Lentiviral Delivery of Small Hairpin RNA Targeting Connective Tissue Growth Factor Blocks Profibrotic Signaling in Tenon's Capsule Fibroblasts

Abstract

Purposes: Trabeculectomy is a surgical procedure for lowering intraocular pressure in glaucoma patients, in which excessive scarring leading to failure of the filtering bleb adversely affects the surgical outcome. Heightened Tenon's capsule fibroblast (TCF) proliferation and extracellular matrix (ECM) deposition are implicated in this process but endogenous factors that regulate TCF functions remain largely elusive. This study sought to elucidate the role of connective tissue growth factor (CTGF) in the regulation of TCF phenotypes and signaling.

Methods: Expression of CTGF in scarring and nonscarring Tenon's capsules was measured by real-time PCR and immunofluorescence. Knockdown of CTGC was achieved by lentivirus delivery of small-hairpin RNA. Cell proliferation was measured by CCK8, cell cycle progression, and apoptosis by flow cytometry, adhesion, migration, and invasion of TCF by functional assays in vitro. Proteins and cytokines related to fibrosis were measured by Western blot and ELISA, respectively.

Results: Expression of CTGF was significantly upregulated in scarring Tenon's capsules and their isolated fibroblasts when compared with the nonfibrotic counterparts. Functionally, targeting CTGF with lentivirus-delivered small-hairpin RNA inhibited the proliferation, adhesion, migration, and invasion of TCFs, accompanied by downregulation of p38 and nuclear factor-κB as well as matrix metalloproteinase-2, cyclin D1, and collagen I. In addition, lentiviral targeting of CTGF reduced the release of fibrosis-related cytokines from TCFs and inhibited TCF-conditioned, medium-induced macrophage chemotaxis.

Conclusions: Our study supports a crucial role of CTGF in the regulation of TCF proliferation and ECM deposition. Targeting CTGF using lentiviral vector may be a promising approach for preventing excessive scarring after trabeculectomy.

Glaucoma, one of the leading causes of blindness, is a group of diseases characterized by aberrant increases in intraocular pressure (IOP) that can damage the optic nerve. Trabeculectomy, also known as glaucoma filtration surgery, is the most common procedure for lowering IOP by allowing drainage of aqueous humor from within the eye into a filtering bleb that is created underneath the conjunctiva and Tenon's capsule.¹ Nevertheless, excessive scarring resulting from increased Tenon's capsule fibroblast (TCF) proliferation and extracellular matrix (ECM) deposition could lead to failure of the filtering bleb and adversely affect the surgical outcome.² In this connection, intraoperative application of mitomycin C or 5-fluorouracil has been widely used in patients at risk for excessive scarring. However, the use of these agents could result in complications, such as cataract formation,³ epithelial toxicity, and bleb leaks that might lead to endophthalmitis.^4,5 This highlighted the need for a better understanding of the mechanism underlying ocular scarring and the development of novel strategies for preventing excessive scarring after trabeculectomy. 

Connective tissue growth factor (CTGF) is an inducible secretory protein known to regulate proliferation and ECM production in fibroblasts of different tissue origins as well as other cell types,^6–8 such as vascular smooth muscle cells and cardiomyocytes.^9,10 Its altered expression has been implicated in normal tissue repair and human diseases characterized by excessive scarring, such as keloid, scleroderma, fibrotic liver diseases, and lung fibrosis.^11–13 

Gene therapy for modulating fibrosis-related genes in TCFs has been promulgated as a novel strategy for controlling scarring after trabeculectomy.^14,15 For effective RNA interference, a small hairpin RNA (shRNA) with a sequence that self-anneals to form a hairpin structure can be employed to silence target gene expression. Importantly, the development of lentiviral vectors for ectopic expression of shRNA has been an advance in the field of gene therapy.¹⁶ In this study, we show that CTGF was remarkably upregulated in scarring Tenon's capsule tissues and their isolated fibroblasts compared with the normal counterparts. We also demonstrate that targeting CTGF with shRNA delivered by a lentiviral vector resulted in sustained inhibition of profibrotic phenotypes and suppression of cytokine expression. Our findings suggest that lentiviral targeting of CTGF in TCFs could be an effective approach to curb scarring after trabeculectomy. 

Materials and Methods

Tissues and Primary TCFs

Primary human TCFs were established from subconjunctival Tenon's capsule isolated from a patient undergone ocular surgery as previously described.¹⁷ In brief, small pieces (∼1 mm³) of human Tenon's capsules from three to five clinical samples were maintained in DMEM medium containing 20% FBS for 5 to 7 days to allow TCFs to adhere to the culture flask. After culture of three to five generations, cells in good state were retained for downstream functional experiments. Healthy Tenon's capsules were isolated from patients with acute ocular trauma who had to clear the Tenon's capsule tissue. Informed consent was obtained from all patients. Acquisition of all tissue specimens had also been approved by the ethical committee at the authors' institution (The Second Affiliated Hospital of Harbin Medical University, Harbin, China; 2014-yan-113). The human macrophage and 293T cell lines were obtained from the Institute of Biochemistry and Cell Biology, CAS (Shanghai, China). The cells were cultured in DMEM (with 2 mM L-glutamine; Invitrogen, San Diego, CA, USA) supplemented with 10% fetal bovine serum (Gibco Laboratories, Gaithersburg, MD, USA) and 1% antibiotics (Invitrogen) at 37°C and 5% CO₂ in a humidified atmosphere. 

Construction of CTGF-shRNA-Expressing Lentiviral Vector and Transduction

Small hairpin RNAs targeting CTGF mRNA were designed online (www.invitrogen.com/rnai): CTGF-homo-936, 5′-GCTGACCTGGAAGAGAACATT-3′; CTGF-homo-1024, 5′-GCACCAGCATGAAGACATACC-3′; CTGF-homo-1135, 5′-GCGAGGTCATGAAGAAGAACA-3′; CTGF-homo-1285, 5′-GCCAGAGAGTGAGAGACATTA-3′. A negative sequence that shared no homology with the mammalian genome was used as control. Successful cloning of these sequences into the pLV-U6-Puro-GFP lentivector was confirmed by Sanger sequencing (Supplementary Table S1). The third-generation HIV lentiviral system (Invitrogen) including the packaging vectors pMD2.G and pRSV-Rev,pMDLg/pRRE was used in this study. We used a 293T cell line for the production of pseudoviral particles according to the manufacturer's instructions (lenti-pseudoviral particles; Genebiology, Shanghai, China). TCFs were then transduced at a multiplicity of infection (MOI) of 50 (i.e., 10 μL of lentivirus at the titer of 10⁸ TU/mL was added to 2 × 10⁵ TCFs). 

Total RNA Extraction and Quantitative Reverse Transcription (RT)-PCR

Total RNA was isolated from frozen tissue or cultured cells with Trizol Reagent (Invitrogen). The quantity and integrity of isolated RNA were determined by A₂₆₀ measurement using a spectrophotometer (ND-1000 NanoDrop; NanoDrop, Wilmington, DE, USA) and 28S/18S ratio detection on an agarose gel. Total RNA was reverse-transcribed into cDNA using M-MLV reverse transcriptase (Promega, Madison, WI, USA) with the random hexamer RT primers. We used GAPDH mRNA as an endogenous control for PCR reactions. We used a commercial kit (SYBR Green Mix Taq Kit; TaKaRa, Tokyo, Japan) to trace the amplified DNA, and the PCR reaction was directly monitored by a genetic analyzer (ABI PRISM 7300 System; Applied Biosystems, Foster City, CA, USA). All primer sequences are provided as the supplementary information (Supplementary Table S2). Reactions were performed in triplicate and relative expression was calculated using the 2^–ΔΔCt method.¹⁸ 

Protein Extraction and Western Blots

Whole cells were lysed in a buffer containing 7 M urea, 1% Triton X-100, 100 mM DTT, 20 mM Tris-HCl, pH 8.5. Protein concentration was determined by Bradford assay (Bio-Rad, Hercules, CA, USA). We separated 10 μg of protein per lane on either 12% or 15% polyacrylamide gels and then transferred to nitrocellulose membranes (Protran, Whatman, Dassel, Germany). Antibodies to CTGF (sc-14939); p38 (sc-7972); p-p38 (sc-7973); Bcl2 (sc-7382); cyclin D1 (sc-8396); matrix metalloproteinase (MMP)-2 (sc-13594); COL1A1 (sc-8784); NF-κB p65 (sc-372); and GAPDH (sc-365062) were purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Horseradish peroxidase-conjugated secondary antibodies were obtained from Jackson Immunoresearch (West Grove, PA, USA). Western blotting was performed using standard methods. The bands were visualized using the enhanced chemiluminescence (ECL) reagents (Amersham Pharmacia, Buckinghamshire, UK). Images were captured with a commercial imaging system (Syngene, Frederick, MD, USA) with densitometry performed using ImageJ software (http://imagej.nih.gov/ij/; provided in the public domain by the National Institutes of Health, Bethesda, MD, USA). 

Immunofluorescence

Cells seeded on cover glass, collagen-coated coverslips, or matrigel were fixed with methanol or 3% paraformaldehyde and probed with primary antibodies followed by incubation with AlexaFluor 594 secondary antibodies (Invitrogen). Images were acquired with a Zeiss LSM 510 microscope (Zeiss, Oberkochen, Germany). 

Cell Proliferation and Adhesion Assays

The number of viable cells was determined by the CCK-8 assay. Briefly, 0.5 × 10⁵ cells per well were seeded in 96-well plates and allowed to attach overnight. On each day of 4 consecutive days, 10 μL CCK-8 (Dojindo, Kumamoto, Japan) solution was added to each well, and the plates were incubated for another 2 hours. Measurements of optical density were obtained at the wavelength of 490 nm using a spectrophotometer. For cell adhesion assay, a polystyrene high-bind 96-well plate (Corning, Inc., Corning, NY, USA) was used. We added 100 mL of cell suspension to each of the collagen I–coated wells. The plate was then incubated at 37°C for 20 minutes to 6 hours to allow the cells to adhere to the surface. We then added 100 mL DMEM to each well to wash off any nonadherent cells for three times. Adherent cells were fixed in methanol, stained with crystal violet (0.1% wt/vol; Sigma-Aldrich, St. Louis, MO, USA), lysed in 10 mM HEPES, pH 7.4, containing 1% (wt/vol) sodium deoxycholate, and quantified using a spectrophotometer (absorbance at 540 nm from which the background at 405 nm was subtracted).¹⁹ 

Cell Cycle and Apoptosis Analyses

For cell cycle analysis, transfected cells in the log phase of growth were harvested by trypsinization, washed with PBS, fixed with 70% ethanol overnight at 4°C, and then incubated with RNase at 37°C for 30 minutes. The nuclei of cells were stained with propidium iodide (5 μg/mL) for an additional 30 minutes and analyzed using a flow-cytometer (FACS Calibur; Becton Dickinson, Franklin Lakes, NJ, USA). We constructed DNA histograms using analysis software (ModFit; Verity Software House, Topsham, ME, USA). Experiments were performed in triplicate. Results were presented as percentage of cell in a particular phase. For apoptosis quantitation, after transfection with the LV-shRNA-CTGF-1024 or negative control lentivirus, both floating and attached cells were collected and subject to annexin V/propidium iodide staining using a detection kit (Annexin V-APC/7-AAD Apoptosis Detection Kit; BioVision, Palo Alto, CA, USA). The resulting fluorescence was measured using a FACS flow cytometer (BD Biosciences, San Jose, CA, USA). 

Cell Invasion and Migration Assays

Human TCFs were transfected with LV-shRNA-CTGF-1024 or negative control lentivirus, cultured for 48 hours, and transferred onto the top of matrigel-coated invasion chambers (24-well insert, 8 μm pore size; BD Biosciences) in serum-free DMEM. We added DMEM containing 10% fetal calf serum to the lower chamber. After 24 hours incubation, noninvaded cells were removed from the inner part of the insert with a cotton swab. Fixation and staining of invaded cells were performed using 0.1% crystal violet. Cells were quantified by fluorescence microscopy (×100). For cell migration assay, TCFs were seeded onto 24-well plates and cultured in complete medium until confluent. The confluent monolayers were washed twice with PBS and incubated in a serum-free medium. Small linear wounds were created by gently striking a pipette tip across the monolayers. The healing of the wounds through cell migration was assessed by measuring the wound distance. 

Determination of Cytokine/Chemokine Levels

The release of cytokines and chemokines at 72 hours postinfection was quantified by ELISA. In brief, TCF homogenates were sonicated and centrifuged at 13,000g for 10 minutes at 4°C, aliquoted and stored at −80°C. Total protein was determined by RC DC Protein Assay (Bio-Rad) before quantification of IL-1/IL-6/TGF-β1/IFN-γ/MCP-1 by ELISA (Catalog: DLB50, D6050, DB100B, DIF50, DCP00; R&D Systems, Minneapolis, MN, USA) following the manufacturer's protocol. Data was expressed as pictograms per microgram of total protein. Absorbance was measured at 492 nm using an ELISA reader (Model 450; Bio-Rad). All samples were analyzed in duplicate. 

Macrophage Chemotaxis Assay

Chemotactic assay was performed using transwell cell migration chambers with 5-μm pore inserts (Cell Biolabs, Inc., San Diego, CA, USA) in a 24-well plate format. Lower chambers conditioned medium of TCFs transfected with LV-shRNA-CTGF-1024 or negative control at different time points (0, 8, 16, and 32 hours) as chemotactic stimulus. Macrophages were placed into upper chambers (10,000 cells per insert) in serum-free medium. The percentage of cells migrated to the lower chamber was determined by commercial dye (CyQuant GR; Invitrogen) staining. Fluorescence was read with a fluorescence plate reader at 480/520 nm and performed as previously described.²⁰ 

Statistical Analysis

Results were analyzed using one-way ANOVA with Fisher's post-hoc test. Data was presented as mean ± SD. Values of P < 0.05 were considered to be significant. 

Results

Upregulation of CTGF in Scarring Tenon's Capsules and Their Isolated Fibroblasts

We first examined the expression of CTGF in human Tenon's capsule tissues with or without scarring. Real-time PCR showed a remarkable upregulation of CTGF mRNA in scarring Tenon's capsule tissues as compared with nonscarring counterparts (Fig. 1A). Concordant with real-time PCR, immunohistochemical staining revealed that intermediate-to-high level of cytoplasmic and nuclear expression of CTGF was observed in fibroblasts isolated from scarring Tenon's capsules but not those isolated from the non-scarring Tenon's capsules (Fig. 1B). 

Figure 1

 

Upregulation of CTGF in scarring Tenon's capsules and their isolated fibroblasts. (A) A significant upregulation of CTGF was demonstrated in human scarring Tenon's capsules (n = 33) when compared with nonscarring Tenon's tissues (n = 5) by real time RT-PCR. **P < 0.01 significantly different from normal Tenon's tissues. (B) Immunofluorescence revealed strong cytoplasmic and nuclear staining of CTGF in fibroblasts isolated from human scarring Tenon's capsules. Fibroblasts isolated from non-scarring Tenon's tissues showed a remarkably lower staining intensity of CTGF.

Effective Repression of CTGF by Lentiviral Delivery of shRNA in TCFs

We next sought to inhibit CTGF expression in primary TCFs using lentivirus expressing CTGF-targeting shRNA. Four shRNA-encoding constructs, namely CTGF-homo-936, CTGF-homo-1024, CTGF-homo-1135 and CTGF-homo-1285 targeting different sites of CTGF transcript, were packaged into pseudotyped lentiviral particles. The transduction efficiencies of all four lentiviral vectors as well as the control vector were confirmed to be more than 99% in TCFs by fluorescence microscopy (Fig. 2A). Transduction of TCFs with CTGF-shRNA-encoding lentiviral vectors significantly reduced the mRNA and protein levels of CTGF (Fig. 2B, 2C). In particular, CTGF-homo-1024 showed the strongest inhibition on both CTGF mRNA and protein expression. The CTGF-homo-1024 encoding lentiviral vector (LV-shRNA-CTGF) was therefore used in subsequent experiments. 

Figure 2

 

Knockdown of CTGF by lentiviral delivery of shRNA in TCF. (A) Transduction efficiencies of lentivirus were confirmed to be over 99% by fluorescence microscopic detection of lentiviral expression of green fluorescence protein (GFP). (B–C) All four CTGF-shRNA-encoding constructs reduced CTGF (B) mRNA and (C) protein expression as determined by real-time PCR and Western blots, respectively. **P < 0.01. ***P < 0.001 significantly different from control lentivirus-transduced group.

Lentiviral Delivery of CTGF-shRNA Inhibits TCF Proliferation Without Affecting Apoptosis

Increased fibroblast proliferation directly contributes to excessive scarring. Effects of LV-shRNA-CTGF on TCF proliferation was measured by CCK8 assay up to 4 days after transduction. Results revealed that LV-shRNA-CTGF reduced the number of viable TCFs in a time-dependent manner. The growth-inhibitory effect peaked on posttransduction days 3 through 4 (Fig. 3A). To determine if decreased number of viable cells was a result of cell cycle arrest, we analyzed the cell cycle distribution of LV-shRNA-CTGF or control lentivirus-transduced TCFs by flow cytometry. At 72 hours posttransfection, LV-shRNA-CTGF increased the proportion of TCFs in G₀/G₁ phase when compared with the control group. A reciprocal reduction of cells in S and G₂/M phases was also observed in the LV-shRNA-CTGF group (Fig. 3B). Loss of phosphatidylserine asymmetry is a molecular hallmark of apoptosis. To determine if LV-shRNA-CTGF induced apoptosis in addition to cell cycle arrest, phosphatidylserine externalization was assayed by flow cytometry of Annexin V-APC/7-AAD double-stained TCFs. As shown in Figure 3C, the percentage of the Annexin V-positive apoptotic cells were not significantly higher in TCFs transduced with LV-shRNA-CTGF, when compared with control lentivirus-transduced and blank controls. These data suggest that LV-shRNA-CTGF exerted its growth-arresting effect without affecting apoptosis in TCFs. Nevertheless, whether prolonged arrest in G₀/G₁ phase induced by CTGF knockdown would result in apoptosis of TCFs at more distal time points remain uncertain. 

Figure 3

 

Knockdown of CTGF suppressed TCF proliferation and cell cycle progression. (A) The number of viable TCFs was reduced upon CTGF knockdown as determined by CCK8 assay. (B) Flow cytometry was performed to analyze cell cycle distribution of TCFs after propidium iodide staining. Knockdown of CTGF induced G₀/G₁-phase cell cycle arrest. (C) Apoptosis was assayed by annexin V-APC/7-AAD staining followed by flow cytometry. Knockdown of CTGF did not significantly alter apoptosis. Data are from three independent experiments. *P < 0.05. **P < 0.01 significantly different from control lentivirus-transduced group.

Lentiviral Delivery of CTGF-shRNA Inhibits TCF Adhesion, Migration, and Invasion

Aside from fibroproliferation, changes in cell adhesion, migration and invasiveness are other phenotypic properties of fibroblast activation. To assess cell adhesion, LV-shRNA-CTGF and control lentivirus-transduced TCFs were allowed to adhere to the surface for 20 minutes to 6 hours. Results showed that LV-shRNA-CTGF strongly reduced TCF adhesion by about 35% at all time points (Fig. 4A). Migration and invasion of TCF were determined by monolayer wound healing and transwell invasion assays, respectively. In this regard, CTGF knockdown substantially reduced the lateral motility (Fig. 4B) and invasiveness (Fig. 4C) of TCFs. To demonstrate the observed shRNA-induced phenotypic changes (i.e., cell proliferation, cell cycle distribution, cell migration) were caused by specific knockdown of CTGF, exogenous CTGF was added to shRNA-transfected cells. In this regard, the effects of CTGF-shRNA on proliferation, cell cycle and migration could no longer be observed in the presence of exogenous CTGF (Supplementary Fig. S1). In this regard, exogenous CTGF significantly increased TCF proliferation but only marginally enhanced TCF migration (Supplementary Fig. S2). 

Figure 4

 

Inhibition of TCF adhesion, migration and invasion upon CTGF knockdown. (A) LV-shRNA-CTGF- and control lentivirus-transduced TCFs were allowed to adhere to in collagen I-coated tissue culture wells for 20 minutes; 40 minutes; and 1, 2, 4, and 6 hours. Adhesion of transduced cells is shown relative to the blank control. (B) The healing of in vitro wounds through cell migration was assessed by measuring the wound distance. (C) We tested LV-shRNA-CTGF and control lentivirus-transduced TCFs for invasiveness in Matrigel transwell chambers. Data are from three independent experiments. **P < 0.01; significantly different from control lentivirus-transduced group.

Lentiviral Delivery of CTGF-shRNA Inhibits the Release of Fibrosis-Related Cytokines From TCFs and Suppresses Macrophage Chemotaxis

Compared with blank control and control lentivirus, LV-shRNA-CTGF significantly lowered the release of a panel of proinflammatory/profibrotic cytokines/chemokines, including IL-1, IL-6, TGF-β1, IFN-γ, and monocyte chemoattractant protein (MCP)-1, from TCFs over 72 hours, in which the concentrations of these cytokines/chemokines were almost reduced by two-thirds (Fig. 5A). Chemotaxis assay was then used assess macrophage chemotaxis towards conditioned medium of TCFs transduced with LV-shRNA-CTGF. As shown in Figure 5B, conditioned medium from LV-shRNA-CTGF-transduced TCFs had a significantly lower ability to chemoattract macrophages as compared with that from blank or control lentivirus-transduced cells. These results suggested that targeting CTGF might reduce the release of pro-inflammatory and pro-fibrotic cytokines/chemokines from TCFs and thereby decreasing macrophage chemotaxis. 

Figure 5

 

Reduced cytokine/chemokine release from TCF and macrophage chemotaxis upon CTGF knockdown. (A) Levels of IL-6, IL-1, TGF-β1, IFN-γ and MCP-1 released by TCFs transduced with LV-shRNA-CTGF or control lentivirus over 72 hours were determined by sandwich ELISA. (B) Macrophages were allowed to migrate toward lower chambers containing conditioned medium from TCFs transduced with LV-shRNA-CTGF or control lentivirus as chemotactic stimulus. We used 5 × 10⁴ cells per well in each assay. Migratory cells were quantified by commercial dye (Invitrogen). Data are from three independent experiments. **P < 0.01; significantly different from control lentivirus-transduced group.

Lentiviral Delivery of CTGF-shRNA Downregulates Profibrotic Mediators and Effectors in TCFs

We next measured the expression of mediators and effectors that are key to fibrosis in TCFs upon CTGF knockdown. Western blots results indicated that LV-shRNA-CTGF significantly reduced the protein expression of nuclear factor (NF)-κB p65 subunit as well as total and phosphorylated p38 (Fig. 6), both of which are known to be involved in fibroblast activation and cytokine release during fibrosis. Our data also showed that CTGF knockdown reduced the expression of proteins related to fibroblast proliferation (cyclin D1), migration (matrix metalloproteinase-2), ECM deposition (Collagen I) but not apoptosis (Bcl-2; Fig. 6). 

Figure 6

 

Downregulation of fibrosis-related proteins in TCFs by CTGF knockdown. Protein expression of p38, phospho-p38, Bcl-2, Cyclin D1, MMP2, collagen I and NF-κB p65 in human TCFs transduced with LV-shRNA-CTGF or control lentivirus was assessed by Western blots. Protein expression in blank control cells was also examined. The expression of GAPDH was used for normalization to ensure uniform protein loading in each lane. **P < 0.01; significantly different from control lentivirus-transduced group.

Discussion

Trabeculectomy is often required when medication fails to control IOP adequately. Although this surgical procedure has an immediate effect on reducing IOP, the postoperative scarring at the level of the subconjunctival Tenon's capsule often impairs its long-term success.¹ Previous studies have shown that human TCFs located in the incision area play a major role in scar formation via the proliferation, migration, and synthesis of ECM.^21,22 This process is mediated by various growth factors and cytokines that activate the profibrotic signaling pathways. In this regard, our study verified CTGF as a therapeutic target for inhibition of profibrotic signaling and phenotypes in TCFs. 

Connective tissue growth factor, a member of the CCN family of matricellular proteins, is highly expressed during wound healing and mediates the pathogenesis of many fibrotic disorders. Previous studies have shown that this cytokine could regulate cell survival,^23,24 proliferation,²⁵ adhesion,²⁶ migration,²⁷ and ECM production in different biological contexts.²⁸ Depending on the cell type, CTGF is also a contextual modulator of a diverse array of other biological and pathological processes, including angiogenesis,²⁹ chondrogenesis,³⁰ osteogenesis and even tumorigenesis.^31,32 Concordantly, our results also demonstrated that CTGF knockdown in TCFs produced drastic phenotypic changes—including inhibition of cell proliferation, adhesion, migration, and invasion—that are relevant to the suppression of fibrosis. Transforming growth factor β and endothelin-1 are two important upstream regulators of CTGF. The former induces CTGF through a complex network of transcriptional interactions requiring Smads, protein kinase C and the Ras/MEK/ERK/Ets-1 pathway,^33,34 whereas the latter acts through JNK and ERK to induce the AP-1 components c-Fos and c-Jun.^35,36 Recently, the regulation of cell proliferation and ECM deposition by CTGF in relation to TGF-β was documented in human TCFs.^37,38 In this regard, targeting CTGF might offer the potential to block only the scarring effects of TGF-β while preserving other important aspects of TGF-β biology. 

Fibroblasts are a rich source of inflammatory cytokines and chemokines that provide important intercellular signals in the microenvironment during tissue injury.^39,40 In this regard, CCN proteins, including CTGF, are considered to play a pivotal role in the promotion of inflammation.⁴¹ The relationship between inflammation and wound healing is complex. While inflammatory response is important for protection against wound infection, timely resolution of inflammation is required for regeneration and scarless healing.^42,43 To this end, CTGF has been shown to activate NF-κB to increase the expression of chemokines (MCP-1 and RANTES) and cytokines (INF-γ, IL-6, and IL-4) that promote the recruitment immune cells.^44,45 Concordantly, our results demonstrated that targeting CTGF in TCFs reduced NF-κB p65 subunit expression and suppressed the release of IL-1, IL-6, TGF-β1, IFN-γ and MCP-1. Functionally, targeting CTGF in TCFs abrogated the chemotaxis of macrophages. In this connection, depletion of macrophages at the initial phase of wound healing has been shown to reduce scar formation.^46,47 

Aside from inflammation, excessive ECM deposition is another important aspect of fibrotic wound healing. In this regard, our findings indicated that targeting CTGF could reduce the expression of collagen I and MMP-2 in TCFs. Nagai and colleagues reported that CTGF could promote the expression of laminin (another ECM component) via p38 activation in retinal pigmented epithelial cells.³⁴ Consistently, we observed a reduction of total and phosphorylated p38 in CTGF-knocked down TCFs. It would be interesting to investigate if p38 is upstream of collagen I in the action of CTGF in human TCFs. Importantly, application of p38 inhibitor has been shown to significantly improve bleb characteristics and IOP control after trabeculectomy in a rabbit model.⁴⁸ It would be encouraging if lentivirus targeting CTGF could be applied in vivo in future studies and achieve the same therapeutic effect. Although some people are concerned with the biological safety of lentivirus, this approach can effectively deliver shRNA sequence into target cells and mediate RNA interference effect for a prolonged period time. The degree of cellular damage was also minimal. 

Taken together, our study demonstrated an orchestrated repression of profibrotic signaling and phenotypes (Fig. 7) upon lentiviral delivery of CTGF-targeting shRNA into TCFs. Our findings not only further our understanding of the molecular mechanism underlying scarring in Tenon's capsule, but also enable clinicians to design and implement novel treatment strategies based on wound healing and scar biology. 

Figure 7

 

A proposed model of CTGF regulation of TCF phenotypes and the expression fibrosis-related signaling mediators and cytokines. Knockdown of CTGF leads to decreased levels of phospho-p38, MMP2, Cyclin D1 and Collagen I, resulting in lowered TCF proliferation, adhesion, migration, invasion and ECM deposition. Downregulation of CTGF also suppressed NF-κB to reduce the release of fibrosis-related cytokines from TCFs and thereby inhibiting macrophage chemotaxis. The overall effect is suppression of fibrosis and scarring.

Acknowledgments

Supported by Heilongjiang Province Science Foundation for Youths (Grant No. QC08C97); the National Natural Science Funds for Young Scholar (Grant No. 81400394); the Overseas Scholars Foundation of Heilongjiang Educational Committee (Grant No. 1254HQ013); the Foundation of Heilongjiang Educational Committee (Grant No. 12531283); the Research Fund for the Postdoctoral Program of Heilongjiang Province of China (Grant No. 2010-2012); and the Research Fund for the Doctoral Program of the Second Affiliated Hospital of Harbin Medical University (Grant No. BS2008-23). 

Disclosure: D. Lei, None; C. Dong, None; W.K.K. Wu, None; A. Dong, None; T. Li, None; M.T.V. Chan, None; X. Zhou, None; H. Yuan None