Expression and Function of Toll-like Receptor-3 and -9 in Human Corneal Myofibroblasts

Abstract

purpose. To investigate the expression and function of toll-like receptor (TLR)-3 and -9 in corneal myofibroblasts.

methods. Two types of human keratocytes were used, which were freshly isolated keratocytes from donor corneas and cultured keratocytes. Expression of the mRNAs for various molecular markers was analyzed in these cells by RT-PCR, and TLR-2, -3, -4, and -9 mRNAs were also analyzed by RT-PCR. Expression of TLR-3 and -9 at the protein level was assessed by flow cytometry. In addition, an antibody array and ELISA were used to detect chemokines and cytokines in the supernatant of cultured keratocytes, with or without stimulation by poly inosine-polycytidylic acid (poly (I:C)) or CpG-DNA. Furthermore, a phagocytosis assay was performed to evaluate whether signaling via TLR-3 and -9 enhances phagocytosis.

results. Keratocytes cultured for three passages underwent differentiation into corneal myofibroblasts. TLR-3 and -9 were detected in corneal myofibroblasts at the mRNA and protein levels, but not in freshly isolated keratocytes. Stimulation of corneal myofibroblasts with poly (I:C) or CpG-DNA enhanced the production of IL-6, IL-8, GRO, ENA-78, and RANTES compared with that by untreated cells. Phagocytic activity of myofibroblasts was upregulated by signaling via TLR-3 and -9.

conclusions. This is the first report on the in vitro expression and function of TLR-3 and -9 in corneal myofibroblasts. The findings suggest that the keratocyte phenotype determines the expression of TLR-3 and -9 and that corneal myofibroblasts may have an important role in bacterial and viral clearance.

The corneal epithelium is the first line of defense against invading organisms. Physical or chemical injury of the cornea can impair the integrity of this epithelial barrier, allowing the entry of organisms and microbial products into the underlying epithelium and the stroma. The corneal stroma consists of abundant extracellular populations of keratocytes and is the second line of defense against invading organisms. In normal corneal tissue, keratocytes are quiescent cells that are derived from the neural crest and have a flattened and stellate morphology. When the integrity of the corneal stroma is disrupted, these quiescent keratocytes differentiate into fibroblasts and/or myofibroblasts, which then proliferate, migrate to the site of injury, and repair the stroma. ^{1 2 3 4 5 6 7} Corneal myofibroblasts differ from keratocytes with respect to morphology, proliferation rate, phagocytic activity, extracellular matrix production, and the expression of various gene products such as CD34, keratocan, and α-smooth muscle actin (α-SMA). ^{8 9 10 11 12 13 14 15 16 17 18} The myofibroblast phenotype is defined by both the shape of these cells and the detection of α-SMA in the cytoplasm. It is well known that fibroblasts/myofibroblasts have a crucial role in corneal defenses. For example, corneal myofibroblasts can engulf bacteria and other foreign bodies, and Lande et al. ¹⁹ reported that human corneal myofibroblasts show significantly more phagocytic activity than do human skin fibroblasts or rabbit chondrocytes. Corneal myofibroblasts also promote contraction of the extracellular matrix after the integrity of the corneal stroma is disrupted. ²⁰ These findings suggest that the phagocytic and contractile properties of corneal fibroblasts/myofibroblasts may be involved in the initial response of this avascular tissue to invading organisms. Recently, Kumagai et al. ²¹ demonstrated that cultured corneal fibroblasts/myofibroblasts expressed mRNA for toll-like receptor (TLF)-4, which is a receptor for lipopolysaccharide (LPS) produced by Gram-negative bacteria. Stimulation with LPS increases the production of IL-8 and monocyte chemotactic protein (MCP)-I by corneal fibroblasts/myofibroblasts. Accordingly, corneal fibroblasts and myofibroblasts may play an important role in corneal defenses by recognizing LPS and producing chemokines that attract leukocytes. 

Although the human TLRs were discovered only 15 years ago, these receptors have since been found to occupy center stage in the innate immune response to invading pathogens. ²² To date, at least 12 TLRs (TLR-1 to -12) have been described. Each TLR acts as the primary sensor of conserved microbial components and drives the induction of specific biological responses. ^{23 24 25} For example, peptidoglycan, lipopolysaccharide, and flagellin are recognized by TLR-2, -4, and -5, respectively, ^{22 25} whereas TLR-3 is involved in the recognition of viral components like double-stranded RNA (dsRNA) and poly (I:C) (poly inosine-polycytidylic acid). ^{26 27 28} In addition, TLR-7 and -8 are involved in the recognition of single-stranded RNA (ssRNA) viruses and ssRNA, ^{29 30 31} whereas TLR-9 recognizes viral and bacterial DNA that has not been methylated at CpG motifs. ^{22 24 32 33 34 35} Recently, Johnson et al. ³⁶ demonstrated that the corneal epithelium expresses functional TLR-9. Activation of TLR-9 stimulates neutrophil recruitment to the corneal stroma in mice and causes a significant increase in corneal thickness and opacity. In addition, Huang et al. ³⁷ have shown that signaling via TLR-9 is important in P. aeruginosa keratitis and in the clearance of bacteria from the cornea. These two reports suggest that TLR-9 has a crucial role in fighting bacterial infection of the cornea. However, there have been no in vitro investigations into the expression and function of TLR-9 in keratocytes or corneal myofibroblasts. To our knowledge, the expression and function of TLR-3 in keratocytes or corneal myofibroblasts also has not been studied. Therefore, we performed the present study to investigate the in vitro expression and function of TLR-3 and -9 in keratocytes and corneal myofibroblasts. 

Materials and Methods

Antibodies and Ligands

Mouse anti-human TLR-3 monoclonal antibody (mAb; IgG₁) was purchased from HyCult Biotechnology b.v. (Uden, The Netherlands) and mouse anti-human TLR-9 mAb (clone 26C593; IgG₁) from Oncogene Research Products (San Diego, CA). These mAbs were used for flow cytometry. A TLR-3 agonist, poly (I:C), was obtained from Calbiochem (La Jolla, CA), and two TLR-9 agonists (CpG-DNA motifs and non-CpG-DNA motifs) were purchased from HyCult Biotechnology b.v. 

Cell Culture

This study was conducted in accordance with the Declaration of Helsinki. Corneas were obtained from the Rocky Mountain Lions’ Eye Bank at 3 to 5 days after the death of the donors (aged 62–70 years) and were maintained in storage medium (Optisol GS; Bausch & Lomb, Rochester, NY) at 4°C until use. Corneas with focal or diffuse stromal opacity or those from patients with a history of corneal disease in the eye bank report were not used. For isolation of the corneal stroma, the epithelium and Descemet’s membrane with the attached endothelium were peeled away in a sheet from the periphery to the center of the cornea by using fine forceps according to a procedure described previously. ³⁸ Before isolation of corneal keratocytes, the peripheral cornea (including the limbal region) was dissected from the stroma to avoid possible contamination by corneal limbal epithelial cells. Next, the stroma was cut into small pieces (∼1 mm in diameter) that were incubated overnight at 37°C in serum-free basal medium containing 0.02% collagenase (Sigma-Aldrich, St. Louis, MO). After the cells were washed three times with phosphate-buffered saline (PBS), a single-cell suspension was obtained by trituration with a fire-polished Pasteur pipette. For flow cytometric analysis, three to five corneas were processed simultaneously. 

Isolated keratocytes were cultured in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal calf serum (FCS). After three passages, the keratocytes were grown to subconfluence in DMEM with 10% FCS, resulting in their differentiation into corneal myofibroblasts. 

Preparation of RNA and RT-PCR

Total RNA was obtained from freshly isolated keratocytes and cultured keratocytes (TRIzol; Invitrogen, San Diego, CA). Reverse transcription was performed (SuperScript First-Strand Synthesis System; Invitrogen). Approximately 0.3 μg of total RNA was used in each reverse transcription reaction. Amplification was performed in a final volume of 20 μL using 5 U/μL of Taq DNA polymerase, 10× PCR buffer (Mg²⁺ free), 50 mM MgCl₂, 10 mM dNTP mix, and 10 μM of each primer (Table 1) . The PCR reaction involved initial denaturation at 94°C for 2 minutes, followed by denaturation at 94°C for 60 seconds, annealing at 54°C and 60°C for 60 seconds, and elongation at 72°C for 60 seconds The reaction was continued for 30 to 35 cycles and was completed by extended elongation at 72°C for 10 minutes After PCR amplification, 10 μL of each PCR product was subjected to electrophoresis on 1.5% agarose gel and the gel was stained with ethidium bromide (0.5 mg/mL). The PCR primers are listed in Table 1 . These primers were selected to discriminate between cDNA and genomic DNA by using specific primers for different exons. The amount of cDNA obtained from freshly isolated or cultured keratocytes was adjusted relative to the band density of glyceraldehyde-3-phosphate dehydrogenase (G3PDH). The amplification curve of the PCR products was drawn from data obtained at four-cycle intervals. Within the linear range of amplification, band densities were compared between different treatments of the keratocytes. 

G3PDH and human TLR-2, -3, or -4 dual PCR kits (Maxim; Biotech, Inc. South San Francisco, CA) were used to detect TLR-2, -3, and -4 mRNA in keratocytes or corneal myofibroblasts, respectively, according to the manufacturer’s instructions. In the dual PCR system, each target gene was coamplified with a specific housekeeping gene to achieve quantitative amplification in a single tube without interfering with the expression of target genes. A human toll-like receptor-9 primer set kit (Maxim, Biotech Inc.) was used to detect TLR-9 mRNA in keratocytes or corneal myofibroblasts. The primer for TLR-3 yielded a 431-bp product and the primer for G3PDH gave a 563-bp product, while the products for TLR-9, -2, and -4 were 260, 302, and 631 bp, respectively. 

Flow Cytometry

To detect the expression of TLR-3 and -9 protein in corneal myofibroblasts, we used flow cytometry (FACScan; BD Biosciences, San Jose, CA). After they were washed with buffer (FACS PBS [pH 7.4], 0.5% BSA, and 0.02% sodium azide), 10⁶ cells were treated with Fc-block (BD Biosciences) for 15 minutes and then were incubated with mAbs targeting human TLR-3 or -9 or with isotype control mouse IgG (BD-PharMingen, San Diego, CA) for 1 hour at room temperature. Cells were then washed twice with the buffer and incubated for 30 minutes with FITC-conjugated anti-mouse IgG (BD PharMingen). Finally, the cells were washed twice more with the buffer and analyzed. To gate out dead cells, staining with a kit containing propidium iodide (PI) was performed according to the manufacturer’s instructions (BD Biosciences). For analysis of intracellular expression, a cell fixation-permeabilization kit (BD PharMingen) was used. Cells were fixed (Cytofix/Cytoperm; BD Biosciences) and then stained with the respective mAbs. Data were analyzed with the accompanying software (Cell Quest; BD Biosciences). 

Antibody Array Technique

The conditioned medium obtained from cultures of corneal myofibroblasts was analyzed with an antibody array (Ray Bio; Human Angiogenesis Antibody I kit; RayBiotech Inc., Norcross, GA) according to the manufacturer’s instructions. 

This kit could simultaneously detect 20 different factors, including chemokines, cytokines, soluble cytokine receptors, and growth factors. 

The antibody array was placed into an eight-well tray, washed twice with Tris-buffered saline, and incubated for 30 minutes at room temperature with 2 mL/well of 1× blocking buffer. Then, 1 mL of supernatant from the myofibroblasts cultured in a six-well plate (35 mm in diameter) was added to each array and incubation was performed overnight at 4°C. After the samples were decanted, the antibody arrays were washed three times (for 5 minutes each) with 2 mL of 1× washing buffer I at room temperature, followed by washing twice (for 5 minutes each) with 1× washing buffer II at room temperature. Then, 1 mL of a 1:250 dilution of biotinylated antibody was added to each array and was incubated overnight at 4°C with shaking. After further washing, the array was incubated for 1 hour at room temperature with 2 mL/well of a 1:1000 dilution of HRP-conjugated streptavidin. After a thorough washing, each membrane was incubated at room temperature for 5 minutes with a mixture of 1× detection buffers C and D. The arrays were then exposed to autoradiograph film (BioMax Light film; Eastman Kodak, Tokyo, Japan) for 1 minute before processing. The signal intensity of individual spots was determined by densitometry with analyzer software (Gel-Pro; Media Cybernetics, Inc., Silver Spring, MD). 

Cytokine and chemokine production by cells incubated with or without CpG-DNA or poly (I:C) stimulation was detected after the cells had been washed twice with PBS and cultured in serum-free DMEM. The culture supernatant was harvested after 24 hours of incubation and was analyzed by using the antibody array. 

Enzyme-Linked Immunosorbent Assay

To detect growth-related oncogene (GRO), IL-6, IL-8, regulated on activation, normal, T-cell–expressed and secreted (RANTES), and epithelial cell-derived neutrophil activating peptide (ENA)-78 in the conditioned medium of corneal myofibroblasts, we used ELISA kits according to the manufacturer’s instructions (Quantikine; R&D Systems, Minneapolis, MN). Corneal myofibroblasts were grown to subconfluence in DMEM with 10% FCS, washed twice with PBS, and then incubated in serum-free DMEM basal medium for 24 hours with or without exposure to poly (I:C) or CpG-DNA or non-CpG-DNA. Then, the supernatant was harvested for analysis by ELISA. 

Phagocytosis Assay

To determine the influence of TLR-3 and -9 on phagocytosis, we pretreated corneal myofibroblasts with poly (I:C) or CpG DNA and measured the uptake of fluorescein-conjugated particles derived from Escherichia coli (K-12 strain; Invitrogen). The phagocytosis assay was performed with a kit (Vybrant; Invitrogen), according to the manufacturer’s instructions, and was performed in serum-free medium, to eliminate the contribution of Fc and/or complement receptors. Corneal myofibroblasts were washed twice in serum-free medium without antibiotics before being challenged with the particles. The optimum time for measurement of phagocytosis was determined to be 45 minutes after particle challenge. At this time, the myofibroblasts were washed with cold PBS to stop additional particle uptake or destruction of particles in the phagolysosomes. Finally, the fluorescence intensity of the myofibroblasts was estimated with a fluorescence plate reader (Fluoroskan-Ascent; Labsystems, Vantaa, Finland). 

Statistical Analysis

Results were expressed as the mean ± SE. Differences were evaluated by Student’s t-test (Excel software; Microsoft, Redmond, WA). 

Results

Expression of CD 34, Keratocan, α-SMA, TGF-β₁, TGF-β₂, TGF-βRI, and TGF-βRII mRNA by Freshly Isolated Keratocytes and Cultured Keratocytes

Keratocytes that had been isolated from donor corneas showed strong expression of the mRNAs for CD34, keratocan, and TGF-βRII, while expression of α-SMA and TGF-β₁ mRNAs was very weak. Expression of mRNAs for TGF-β₂ and TGF-βRI was not detected. In contrast, keratocytes cultured for three passages with 10% FCS showed strong expression of α-SMA, TGF-β₁, TGF-β₂, TGF-βRI, and TGF-βII mRNAs, but CD34 and keratocan mRNA expression was weak (Fig. 1) . Decreased expression of CD34 and keratocan with increased expression of α-SMA, TGF-β1, TGF-β2, and TGF-βRI were observed in cultured keratocytes. These results indicate that the keratocytes had differentiated into corneal myofibroblasts during culture in medium containing FCS. 

Expression of TLR-2, -3, -4, and -9 mRNAs by Freshly Isolated Keratocytes and Corneal Myofibroblasts

Both types of cells expressed TLR-2 and -4 mRNAs. However, TLR-3 and -9 mRNAs were expressed only by corneal myofibroblasts and not by freshly isolated keratocytes (Figs. 2A 2B) . Upregulation of TLR-3 and -9 was observed along with the expression of α-SMA. These results indicated that TLR-3 and -9 were expressed only after keratocytes underwent differentiation into myofibroblasts. 

Flow Cytometric Analysis of TLR-3 and -9 Expression by Corneal Myofibroblasts

Cell surface expression of TLR-3 by corneal myofibroblasts was examined with flow cytometry, but no surface expression was detected. After permeabilization of the cells, however, flow cytometry showed intracellular expression of TLR-3. Surface expression of TLR-9 on corneal myofibroblasts was also examined by flow cytometry analysis, and weak expression was detected. In contrast, strong intracellular expression of TLR-9 was detected by flow cytometry after permeabilization of the cells (Fig. 3) . Cell surface expression of TLR-2 and -4 was not detected by flow cytometry (data not shown). Therefore, no further analysis of these TLRs was performed in this study. 

Chemokine Production in Corneal Myofibroblasts Detected by Antibody Array Analysis

To investigate production of chemokines and other factors by corneal myofibroblasts, we analyzed 20 factors by antibody array analysis. The appearance of representative arrays used to analyze the culture supernatant of corneal myofibroblasts is shown in Figure 4GRO, IL-8, and MCP-1 proteins were detected in corneal myofibroblast culture medium. The myofibroblasts also produced cytokines and various growth factors such as IL-6, tissue inhibitor of metalloproteinase (TIMP)-1, TIMP-2, and angiogenin. 

To examine whether signaling via TLR-3 or -9 enhanced chemokine/cytokine production, we compared myofibroblasts treated with CpG-DNA or poly (I:C) and untreated cells. The mean optical intensity of positive spots from the culture supernatants of cells incubated with CpG-DNA or poly (I:C) were compared with those from the culture supernatants of unstimulated cells. This showed that stimulation of myofibroblasts with CpG-DNA enhanced the production of GRO, IL-6, and IL-8 (Fig. 4A) . In contrast, stimulation with poly (I:C) enhanced the production of IL-6, IL-8, GRO, RANTES, ENA-78, and b-FGF (Fig. 4B) . 

Chemokine Production in Corneal Myofibroblasts Detected by ELISA

Corneal myofibroblasts produced GRO, IL-6, IL-8, RANTES, and ENA-78 constitutively, while stimulation by poly (I:C) enhanced the production of GRO, IL-6, IL-8, RANTES, and ENA-78 at 24 hours, but not at 6 hours. In contrast, stimulation of corneal myofibroblasts with CpG-DNA induced GRO, IL-6, and IL-8 production, but RANTES and ENA-78 production remained the same as that of untreated cells at both 6 and 24 hours. These results corresponded to those obtained by antibody array analysis (Fig. 5) . 

Poly (I:C) and CpG-DNA Enhancement of Phagocytosis by Corneal Myofibroblasts

To examine whether TLR-3 or -9 signaling regulates phagocytosis, we compared myofibroblasts incubated with CpG-DNA or poly (I:C) and cells without treatment. Corneal myofibroblasts stimulated with CpG-DNA or poly (I:C) showed much greater uptake of fluorescein-conjugated particles than did untreated cells, as indicated by their higher fluorescence intensity. Figure 6shows that CpG-DNA or poly (I:C) pretreatment resulted in a marked increase in the phagocytosis of fluorescent particles. 

Discussion

In this study, we found that three passages of keratocytes in culture medium containing 10% FCS caused a decrease in the expression of CD34 and keratocan, whereas that of α-SMA was increased. CD34 and keratocan are keratocyte markers, whereas α-SMA is a marker of myofibroblasts. These results indicate that the cultured keratocytes were activated and underwent differentiation into myofibroblasts. It is already well known that TGF-β promotes keratocyte differentiation into myofibroblasts. ^{39 40 41 42 43 44 45} In this study, when keratocytes were cultured in medium with 10% FCS, these cells spontaneously expressed TGF-β₁ and -β₂ mRNAs, as well as showing increased expression of α-SMA mRNA. Furthermore, TGF-βRI was more highly expressed in myofibroblasts than in keratocytes. Our findings may indicate that keratocytes differentiate into myofibroblasts after these cells are stimulated to produce TGF-β. In this study, myofibroblasts showed expression of TLR-3 and -9 at both the mRNA and protein levels, but freshly isolated keratocytes did not. This observation is consistent with the concept that TLR-3 and -9 are expressed only when keratocytes differentiate into myofibroblasts. Why expression of TLR-3 and -9 is detected in myofibroblasts but not keratocytes remains unknown, but there may be a regulatory mechanism in keratocytes to avoid excessive inflammation that could lead to corneal opacity, as is the case in corneal epithelial cells. ^{46 47} However, when physical or chemical injury affects the cornea, the risk of bacterial or viral infection is increased, and so keratocytes may be activated and differentiate into corneal myofibroblasts that express TLR-3 and -9. 

TLR-3 is a receptor for dsRNA, which is a molecular marker of viral infection that is produced by most viruses at some point during the replication cycle. In this study, flow cytometry revealed that TLR-3 was expressed intracellularly by myofibroblasts, but not on the cell surface. This finding indicates that dsRNA binds to intracellular TLR-3. It is well known that activation of TLR-3 induces the activation of NF-κB and thus leads to the production of several inflammatory cytokines. ^{26 48} In our study, we showed that poly (I:C) could enhance myofibroblast secretion of IL-6, IL-8, GRO, RANTES, ENA-78, and b-FGF. It is known that IL-8, GRO, and ENA-78 belong to the C-X-C chemokine family and are strong neutrophil chemoattractants, whereas RANTES belongs to the C-C chemokine family and is a strong lymphocyte chemoattractant. These results suggest that chemokines produced by corneal myofibroblasts through TLR-3 signaling induce neutrophil or lymphocyte infiltration into the corneal stroma that clears viral infection. 

TLR-9 is a receptor for bacterial and viral CpG-DNA and has an important role in host defenses against infection. ^{36 37} In the present study, CpG-DNA enhanced the production of GRO, IL-6, and IL-8 by myofibroblasts. Such findings suggest that TLR-9 signaling, similar to TLR-3 signaling, induces myofibroblasts to release substances that promote neutrophil infiltration into the corneal stroma to clear bacteria and viruses effectively. However, CpG-DNA did not enhance the production of RANTES, ENA-78, or b-FGF. Although the exact mechanism is unknown, differences of the signal transduction pathway between TLR-3 and -9 may explain these differences of chemokine and growth factor expression. ⁴⁹ Recently, Latz et al. ⁵⁰ reported that TLR-9 was localized on the endoplasmic reticulum in dendritic cells and macrophages. In this study, flow cytometric analysis revealed that TLR-9 was weakly expressed on the cell surface, but showed strong intracellular expression. Recently, Barton et al. ⁵¹ reported that TLR-9 localized on the cell surface responded normally to synthetic TLR-9 ligands, but not to viral nucleic acids, while recognition of the nucleic acid in viral capsids was performed by intracellular TLR-9. Therefore, strong intracellular expression of TLR-9 in myofibroblasts may play an important role in the recognition of bacterial or viral nucleic acids. 

Before intracellular TLR-3 and -9 in myofibroblasts can bind with poly (I:C) and CpG-DNA, these products must undergo phagocytosis. Recently, Blander and Medzhitov ⁵² demonstrated that activation of the TLR-2 and -4 signaling pathways by bacteria regulates multiple steps of phagocytosis, including internalization and phagosome maturation, whereas Doyle et al. ⁵³ reported that numerous TLR ligands specifically enhance macrophage phagocytosis. To examine whether TLR-3 or -9 signaling could regulate phagocytosis, we pretreated myofibroblasts with CpG DNA or poly (I:C) and measured the uptake of fluorescent particles derived from E. coli. We found that TLR-3 or -9 activation led to increased uptake of the particles, suggesting that TLR-3 and -9 may play an important role in promoting the clearance of bacteria or viruses through upregulation of myofibroblast phagocytic activity and induction of chemokine production that promotes neutrophil infiltration into the corneal stroma. 

In summary, we demonstrated that human corneal myofibroblasts, but not keratocytes, could respond to poly (I:C) or CpG-DNA due to intracellular expression of TLR-3 and -9 Activation of TLR-3 or -9 caused an increase in the production of various chemokines, whereas ligands for TLR-3 or -9 enhanced phagocytosis by myofibroblasts. These results may suggest that TLR-3 and -9 in corneal myofibroblasts play a vital role in the inflammatory response that clears viruses or bacteria from the corneal stroma. 

 Submitted for publication August 14, 2006; revised January 12 and February 28, 2007; accepted May 7, 2007.

 Disclosure: N. Ebihara, None; S. Yamagami, None; L. Chen, None; T. Tokura, None; M. Iwatsu, None; H. Ushio, None; A. Murakami, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Primers for Detecting mRNA Expression

Gene	Primers	Annealing Temp (°C)	Product Length (bp)	Cycling
3PDH	F: ACCACAGTCCATGCCATCAC	60	452	38
	R: TCCACCACCCTGTTGCTGTA
CD-34	F: GCAAGCCACCAGAGCTATTC	54	195	38
	R: GCAAGCCACCAGAGCTATTC
Keratocan	F: ATGGCAGGCACAATCTGTTTCATC	60	1095	38
	R: TTAAATAATGACAGCCTGCAGAAG
a-SMA	F: CCAGCTATGTGAAGAAGAAGAGG	60	965	38
	R: GTGATCTCCTTCTGCATTCGGT
TGF-b1	F: CTCCGAGAAGCGGTACCTGAAC	60	288	38
	R: CACTTGCAGTGTGTTATCCCT
TGF-b2	F: ACTTTCTACAGACCCTACTT	60	405	38
	R: GCCATCAATACCTGCAAATC
TGF-bRI	F: ACGGCGTTACAGTGTTCTG	63	358	38
	R: GGTGTGGCAGATATAGACC
TGF-bRII	F: AGCAACTGCAGCATCACCTC	55	688	38
	R: TGATGTCTGAGAAGATGTCC

Figure 1.

 

Expression of CD 34, keratocan, α-SMA, TGF-β₁, TGF-β₂, TGF-βRI, and TGF-βRII mRNA by freshly isolated keratocytes and cultured keratocytes. The quantity of cDNA4 from the freshly isolated keratocytes and cultured keratocytes was adjusted by the band density of glyceraldehyde-3-phosphate dehydrogenase (G3PDH). Within the linear range of amplification, the band densities were compared between the both groups. Keratocytes isolated freshly from donor corneas (A) strongly expressed the mRNA for CD 34, keratocan, and TGF-βRII, whereas the expression of α-SMA and TGF-β₁ mRNA was very weak. Expression of mRNA for TGF-β₂ and TGF-βRI was not detected. In contrast, keratocytes cultured for three passages in medium with 10% FCS (B) showed strong expression of α-SMA, TGF-β₁, TGF-β₂, TGF-βRI, and TGF-βII, but CD 34 and keratocan were expressed weakly. (A) freshly isolated keratocytes; (B) cultured keratocytes (three passages).

Figure 2.

 

Expression of TLR-2, -3, -4, and -9 mRNAs by freshly isolated keratocytes and cultured keratocytes. Both types of cells expressed TLR-2 and -4 mRNAs. However, TLR-3 and -9 mRNAs were expressed only by cultured keratocytes, not by freshly isolated keratocytes. (A) freshly isolated keratocytes; (B) cultured keratocytes (three passages). C, positive control; S, sample; G3PDH, 563 bp; TLR2, 302 bp; TLR3, 431 bp; TLR4, 631 bp; TLR9, 260 bp.

Figure 3.

 

Flow cytometric analysis of TLR-3 and -9 expression by corneal myofibroblasts. Cell surface expression of TLR-3 by corneal myofibroblasts was examined using flow cytometry, but the surface expression of TLR-3 was not detected. After permeabilization of the cells, however, flow cytometry showed intracellular expression of TLR-3. Surface expression of TLR-9 on corneal myofibroblasts was also examined by flow cytometry, and weak expression was detected. In contrast, strong intracellular expression of TLR-9 was detected by flow cytometry after permeabilization of the cell. Black trace: data from control mouse IgG; red trace: data with anti-TLR-3 or -9 mAb.

Figure 4.

 

Chemokine production in corneal myofibroblasts detected by antibody array analysis. The appearance of representative arrays for factors in the culture supernatant of corneal myofibroblasts are shown in (A) and (B). GRO, IL-8, and MCP-1 proteins were detected in corneal myofibroblast culture medium. The corneal myofibroblasts also produced cytokines and growth factors such as IL-6, tissue inhibitor of metalloproteinase (TIMP)-1, TIMP-2, and angiogenin. To examine whether signaling via TLR-9 or -3 enhanced chemokine-cytokine production, we compared myofibroblasts treated with CpG-DNA or poly (I:C) and cells without any treatment. The mean optical intensity of positive spots from the culture supernatants of cells incubated with CpG-DNA or poly (I:C) were compared with those from the culture supernatants of unstimulated cells. Stimulation of myofibroblasts with CpG-DNA enhanced the production of GRO, IL-6, and IL-8 (A). In contrast, stimulation with poly (I:C) enhanced the production of IL-6, IL-8, GRO, RANTES, ENA-78, and b-FGF (B).

Figure 5.

 

Chemokine production in corneal myofibroblasts detected by ELISA. Corneal myofibroblasts produced GRO, IL-6, IL-8, RANTES, and ENA-78 constitutively, and stimulation by poly (I:C) enhanced GRO, IL-6, IL-8, RANTES, and ENA-78 production at 24 hours, but not 6 hours. In contrast, stimulation of corneal myofibroblasts with CpG-DNA induced GRO, IL-6, and IL-8 productions, but RANTES and ENA-78 production of CpG-DNA-treated cells remained the same as that of untreated cells at 6 and 24 hours.

Figure 6.

 

CpG-DNA and poly (I:C) enhanced phagocytosis by corneal myofibroblasts. To examine whether TLR-9 or -3 signaling regulates phagocytosis, we compared myofibroblasts incubated with CpG-DNA or poly (I:C) to cells without treatment. Corneal myofibroblasts stimulated with CpG-DNA or poly (I:C) showed much uptake of fluorescein-conjugated particles compared with untreated cells, as indicated by their higher fluorescence intensity. The vertical axis indicates the average percentage intensity relative to the control.