Portal de Periódicos da CAPES

Expression of myofibroblast activation molecules in proliferative vitreoretinopathy epiretinal membranes

2010; Wiley; Volume: 89; Issue: 2 Linguagem: Inglês

10.1111/j.1755-3768.2010.01916.x

1755-3768

Ahmed M. Abu El‐Asrar, L Missotten, Karel Geboes,

Signaling Pathways in Disease

Purpose: Fibrotic disorders are associated with activation of fibroblasts into extracellular matrix-secreting myofibroblasts expressing α-smooth muscle actin (α-SMA). Myofibroblasts are the predominant cellular component of proliferative vitreoretinopathy (PVR) epiretinal membranes. We investigated the expression of molecules involved in myofibroblast activation, migration and proliferation in PVR epiretinal membranes. Methods: Fifteen membranes were studied by immunohistochemical techniques using monoclonal and polyclonal antibodies directed against snail, fibroblast activation protein (FAP), CD44, hydrogen peroxide-inducible clone-5 (Hic-5), galectin-3, interleukin-13 receptor α2 (IL-13Rα2) and receptor for advanced glycation end products (RAGE). Results: Myofibroblasts expressing α-SMA were present in all membranes. Myofibroblasts expressed nuclear immunoreactivity for Snail and Hic-5, cytoplasmic immunoreactivity for FAP, IL-13Rα2 and RAGE and membranous immunoreactivity for CD44. There was no immunoreactivity for galectin-3. The number of cells expressing α-SMA correlated significantly with the number of cells expressing Snail (r = 0.56; p = 0.03), Hic-5 (r = 0.526; p = 0.044), IL-13Rα2 (r = 0.773; p = 0.001) and RAGE (r = 0.734; p = 0.002). Conclusions: Snail, FAP, CD44, Hic-5, IL13Rα2 and RAGE may be involved in proliferative events occurring in PVR. The biological process of fibrosis, typically associated with an abnormal accumulation of extracellular matrix, occurs in response to various stimuli in many biological systems. During some repair processes, fibrosis functions as necessary homeostatic mechanism. In other cases, however, fibrosis may occur at critical anatomic sites, thereby adversely affecting organ function (Wynn 2008). One example of the pathological fibrosis is proliferative vitreoretinopathy (PVR). Proliferative vitreoretinopathy is a process of fibrocellular proliferation on either side of the retina that may complicate rhegmatogenous retinal detachment. The formation and gradual contraction of epiretinal membranes causes a marked distortion of the retinal architecture and results in complex retinal detachments that are difficult to repair. The key cellular mediator of fibrosis is the myofibroblast, a cell type differentiated from quiescent fibroblasts. These are contractile cells, characterized by the expression of α-smooth muscle actin (α-SMA), and their presence is a marker of progressive disease. They have the capacity to produce several extracellular matrix components including collagen resulting in fibrosis (Wynn 2008). Such myofibroblasts, sometimes termed 'activated fibroblasts', are known to be present in the stromal compartment of most invasive human cancers, in areas of chronic inflammation and in tissues undergoing the remodelling seen during wound healing (Wynn 2008). Previous studies have shown that α-SMA-expressing myofibroblasts are the principal cellular component of PVR epiretinal membranes (Abu El-Asrar et al. 2006, 2007, 2008). In addition, we demonstrated that circulating fibrocytes may function as precursors of myofibroblasts in PVR epiretinal membranes (Abu El-Asrar et al. 2008). The activation and proliferation of myofibroblasts, when unchecked, invariably results in tissue fibrosis. Little is known about the molecular changes that accompany activation of myofibroblasts during PVR progression. Several molecules are involved in myofibroblast activation, migration and proliferation such as Snail (Francí et al. 2006; Boutet et al. 2007; Franz et al. 2009; Rowe et al. 2009), fibroblast activation protein (FAP) (Park et al. 1999), CD44 (Messadi & Bertolami 1993; Svee et al. 1996; Florquin et al. 2002; Pouschop et al. 2004; Acharya et al. 2008; Huebener et al. 2008), hydrogen peroxide-inducible clone-5 (Hic-5) (Dabiri et al. 2008a,b;), galectin-3 (Henderson et al. 2006; Lippert et al. 2008), interleukin-13 receptor α2 (IL-13Rα2) (Shimamura et al. 2008) and receptor for advanced glycation end products (RAGE) (Fehrenback et al. 2001; De Vriese et al. 2006). This study was, therefore, undertaken to investigate the expression and distribution of Snail, FAP, CD44, Hic-5, galectin-3, IL-13Rα2 and RAGE in PVR epiretinal membranes. Immunostaining for α-SMA was performed to reveal the presence and distribution of myofibroblasts within PVR epiretinal membranes. Epiretinal membranes were obtained from 15 eyes undergoing vitreoretinal surgery for the treatment of retinal detachment complicated by PVR. All eyes had had previous vitrectomy for rhegmatogenous retinal detachment. Membranes were fixed in 10% formalin solution and embedded in paraffin. The study was conducted according to the tenets of the Declaration of Helsinki, and informed consent was obtained from all patients. The study was approved by the Research Centre, College of Medicine, King Saud University. Endogenous peroxidase was abolished with 2% hydrogen peroxide in methanol for 20 min, and nonspecific background staining was blocked by incubating the sections for 5 min in normal swine serum. For FAP, CD44, Galectin-3, IL-13Rα2 and RAGE, antigen retrieval was performed by boiling the sections in 10 mm Tris-ethylene-diaminetetraacetic acid (EDTA) buffer [pH 9] for 30 min. For Snail, and Hic-5 detection, antigen retrieval was performed by boiling the sections in 10 mm citrate buffer [pH 6] for 30 min. Subsequently, the sections were incubated with the monoclonal and polyclonal antibodies listed in Table 1. Optimal working concentration and incubation time for the antibodies were determined earlier in pilot experiments. For RAGE, a second step was introduced using 1/20 rabbit anti-goat-peroxidase + 1/10 normal human serum. The sections were then incubated for 30 min with immunoglobulin conjugated to peroxidase-labelled dextran polymer [EnVision (Flex); Dako, Carpinteria, CA, USA]. The reaction product was visualized by incubation for 10 min in 0.05 m acetate buffer at pH 4.9, containing 0.05% 3-amino-9-ethylcarbazole (Sigma-Aldrich, Bornem, Belgium) and 0.01% hydrogen peroxide, resulting in bright-red immunoreactive sites. The slides were then faintly counterstained with Harris haematoxylin. To identify the phenotype of cells expressing FAP, sequential double immunohistochemistry was performed as previously described (Abu El-Asrar et al. 2008). Omission or substitution of the primary antibody with an irrelevant antibody of the same species and staining with chromogen alone were used as negative controls. Sections from patients with colorectal carcinoma were used as positive controls. The sections from the control patients were obtained from patients treated at the University Hospital, University of Leuven, Belgium, in full compliance with tenets of the Declaration of Helsinki. Immunoreactive cells were counted in five representative fields, using an eyepiece calibrated grid in combination with the 40× objective. With this magnification and calibration, the cells present in an area of 0.33 mm × 0.22 mm were counted. Pearson correlation coefficients were computed to investigate the linear relationship between the variables investigated. A p-value < 0.05 indicated statistical significance. bmdp 2007 Statistical Package (BMDP Statistical Software, Inc., Los Angles, CA, USA) was used for the statistical analysis. There was no staining in the negative control slides (Fig. 1A). All PVR membranes showed myofibroblasts expressing a strong cytoplasmic immunoreactivity for α-SMA (Fig. 1B, C), with a mean number of 62 ± 46.9 (range, 10–200). Immunostaining for Snail revealed cells expressing nuclear immunoreactivity (Fig. 1D, E) in all membranes, with a mean number of 59.1 ± 29.1 (range, 5–100). Cells expressing cytoplasmic immunoreactivity for FAP (Fig. 2A, B) were noted in 14 (93.3%) membranes, with a mean number of 59.3 ± 49.0 (range, 0–200). Cells expressing membranous immunoreactivity for CD44 (Fig. 2C, D) were noted in 14 (93.3%) membranes, with a mean number of 57.4 ± 44.2 (range, 0–150). Cells expressing nuclear immunoreactivity for Hic-5 (Fig. 2E) were noted in 14 (93.3%) membranes, with a mean number of 43.3 ± 22.9 (range, 0–80). Cells expressing strong cytoplasmic immunoreactivity for IL-13Rα2 (Fig. 3A, B) were noted in all membranes, with a mean number of 57.5 ± 29.2 (range, 8–105). Cells expressing strong cytoplasmic immunoreactivity for RAGE (Fig. 3C, D) were noted in all membranes, with a mean number of 57.0 ± 35.1 (range, 15–120). There was no immunoreactivity for galectin-3. In serial sections, the distribution of myofibroblasts expressing α-SMA was similar to the distribution of cells expressing Snail, FAP, CD44, Hic-5, IL-13Rα2 and RAGE. Double immunohistochemistry confirmed that α-SMA-expressing myofibroblasts also co-expressed FAP (Fig. 3E). Negative control slide that was treated identically with an irrelevant antibody showing no labelling (A) (original magnification ×40). Immunohistochemical staining for α-smooth muscle actin showing cytoplasmic immunoreactivity in myofibroblasts. Low-power (B) (original magnification ×40) and high-power (C) (original magnification ×100). Immunohistochemical staining for Snail showing cells expressing nuclear immunoreactivity. Low-power (D) (original magnification ×40) and high-power (E) (original magnification ×100). Immunohistochemical staining for fibroblast activation protein. Low-power (A) (original magnification ×40) and high-power (B) (original magnification ×100) showing cells expressing cytoplasmic immunoreactivity for fibroblast activation protein. Immunohistochemical staining for CD44. Low-power (C) (original magnification ×40) and high-power (D) (original magnification ×100) showing cells expressing membranous immunoreactivity for CD44. Immunohistochemical staining for Hic-5 showing cells expressing nuclear immunoreactivity (E) (original magnification ×40). Immunohistochemical staining for interleukin-13 receptor α2 (IL-13Rα2). Low-power (A) (original magnification ×40) and high-power (B) (original magnification ×100) showing cells expressing cytoplasmic immunoreactivity for IL-13Rα2. Immunohistochemical staining for the receptor for advanced glycation end products (RAGE). Low-power (C) (original magnification ×40) and high-power (D) (original magnification ×100) showing cells expressing cytoplasmic immunoreactivity for RAGE. Double immunohistochemistry for α-smooth muscle actin (α-SMA) (blue) and fibroblast activation protein (red) showing cells co-expressing α-SMA and fibroblast activation protein (arrows) (E). No counterstain was applied (original magnification ×100). Table 2 shows Pearson correlation coefficients between the numbers of studied variables. The number of cells expressing α-SMA correlated significantly with the number of cells expressing Snail, Hic-5, IL-13Rα2 and RAGE. The number of cells expressing FAP correlated significantly with the number of cells expressing Snail, CD44, Hic-5 and RAGE. The number of cells expressing Snail correlated significantly with the number of cells expressing FAP, Hic-5, IL-13Rα2 and RAGE. The number of cells expressing CD44 correlated significantly with the number of cells expressing FAP, Hic-5 and RAGE. The number of cells expressing Hic-5 correlated significantly with the number of cells expressing FAP, Snail, CD44, IL-13Rα2 and RAGE. The number of cells expressing RAGE correlated significantly with the number of cells expressing FAP, Snail, CD44, Hic-5 and IL-13Rα2. The number of cells expressing IL-13Rα2 correlated significantly with the number of cells expressing Snail, Hic-5 and RAGE. Snail, a zinc finger transcription factor, plays a crucial role in the process of fibroblast activation and myofibroblast transdifferentiation (Francí et al. 2006; Boutet et al. 2007; Franz et al. 2009; Rowe et al. 2009). Under serum-free conditions, fibroblasts do not express detectable levels of Snail mRNA or protein. In contrast, in the presence of serum or growth factors, both Snail mRNA and intranuclear protein levels are strongly induced in fibroblasts. When challenged within a tissue-like, three-dimensional extracellular matrix, Snail-deficient fibroblasts exhibit global alterations in gene expression, which include defects in membrane type-1 matrix metalloproteinase-dependent invasive activity. Snail-deficient fibroblasts explanted atop the live chick chorioallantoic membrane lack tissue-invasive potential and fail to induce neovessel formation (Rowe et al. 2009). Francí et al. (2006) demonstrated that expression of Snail was not constitutive. Whereas, Snail expression was observed in activated fibroblasts involved in wound healing. When healing had progressed, and re-epithelization was complete, Snail immunoreactivity in fibroblasts fell down to the initial undetectable levels. Moreover, Snail was detected in pathological conditions causing hyperstimulation of fibroblasts, such as fibromatosis, sarcomas and fibrosarcomas. Snail immunoreactivity was mainly observed in the nuclei. Franz et al. (2009) also showed that most of the nuclear Snail-positive cells in oral squamous cell carcinoma also exhibited α-SMA immunoreactivity. The majority of these Snail-positive cells were stromal cells detectable in the invasive front in the vicinity of the tumour nests. They also showed that the expression of Snail was increased with raising α-SMA positivity, suggesting that there was a functional linkage to myofibroblast development. Our findings of nuclear immunoreactivity for Snail in myofibroblasts and that there was a significant correlation between the number of α-SMA-positive myofibroblasts and the number of myofibroblasts expressing Snail in PVR epiretinal membranes are consistent with these previous studies. In agreement with these in vitro and in vivo studies, Boutet et al. (2007) demonstrated that renal fibrosis was linked to the activity of Snail genes and reactivation of Snail in the adult kidney was sufficient to induce fibrosis. Fibroblast activation protein is a 97-kDa type II transmembrane glycoprotein with gelatinase/collagenase and dipeptidyl peptidase activities belonging to the serine protease family. FAP was demonstrated to cleave gelatin and type I collagen (Park et al. 1999). Recently, Christiansen et al. (2007) demonstrated that FAP works in synchrony with other proteinases to cleave partially degraded or denatured collagen types I and III. Fibroblast activation protein is expressed by activated stromal fibroblasts associated with rapid tissue growth as in embryogenesis, wound healing and epithelial-derived malignancies, but not by epithelial carcinoma cells, quiescent fibroblasts or other normal tissues (Park et al. 1999). Fibroblast activation protein, therefore, has been implicated in proteolytic degradation and remodelling of extracellular matrix as activated fibroblasts simultaneously multiply, migrate, synthesize and deposit extracellular matrix constituents during rapid tissue growth and expansion (Park et al. 1999). Overexpression of FAP reportedly leads to promotion of tumour growth and increases in metastatic potential (Cheng et al. 2002; Henry et al. 2007), whereas targeting FAP enzymatic activity inhibits tumour growth (Cheng et al. 2002; Lee et al. 2005; Ostermann et al. 2008). Our findings indicate that myofibroblasts are the major cell type expressing FAP in PVR epiretinal membranes. Similarly, activated stromal fibroblasts of benign melanocytic nevi (Huber et al. 2003) and rheumatoid myofibroblast-like synoviocytes (Bauer et al. 2006) co-expressed α-SMA and FAP. Matrix metalloproteinase-9 (MMP-9) is also instrumental in extracellular matrix remodelling in pathological processes of invasive growth and is also known to be expressed by myofibroblasts in PVR epiretinal membranes (Abu El-Asrar et al. 2007). In addition, Symeonidis et al. (2009) demonstrated a significant correlation between PVR grade and the levels of proMMP-9 in the vitreous. The function of FAP in PVR microenvironment is unknown. However, based on the selective expression of FAP by PVR myofibroblasts, we hypothesize that FAP is implicated in PVR propagation and inhibition of FAP proteolytic activity might attenuate the invasive capabilities of PVR myofibroblasts, leading to attenuation of PVR propagation. CD44 is a cell surface matrix adhesion receptor for hyaluronic acid on the surface of fibroblasts involved in mediating fibroblast migration and invasion of the wound provisional matrix resulting in the formation of fibrotic tissue (Svee et al. 1996). In this study, immunostaining for CD44 revealed membranous immunoreactivity in myofibroblasts in PVR epiretinal membranes. Our observations are consistent with previous reports showing expression of CD44 by fibroblasts and myofibroblasts in other fibrotic disorders. Cell surface CD44 expression has been found to be increased in hypertrophic scar fibroblasts derived from human cutaneous scar tissue when compared to normal skin fibroblasts (Messadi & Bertolami 1993). Immunohistochemical analysis of lung tissue derived from patients with acute alveolar fibrosis after lung injury revealed CD44-expressing fibroblasts throughout newly formed fibrotic tissue (Svee et al. 1996). CD44 colocalized with α-SMA-positive myofibroblasts in the interstitum of renal biopsy specimens from patients with IgA nephropathy and correlated significantly with the degree of glomerular interstitial damage (Florquin et al. 2002). In the infarcted mouse myocardium, CD44 expression was markedly increased and was localized on infiltrating α-SMA-positive myofibroblasts (Huebener et al. 2008). Several lines of evidence link CD44 with fibroblast adhesion and migration to sites of injury and the development of fibrosis. By blocking CD44 function with an anti-CD44 antibody, acute lung injury fibroblast adhesion and migration on the provisional matrix proteins fibronectin, fibrinogen and hyaluronic acid and invasion in a fibrin matrix were inhibited (Svee et al. 1996). Recently, Acharya et al. (2008) demonstrated that CD44-deficient fibroblasts exhibited fewer stress fibres and focal adhesion complexes, and their migration was characterized by increased velocity and loss of directionality, compared with wild-type fibroblasts. In addition, they demonstrated decreased levels of α-SMA in CD44-deficient fibroblasts compared with wild-type fibroblasts, suggesting that the presence of CD44 may affect fibroblast activation. In vivo studies demonstrated that CD44-deficient mice had significantly lower α-SMA-positive myofibroblast density and reduced collagen content in the infarcted myocardium (Huebener et al. 2008), and in the obstructed kidney (Pouschop et al. 2004) compared with wild-type mice. Collectively, these data suggest a role for CD44 in mediating myofibroblast invasion and subsequent tissue fibrosis in PVR. Hic-5 is a transforming growth factor-β- and H202-inducible focal adhesion protein that shuttles between focal adhesions and the nucleus through an oxidant-sensitive nuclear export signal. Hic-5 accumulates in the nucleus in response to oxidants, such as H202, where it may serve as a transcription factor (Shibanuma et al. 2003). In the present study, immunostaining for Hic-5 revealed nuclear positivity in myofibroblasts in PVR epiretinal membranes. The number of α-SMA-expressing myofibroblasts correlated significantly with the number of myofibroblasts expressing Hic-5. This is in agreement with studies showing that Hic-5 is markedly upregulated and localizes to the nucleus of myofibroblasts from hypertrophic scars (Dabiri et al. 2008b). Recently, Dabiri et al. (2008a) demonstrated that Hic-5 supports the persistent myofibroblast phenotype seen in hypertrophic scar fibroblasts. Genetic silencing of Hic-5 in hypertrophic scar myofibroblasts dramatically reduces transforming growth factor-β1 production, decreases the generation of supermature focal adhesions, reduces the expression of α-SMA and decreases collagen contraction and extracellular matrix synthesis (Dabiri et al. 2008a). These data suggest that Hic-5 is an important target in pathogenic myofibroblasts that could be inhibited therapeutically to help in the treatment of fibrotic disorders. Galectin-3 is critical for myofibroblast activation and is considered to be an immediate early gene which is upregulated rapidly in response to tissue injury. In a rat model of liver fibrosis, galectin-3 expression was temporally and spatially associated with fibrosis and myofibroblast activation and was maximal at peak of fibrosis (Henderson et al. 2006). In addition, Lippert et al. (2008) demonstrated loss of galectin-3 mRNA and protein expression in Crohn's disease stenoses and fistulae correlating with prolonged activation of fibroblasts. Therefore, the absence of galectin-3 immunoreactivity in PVR epiretinal membranes might be explained by the chronicity of the disease process. Several studies demonstrated that fibrosis is driven by the Th2-type cytokine IL-13 signalling via a unique IL-13 receptor, IL-13Rα2, that induces production of transforming growth factor- β1, insulin-like growth factor and early growth response gene 1, and myofibroblast production of collagen (Fichtner-Feigl et al. 2007, 2008; Shimamura et al. 2008). In this study, PVR epiretinal membranes showed strong immunoreactivity for IL-13Rα2 in myofibroblasts. In addition, there was a significant correlation between the number of α-SMA-expressing myofibroblasts and the number of myofibroblasts expressing IL-13Rα2. Our observations are consistent with previous reports showing increased IL-13Rα2 expression in fibrotic diseases such as human idiopathic interstitial pneumonia (Jakubzick et al. 2004), human chronic liver fibrosis induced by nonalcoholic steatohepatitis (Shimamura et al. 2008) and in skin lesions of a murine model of bleomycin-induced scleroderma (Matsushita et al. 2004). Increased IL-13Rα2 expression was reported on fibroblasts in human interstitial pneumonia (Jakubzick et al. 2004) and colocalized with α-SMA in hepatic stellate cells in patients with chronic liver fibrosis (Shimamura et al. 2008). Shimamura et al. (2008) demonstrated that targeting IL-13Rα2 caused a substantial decline in experimental liver fibrosis. Thus, IL-13Rα2 could be one of the key molecules that participate in the development of PVR and could be an important therapeutic target for the prevention of PVR. Receptor for advanced glycation end products is a member of the immunoglobulin superfamily of cell surface molecules and interacts with diverse ligands, including advanced glycation end products, S100/calgranulins, amphoterins and amyloid- β-peptide. This ligand–receptor interaction activates multiple cellular signal transduction pathways leading to the secretion of inflammatory cytokines, chemokines, upregulation of adhesion molecules and the production of growth factors (Yan et al. 2003). Several studies demonstrated that RAGE plays a key role in fibrotic disorders. Expression of RAGE was upregulated during transdifferentiation of hepatic stellate cells into myofibroblasts (Fehrenback et al. 2001). Myofibroblast transdifferentiation of mesothelial cells and peritoneal fibrosis in chronic uraemia are mediated by the ligand engagement of RAGE (De Vriese et al. 2006). Inhibition of RAGE had prominent antifibrotic effects confirming the major role of RAGE in the progression of fibrosis (De Vriese et al. 2006; Xia et al. 2008). In this study, we demonstrated that strong immunoreactivity for RAGE was specifically localized in α-SMA-positive myofibroblasts in PVR epiretinal membranes and there was a significant correlation between the number of α-SMA-expressing myofibroblasts and the number of myofibroblasts expressing RAGE. This is in agreement with studies showing overexpression of RAGE in other fibrotic disorders such as systemic sclerosis fibrotic skin lesions (Yoshizaki et al. 2009) and experimental hepatic fibrosis (Xia et al. 2008), and increased RAGE expression was detected in fibroblasts in systemic fibrotic skin lesions (Yoshizaki et al. 2009). In conclusion, the restriction of co-expression of Snail, FAP, CD44, Hic-5, IL-13Rα2, and RAGE to myofibroblasts, which are considered to be the proliferating cellular components of PVR, suggests that these molecules may be involved in proliferative events occurring in PVR. Inhibitors to these molecules may interrupt development, recruitment and/or activation of myofibroblasts and could provide a unique therapeutic approach to prevent/attenuate PVR progression. The authors thank Mr Dustan Kangave, MSc for statistical assistance, Ms Lieve Ophalvens, Ms Christel Van den Broeck and Mr Johan Van Evan for technical assistance and Ms Connie B. Unisa-Marfil for secretarial work. This work was supported by the College of Medicine Research Center, King Saud University, Medical Research Chair Funded by Dr Nasser Al-Rasheed.