Effect of Metalloproteinase Inhibitor on Corneal Cytokine Expression after Alkali Injury

Abstract

purpose. Interleukin (IL)-1α and IL-6 levels in the cornea are greatly elevated during the early stages after an alkali burn in mice. The authors investigated the effect of synthetic inhibitor of matrix metalloproteinase (SIMP) on the expression of inflammatory cytokines in alkali-burned murine corneas and evaluated the clinical appearance of the eyes.

methods. After 0.5N NaOH-alkali burns to 400 corneas of ICR mice, 200 received 400 μg/ml of SIMP topically 4 times a day while 200 corneas were similarly treated with vehicle only. At days 4, 7 and 14 after injury, each cornea was assigned a clinical score for corneal opacity, corneal epithelial defect, hyphema and cataract. Extracts of injured corneas in each group were then assayed for cytokine production using ELISA systems for IL-1α, IL-1β, IL-6 and tumor necrosis factor-α (TNF-α).

results. The levels of IL-1α, IL-1β and IL-6 were significantly lower in the SIMP-treated group than in the vehicle-treated group 7 days after the burn. However, levels of these cytokines were similar in the SIMP and non-SIMP groups at days 4 and 14. Levels of TNF-α did not differ between both groups at any postinjury time. In the SIMP-treated corneas, there was less opacification and hyphema formation and epithelial regeneration was faster.

conclusions. Topical application of SIMP in alkali-burned murine corneas reduced the expression of IL-1α, IL-1β, and IL-6 and lessened the severity of the injury.

The matrix metalloproteinase (MMP) family of enzymes are found in cornea. ^{1 2 3 4 5} An increased release of MMPs after an alkali burn is thought to be responsible for the destruction of the alkali-burned cornea, ^{6 7 8 9} a condition that is typically uncontrollable. Previous reports have revealed that matrix metalloproteinase inhibitors (MMPI), such as a thiol peptide, ¹⁰ tissue inhibitor of matrix metalloproteinase (TIMP), ¹¹ and synthetic inhibitor of matrix metalloproteinase (SIMP) ^{12 13} are effective in the treatment of alkali injuries. These reports demonstrated that MMPI-treated eyes have less inflammatory cell infiltration and less destruction of corneal collagen compared to vehicle-treated control eyes. However, the mechanisms of effect of MMPIs have not been fully clarified. 

Our recent research has demonstrated clearly that interleukin (IL)-1α and IL-6 are highly expressed in the cornea during the early stages after an alkali burn and may play an essential role in polymorphonuclear leukocyte (PMN) infiltration or associated corneal damage after the burn. ^{14 15} To more fully understand the mechanisms of MMPI action, we investigated the effect of MMPI on cytokine expression in the alkali-burned murine cornea. ¹⁴  

Materials and Methods

Synthetic Inhibitor of Matrix Metalloproteinase

We used SIMP developed by Galardy and associates. ^{12 16} This is an extremely potent inhibitor of fibroblast collagenase, gelatinase, and stromelysin and is also known as Galardin or GM6001. The compound is a dipeptide analogue with the structure N-[2(R)-2-(hydroxamido carbonylmethyl)-4-methylpentanoyl]-l-tryptophane methylamide. For topical application, SIMP was dissolved at a concentration of 400 μg/ml in N-2-hydroxyethylpiperazine-N′-2-ethane sulfonic acid (HEPES) buffer containing 2% dimethyl sulfoxide (pH 7.4). SIMP was gifted from Sankyo Co., Ltd. (Tokyo, Japan). For use as the control vehicle (pH 7.4), HEPES buffer containing 2% dimethyl sulfoxide was prepared. The solutions were stored at 4°C until use. 

Animals

Seven-week-old female ICR mice were used in this experiment. The animals were housed in plastic cages in a room with a 12-h light/12-h dark cycle. They were free from ocular disease. All mice were treated in compliance with the guidelines of the ARVO Resolution on the Use of Animals in Ophthalmic and Vision Research, and the experimental protocol was approved by the Committee for Animal Research, Kyoto Prefectural University of Medicine. 

Corneal Alkali Burn

An alkali burn was created as described previously. ¹⁴ Briefly, 2 μl of 0.5N NaOH was instilled to 400 corneas of ICR mice, with no subsequent eye-washing; 60 unburned corneas were used to confirm normal cytokine levels. 

Treatment of Mouse Alkali Burns with SIMP

After alkali injuries to 400 corneas, 200 were randomly assigned for treatment with vehicle containing 400 μg/ml of SIMP, whereas the other 200 controls were assigned for treatment with vehicle only. Treatment was started 2 hours after the burn, and one drop of either SIMP solution or vehicle was applied topically four times a day, every 3.5 hours from 8 AM to 10 PM. This treatment regimen was continued for 14 days after the alkali burn. 

At 4, 7, and 14 days after the burn, animals were sacrificed by severing the spinal cord. After the eyes were extracted, external examinations of each eye were performed by two independent examiners who were unaware of whether the animal had received SIMP or vehicle. Eyes were assessed for the presence of corneal opacity, corneal epithelial defects, hyphema, and cataracts. Each cornea was assigned a clinical score according to the area of corneal opacity using the after classification: 0, no opacity; 1, opacity covers less than one third of the corneal surface; 2, opacity covers more than one third and less than two thirds of the corneal surface; and 3, opacity covers more than two thirds of the corneal surface. The corneal epithelial defects were assigned a clinical score as follows: 0, no defects or superficial puctate keratitis; 1, epithelial defects cover less than one third of the corneal surface; 2, epithelial defects cover more than one third and less than two thirds of the corneal surface; 3, epithelial defects cover more than two thirds of the corneal surface. The presence of hyphema or cataract was classified as absent or present: 0, no hyphema; 1, existence of hyphema; 0, no cataract; 1, existence of cataract. 

Histologic analyses also were performed for six SIMP-treated and six vehicle-treated eyes 14 days after the burn. These eyes were fixed with 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS; pH 7.4) and then dehydrated with 20% sucrose in 0.1 M PBS. Specimens were embedded in O.C.T. compounds (Miles Inc., Elkhart, IN), after which 8-μm cryostat sections were cut, air-dried, and stained with hematoxylin-eosin. 

Cytokine Quantitation

To examine changes in cytokine expression with topical application of SIMP, the concentrations of IL-1α, IL-1β, IL-6, and TNF-α were measured in both SIMP- and vehicle-treated corneas 4, 7, and 14 days after the burn, using enzyme-linked immunosorbant assay (ELISA) systems for IL-1α, IL-1β (Genzyme Co., Cambridge, MA), IL-6 (Endogen, Inc., Cambridge, MA), and TNF-α (Genzyme Co.). Uninjured corneas were used to confirm normal cytokine levels. At each time point, the eyes were extracted, and the supernatants of corneal lysates were prepared as described previously. ^{14 17} Briefly, excised corneas 3 mm in diameter were frozen in liquid nitrogen and then smashed and homogenized in PBS at a ratio of 100 μl per cornea. The supernatants were collected by centrifugation at 1500g for 10 minutes and stored at −80°C until use. Each test sample comprised 6 to 7 whole corneas, with 9 to 12 samples (60–80 corneas) used at each time point in both groups. 

Statistics

Clinical scores were compared for significance using the Mann–Whitney U test. ELISA values were compared using Student’s t-test. 

Results

Treatment of Mouse Alkali Burn with SIMP

Corneal opacification was graded as moderate to severe 4 days after an alkali burn in SIMP- and non-SIMP–treated eyes. Thereafter, it gradually receded until 14 days after the burn in both groups. The score for corneal opacification was significantly lower at postinjury days 7 and 14 in the SIMP-treated group than that in controls (Fig. 1 A). The score for corneal epithelial defects was significantly lower than controls at day 7 in the SIMP-treated group. At postinjury day 14, the corneal epithelium had almost healed in both groups (Fig. 1B) . In the vehicle-only group, hyphema was observed in 22% (average score, 0.22) of eyes 4 days after injury, in 49% (average score, 0.49) of eyes 7 days after injury, and in 43% (average score, 0.43) of eyes 14 days after injury. In contrast, in the SIMP-treated group, hyphema was observed in 28% (average score, 0.28) of eyes at day 7. This was significantly lower than in the control group (Fig. 1C) . The scores for cataract were lower in the SIMP-treated group than in controls at all time points, but these were not statistically significant (Fig. 1D) . Figure 2 shows the typical appearance of corneas 14 days after injury. Although SIMP-treated corneas (Figs. 2A 2C) contained few inflammatory cells and little tissue destruction, control corneas treated with vehicle only (Figs. 2B 2D) disclosed massive inflammatory cell infiltration and severe stromal destruction. 

Cytokine Quantitation

In vehicle-treated corneas, IL-1α levels were greatly increased 4 and 7 days after the burn (its value was approximately 8.5 times the baseline value). Although the expression of IL-1α 4 days after injury was not different between SIMP- and vehicle-treated corneas, the level of IL-1α in the SIMP-treated group at day 7 was 32.5 ± 5.5 pg/cornea (mean ± SEM), significantly lower than in vehicle-treated corneas (45.2 ± 2.7 pg/cornea) (Fig. 3) . At day 14 after the burn, the levels of IL-1α had returned almost to normal levels in both groups. The level of IL-1β at day 7 also was significantly lower in the SIMP-treated group (5.6 ± 0.4 pg/cornea) than in controls (8.0 ± 1.0 pg/cornea) (Fig. 4) . At days 4 and 14 after injury, the levels of IL-6 did not differ between the two groups; however, the concentration of IL-6 in the SIMP-treated group at day 7 was 14.1 ± 2.7 pg/cornea, significantly lower than in controls (29.3 ± 6.6 pg/cornea) (Fig. 5) . Levels of TNF-α were elevated approximately 1.2 times the baseline value in both the SIMP- and vehicle-treated corneas at day 7; however, the levels did not differ between the two groups (Fig. 6) . 

Discussion

Alkali injury of the cornea is characterized by the massive infiltration of PMNs into the stroma and the severe destruction of corneal collagens. In our recent experiment, among the 10 cytokines examined, IL-1α and IL-6 were found to be induced strongly in alkali-burned corneas, the peak production of IL-1α being 3 days after injury and the peak production of IL-6 being 7 days after injury. ^{14 15} IL-1 and IL-6 can lead to infiltration by PMNs, after which these cells release superoxide, prostaglandins, lysosomal enzymes, or MMPs, which may cause corneal stromal melting. IL-1 also induces MMPs in cultured corneal stromal cells ¹⁸ and is known to induce corneal epithelial and stromal cells to express IL-6 and IL-8. ^{19 20 21} In addition, a recent report has demonstrated clearly that IL-1α induces MMPs and IL-8 in corneal fibroblasts extracted from wounds. ²² In view of this, inflammatory cytokines upregulated after an alkali burn are considered to be responsible for cell infiltration and collagen destruction. 

The present study has demonstrated that the topical application of SIMP significantly reduces the expression of IL-1α, IL-1β, and IL-6 in alkali-burned mouse corneas 7 days after the injury. It is noteworthy that at 4 days after injury, the levels of IL-1α, IL-1β, and IL-6 were the same in SIMP- and vehicle-treated corneas, implying that the reduction of these cytokines that are seen 7 days after injury probably is a result of the indirect action of the SIMP. One kind of SIMP (thiol peptide) recently has been shown to have an inhibitory effect on the chemotaxis of PMNs. ²³ However, on the basis of our data demonstrating that IL-1α and IL-6 after an alkali burn were mainly expressed in regenerating epithelium as opposed to infiltrating cells, ¹⁴ we consider that the reduction of inflammatory cytokines by the SIMP used in these experiments is not the result of the inhibitory effect on the chemotaxis of PMNs. Previous work has shown that the levels of IL-1α and IL-6 after an alkali burn correlate well with the concentration of alkali solution. ^{14 15} In view of this, it is likely that the expression of inflammatory cytokines in regenerating epithelium ¹⁴ correlates with the severity of the destruction of corneal collagens. We speculate that degraded corneal collagens induce regenerating epithelium to express IL-1α, IL-1β, and IL-6 and that the suppression of stromal degradation by SIMP results in less expression of cytokines in alkali-burned cornea. However, we cannot completely exclude the possibility that SIMP may prevent the activation or inhibit the expression of these cytokines by a mechanism(s) independent of stromal collagens. 

Previous reports have demonstrated that MMPI prevents the ulceration of rabbit corneas after an alkali burn. ^{11 12 13} In this study, using a mouse alkali-burn model, we confirmed the beneficial effect of SIMP on corneal opacification, epithelial regeneration, and hyphema formation after the burn, although the effect is not immediate. Recently, Saika and associates ²⁴ demonstrated that TIMPs enhance the spreading of the corneal epithelium ex vivo and the proliferation of corneal epithelial cells in vitro. Our data indicated that MMPIs also promote epithelial regeneration in vivo. Interestingly, we found that the incidence of hyphema formation was significantly lower in SIMP-treated eyes than it was in non-SIMP treated eyes 7 days after an alkali burn. Because trabecular meshwork cells and ciliary muscle cells express MMPs, ^{25 26} our results suggest that hyphema formation in alkali-burned eyes may be the result of the destruction of trabecular meshwork, iris, and ciliary body by MMPs. Hyphema complications are clinical features of some cases of alkali-burned eyes, and it is worth considering the possibility that MMPI may be useful in preventing the occurrence of hyphema after alkali injury. 

Although IL-1α, IL-1β, and IL-6 are cytokines lacking a transmembrane domain precursor, TNF-α contains a transmembrane domain, and membrane-anchored pro–TNF-α is proteolytically processed to the mature TNF-α. ²⁷ Several MMPs are responsible for the processing of mature TNF-α, ²⁷ a processing that MMPIs can inhibit. ^{28 29} We found that the protein level of TNF-α after our alkali burn did not differ between the SIMP-treated and control groups. This may be because both precursor and mature TNF-α were contained in the supernatants of each sample, and both were detected by ELISA systems. Alternatively, it could be because the upregulation of TNF-α after the burn is only slight, making it impossible to detect the effect of SIMP on TNF-α expression. We predict that in conditions in which the expression or the processing of TNF-α is excessively upregulated, MMPI may have an inhibitory effect on TNF-α expression. Interestingly, Fas ligand, a major inducer of apoptosis, belongs to the TNF family and is prevented from shedding by MMPI. ³⁰ We predict that further mechanisms may exist in the action of MMPI that need to be established. 

The anticollagenase activity of MMPI has been considered to be the main mechanism by which it protects alkali-burned eyes from becoming ulcerated; however, the inhibitory effect of MMPI on the expression of inflammatory cytokines after the burn also may be responsible for the reduced damage seen in MMPI-treated eyes. 

 Supported in part by a grant in-aid for scientific research (09877336, 08457467, 09771459) from the Japanese Ministry of Education, Culture and Science; a research grant from the Japanese Ministry of Health and Welfare; and the intramural research fund of the Kyoto Prefectural University of Medicine.

 Submitted for publication January 7, 1999; revised April 6, 1999; accepted May 5, 1999.

 Proprietary interest category: N.

 Corresponding author: Chie Sotozono, Department of Ophthalmology, Kyoto Prefectural University of Medicine, Hirokoji, Kawaramachi, Kamigyoku, Kyoto 602-0841, Japan. E-mail: [email protected]

 

Figure 1.

 

Time course for clinical scores after alkali burn. Two hours after 2μ l of 0.5N NaOH was applied to corneas of ICR mice to create an alkali burn, eyes were administered 400 μg/ml SIMP or vehicle only 4 times a day. (A) Area of corneal opacity; (B) area of corneal epithelial defect; (C) hyphema; (D) cataract. The details of how these clinical scores were arrived at are described in the Methods section. *P < 0.05, **P < 0.005 (Mann–Whitney U test).

Figure 2.

 

Mouse corneas treated for 14 days after an alkali burn with SIMP or vehicle only. Representative appearance of 400 μg/ml SIMP-treated eye (A, C) and vehicle-treated eye (B, D) 14 days after a 0.5N NaOH-alkali burn. (C, D) Histology of the central corneas seen in (A) and (B). Calibration bar, 100 μm.

Figure 3.

 

IL-1α concentrations in corneas treated with 400 μg/ml SIMP or vehicle only after a 0.5N NaOH-alkali burn. Uninjured corneas were used to confirm normal cytokine levels. The corneas used for the determination of the clinical score (Fig. 1) also were assayed for the presence of IL-1α. Each point represents mean ± SEM of 9 to 12 pooled samples, each sample comprising 6 to 7 corneas. Each column, therefore, represents the mean of 60 to 80 corneas.* P < 0.025 (Student’s t-test).

Figure 4.

 

IL-1β concentration in corneas treated with 400 μg/ml SIMP or vehicle only after a 0.5N NaOH-alkali burn. The same samples used for the IL-1α measurement (Fig. 3) also were assayed for the presence of IL-1β. *P < 0.05 (Student’s t-test).

Figure 5.

 

IL-6 concentration in corneas treated with 400 μg/ml SIMP or vehicle only after a 0.5N NaOH-alkali burn. The same samples used for the IL-1α measurements (Fig. 3) also were assayed for the presence of IL-6. **P < 0.025 (Student’s t-test).

Figure 6.

 

TNF-α concentration in corneas treated with 400 μg/ml SIMP or vehicle only after a 0.5N NaOH-alkali burn. The same samples used for the IL-1α measurements (Fig. 3) also were assayed for the presence of TNF-α.

The authors thank Andrew J. Quantock for helpful comments and critical reading of the manuscript.