cells (f). This additional case of subacute cutaneous lupus erythematosus demonstrates the architectural disposition of the infiltrate being superficial and band-like. Illustrated is MXA, the surrogate type I interferon marker, demonstrating a marked upregulation in the type I interferon signature (g). Rock2 is markedly upregulated in the epidermis, amidst inflammatory cells and in the endothelium (h).
antibody syndrome (1 case), actinic keratosis (1 case), and normal skin (1 case).(Table 5). In each case, 4-μm sections were stained with the ROCK2 isoform for Rho kinase. The image of immunohistochemical ROCK2 staining in formalin fixed paraffin embedded tissue section was performed on a Leica Bond system using the standard protocol F. The section was pretreated using heat mediated antigen retrieval with Sodium Citrate buffer (pH 6, epitope retrieval solution 1) for 30 mins. The section was then incubated with Sigma antibody (Catalog: HPA007459, 1:100
(Table 4): 1 biopsy of Degos disease-like coronary vessel changes and 6 biopsies of Kohlmeier-Degos disease. In each of the cases, there was corroborative clinical evidence that supported the histologic diagnoses. Additional 14 control cases were examined comprising random skin biopsies of the lower extremity procured for small fiber neuropathy (2 cases), lichen planus-like infiltrates (3 cases), atypical hemolytic uremic syndrome where biopsies of normal skin were procured (2 cases), a bullous arthropod hypersensitivity reaction (1 case), scar (1 case), chemotherapy effect (1 case), angiolipoma (1 case), antiphospholipid 3
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
event with resultant cardiac arrest, from which she was resuscitated. She is followed annually with biopsies of normal skin, which have always shown some degree of expression of ROCK2 in the microvasculature of lesser intensity than at initial presentation (Table 4) (Fig. 3a–d). Given the typical lack of any staining for ROCK2 in normal skin, one could postulate that endothelial cell staining for ROCK2 in the context of normal skin is an abnormal finding that could potentially signify a systemic inflammatory state. In the 11 cases of scleroderma studied, including one case of sclerodermatomyositis and 1 case of morphea, the results were varied. Four cases showed moderate to strong expression of ROCK2 in endothelium: one case of morphea showed strong expression, while moderate expression was seen in one case of sclerodermatomyositis and two cases of systemic scleroderma. In 6 of the 10 cases there was weak expression within the endothelium. All cases were cutaneous biopsies of affected skin in patients with advanced systemic sclerosis. With the exception of the sclerodermatomyositis case, the staining was largely confined to the endothelium without any discernible immunoreactivity in the epidermis and amidst inflammatory cells (Fig. 4a–d). One case of systemic scleroderma showed no expression. The control biopsies that were devoid of inflammation were negative for ROCK2 expression, specifically in normal skin, scar tissue, cell poor urticaria, actinic keratosis, angiolipoma, a toxic irritant chemotherapy reaction, normal skin in the context of hemolytic-uremic syndrome, and small fiber neuropathy. However, in biopsies where there was some degree of inflammation such as lichenoid dermatitis and bullous arthropod bite hypersensitivity reaction, there was upregulation of staining for ROCK2 but not of the magnitude seen in inflammatory cell rich forms of lupus erythematosus and dermatomyositis. In biopsies of lichenoid inflammation, there was disparity between the minimal staining of the inflammatory cell infiltrate and the stronger signal in endothelium, defining a pattern once again at disparity with that observed in the setting of lupus erythematosus (Fig. 5a–d).
dilution) for 15 mins at room temperature and detected using an HRP conjugated compact polymer system. DAB was used as the chromogen. The section was then counterstained with haematoxylin and mounted with Leica Micromount. A qualitative and quantitative assessment of expression of Rho kinase (i.e. ROCK2 expression) was made. The distribution of staining was assessed, specifically in regards to epidermal and dermal staining including endothelial cell localization. Intensity of staining was graded out of three. This study was presented to the IRB of Weill Cornell Medicine and was considered exempt. 3. Results The biopsies of lupus erythematosus independent of the subtype showed significant endothelial cell staining that was comparable in intensity to that observed in the setting of dermatomyositis.(Table 1) In addition, there was ROCK2 expression in inflammatory cells and amidst epithelial structures, although less than in endothelial expression. The combined staining pattern in epithelial structures, inflammatory cells and/or endothelium was seen in 17 out of 24 cases (Fig. 1a–h). In 7 cases the staining pattern was confined to the endothelium represented by 1 case of DLE, 3 cases of SLE, and 3 cases of SCLE. In approximately half of the cases there was concurrent staining in epithelial structures, endothelium and inflammatory cells (Fig. 1a–h). Overall, extent of immunoreactivity was greater in biopsies of lupus erythematosus compared to dermatomyositis in regards to the various cell types highlighted by ROCK2, predominantly reflective of its enhanced expression in epithelial structures (i.e. epidermis and adnexae) and inflammatory cells as opposed to any significant quantitative difference in endothelial cell expression (Fig. 1a–d). The biopsies of dermatomyositis showed significant expression of ROCK2 within the endothelium. In the half of cases (i.e. 7 out of the 14 cases) ROCK2 expression was confined to the endothelium (Fig. 2b, c). In general, the biopsies were pauci-inflammatory. However, in four of the cases the inflammatory cells present stained positively for ROCK2; in one of the cases the staining pattern was weak while in the other three it was moderately intense. In 6 cases there was staining of the epidermis, but it was primarily localized to the lower third of the epidermis and or was weak (Fig. 2d). In only 3 cases was the staining intensity in the epidermis moderate. There was a gradation of staining with the greatest intensity in the superficial dermis (Table 2). In comparing dermatomyositis and lupus erythematosus, in general a greater extent of staining was noted within the epidermis and hair follicle in the setting of lupus erythematosus. A similar intensity of endothelial cell staining was observed in both diseases. The concurrent staining of epithelial structures, endothelium and inflammatory cells was in lupus erythematosus than dermatomyositis. Cases of dermatomyositis were pauci-inflammatory compared to lupus erythematosus; however, in cases of lymphocyte rich dermatomyositis the extent of staining in the inflammatory cells and in the epithelial structures was similar to lupus erythematosus. In both the skin and intestinal biopsies of patients with KohlmeierDegos disease, there was a very strong nuclear staining pattern within the endothelium of vessels throughout the samples in all cases examined, excluding a biopsy from one patient where there was no significant expression or only weak expression. A significant degree of staining was not observed amidst inflammatory cells and in the epidermis or hair follicle in most of the cases. However, in two cases there was a weak staining pattern observed in the epidermis and inflammatory cells, and in one of the two weak staining in stromal fibroblasts. The dominant staining pattern however was within the endothelium. In this regard the staining pattern resembles that of dermatomyositis. In a patient with a Kohlmeier Degos-like reaction involving the coronary vessels, a biopsy of normal skin showed a striking upregulation of staining in the endothelium at the time of the patient's acute presentation heralded by an occlusive coronary vascular
4. Discussion We examined the expression of ROCK2 in a variety of autoimmune/ autoinflammatory conditions affecting the skin. We compared this to ROCK2 expression in normal skin and other inflammatory dermatoses not reflective of autoimmune processes. There was extensive expression in the microvasculature of the skin in lupus erythematosus, dermatomyositis, Kohlemeir-Degos disease, and scleroderma. There were no significant qualitative or quantitative differences in ROCK2 expression between these conditions. In general, biopsies of lupus erythematous showed a greater degree of ROCK2 staining in the epidermis, hair follicle and within inflammatory cells as compared to other autoimmune diseases examined. The exception to this was one case of lymphocyterich dermatomyositis, where a similar pattern of upregulated ROCK2 expression occurred in epithelium, endothelium and perivascular inflammatory cells. ROCK2 expression is not unique to these connective tissue diseases and in general is upregulated in any biopsy showing inflammation. For this reason, it is unlikely to be a discriminatory stain; there are however some qualitative and quantitative staining differences, particularly prominence of staining in the endothelium and attenuated staining in epithelial structures or inflammatory cells. The pathophysiology underlying ROCK2 upregulation in autoimmune conditions and the significance of ROCK2 expression remain as areas for future inquiry. ROCK2 and the other membrane of the Rho kinase pathway, ROCK1, are two serine threonine kinases whose activity is controlled by the binding of activated RhoA [6]. RhoA, a member of the serine threonine kinase family, modulates several critical functions including inflammatory and cytoskeletal mechanisms such as cell motility, proliferation, and differentiation. ROCK1 and ROCK2 are downstream effector proteins that phosphorylate myosin light chains resulting in the formation of stress fibers, contributing to cellular adhesion. There is anatomic variation in the expression of 4
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
Fig. 2. ROCK2 expression in dermatomyositis. Dermatomyositis is the prototypic anti-endothelial cell antibody syndrome and hence microvascular changes are a conspicuous aspect of the cutaneous pathology. This particular image shows the striking compensatory vascular ectasia along with subepidermal vascular drop out that is observed in biopsies of dermatomyositis (a). In the majority of cases the staining for Rock2 was largely confined to the endothelium whereby the greatest extent of expression was observed in the superficial vascular plexus, also defining the microanatomic site associated with the greatest extent of vascular injury (b, c). In a few cases however, there was significant staining for Rock2 in the lower third of the epidermis and the typical strong endothelial cell signature (d).
expression of ROCK2 as a marker of endothelial cell dysfunction or a contributing element of endothelial cell dysfunction. One potential adverse effect on endothelium is a down-regulation of nitric oxide expression attributable to enhanced ROCK2 expression. The effects of thrombin on the downregulation of nitrous oxide is also intimately linked with the Rho kinase pathway [10]. Overexpression of ROCK1 and ROCK2 results in impairment of intercellular adhesion between endothelial cells. Although basal levels of ROCK activity are critical for vascular endothelial cadherin and endothelial based adhesion, overexpression can impair endothelial adhesion important for normal vascular integrity. Endothelial adhesion is mediated by vascular cadherin and the intracellular actin-based cytoskeleton which in turn is critically linked with Rho kinase activity. One of the important cardioprotective effects of statins relates to inhibition of ROCK2 expression [11]. Rho kinase elevation in the studied autoimmune diseases may be due to a single nucleotide polymorphism. It could be analogous to the enhanced expression of type I interferon seen in patients with systemic lupus erythematosus, who have interferon alpha gene polymorphisms. Significant association was found between with systemic scleroderma and a polymorphism in the ROCK1 gene (rs35996865), a polymorphism in ROCK2 gene (rs10178332), a polymorphism in RhoA gene
ROCK1 and ROCK2, the former being found primarily in liver, lung, spleen, kidney, and intestine and the latter identified in the brain and heart [7]. ROCK2 expression in lymphocytes and epithelial structures is a consistent feature in skin biopsies of patients with lupus erythematosus. T-cell dysfunction in lupus erythematosus is reflective of enhanced TH1 activity with high levels of interleukin 17 and 21. This TH1 activity is in turn upregulated by interferon regulatory factor 4 (IRF4) via phosphorylation by ROCK2 [8]. Not surprisingly, ROCK inhibition is either non-selectively or selectively associated with a reduction in interleukin 17 and interleukin 21. In one study, the authors gave inhibitors of Rho/ kinase which resulted in significant reductions in a number of proinflammatory cytokines including IL-6, IL-1, and TNF-alpha, while increasing IL-10 [9]. Another mechanism by which excessive upregulation of ROCK2 results in a proinflammatory state is via NF-κB activation. The enhanced expression within the endothelium in dermatomyositis, lupus erythematosus and scleroderma could be pathogenetically important in light of microvascular complications intrinsic to these conditions. The striking degree of endothelial cell staining in biopsies of dermatomyositis and lupus erythematosus suggests a role for enhanced 5
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
Fig. 3. ROCK2 expression in Kohlemeier-Degos disease. The biopsy is one procured from the classic porcelain telangietactic plaque that defines classic KohlemeierDegos disease showing a zone of striking epidermal attenuation accompanied by marked a vascular hyalinizing fibrosis with concomitant striking vascular drop out. (a) Corroborative of the diagnosis of Kohlemeier-Degos disease is prominent C5b-9 deposition in vessels (b) in concert with significant upregulation of MXA in the epidermis, amidst inflammatory cells and within the endothelium (c). There is an extensive prominent endothelial cell signature for ROCK2 (d–f). However, a significant degree of immunoreactivity is not observed amidst the lymphocytes (f).
Fig. 4. ROCK2 expression in Systemic Scleroderma. In this case of systemic scleroderma, there is deep seated hyalinizing fibrosis associated with vascular drop out (a). There is acquisition of smooth muscle actin staining amidst the fibroblasts compatible with a procollagen phenotype (b). There is significant endothelial cell staining; the staining intensity is +2/3. ROCK2 staining is not observed in the epidermis, adnexal structures, or inflammatory cells (c).
scleroderma, another disease where upregulation of the Rho kinase pathway may be pathogenetically relevant. Intramural lipophage accumulation, a characteristic finding, could also be a ROCK2 effect since decreased foam cell formation and increase in cholesterol efflux has been demonstrated in ROCK2 deficient mice [15,16]. Although Kohlmeier-Degos disease is not a primary autoimmune disease, the extent of vasocentric lymphocytic infiltration along with the upregulated type I interferon profile defines a pathologic reaction pattern reminiscent of vascular reaction in primary autoimmune vasculitic syndromes such as dermatomyositis and lupus erythematosus. Moreover, Degos disease-like features are described in both lupus erythematosus and dermatomyositis. Thrombin increases the expression of the monocyte chemoattractant protein 1 of endothelial cells likely through Rho kinase. This may provide an explanation for the mononuclear cell infiltrates seen in the primary vascular lesion of KohlmeierDegos disease [17]. In summary, lipophage accumulation, smooth muscle hyperplasia, and endothelial cell dysfunction are all characteristic features of the vascular changes seen in the setting of Degos disease, dermatomyositis,
(rs2177268) and two polymorphisms in RhoC gene (rs11102522 and rs11538960). A similar gene polymorphism has been implicated in Behcet's disease [12]. Mechanisms underlying Rho kinase activation in Kohlemeier-Degos disease are likely complicated and multifactorial in etiology. Both an upregulated type I interferon microenvironment and C5b-9 deposition in vessels are features of the microvascular and arteriopathic changes seen in Kohlmeier-Degos disease [13]. Enhanced hyaluronic acid deposition in the intima (a probable interferon alpha effect) and C5a as a component of complement activation are known inducers of the Rho kinase pathway. The upregulation of Rho kinase may in turn contribute at least in part to the distinctive pathology observed in this rare and at times fatal primary vascular disorder. One of the critical components of the vascular pathology is a striking neointimal proliferative lesion that affects small and medium-sized arteries of the gastrointestinal tract. Other organ systems can show the same obliterative neointimal arteriopathy. Rho kinase may alter the properties of intramural smooth muscle cells including promotion of myofibroblastic proliferation in the intima [14]. Similar obliterative arteriopathic changes occur in 6
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
Fig. 5. ROCK2 expression in control biopsies. The biopsy is one of normal skin procured to assess for vascular C5b-9 deposition in the setting of atypical hemolytic uremic syndrome. Significant vascular deposition is not seen (a). In this case of cell poor urticaria only a weak endothelial cell staining pattern is observed (b). In this case of lichen planus pigmentosus there is no significant staining for ROCK2 (c). A biopsy was performed showing a lichenoid dermatitis thought to be clinically and histologically compatible with a lichenoid keratosis given its solitary nature. There is weak staining of endothelium for ROCK2 while the lymphocytic infiltrate is without significant staining (d).
scleroderma and lupus erythematosus. All of these conditions can be associated with a significant degree of inflammation, oftentimes linked with enhanced interferon alpha expression. Both the inflammatory and arteriopathic changes could in part be based on excessive Rho kinase activity. The Rho kinase pathway activation may also be involved in the recruitment of inflammatory cells in other forms of cutaneous inflammation given its frequent upregulation in any state of cutaneous inflammation.
org/10.1309/AJCP66QIMFARLZKI. [6] Rozo C, Chinenov Y, Maharaj RK, et al. Targeting the RhoA-ROCK pathway to reverse T-cell dysfunction in SLE. Ann Rheum Dis 2017;76(4):740–7. https://doi.org/ 10.1136/annrheumdis-2016-209850. [7] Hartmann S, Ridley AJ, Lutz S. The function of rho-associated kinases ROCK1 and ROCK2 in the pathogenesis of cardiovascular disease. Front Pharmacol 2015;6:276. https://doi.org/10.3389/fphar.2015.00276. [8] Isgro J, Gupta S, Jacek E, et al. Enhanced rho-associated protein kinase activation in patients with systemic lupus erythematosus. Arthritis Rheum 2013;65(6):1592–602. https://doi.org/10.1002/art.37934. [9] Wang Y, Lu Y, Chai J, et al. Y-27632, a rho-associated protein kinase inhibitor. Inhibits Systemic Lupus Erythematosus Biomed Pharmacother 2017;88:359–66. https://doi.org/10.1016/j.biopha.2017.01.069. [10] Tratsiakovich Y, Kiss A, Gonon AT, Yang J, Sjöquist P-O, Pernow J. Inhibition of rho kinase protects from ischaemia–reperfusion injury via regulation of arginase activity and nitric oxide synthase in type 1 diabetes. Diab Vasc Dis Res 2017;14(3):236–45. https://doi.org/10.1177/1479164116687935. [11] Yi R, Xiao-Ping G, Hui L. Atorvastatin prevents angiotensin II-induced high permeability of human arterial endothelial cell monolayers via ROCK signaling pathway. Biochem Biophys Res Commun 2015;459(1):94–9. https://doi.org/10. 1016/j.bbrc.2015.02.076. [12] Pehlivan Y, Yolbas S, Cetin GY, et al. Investigation of the association between rho/ rho-kinase gene polymorphisms and systemic sclerosis. Rheumatol Int 2016;36(3):421–7. https://doi.org/10.1007/s00296-015-3400-4. [13] Singleton PA, Dudek SM, Ma S-F, Garcia JGN. Transactivation of Sphingosine 1phosphate receptors is essential for vascular barrier regulation. J Biol Chem 2006;281(45):34381–93. [14] Shimokawa H, Sunamura S, Satoh K. RhoA/rho-kinase in the cardiovascular system. Circ Res 2016;118(2):352–66. https://doi.org/10.1161/CIRCRESAHA.115.306532. [15] Pellegrino PR, Schiller AM, Haack KK, Zucker IH. Central angiotensin-II increases blood pressure and sympathetic outflow via rho kinase activation in conscious rabbits. Hypertension. 2016;68(5):1271–80. https://doi.org/10.1161/ HYPERTENSIONAHA.116.07792. [16] Liu L-J, Yao F-J, Lu G-H, et al. The role of the rho/ROCK pathway in Ang II and TGFβ1-induced atrial remodeling. PLoS One 2016;11(9):e0161625https://doi.org/10. 1371/journal.pone.0161625. [17] Kawanami D, Matoba K, Kanazawa Y, Ishizawa S, Yokota T, Utsunomiya K. Thrombin induces MCP-1 expression through rho-kinase and subsequent p38MAPK/NF-κB signaling pathway activation in vascular endothelial cells. Biochem Biophys Res Commun 2011;411(4):798–803. https://doi.org/10.1016/j. bbrc.2011.07.031.
Sources of funding None. Declaration of competing interest None. References [1] Nakagawa O, Fujisawa K, Ishizaki T, Saito Y, Nakao K, Narumiya S. ROCK-I and ROCK-II, two isoforms of rho-associated coiled-coil forming protein serine/threonine kinase in mice. FEBS Lett 1996;392(2):189–93. https://doi.org/10.1016/00145793(96)00811-3. [2] Sawada N, Liao JK. Rho/Rho-associated coiled-coil forming kinase pathway as therapeutic targets for statins in atherosclerosis. Antioxid Redox Signal 2014;20(8):1251–67. https://doi.org/10.1089/ars.2013.5524. [3] Zhang X, Zhang T, Gao F, et al. Fasudil, a rho-kinase inhibitor, prevents intimamedia thickening in a partially ligated carotid artery mouse model: effects of fasudil in flow-induced vascular remodeling. Mol Med Rep 2015;12(5):7317–25. https:// doi.org/10.3892/mmr.2015.4409. [4] Yuan T-Y, Yan Y, Wu Y-J, et al. Vasodilatory effect of a novel rho-kinase inhibitor, DL0805-2, on the rat mesenteric artery and its potential mechanisms. Cardiovasc Drugs Ther 2014;28(5):415–24. https://doi.org/10.1007/s10557-014-6544-7. [5] Magro CM, Poe JC, Kim C, et al. Degos disease: a C5b-9/interferon-α–mediated endotheliopathy syndrome. Am J Clin Pathol 2011;135(4):599–610. https://doi.
7
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology journal homepage: www.elsevier.com/locate/anndiagpath
Enhanced cutaneous Rock2 expression as a marker of Rho Kinase pathway activation in autoimmune disease and Kohlemeier-Degos disease
T
⁎
Cynthia Magro , Ziv Schwartz, Jad Saab, Amin Hedayat Weill Cornell Medicine, New York, United States of America
A R T I C LE I N FO
A B S T R A C T
Keywords: ROC vK2 Rho Kinase Scleroderma Lupus Dermatomyositis Degos Autoimmune
The small guanosine triphosphatase Rho and its target Rho kinase are involved in a heterogeneous spectrum of cellular activities, many of which are integral to cytoskeletal organization. Furthermore, the Rho kinases result in NF kappa beta activation and hence the induction of various pro-inflammatory cytokines including TNF-alpha, IL-1B and IL-6. ROCK2 is a downstream protein, whose expression is indicative of Rho Kinase activation. Given the diverse effects of Rho-kinase, including a potentially critical role in augmenting inflammation, ROCK2 expression was examined in biopsies of select autoimmune connective tissue diseases as compared to control diagnoses. Select cases of lupus erythematosus, dermatomyositis, autoimmune sclerodermoid disorders and Kohlmeier-Degos disease (a distinctive vasculopathy that occurs in the other aforesaid conditions but also as a forme fruste microvascular and arteriopathic syndrome) were studied. Control biopsies included normal skin and cutaneous inflammatory conditions unrelated to collagen vascular disease/autoimmune disease. We found ROCK2 expression significantly increased in biopsies of lupus erythematosus, dermatomyositis, scleroderma and Kohlmeier-Degos disease. A pattern emerged of consistent marked ROCK2 upregulation in endothelium and variable expression in inflammatory cells and epithelium. While expression was undetectable in normal skin, it was found in inflamed skin unrelated to specific autoimmune disease. The staining pattern could approach that seen in study group cases but was less pronounced and preferentially upregulated in the endothelium, with a lesser extent of staining in the epidermis and inflammatory cells. Rho kinase is a driving factor in diverse cutaneous diseases especially autoimmune disease and Kohlmeier-Degos disease. This significantly upregulated pathway defines a potential target for biologic therapy.
1. Introduction The small guanosine triphosphatase (GTP) Rho and its target Rho kinase are involved in a heterogeneous spectrum of cellular activities many of which are integral to cytoskeletal organization. The Rho GTPase, RhoA, demonstrates a specific conformational alteration, which following binding to GTP leads to intracellular protein activation. P160 ROCK is a protein serine/threonine kinase, whereby it has a kinase domain associated with Rho-binding sequences. There are two related isoforms referred to as ROCK1 and ROCK2, both being activated by Rho-kinase. When considering the varied functions of the ROCK proteins, cell migration, adhesion, cellular proliferation, apoptosis, and cellular differentiation are potentially modulated by these proteins. Both ROCK1 and ROCK2 are considered critical molecular switches integral to actin cytoskeleton organization, apoptosis, reactive oxygen species, cell migration and adhesion. They can promote a pro-inflammatory milieu with upregulation of various cytokines including
⁎
Corresponding author. E-mail address: [email protected] (C. Magro).
https://doi.org/10.1016/j.anndiagpath.2019.151414
1092-9134/ © 2019 Elsevier Inc. All rights reserved.
TNF-alpha, IL-1B and IL-6 [1]. In addition to its role in inflammation, ROCK2 has been linked with vascular remodeling. Vascular smooth muscle motility and smooth muscle contraction occurs through actin activation of myosin ATPase via ROCK2. Arterial wall smooth muscle migration, which is highly dependent on Rho kinase activation, is a critical impetus to neointimal proliferation seen in diverse settings ranging from atherosclerosis to accelerated vascular rejection post-transplantation to the fibrosing intimal vascular lesions that are observed in the setting of lupus erythematosus and Kohlmeier-Degos disease [2-4]. The Rho pathway has not been explored in the cutaneous lesions of select autoimmune conditions where inflammation and vascular disease are integral players in the clinical phenotype. Dermatomyositis, scleroderma, lupus erythematosus and Degos disease are among these disorders. Degos disease is of particular interest since it can be seen either as a forme fruste of vascular disease and or seen in the setting of select autoimmune connective disease syndromes such as dermatomyositis
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
Table 1 Lupus: ROCK 2 immunoreactivity among biopsies of lupus patients. Case
Age/sex
Diagnosis
Microscopic description
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
32/F 3 mths/F 51/F 58/F 26/F 54/M 2/F 48/F 65/F 25/F 64/M 64/F 67/M 64/F 45/F 48/M 48/M 36/F 58/F 30/F 44/F 43/M 55/F 54/F
Rowell's syndrome Neonatal lupus DLE DLE Early DLE in the setting of SLE SLE SLE SLE SLE SLE SLE/SCLE SCLE SCLE SCLE SCLE SCLE SCLE SCLE SCLE SCLE SCLE Proton pump inhibitor therapy induced SCLE Lupus profundus with overlying changes of SCLE Lupus Profundus
+1–2/3 EC; +1/3 epidermis +3/3 EC; +1/3 epidermis; +1/3 inflammatory cells +3/3 EC; +1/3 basal layer of epidermis; +1/3 inflammatory cells +2/3 EC +3/3 EC; 2/3 lower portion of epidermis; +1/3 inflammatory cells +1/3 EC +3/3 EC +3/3 EC; +2/3 epidermis; +1/3 inflammatory cells +2/3 EC +3/3 EC; +1/3 epidermis; 0/3 inflammatory cells +3/3 EC; +2/3 epidermis; +1/3 inflammatory cells +3/3 EC +3/3 EC; +2/3 epidermis; +2/3 inflammatory cells +3/3 EC +3/3 EC; +3/3 epidermis +3/3 EC; +2/3 epidermis; +2/3 inflammatory cells +3/3 EC; +2/3 epidermis; +1/3 inflammatory cells +3/3 EC; +2/3 epidermis, +2/3 inflammatory cells; +2/3 follicle +3/3 EC; +3/3 perivascular inflammatory cells +2/3 EC; +1/3 epidermis +3/3 EC +3/3 EC; +2/3 epidermis +3/3 EC; +2/3 epidermis; +3/3 inflammatory cells +3/3 EC; +2/3 inflammatory cells
Mths – Months, F – Female, M- Male, SCLE – Subacute Cutaneous Lupus Erythematosus, DLE - Discoid Lupus Erythematosus, SLE - Systemic lupus erythematosus; EC = endothelial cell staining.
[5]. We assessed ROCK2 expression in biopsies of lupus erythematosus, dermatomyositis, autoimmune sclerodermoid conditions and Kohlmeier-Degos disease. We also examined the expression of Rho kinase in normal skin, in the setting of other primary cutaneous inflammatory conditions, and in certain systemic diseases such as atypical hemolytic uremic syndrome where clinically normal skin was biopsied.
Table 3 Scleroderma: ROCK 2 immunoreactivity among biopsies of scleroderma patients.
2. Materials and methods Select cases diagnosed as lupus erythematosus, dermatomyositis, autoimmune sclerodermoid disorders were examined. Lupus erythematosus was represented by 24 cases (Table 1): 3 cases of discoid lupus erythematosus (DLE), 13 cases of subacute cutaneous lupus erythematosus (SCLE), 2 cases of lupus profundus, 7 cases of systemic lupus erythematosus (SLE), 1 case of Rowell's syndrome and 1 case of neonatal lupus erythematosus. Dermatomyositis was represented by 14 cases.(Table 2) Scleroderma was represented by 11 cases (Table 3): 9 cases of systemic scleroderma, 1 case of sclerodermatomyositis, and 1 case of morphea. Kohlmeier-Degos disease was represented by 7 cases
Case
Age/Sex
Diagnosis
Microscopic description
1 2 3 4 5 6 7 8 9 10
41/F 69/M NA NA NA NA NA NA NA 60/F
SSCL SSCL (Anti-Ro positive) SSCL SSCL SSCL SSCL SSCL SSCL SSCL Sclerodermatomyositis
11
33/F
Morphea
0/3 EC +1/3 EC +1/3 EC +1/3 EC +1/3 EC +1/3 EC +2/3 EC +1/3 EC +2/3 EC +2/3 EC; +1/3 basal layer of epidermis +3/3 EC
F – female; M – male; NA – not available; SSCL – systemic scleroderma; EC endothelial cell staining.
Table 2 Dermatomyositis: ROCK 2 immunoreactivity among biopsies of dermatomyositis patients. Case
Age/Sex
Diagnosis
Microscopic description
1 2 3 4 5 6 7 8
56/M 52/F 18/M 50/F 70/F 48/F 58/M 77/F
DM DM DM DM DM DM DM DM
9 10 11 12 13 14
76/M 74/F 26/F 48/M 57/M 46/M
DM End-Stage, Gottron's papule, DM Quiescent phase, DM Gottron's papule, DM Lymphocyte-rich DM Anti-Mi-2 Dermatomyositis
+3/3 EC; +2/3 epidermis; +2/3 inflammatory cells +3/3 EC; +1/3 inflammatory cells +2/3 EC +3/3 EC +1/3 EC; +1/3 basal layer of epidermis +2/3 EC +3/3 EC +3/3 EC; +1/3 epidermis (basilar and parabasilar keratinocytes); greatest staining superficially within the dermis; +2/3 staining of fibroblasts +2/3 EC +3/3 EC; +1/3 epidermis +2/3 EC +1–2/3 EC (most apparent superficially) +3/3 EC; +2/3 epidermis; +2/3 inflammatory cells +3/3 EC; +2/3 epidermis; +2/3 inflammatory cells
F – female; M – male; DM – dermatomyositis; EC - endothelial cell staining. 2
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
Table 4 Kohlmeier Degos: ROCK 2 immunoreactivity among biopsies of Kohlmeier Degos patients. Case
Age/Gender
Diagnosis
Microscopic findings
1 2 3 4 5 6 7
31/F 19/M 7/F NA 43/M 43/F 57/M
Degos disease like changes in the coronary vessels Multiorgan Kohlmeier Degos Disease Multiorgan Kohlmeier Degos Disease Multiorgan Kohlmeier Degos Disease Kohlmeier Degos disease Late Stage Kohlmeier Degos disease Kohlmeier Degos Disease
+3/3 EC before treatment; +1/3 EC after treatment +3/3 EC +2/3 EC +3/3 EC; weak staining of epidermis, inflammatory cells, and stromal cells 0/3 EC +2/3 EC; +0–1/3 epidermis; +0–1/3 inflammatory cells +2/3 EC; 0/3 epidermis; 0/3 inflammatory cells
F – female; M – male; NA – not available; EC - endothelial cell staining. Table 5 Controls: ROCK 2 immunoreactivity among biopsies of control subjects. Case
Age/Gender
Diagnosis
Microscopic findings
1 2 3 4 5 6 7 8 9 10
42/M 33/F 65/F 66/F 50/F 24/F 52/M 57/F 48/F 71/M
+2/EC; +0–1/3 inflammatory cells +1/3 inflammatory cells +1/3 EC; +1/3 inflammatory cells +2/3 EC; +0–1/3 inflammatory cells 0/3 EC +0–1/3 EC +3/3 EC 0/3 EC 0/3 EC +2/3 EC
11 12 13 14
43/M 46/M 91/F 59/F
Bullous eosinophilic cellulitis-like arthropod bite hypersensitivity reaction Lichenoid dermatitis Lichenoid dermatitis Lichenoid dermatitis Angiolipoma Normal skin in the setting of small fiber neuropathy Antiphospholipid antibody syndrome Normal skin biopsy in the setting of atypical Hemolytic uremic syndrome (HUS) Small fiber neuropathy, normal skin biopsy sampled Superficial fibrosing dermopathy with chronic microvascular changes and irritant epidermal changes suggestive of toxic chemotherapy effect Atypical HUS, biopsy of normal skin. Normal skin Actinic keratosis Scar
0/3 0/3 0/3 0/3
EC EC EC EC
F – female; M – male; EC - endothelial cell staining; NOS – not otherwise specified.
Fig. 1. ROCK2 expression in lupus. This biopsy shows a mild lymphocytic mediated interface dermatitis involving the acrosyringium and hair follicle compatible with early discoid lupus erythematosus in the setting of systemic lupus erythematosus (a). The Rock2 studies show significant upregulation in the epithelium, amidst inflammatory cells and in endothelium (b). In this additional case of discoid lupus erythematosus the lower power image captures the extent of staining for Rock2 in the epidermis, inflammatory cells and in endothelium (c). A higher power image is provided as well (d). In this case of subacute cutaneous lupus erythematosus, the typical cellular band like lymphocytic infiltrate is seen (e). Rock2 shows a significantly upregulated pattern of staining involving the lower portion of the epidermis, endothelium and many inflammatory cells (f). This additional case of subacute cutaneous lupus erythematosus demonstrates the architectural disposition of the infiltrate being superficial and band-like. Illustrated is MXA, the surrogate type I interferon marker, demonstrating a marked upregulation in the type I interferon signature (g). Rock2 is markedly upregulated in the epidermis, amidst inflammatory cells and in the endothelium (h).
antibody syndrome (1 case), actinic keratosis (1 case), and normal skin (1 case).(Table 5). In each case, 4-μm sections were stained with the ROCK2 isoform for Rho kinase. The image of immunohistochemical ROCK2 staining in formalin fixed paraffin embedded tissue section was performed on a Leica Bond system using the standard protocol F. The section was pretreated using heat mediated antigen retrieval with Sodium Citrate buffer (pH 6, epitope retrieval solution 1) for 30 mins. The section was then incubated with Sigma antibody (Catalog: HPA007459, 1:100
(Table 4): 1 biopsy of Degos disease-like coronary vessel changes and 6 biopsies of Kohlmeier-Degos disease. In each of the cases, there was corroborative clinical evidence that supported the histologic diagnoses. Additional 14 control cases were examined comprising random skin biopsies of the lower extremity procured for small fiber neuropathy (2 cases), lichen planus-like infiltrates (3 cases), atypical hemolytic uremic syndrome where biopsies of normal skin were procured (2 cases), a bullous arthropod hypersensitivity reaction (1 case), scar (1 case), chemotherapy effect (1 case), angiolipoma (1 case), antiphospholipid 3
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
event with resultant cardiac arrest, from which she was resuscitated. She is followed annually with biopsies of normal skin, which have always shown some degree of expression of ROCK2 in the microvasculature of lesser intensity than at initial presentation (Table 4) (Fig. 3a–d). Given the typical lack of any staining for ROCK2 in normal skin, one could postulate that endothelial cell staining for ROCK2 in the context of normal skin is an abnormal finding that could potentially signify a systemic inflammatory state. In the 11 cases of scleroderma studied, including one case of sclerodermatomyositis and 1 case of morphea, the results were varied. Four cases showed moderate to strong expression of ROCK2 in endothelium: one case of morphea showed strong expression, while moderate expression was seen in one case of sclerodermatomyositis and two cases of systemic scleroderma. In 6 of the 10 cases there was weak expression within the endothelium. All cases were cutaneous biopsies of affected skin in patients with advanced systemic sclerosis. With the exception of the sclerodermatomyositis case, the staining was largely confined to the endothelium without any discernible immunoreactivity in the epidermis and amidst inflammatory cells (Fig. 4a–d). One case of systemic scleroderma showed no expression. The control biopsies that were devoid of inflammation were negative for ROCK2 expression, specifically in normal skin, scar tissue, cell poor urticaria, actinic keratosis, angiolipoma, a toxic irritant chemotherapy reaction, normal skin in the context of hemolytic-uremic syndrome, and small fiber neuropathy. However, in biopsies where there was some degree of inflammation such as lichenoid dermatitis and bullous arthropod bite hypersensitivity reaction, there was upregulation of staining for ROCK2 but not of the magnitude seen in inflammatory cell rich forms of lupus erythematosus and dermatomyositis. In biopsies of lichenoid inflammation, there was disparity between the minimal staining of the inflammatory cell infiltrate and the stronger signal in endothelium, defining a pattern once again at disparity with that observed in the setting of lupus erythematosus (Fig. 5a–d).
dilution) for 15 mins at room temperature and detected using an HRP conjugated compact polymer system. DAB was used as the chromogen. The section was then counterstained with haematoxylin and mounted with Leica Micromount. A qualitative and quantitative assessment of expression of Rho kinase (i.e. ROCK2 expression) was made. The distribution of staining was assessed, specifically in regards to epidermal and dermal staining including endothelial cell localization. Intensity of staining was graded out of three. This study was presented to the IRB of Weill Cornell Medicine and was considered exempt. 3. Results The biopsies of lupus erythematosus independent of the subtype showed significant endothelial cell staining that was comparable in intensity to that observed in the setting of dermatomyositis.(Table 1) In addition, there was ROCK2 expression in inflammatory cells and amidst epithelial structures, although less than in endothelial expression. The combined staining pattern in epithelial structures, inflammatory cells and/or endothelium was seen in 17 out of 24 cases (Fig. 1a–h). In 7 cases the staining pattern was confined to the endothelium represented by 1 case of DLE, 3 cases of SLE, and 3 cases of SCLE. In approximately half of the cases there was concurrent staining in epithelial structures, endothelium and inflammatory cells (Fig. 1a–h). Overall, extent of immunoreactivity was greater in biopsies of lupus erythematosus compared to dermatomyositis in regards to the various cell types highlighted by ROCK2, predominantly reflective of its enhanced expression in epithelial structures (i.e. epidermis and adnexae) and inflammatory cells as opposed to any significant quantitative difference in endothelial cell expression (Fig. 1a–d). The biopsies of dermatomyositis showed significant expression of ROCK2 within the endothelium. In the half of cases (i.e. 7 out of the 14 cases) ROCK2 expression was confined to the endothelium (Fig. 2b, c). In general, the biopsies were pauci-inflammatory. However, in four of the cases the inflammatory cells present stained positively for ROCK2; in one of the cases the staining pattern was weak while in the other three it was moderately intense. In 6 cases there was staining of the epidermis, but it was primarily localized to the lower third of the epidermis and or was weak (Fig. 2d). In only 3 cases was the staining intensity in the epidermis moderate. There was a gradation of staining with the greatest intensity in the superficial dermis (Table 2). In comparing dermatomyositis and lupus erythematosus, in general a greater extent of staining was noted within the epidermis and hair follicle in the setting of lupus erythematosus. A similar intensity of endothelial cell staining was observed in both diseases. The concurrent staining of epithelial structures, endothelium and inflammatory cells was in lupus erythematosus than dermatomyositis. Cases of dermatomyositis were pauci-inflammatory compared to lupus erythematosus; however, in cases of lymphocyte rich dermatomyositis the extent of staining in the inflammatory cells and in the epithelial structures was similar to lupus erythematosus. In both the skin and intestinal biopsies of patients with KohlmeierDegos disease, there was a very strong nuclear staining pattern within the endothelium of vessels throughout the samples in all cases examined, excluding a biopsy from one patient where there was no significant expression or only weak expression. A significant degree of staining was not observed amidst inflammatory cells and in the epidermis or hair follicle in most of the cases. However, in two cases there was a weak staining pattern observed in the epidermis and inflammatory cells, and in one of the two weak staining in stromal fibroblasts. The dominant staining pattern however was within the endothelium. In this regard the staining pattern resembles that of dermatomyositis. In a patient with a Kohlmeier Degos-like reaction involving the coronary vessels, a biopsy of normal skin showed a striking upregulation of staining in the endothelium at the time of the patient's acute presentation heralded by an occlusive coronary vascular
4. Discussion We examined the expression of ROCK2 in a variety of autoimmune/ autoinflammatory conditions affecting the skin. We compared this to ROCK2 expression in normal skin and other inflammatory dermatoses not reflective of autoimmune processes. There was extensive expression in the microvasculature of the skin in lupus erythematosus, dermatomyositis, Kohlemeir-Degos disease, and scleroderma. There were no significant qualitative or quantitative differences in ROCK2 expression between these conditions. In general, biopsies of lupus erythematous showed a greater degree of ROCK2 staining in the epidermis, hair follicle and within inflammatory cells as compared to other autoimmune diseases examined. The exception to this was one case of lymphocyterich dermatomyositis, where a similar pattern of upregulated ROCK2 expression occurred in epithelium, endothelium and perivascular inflammatory cells. ROCK2 expression is not unique to these connective tissue diseases and in general is upregulated in any biopsy showing inflammation. For this reason, it is unlikely to be a discriminatory stain; there are however some qualitative and quantitative staining differences, particularly prominence of staining in the endothelium and attenuated staining in epithelial structures or inflammatory cells. The pathophysiology underlying ROCK2 upregulation in autoimmune conditions and the significance of ROCK2 expression remain as areas for future inquiry. ROCK2 and the other membrane of the Rho kinase pathway, ROCK1, are two serine threonine kinases whose activity is controlled by the binding of activated RhoA [6]. RhoA, a member of the serine threonine kinase family, modulates several critical functions including inflammatory and cytoskeletal mechanisms such as cell motility, proliferation, and differentiation. ROCK1 and ROCK2 are downstream effector proteins that phosphorylate myosin light chains resulting in the formation of stress fibers, contributing to cellular adhesion. There is anatomic variation in the expression of 4
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
Fig. 2. ROCK2 expression in dermatomyositis. Dermatomyositis is the prototypic anti-endothelial cell antibody syndrome and hence microvascular changes are a conspicuous aspect of the cutaneous pathology. This particular image shows the striking compensatory vascular ectasia along with subepidermal vascular drop out that is observed in biopsies of dermatomyositis (a). In the majority of cases the staining for Rock2 was largely confined to the endothelium whereby the greatest extent of expression was observed in the superficial vascular plexus, also defining the microanatomic site associated with the greatest extent of vascular injury (b, c). In a few cases however, there was significant staining for Rock2 in the lower third of the epidermis and the typical strong endothelial cell signature (d).
expression of ROCK2 as a marker of endothelial cell dysfunction or a contributing element of endothelial cell dysfunction. One potential adverse effect on endothelium is a down-regulation of nitric oxide expression attributable to enhanced ROCK2 expression. The effects of thrombin on the downregulation of nitrous oxide is also intimately linked with the Rho kinase pathway [10]. Overexpression of ROCK1 and ROCK2 results in impairment of intercellular adhesion between endothelial cells. Although basal levels of ROCK activity are critical for vascular endothelial cadherin and endothelial based adhesion, overexpression can impair endothelial adhesion important for normal vascular integrity. Endothelial adhesion is mediated by vascular cadherin and the intracellular actin-based cytoskeleton which in turn is critically linked with Rho kinase activity. One of the important cardioprotective effects of statins relates to inhibition of ROCK2 expression [11]. Rho kinase elevation in the studied autoimmune diseases may be due to a single nucleotide polymorphism. It could be analogous to the enhanced expression of type I interferon seen in patients with systemic lupus erythematosus, who have interferon alpha gene polymorphisms. Significant association was found between with systemic scleroderma and a polymorphism in the ROCK1 gene (rs35996865), a polymorphism in ROCK2 gene (rs10178332), a polymorphism in RhoA gene
ROCK1 and ROCK2, the former being found primarily in liver, lung, spleen, kidney, and intestine and the latter identified in the brain and heart [7]. ROCK2 expression in lymphocytes and epithelial structures is a consistent feature in skin biopsies of patients with lupus erythematosus. T-cell dysfunction in lupus erythematosus is reflective of enhanced TH1 activity with high levels of interleukin 17 and 21. This TH1 activity is in turn upregulated by interferon regulatory factor 4 (IRF4) via phosphorylation by ROCK2 [8]. Not surprisingly, ROCK inhibition is either non-selectively or selectively associated with a reduction in interleukin 17 and interleukin 21. In one study, the authors gave inhibitors of Rho/ kinase which resulted in significant reductions in a number of proinflammatory cytokines including IL-6, IL-1, and TNF-alpha, while increasing IL-10 [9]. Another mechanism by which excessive upregulation of ROCK2 results in a proinflammatory state is via NF-κB activation. The enhanced expression within the endothelium in dermatomyositis, lupus erythematosus and scleroderma could be pathogenetically important in light of microvascular complications intrinsic to these conditions. The striking degree of endothelial cell staining in biopsies of dermatomyositis and lupus erythematosus suggests a role for enhanced 5
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
Fig. 3. ROCK2 expression in Kohlemeier-Degos disease. The biopsy is one procured from the classic porcelain telangietactic plaque that defines classic KohlemeierDegos disease showing a zone of striking epidermal attenuation accompanied by marked a vascular hyalinizing fibrosis with concomitant striking vascular drop out. (a) Corroborative of the diagnosis of Kohlemeier-Degos disease is prominent C5b-9 deposition in vessels (b) in concert with significant upregulation of MXA in the epidermis, amidst inflammatory cells and within the endothelium (c). There is an extensive prominent endothelial cell signature for ROCK2 (d–f). However, a significant degree of immunoreactivity is not observed amidst the lymphocytes (f).
Fig. 4. ROCK2 expression in Systemic Scleroderma. In this case of systemic scleroderma, there is deep seated hyalinizing fibrosis associated with vascular drop out (a). There is acquisition of smooth muscle actin staining amidst the fibroblasts compatible with a procollagen phenotype (b). There is significant endothelial cell staining; the staining intensity is +2/3. ROCK2 staining is not observed in the epidermis, adnexal structures, or inflammatory cells (c).
scleroderma, another disease where upregulation of the Rho kinase pathway may be pathogenetically relevant. Intramural lipophage accumulation, a characteristic finding, could also be a ROCK2 effect since decreased foam cell formation and increase in cholesterol efflux has been demonstrated in ROCK2 deficient mice [15,16]. Although Kohlmeier-Degos disease is not a primary autoimmune disease, the extent of vasocentric lymphocytic infiltration along with the upregulated type I interferon profile defines a pathologic reaction pattern reminiscent of vascular reaction in primary autoimmune vasculitic syndromes such as dermatomyositis and lupus erythematosus. Moreover, Degos disease-like features are described in both lupus erythematosus and dermatomyositis. Thrombin increases the expression of the monocyte chemoattractant protein 1 of endothelial cells likely through Rho kinase. This may provide an explanation for the mononuclear cell infiltrates seen in the primary vascular lesion of KohlmeierDegos disease [17]. In summary, lipophage accumulation, smooth muscle hyperplasia, and endothelial cell dysfunction are all characteristic features of the vascular changes seen in the setting of Degos disease, dermatomyositis,
(rs2177268) and two polymorphisms in RhoC gene (rs11102522 and rs11538960). A similar gene polymorphism has been implicated in Behcet's disease [12]. Mechanisms underlying Rho kinase activation in Kohlemeier-Degos disease are likely complicated and multifactorial in etiology. Both an upregulated type I interferon microenvironment and C5b-9 deposition in vessels are features of the microvascular and arteriopathic changes seen in Kohlmeier-Degos disease [13]. Enhanced hyaluronic acid deposition in the intima (a probable interferon alpha effect) and C5a as a component of complement activation are known inducers of the Rho kinase pathway. The upregulation of Rho kinase may in turn contribute at least in part to the distinctive pathology observed in this rare and at times fatal primary vascular disorder. One of the critical components of the vascular pathology is a striking neointimal proliferative lesion that affects small and medium-sized arteries of the gastrointestinal tract. Other organ systems can show the same obliterative neointimal arteriopathy. Rho kinase may alter the properties of intramural smooth muscle cells including promotion of myofibroblastic proliferation in the intima [14]. Similar obliterative arteriopathic changes occur in 6
Annals of Diagnostic Pathology 44 (2020) 151414
C. Magro, et al.
Fig. 5. ROCK2 expression in control biopsies. The biopsy is one of normal skin procured to assess for vascular C5b-9 deposition in the setting of atypical hemolytic uremic syndrome. Significant vascular deposition is not seen (a). In this case of cell poor urticaria only a weak endothelial cell staining pattern is observed (b). In this case of lichen planus pigmentosus there is no significant staining for ROCK2 (c). A biopsy was performed showing a lichenoid dermatitis thought to be clinically and histologically compatible with a lichenoid keratosis given its solitary nature. There is weak staining of endothelium for ROCK2 while the lymphocytic infiltrate is without significant staining (d).
scleroderma and lupus erythematosus. All of these conditions can be associated with a significant degree of inflammation, oftentimes linked with enhanced interferon alpha expression. Both the inflammatory and arteriopathic changes could in part be based on excessive Rho kinase activity. The Rho kinase pathway activation may also be involved in the recruitment of inflammatory cells in other forms of cutaneous inflammation given its frequent upregulation in any state of cutaneous inflammation.
org/10.1309/AJCP66QIMFARLZKI. [6] Rozo C, Chinenov Y, Maharaj RK, et al. Targeting the RhoA-ROCK pathway to reverse T-cell dysfunction in SLE. Ann Rheum Dis 2017;76(4):740–7. https://doi.org/ 10.1136/annrheumdis-2016-209850. [7] Hartmann S, Ridley AJ, Lutz S. The function of rho-associated kinases ROCK1 and ROCK2 in the pathogenesis of cardiovascular disease. Front Pharmacol 2015;6:276. https://doi.org/10.3389/fphar.2015.00276. [8] Isgro J, Gupta S, Jacek E, et al. Enhanced rho-associated protein kinase activation in patients with systemic lupus erythematosus. Arthritis Rheum 2013;65(6):1592–602. https://doi.org/10.1002/art.37934. [9] Wang Y, Lu Y, Chai J, et al. Y-27632, a rho-associated protein kinase inhibitor. Inhibits Systemic Lupus Erythematosus Biomed Pharmacother 2017;88:359–66. https://doi.org/10.1016/j.biopha.2017.01.069. [10] Tratsiakovich Y, Kiss A, Gonon AT, Yang J, Sjöquist P-O, Pernow J. Inhibition of rho kinase protects from ischaemia–reperfusion injury via regulation of arginase activity and nitric oxide synthase in type 1 diabetes. Diab Vasc Dis Res 2017;14(3):236–45. https://doi.org/10.1177/1479164116687935. [11] Yi R, Xiao-Ping G, Hui L. Atorvastatin prevents angiotensin II-induced high permeability of human arterial endothelial cell monolayers via ROCK signaling pathway. Biochem Biophys Res Commun 2015;459(1):94–9. https://doi.org/10. 1016/j.bbrc.2015.02.076. [12] Pehlivan Y, Yolbas S, Cetin GY, et al. Investigation of the association between rho/ rho-kinase gene polymorphisms and systemic sclerosis. Rheumatol Int 2016;36(3):421–7. https://doi.org/10.1007/s00296-015-3400-4. [13] Singleton PA, Dudek SM, Ma S-F, Garcia JGN. Transactivation of Sphingosine 1phosphate receptors is essential for vascular barrier regulation. J Biol Chem 2006;281(45):34381–93. [14] Shimokawa H, Sunamura S, Satoh K. RhoA/rho-kinase in the cardiovascular system. Circ Res 2016;118(2):352–66. https://doi.org/10.1161/CIRCRESAHA.115.306532. [15] Pellegrino PR, Schiller AM, Haack KK, Zucker IH. Central angiotensin-II increases blood pressure and sympathetic outflow via rho kinase activation in conscious rabbits. Hypertension. 2016;68(5):1271–80. https://doi.org/10.1161/ HYPERTENSIONAHA.116.07792. [16] Liu L-J, Yao F-J, Lu G-H, et al. The role of the rho/ROCK pathway in Ang II and TGFβ1-induced atrial remodeling. PLoS One 2016;11(9):e0161625https://doi.org/10. 1371/journal.pone.0161625. [17] Kawanami D, Matoba K, Kanazawa Y, Ishizawa S, Yokota T, Utsunomiya K. Thrombin induces MCP-1 expression through rho-kinase and subsequent p38MAPK/NF-κB signaling pathway activation in vascular endothelial cells. Biochem Biophys Res Commun 2011;411(4):798–803. https://doi.org/10.1016/j. bbrc.2011.07.031.
Sources of funding None. Declaration of competing interest None. References [1] Nakagawa O, Fujisawa K, Ishizaki T, Saito Y, Nakao K, Narumiya S. ROCK-I and ROCK-II, two isoforms of rho-associated coiled-coil forming protein serine/threonine kinase in mice. FEBS Lett 1996;392(2):189–93. https://doi.org/10.1016/00145793(96)00811-3. [2] Sawada N, Liao JK. Rho/Rho-associated coiled-coil forming kinase pathway as therapeutic targets for statins in atherosclerosis. Antioxid Redox Signal 2014;20(8):1251–67. https://doi.org/10.1089/ars.2013.5524. [3] Zhang X, Zhang T, Gao F, et al. Fasudil, a rho-kinase inhibitor, prevents intimamedia thickening in a partially ligated carotid artery mouse model: effects of fasudil in flow-induced vascular remodeling. Mol Med Rep 2015;12(5):7317–25. https:// doi.org/10.3892/mmr.2015.4409. [4] Yuan T-Y, Yan Y, Wu Y-J, et al. Vasodilatory effect of a novel rho-kinase inhibitor, DL0805-2, on the rat mesenteric artery and its potential mechanisms. Cardiovasc Drugs Ther 2014;28(5):415–24. https://doi.org/10.1007/s10557-014-6544-7. [5] Magro CM, Poe JC, Kim C, et al. Degos disease: a C5b-9/interferon-α–mediated endotheliopathy syndrome. Am J Clin Pathol 2011;135(4):599–610. https://doi.
7