Effects of TGF-β2 on Immune Response–Related Gene Expression Profiles in the Human Corneal Endothelium

Abstract

purpose. To determine the effects of transforming growth factor (TGF)-β2 on immune-response–related gene expression profiles in the stimulated human corneal endothelium (HCE).

methods. A human complementary DNA (cDNA) expression array analysis was used to investigate the effects of TGF-β2 on cultured HCE incubated with interleukin (IL)-1α and tumor necrosis factor (TNF)-α. Gene-specific semiquantitative reverse transcription–polymerase chain reaction (RT-PCR) and enzyme-linked immunosorbent assay (ELISA) were performed to examine the gene expression patterns revealed by the cDNA expression array analysis. Moreover, the expression of newly identified genes in HCE was confirmed by RT-PCR in human donor corneas.

results. cDNA expression array analysis and semiquantitative RT-PCR revealed that TGF-β2 downregulated the expression of IL-6, growth-related (Gro)-α (CXCL1), monocyte chemotactic protein (MCP)-1 (CCL2), granulocyte-colony stimulating factor (G-CSF), and insulin-like growth factor binding protein (IGFBP)-5 and upregulated the expression of tissue inhibitor of metalloproteinase (TIMP)-1. ELISA confirmed TGF-β2–mediated changes in the expression of IL-6, CXCL1, CCL2, G-CSF, IGFBP-5, and TIMP-1 at the protein level. CXCL1, G-CSF, and IGFBP-5 mRNAs were detected for the first time in the HCE of donor corneas.

conclusions. TGF-β2 downregulates IL-6, CXCL1, CCL2, G-CSF, and IGFBP-5, and upregulates TIMP-1 in cultured HCE stimulated with proinflammatory cytokines, suggesting the immunomodulatory role of TGF-β2 in the aqueous humor and the pathophysiological significance of TGF-β2 in the anterior chamber of the eye.

The corneal endothelium (CE) faces the aqueous humor (AH) as a part of the inner side of the anterior chamber of the eye. The CE constitutively expresses various genes to maintain fundamental corneal functions, such as stromal dehydration, corneal transparency, metabolic activity, and signal transduction. ^{1 2} Moreover, ex vivo and cultured human CE (HCE) cells express abundant cytokine and chemokine genes by cytokine stimulation. ³ Because it has long been recognized that the anterior chamber of the eye has an immune privilege, analysis of the interaction between CE and AH can provide a theoretical rationale for the immune-privileged status in the anterior segment of the eye. 

Transforming growth factor (TGF)-β is a stable, multifunctional polypeptide growth factor. ⁴ Specific receptors for this protein have been found on various cells, but the effect of the molecule varies with the cell type and growth conditions. There are three isoforms of TGF-β. They have similar, but not identical, functions on target cells. ⁴ In the AH of the eye, various immunosuppressive substances, such as TGF-β, α-melanocyte stimulating hormone, ⁵ vasoactive intestinal peptide, ⁶ and macrophage migration inhibitory factor, ⁷ have the capacity to modulate immune responses. ⁸ Among them, abundant TGF-β2, which is produced in the ciliary epithelium, ⁹ is secreted into the normal AH ^{9 10} and is considered to play a critical role in the immune privilege of the anterior chamber. ¹¹ It is not fully understood, however, how TGF-β2 affects immunoreactive factors produced in the AH. 

In this study, we prepared HCE cells stimulated with interleukin (IL)-1 and tumor necrosis factor (TNF)-α, because the cornea synthesizes these proinflammatory cytokines in inflamed conditions such as uveitis, ^{12 13} herpetic stromal keratitis, ¹⁴ and corneal allograft rejection. ^{15 16} We then investigated the effects of the active form of TGF-β2 on the immune-response–related gene expression profiles in the HCE with a complementary DNA (cDNA) expression array analysis. Our findings revealed that TGF-β2 regulates immune-response–related genes in the HCE that are associated with the correlation between the HCE and AH in the anterior chamber of the eye. 

Materials and Methods

Cell Culture and Cytokine Treatment

Normal human donor corneas were supplied by the Central Florida Lions Eye and Tissue Bank. The HCE and Descemet membrane were peeled away in a sheet from the periphery to the center of the cornea with fine forceps. To avoid the inclusion of posterior stromal tissue, only strips of HCE-Descemet tissue that were excised smoothly from the stroma, without residual stroma, were used in the experiments. Primary HCE cells were cultured as described elsewhere. ¹⁷ Recombinant human IL-1α, TNF-α, and TGF-β2 (R&D Systems, Minneapolis, MN) were diluted with phosphate-buffered saline (PBS). Fifth-passage HCE cells were transferred to 2 mL of DMEM containing 1% fetal bovine serum in 50-mm culture dishes and then treated for 24 hours with 20 ng/mL each of human IL-1α and TNF-α (IL-1α/TNF-α group), 20 ng/mL each of human IL-1α and TNF-α plus 20 ng/mL of recombinant human TGF-β2 (TGF-β2 group), or PBS only (control group). Two independent culture dish sets were prepared. RNAs isolated from those groups were used for gene array analysis or semiquantitative RT-PCR. The supernatants were stored at −70°C until use in an enzyme-linked immunosorbent assay (ELISA). 

Human Immunology Gene Array

A human cDNA expression array (Toyobo, Osaka, Japan), in which 648 known immunology-related genes are represented, was used in this experiment. A complete list of the 648 immunology-related genes included in this human array can be found at http://www.toyobo.co.jp/seihin/xr/index.html. Total RNA was isolated using an RNA extraction reagent (Isogen; Nippon Gene, Tokyo, Japan) according to the manufacturer’s instructions. After DNase treatment, an α-³²P-labeled cDNA probe was synthesized from the total RNA. Each cDNA probe was purified (Chroma-spin columns; Clontech, Palo Alto, CA). Incorporation of the label was assessed by scintillation counting. Equal counts per minute of the cDNA probe for each group was hybridized in a hybridization solution (Express Hyb; Clontech) with the human gene immunology array membranes overnight at 68°C with continuous agitation. The array membranes were washed in wash solution 1 (2× SSC, 1% SDS) and then in wash solution 2 (0.1× SSC, 0.5% SDS) at 68°C. The array membranes were exposed to a phosphorescence imager (Eastman Kodak, Rochester, NY) at room temperature. Two sets of phosphorescent imaging results on separate culture experiments were analyzed (four hybridizations of four cDNA array membranes) and compared with those from another phosphorescence imager (Storm PhosphorImager; Amersham Biosciences, Sunnyvale, CA). The data obtained from the phosphorescence imaging analysis system reflects the accumulated x-ray energy and provides us with accurate quantification in comparison with densitometric analysis with spot size. After normalization to the intensity levels of all housekeeping genes included in the membrane, the average intensity differences among genes upregulated or downregulated in the two hybridizations were calculated (Vision Array, ver. 1.1 software; Imaging Research Inc., Ontario, Canada). 

RNA Preparation and RT-PCR

Total RNA was isolated from cultured HCE cells and from HCE peeled from donor corneas, by use of an extraction reagent (Isogen; Nippon Gene) according to the manufacturer’s instructions. Water was used as a negative control. After DNase treatment, first-strand cDNA was synthesized from the total RNA with a reverse-transcription system (Promega Corp., Tokyo, Japan). The PCR reaction mixtures comprised 1% cDNA, 10 mM Tris-Cl (pH 8.3), 50 mM KCl, 1.5 mM MgCl₂, 0.2 mM dNTPs, 20 pmol oligonucleotides, and 2.5 U Taq polymerase (AmpliTaq Gold; Perkin Elmer, Wellesley, MA), in a 50-μL reaction volume. After incubation at 95°C for 9 minutes, amplification was performed at 94°C for 30 seconds and then at 60°C for 30 seconds in a thermocycler (I Cycler; Bio-Rad Laboratories, Hercules, CA). Samples were separated in a 2% agarose gel, and the products were visualized with ethidium bromide. An optical scanner was used to determine the densities of the gel bands of the PCR products. The band densities were normalized to those for glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The linear amplified curve of the PCR product of each sample was examined at four-cycle intervals. Within the linear range of amplification, four sets of PCR products were prepared under appropriate cycling conditions, and the band densities were compared among the IL-1α/TNF-α, TGF-β2, and control groups. The control group was analyzed by RT-PCR and then ELISA, to reveal whether TGF-β2 antagonizes or enhances the effects of IL-1α/TNF-α. 

ELISA of Supernatants of Cultured HCE with or without TGF-β2 Stimulation

Culture supernatants were used for ELISA. The protein concentrations of IL-6, monocyte chemotactic protein 1 (MCP-1, CCL2), granulocyte-colony stimulating factor (G-CSF; Biosource International, Camarillo, CA), and Gro-α (CXCL1; R&D Systems) in the culture supernatant were measured with ELISA kits, according to the manufacturer’s protocol. The insulin-like growth factor binding protein (IGFBP)-5 concentration was measured by direct ELISA. Recombinant IGFBP-5 (R&D Systems) and diluted serum were incubated at room temperature for 2 hours in a 96-well coated plate. Polyclonal goat anti-human IGFBP-5 antibody (1 μg/mL; R&D Systems) was applied for 1 hour as a detection antibody after blocking was performed with 3% bovine serum albumin at room temperature for 30 minutes. A horseradish peroxidase (HRP)–conjugated rabbit anti-goat IgG (H+L) antibody (1 μg/mL; Bethyl Laboratories, Montgomery, TX) was incubated for 30 minutes at room temperature. The concentration of tissue inhibitor of matrix metalloproteinase (TIMP)-1 was quantitated with an ELISA kit (Duoset ELISA Development Kit; R&D Systems) and plates read with a microplate reader (Molecular Bioscience Group, Hercules, CA) at an optical density of 450 nm, with the correction wavelength set at 550 nm. 

Detection of CXCL1, G-CSF, and IGFBP-5 Expression in Ex Vivo HCE Cells from Donor Corneas

To confirm the expression of newly identified genes CXCL1, G-CSF, and IGFBP-5 expression in HCE cells, RT-PCR was performed in HCE of donor corneas immersed in storage medium (Optisol GS; Chiron Vision, Irvine, CA). The HCE was peeled off as a sheet and subjected to mRNA extraction, DNase treatment, and subsequent RT-PCR. Water was used as a negative control for RT-PCR. 

Statistical Analysis

The Fisher protected least-significant-difference test was used to compare the band densities on RT-PCR, and the protein concentrations on ELISA. P < 0.05 was considered significant. 

Results

Screening of Immune-Response–Related Gene Expression by cDNA Expression Array

A series of experiments was performed twice using RNA isolated from two separate cultures of either IL-1α/TNF-α or TGF-β2 group (four hybridizations of four cDNA array membranes). The far left and far right duplicate lanes of these membranes contained 11 housekeeping gene controls and two negative controls. No signals were observed for the negative control spots (pUC and luciferase), indicating that the hybridization was highly specific (Fig. 1) . Among the tested 648 immune-response–related genes, 15 genes were downregulated and 2 were upregulated in the IL-1α/TNF-α/TGF-β2– compared with the IL-1α/TNF-α–treated HCE cells. The average positive and negative intensity differences greater than 1.5 (in arbitrary units) between the IL-1α/TNF-α group and IL-1α/TNF-α plus TGF-β2 group are listed in Table 1 . TGF-β2 stimulation resulted in a significantly decreased expression of IL-6, Groα (CXCL1), CCL2, CXCL8, G-CSF, IGFBP-5, and CCAAT/enhancer–binding protein (C/EBP)-β and increased expression of a tissue inhibitor of matrix metalloproteinase (TIMP)-1 in the HCE (Fig. 1 ; Table 1 ). The average intensity differences greater than 7 (in arbitrary units) between the IL-1α/TNF-α and TGF-β2 groups were accepted as significant and applied for further study. 

Immune-Response–Related Gene Expression by Semiquantitative RT-PCR Analysis

To confirm the results of gene array analysis and examine whether TGF-β2 antagonizes or enhances effects of IL-1α and TNF-α, semiquantitative RT-PCR was performed. The oligonucleotide primers used for RT-PCR are listed in Table 2 . GAPDH was detected on RT-PCR in the control, IL-1α/TNF-α, and TGF-β2 groups. There were no significant differences in GAPDH among the groups (Fig. 2A) . After confirmation of linear amplification of the PCR products, as shown in Figure 2A , the gene expression levels were compared. Figure 2B shows that the mRNA levels of all eight genes examined in the IL-1α/TNF-α group were significantly higher than those in the control group. The mRNA levels of IL-6, CXCL1, CCL2, G-CSF, and IGFBP-5 in the TGF-β2 group were significantly lower than those in the IL-1α/TNF-α group. On the contrary, the gene expression level of TIMP-1 in the TGF-β2 group was significantly higher than that in the IL-1α/TNF-α group. These findings were consistent with the results of cDNA array analysis. There was no distinguishable difference in mRNA levels of CXCL8 and C/EBP-β between the IL-1α/TNF-α and TGF-β2 groups. Therefore, we further determined the protein concentrations of IL-6, CXCL1, CCL2, G-CSF, IGFBP-5, and TIMP-1 in the cultured HCE supernatant. 

Immune-Response–Related Protein Production

We next determined whether the change in IL-6, CXCL1, CCL2, G-CSF, IGFBP-5, and TIMP-1 mRNA expression results in altered protein production. Figure 3 shows that six proteins in the IL-1α/TNF-α group were significantly higher than those in the control group, consistent with the results of semiquantitative RT-PCR. The high amount of IL-6 produced in the IL-1α/TNF-α group was significantly suppressed in the TGF-β2 group. The levels of CXCL1, CCL2, G-CSF, and IGFBP-5 production in the TGF-β2 group were significantly lower than those in the IL-1α/TNF-α group. On the contrary, the level of TIMP-1 in the TGF-β2 group was significantly higher than that in the IL-1α/TNF-α group. 

CXCL1, G-CSF, and IGFBP-5 Expression in Ex Vivo HCE by RT-PCR Analysis

RT-PCR was performed to detect mRNA expression of the newly identified CXCL1, G-CSF, and IGFBP-5 gene expression in ex vivo HCE cells from donor corneas. Two donor corneas were examined separately. GAPDH was detected in ex vivo HCE of donor corneas, but not in a negative control sample. CXCL1, G-CSF, and IGFBP-5 mRNAs were all detected in ex vivo HCE cells of two donor corneas (Fig. 4) . 

Discussion

In the present study, we conducted a comprehensive examination of the effects of TGF-β2 on the expression of immune-response–related genes in HCE cells stimulated with the proinflammatory cytokines IL-1α and TNF-α, using a combination of cDNA array and semiquantitative RT-PCR analysis. We demonstrated that TGF-β2 downregulated the expression of IL-6, Groα (CXCL1), CCL2, G-CSF, and IGFBP-5, and upregulated the expression of TIMP-1. The number of TGF-β2–downregulated genes was higher than those upregulated by TGF-β2, consistent with the effect of TGF-β1 on the gene expression profiles in cultured human corneal epithelium. ¹⁸ Among the TGF-β2–downregulated genes, detected by cDNA array and RT-PCR analysis, the expression levels of IL-6, CXCL1, CCL2, G-CSF, and IGFBP-5 were also confirmed by ELISA analysis. The TIMP-1 gene and protein were upregulated with TGF-β2 treatment. Moreover, in addition to the expression of IL-6, ³ CCL2, ³ CXCL8, ¹⁹ and TIMP-1 ²⁰ in HCE, as previously described, our data showed for the first time that HCE expresses CXCL1, G-CSF, and IGFBP-5 mRNAs. Although the donor corneas were stored in a storage medium (Optisol GS; Chiron), and the removed HCE samples were acquired from the donor corneas within 10 days of death, the data may not reflect 100% of the in vivo expression, due to the possible effect of the storage medium. However, expression levels of these three mRNAs were enhanced by stimulation of HCE cells from ex vivo donor corneas with IL-1α and TNF-α (Yamagami S, unpublished observation, 2003), suggesting the similarity between the ex vivo and in vitro studies. 

The cornea, including CE, is an avascular tissue and requires rapid induction of an innate immune response to infection by chemoattractants for neutrophils. Both CXCL1 and CXCL8 in the HCE are chemoattractants for neutrophils. ^{21 22} G-CSF also has the ability to promote the maturation, mobilization, and adhesion of neutrophils. ²³ Previous studies have shown decreased chemotaxis to CXCL8 in polymorphonuclear cells (PMNs) isolated from G-CSFR–deficient mice, supporting a regulatory role of G-CSF in neutrophil chemokine responsiveness. ²⁴ However, neutrophils by themselves exhibit direct cytotoxicity toward FasL-bearing CE, and TGF-β in the AH suppresses the cytotoxicity of neutrophils toward FasL-bearing CE. ²⁵ Our findings that TGF-β2 downregulates neutrophil chemoattractants suggest that TGF-β can protect HCE not only through direct suppression of neutrophil cytotoxicity, but also through prevention of excessive neutrophil infiltration. 

G-CSF was originally discovered as a hematopoietic growth factor that stimulates the proliferation and maturation of neutrophil precursors. ²⁶ G-CSF attenuates the release capacity for inflammatory cytokines such as TNF-α, interferon-γ, and IL-2, when stimulated with lipopolysaccharide, suggesting an inhibitory effect of G-CSF on inflammatory immune responses. ^{27 28} Moreover, G-CSF inhibits some allogeneic immune responses on hematopoietic allograft rejection in mice. ^{29 30 31} Similar immunosuppressive effects on allogeneic immune responses have been observed in human G-CSF–mobilized peripheral blood mononuclear cells. ³² In addition, CD4 T cells exposed to G-CSF in vivo have the characteristics of regulatory T-cells possessing an intrinsic low proliferative capacity, and contact-independent suppression of antigen-driven proliferation. ³³ These findings indicate that G-CSF may, at least in part, contribute to the immune privilege of the anterior chamber of the eye in an inflammatory situation. The fact that TGF-β2 suppresses G-CSF expression in HCE cells, however, may suggest the existence of unidentified immunologic role of G-CSF in the anterior chamber environment in addition to the modulation of neutrophil infiltration. 

A family of IGF-binding proteins (IGFBP-1-6) modulates the multiple activities of IGF-I and -II. In many situations, IGFBP-5 shows biological activity in the absence of IGFs, indicating the existence of IGF-independent actions. ³⁴ IGFBP-5 is involved in the control of cell proliferation and differentiation, which depends on the cell species. ³⁴ IL-1α ³⁵ and IL-6 ³⁶ increase IGFBP-5 expression in ovine articular chondrocytes and osteoblast cultures, respectively. TGF-β has no effect on ovine articular chondrocyte IGFBP-5. ³⁵ In the anterior segment of the eye, IGFBP-5 has been detected in the rat ciliary process ³⁷ and the human trabecular meshwork. ³⁸ In our study, IGFBP-5 was detected in cultured and ex vivo HCE cells. The exact role of IGFBP-5 in HCE is, however, still unknown in the anterior chamber of the eye. Further study is needed to reveal the function of IGFBP-5 in the AH. 

IL-6 is a pleiotropic cytokine whose synthesis can be induced by IL-1α and TNF-α. IL-6 was detected in the AH of patients and rodents with uveitis, and thus it has been considered a major mediator of uveitis. ^{39 40 41} IL-6 is not by itself mitogenic for T-cell proliferation, but can antagonize the immunosuppressive effect of the active form of TGF-β2, which potently suppresses T-cell proliferation. ^{42 43} The suppression of IL-6 secretion with the active form of TGF-β2 in HCE implies that TGF-β2 can indirectly regulate IL-6 in the AH, although secreted IL-6 by itself antagonizes the immunosuppressive effect of TGF-β2. 

Matrix metalloproteinase (MMP) activity is controlled through interaction with TIMPs, principal natural inhibitors of MMPs. ⁴⁴ PMNs lead to tissue damage through the activation and secretion of MMPs, ⁴⁵ while promoting bacterial clearance. TIMP-1 prevents infection by protecting against tissue destruction, ⁴⁶ and TGF-β1 and -β2 have been reported to induce expression of the TIMP-1 gene. ^{47 48} Upregulation of TIMP-1 in the HCE by TGF-β2 in the AH may indicate a mechanism that minimizes the destructive damage in the HCE. 

It is worth commenting on the modulation of angiogenesis by cytokines, a central process in physiological and pathophysiological situations. Angiogenesis, a common feature of inflammatory, infectious, and traumatic diseases of the cornea, leads to visual impairment and failure of the immune privilege of the cornea. ⁴⁹ In the anterior chamber of the eye, TGF-β2 is regarded as a critical factor that protects the corneal tissue against abnormal vascularization. ⁵⁰ G-CSF ⁵¹ and CXCL1, ^{52 53 54} which are expressing abundantly in the HCE, exhibit potent angiogenic activity. On the contrary, TIMP-1 suppresses angiogenesis in the cornea. ⁵⁵ Our findings that TGF-β2 downregulates G-CSF and CXCL1, and upregulates TIMP-1 expression may explain the antiangiogenic effect of TGF-β2 in the AH. 

In summary, we investigated the effect of TGF-β2 on the immune-response–related genes expressed in the HCE, using cDNA array technology, semiquantitative RT-PCR, and ELISA. TGF-β2 downregulates IL-6, CXCL1, CCL2, G-CSF, and IGFBP-5, and upregulates TIMP-1 in cultured HCE stimulated with proinflammatory cytokines, indicative of immunomodulatory roles of TGF-β2 in the AH. Furthermore, we identified, for the first time, CXCL1, G-CSF, and IGFBP-5 expressed in cultured and ex vivo HCE cells. Our results have pathophysiological implications in the anterior chamber immunology of the eye. 

 Supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.

 Submitted for publication August 20, 2003; revised September 16, 2003; accepted September 26, 2003.

 Disclosure: S. Yamagami, None; S. Yokoo, None; T. Mimura, None; S. Amano, 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.

 

Figure 1.

 

Representative human immunology-related gene arrays hybridized with α-³²P-labeled cDNA probes synthesized from total RNA isolated from cultured HCE exposed to (A) 20 ng/mL each of IL-1α and TNF-α (IL-1α/TNF-α group) and to (B) 20 ng/mL each of IL-1α, TNF-α, and TGF-β2 (TGF-β2 group) for 24 hours. The TGF-β2–downregulated genes were IL-6, Groα (CXCL1), MCP-1 (CCL2), IL-8 (CXCL8), G-CSF, IGFBP-5, and C/EBP-β. The TGF-β2–upregulated gene was TIMP-1. 1: IL-6, 2: CXCL1, 3: CCL2, 4: CXCL8, 5: G-CSF, 6: IGFBP-5, 7: C/EBP-β, 8: TIMP-1. Data are representative of two independent experiments.

Table 1.

 

Genes Downregulated or Upregulated on TGF-β2 Treatment

Gene	Intensity Difference	Accession Number
Downregulated
Interleukin 6 (IL-6)	−53.0	M29150
CXCL1, growth-related α (GROα)	−38.4	NM001511
CCL2, monocyte chemotactic protein (MCP)-1	−19.6	X14768
CXCL8, IL-8	−15.0	Y00787
Granulocyte-colony stimulating factor (G-CSF)	−12.1	X03655
Insulin-like growth factor binding protein (IGFBP)-5	−8.0	NM00599
CCAAT/enhancer binding protein (C/EBP) β	−7.1	X52560
Siva	−2.3	U82938
Prolactin	−2.2	V00566
Indian hedgehog protein (Ihh)	−2.0	L38517
Fibroblast growth factor (FGF)-7	−1.8	M60828
IGFBP-4	−1.8	M62403
MMP2 (type IV collagenase, gelatinase A)	−1.7	J03210
Insulin receptor substrate (IRS)-2	−1.6	AF073310
Glucose-regulated protein (GRP) 94, endoplasmin	−1.5	X15187
Upregulated
Tissue inhibitor of matrix metalloproteinase (TIMP)-1	10.6	X03124
Extracellular signal-regulated kinase (ERK) 5	3.2	U25278

 The two sets of autoradiograph results were analyzed and compared. The average intensity differences greater than 1.5 (in arbitrary units) between the IL-1α/TNF-α–treated and IL-1α/TNF-α plus TGF-β2–treated groups are listed.

Table 2.

 

Oligonucleotide Primers for RT-PCR

Gene	PCR Primers	Product Size (bp)	GenBank Accession Number
GAPDH	5′-GGTGAAGGTCGGTGTGAACGGA-3′	223	P04406
	5′-TGTTAGTGGGGTCTCGCTCCTG-3′
IL-6	5′-AATTCGGTACATCCTCGACGG-3′	522	X04602
	5′-TGACCAGAAGAAGGAATGCCC-3′
CXCL1 (Groα)	5′-TTCACCCCAAGAACATCCAA-3′	306	NM001511
	5′-CTCCTAAGCGATGCTCAAACAC-3′
CCL2 (MCP-1)	5′-TCTCGCCTCCAGCATGAAA-3′	267	M24545
	5′-TCCTGAACCCACTTCTGCTTG-3′
CXCL8 (IL-8)	5′-AAGAGCCAGGAAGAAACCACC-3′	466	Y00787
	5′-ATTGCATCTGGCAACCCTACA-3′
G-CSF	5′-TGGCAGCAGATGGAAGAACT-3′	263	X03655
	5′-CGTTCTGCTCTTCCCTGTCTTT-3′
IGFBP-5	5′-ATCTCCGAGCTGAAGGCTGAA-3′	371	NM00599
	5′-TCGAAGGTGTGGCACTGAAAG-3′
C/EBPβ	5′-GGAGCCCGTCGGTAATTTT-3′	309	X52560
	5′-CAACAGCAACAAGCCCTAGAAC-3′

Figure 2.

 

Semiquantitative gene expression levels in the control, IL-1α/TNF-α and TGF-β2 groups. The linear amplified curve of the PCR product of each sample was examined at four-cycle intervals. PCR products were prepared under appropriate cycling conditions, and the band densities were compared. (A) There was no significant difference in GAPDH among the groups. (B) All gene expression levels in the IL-1α/TNF-α group were significantly higher than those in the control group. The expression levels of IL-6, CXCL1, CCL2, G-CSF, and IGFBP-5, but not CXCL8 and C/EBP-β, in the TGF-β2 group are significantly lower than those in the IL-1α/TNF-α group. The expression level of TIMP-1 in the TGF-β2 group is significantly higher than that in the IL-1α/TNF-α group. The data represent the mean ± SD for four sets of PCR products. Data are representative of two independent experiments. Con, Control group; Cyto, IL-1α/TNF-α group; TGF, TGF-β2 group. *P < 0.001; **P < 0.0001.

Figure 3.

 

Protein productions in the supernatants harvested from 1 × 10⁶ cultured HCE cells. The concentrations of all six proteins measured in the IL-1α/TNF-α group were higher than those in the control group. The protein concentrations of CCL2, CXCL2, IL-6, and CXCL2 in the TGF-β2 group were significantly lower than those in the IL-1α/TNF-α group. TIMP-1 protein concentration in the TGF-β2 group was significantly higher than that in the IL-1α/TNF-α group. The results are the mean ± SD of quadruplicate determinations in a representative experiment. Data are representative of two independent experiments. Con, Control group; Cyto, IL-1α/TNF-α group; TGF, TGF-β2 group. *P < 0.001; **P < 0.0001.

Figure 4.

 

mRNA expression in ex vivo donor HCE cells. RT-PCR was performed to detect mRNA expression in ex vivo HCE of donor corneas immersed in preservative. Two donor corneas were examined separately. Water was used as a negative control sample. GAPDH was detected, but not in a negative control sample (21 cycles). CXCL1, G-CSF, and IGFBP-5 genes are constitutively expressed in HCE of donor corneas (30 cycles). Numbers 1, 2, and 3 denote the HCE from donors 1 and 2, and a negative control, respectively.

The authors thank Chinatsu Kamikokuryo-Yokota for excellent technical support, Koichi Ohta and Nobuyuki Ebihara for helpful suggestions on TGF-β stimulation experiments, and IOVS volunteer editor Dongli Yang (University of Michigan, Ann Arbor, MI) for editing the manuscript.