Desiccating Stress Stimulates Expression of Matrix Metalloproteinases by the Corneal Epithelium

Abstract

purpose. To determine the effects of experimental dryness on production of matrix metalloproteinases (MMPs) and their physiological inhibitors (TIMPs) by the corneal epithelium.

methods. Experimental dry eye (EDE) was created in two strains of mice: BALB/c and C57BL/6. Real-time PCR was performed with MMP and TIMP probes, and the results were analyzed by the comparative C_T method, selecting the relative mRNA levels in untreated control samples as calibrator. Immunofluorescent staining with specific antibodies immunolocalized MMP proteins in situ. MMP enzymatic activity was evaluated in tears and corneal lysates. Corneal permeability to Oregon green dextran (OGD) and sodium fluorescein was measured. Corneal smoothness was evaluated by graded regularity of a ring reflected off the corneal surface.

results. Desiccating stress significantly increased levels of MMP-1, -3, -9, and -10 transcripts in the corneal epithelium in C57BL/6 mice, but had no effect on levels of MMP transcripts in the corneal epithelium of BALB/c mice. There was no change in levels of TIMP transcripts except for TIMP-4 which significantly increased on day 10 in C57BL/6 mice. The MMP-1, -3, and -9 concentration in tears significantly increased compared with control levels after EDE for 4 and 6 days, respectively, in C57BL/6 and BALB/c. Changes in MMP protein expression detected by immunofluorescent staining were similar to changes in gene transcripts for most MMPs. EDE increased corneal permeability to OGD and fluorescein and corneal surface irregularity.

conclusions. Corneal dryness stimulates production of certain MMPs in a strain-dependent fashion and causes the disruption of the corneal barrier, thus increasing permeability and corneal irregularity.

Matric metalloproteinases (MMPs) are a family of proteolytic enzymes that have an integral physiological role in angiogenesis, inflammation, and wound repair through their ability to degrade extracellular matrix components. At least 23 MMPs have been identified and they are subclassified according to their substrates: the collagenases (MMP-1, -8, and -13) degrade fibrillar collagen types I, II, and III; the gelatinases (MMP-2 and -9) degrade types IV and VII collagen that are found in basement membranes; the stromelysins (MMP-3 and -10) and the matrilysins (MMP-7 and -26) degrade proteoglycans, laminin, and glycoproteins; and the membrane-type (MT-MMP-1, -2, and -3), which are bound to epithelial cell membranes, can activate other MMPs. ^{1 2 3 4 5 6 7 8 9}  

MMPs have been found to have pathogenic roles in inflammatory diseases, such as arthritis, periodontitis, glomerulonephritis, and atherosclerosis, and in corneal epithelial and stromal ulceration. ^{10 11 12 13 14 15} There is increasing evidence that dry eye induces inflammation on the ocular surface that is responsible in part for the ocular surface epithelial disease and irritation symptoms that develop. Increased levels of inflammatory cytokines (IL-1β and TNF-α) and concentrations and activities of MMPs have been detected in the tear fluid and/or conjunctival epithelia of patients with keratoconjunctivitis sicca (KCS). ^{16 17 18 19 20}  

Our previous studies have demonstrated that IL-1β and TNF-α upregulate production of gelatinase B (MMP-9), collagenases and stromelysins by the human corneal epithelial cells. ^{21 22 23} It is also known that the desiccating and hyperosmolar stress of dry eye activates MAP kinase (MAPK) stress-signaling pathways, that are key regulators of MMP transcription. ^{24 25 26 27} The dynamic interplay between inflammatory cytokines and MMPs appears to play a key role in the pathogenesis of KCS. 

We have reported that corneal epithelial barrier function is altered in our murine experimental dry eye (EDE) model ²⁸ and that there is an increased production of inflammatory mediators on the ocular surface that is similar to human KCS. ^{24 29} MMP-9 was found to be responsible, at least in part, for the altered corneal epithelial permeability in KCS, because corneal epithelial barrier function was preserved in MMP-9 knockout mice in response to EDE. ²⁸  

There are three levels of MMP regulation: transcriptional control, activation of the secreted pro enzyme, and inhibition of the enzyme by endogenous inhibitors known as tissue inhibitors of metalloproteinase (TIMPs). ^{3 30 31} Each MMP is inhibited by at least one of four recognized TIMPs, and it has been proposed that MMP activity is related to the ratio of the concentration of an MMP to its TIMP. An altered balance of MMPs and TIMPs has been implicated in the pathogenesis of corneal ulceration and wound healing. ^{32 33}  

The expression of MMPs, other than MMP-9, and TIMPs on the ocular surface in response to desiccating stress has not been evaluated. The purpose of the present study was to evaluate the expression of MMP-1, -3, -7, -10, and -13, as well as MMP-9, and TIMPs 1 to 4 in the corneal epithelium before and at two sequential time points after exposure to desiccating ocular surface stress in C57BL/6 and BALB/c mouse strains that have been found to respond differently to experimental pseudomonas keratitis. ³⁴ Corneal barrier function and surface smoothness, clinical measures of keratitis sicca were also evaluated in this mouse model. 

Methods

Experimental Desiccating Ocular Surface Stress

This research protocol was approved by the Baylor College of Medicine, Center for Comparative Medicine, and it conformed to the standards in the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Experimental dry eye (EDE) was produced in 6- to 8-week-old C57BL/6 and BALB/c mice of both genders, by subcutaneous injection of 0.5 mg/0.2 mL scopolamine hydrobromide (Sigma-Aldrich, St. Louis, MO) in alternating hindquarters four times a day (8 AM, 11 AM, 2 PM, and 5 PM) and by exposure to an air draft and 30% to 40% ambient humidity for 18 hours per day, as has been reported. ^{24 25 27} The mice were treated in this fashion for various lengths of time—2, 5, 10, and 12 days—depending on the parameter that was evaluated. Untreated age and gender-matched mice were used as normal control subjects. 

RNA Isolation and Real-Time PCR

Total RNA was isolated from corneal epithelia that was collected and pooled from 10 eyes at each time point by acid guanidium thiocyanate-phenol-chloroform extraction. ³⁵ The RNA concentration was measured by its absorption at 260 nm, and samples were stored at −80°C before use. 

First-strand cDNA was synthesized from 1 μg of total RNA with random hexamers by M-MuLV reverse transcription (Ready-To-Go You-Prime First-Strand Beads; GE Healthcare, Inc., Arlington Heights, NJ), as previously described. ^{24 25 35} Real-time PCR was performed with specific MGB probes (Table 1 ; Taqman; Applied Biosystems, Inc. [ABI], Foster City, CA) a PCR master mix (Taqman, AmpErase UNG; ABI), and a commercial thermocycling system (Smart Cycler; Cepheid, Sunnyvale, CA), according to the manufacturers’ recommendations. Assays were performed in duplicate in each experiment, and they were repeated in four different sets of mice. A nontemplate control was included in all the experiments, to evaluate DNA contamination of the reagent used. The GAPDH gene was used as an endogenous reference for each reaction, to correct for differences in the amount of total RNA added. The results of quantitative PCR were analyzed by the comparative C_T method (ABI Bulletin, No.2 (P/N 4303859) ^{24 36} ; where target fold = 2^−ΔΔC_t . The cycle threshold (C_t) was determined with the primary (fluorescent) signal; it is the cycle at which the signal crosses a user-defined threshold. The results were normalized by the C_t value of GAPDH and the relative mRNA level in the C57BL/6 untreated group was used as the calibrator. 

Tear Fluid Washings

Tear fluid washings were collected by a previously reported method. ²⁴ Briefly, 1.5 μL of PBS containing 0.1% bovine serum albumin (BSA) was instilled into the conjunctival sac. The tear fluid and buffer were collected with a 1-μL volume glass capillary tube (Drummond Scientific Co., Broomall, PA) by capillary action from the tear meniscus in the lateral canthus. The tear washings from both eyes of each mouse (six mice per group) were pooled (2 μL) and stored at −80°C until activity assays were performed. 

MMP Activity Assay

The total enzyme activity levels of MMP-1, -3, and -9 protein in the corneal epithelium (two mice per group) and tears were determined with MMP activity assay systems (Biotrack; GE Healthcare), according to the manufacturer’s protocol. In brief, 100 μL of each pro-MMP standard (0.5–32 ng/mL), tissue lysate or tears, or assay buffer (for blanks) were incubated at 4°C overnight in microtiter wells precoated with the respective anti-MMP antibody. MMPs present in these samples bound to the wells. Other components of the sample were removed by washing four times with 0.01 M sodium phosphate buffer (pH 7.0) containing 0.05% Tween-20. To measure the total activity of MMP-1, -3 and -9, we activated bound pro-MMPs activated with 50 μL of 1 mM p-aminophenylmercuric acetate (APMA) in assay buffer at 37°C for 2 hours. Detection reagent (50 μL) was then added to each well and incubated at 37°C for 6 hours. Active MMPs were detected through their ability to activate a modified prodetection enzyme that subsequently cleaved its chromogenic peptide substrate. The resultant color was read at 405 nm (Versamax; Molecular Devices, Sunnyvale, CA) microplate reader. The activity of MMP-1, -3, and -9 in a sample was determined by interpolation from a standard curve. 

Immunofluorescent Staining

Immunofluorescent staining was performed with polyclonal antibodies to immunolocalize MMP and TIMP proteins in corneal tissue sections from normal control and dry eye mice in situ. 

In brief, eyes and adnexa from mice in each group were surgically excised, embedded (OCT compound; VWR, Suwannee, GA), and flash frozen in liquid nitrogen. Sagittal 8-μm sections were cut with a cryostat (model HM 500; Micron, Waldorf, Germany) and placed on glass slides that were stored at −80°C. The tissue sections for immunofluorescent staining were fixed with acetone at −20°C for 5 minutes, then permeabilized with PBS containing 0.1% Triton X-100 for 10 minutes. After blocking with 20% normal horse serum in PBS for 45 to 60 minutes (except for MMP-1 where 20% goat normal serum was used), primary polyclonal goat antibodies against MMPs (Santa Cruz Biotechnology, Santa Cruz, CA; except for MMP-1 where a polyclonal rabbit antibody was used; Chemicon International, Temecula, CA) were applied at a 1:100 dilution and incubated for 1 hour at RT. Secondary antibodies, Alexa-Fluor 488–conjugated goat anti-rabbit IgG or Alexa-Fluor 488–conjugated donkey anti-goat IgG (1:300) was then applied and the sections incubated in a dark chamber for 1 hour, followed by counterstaining with propidium iodide (PI; 2 μg/mL in PBS) for 5 minutes. 

Digital confocal images (512 × 512 pixels) were captured with a laser-scanning confocal microscope (LSM 510; with krypton-argon and He-Ne laser; Carl Zeiss Meditec, Inc., Thornwood, NY) with 488-nm excitation and 543-nm emission filters (LP505 and LP560, respectively; Carl Zeiss Meditec, Inc.) and were acquired with a 40/1.3× oil-immersion objective. Samples from untreated control and EDE samples were captured, by using identical photomultiplier tube gain settings, and then were processed (LSM-PC; Carl Zeiss Meditec, Inc.; and Photoshop 7.0 software; Adobe Systems, Mountain View, CA). Staining intensities in the superficial and basal corneal epithelia were graded by consensus of two masked observers who used a previously reported scale; grade 0, no different from the secondary antibody control; +, slightly greater than the secondary antibody control; ++, moderate staining; and +++, intense staining. ³⁷  

Corneal Permeability to Oregon Green Dextran

Corneal epithelial permeability to the fluorescent molecule Oregon green dextran (OGD 70,000 MW; Invitrogen, Eugene, OR) was assessed in 10 eyes of both strains, before and after 5 and 10 days of EDE. Briefly, 0.5 μL of 50 μg/mL OGD was instilled onto the ocular surface 1 minute before euthanatization. Corneas were rinsed with 2 mL of PBS and photographed with a stereoscopic zoom microscope (model SMZ 1500; Nikon, Melville, NY), with a fluorescence excitation at 470 nm. Images were obtained 2 hours after the last scopolamine injection and were processed (Metavue 6.24r software). The severity of corneal OGD staining was graded in digital images, according to the Baylor grading scheme for corneal fluorescent staining, ³⁸ by consensus of two masked observers. Briefly, the number of dots of fluorescein staining was graded in the 1-mm central cornea zone of each eye according to a standardized 5-point scale (0 dots, 0; 1–5 dots, 1; 6–15 dots, 2; 16–30 dots, 3; >30 dots, 4). One point was added to the score if there was one area of confluent staining and two points were added for two or more areas of confluence. 

Corneal Permeability to Fluorescein

Corneal epithelial permeability to fluorescein was assessed in 10 eyes of both strains before and after 5 and 10 days of EDE with a fluorophotometer (Fluorotron Master; Ocumetrics, Mountain View, CA), an instrument that has been used for objective measurement of fluorescein permeability in human corneas. ³⁹ Tungsten light is passed through two different filters that allow only wavelengths in the excitation bandwidth of fluorescein. The light is then focused in the eye by the optics of the fluorophotometer. With the anterior chamber adapter attached, the fluorophotometer measures fluorescence in a stepwise fashion toward the lens until it finds a fluorescein maximum, which it outputs to the computer screen. 

Briefly, the mice were euthanatized, the eyes were removed and washed three times with 1 mL normal saline, and the baseline fluorescence of the eye was measured by placing the eye in a specially fitted cap on the anterior chamber adapter of the fluorophotometer. The eye was then placed cornea side down in 0.5 mL of 0.5% 5(6)-carboxyfluorescein (Sigma-Aldrich) for 40 seconds. The eye was washed three times with 1 mL of normal saline from the optic nerve side of the eye. Six measurements of the corneal permeability were then recorded in 360°, rotating the eye 60° between each measurement. 

Evaluation of Corneal Smoothness

Corneal smoothness was assessed in 10 eyes of both strains before and after 5 or 10 days of EDE. Reflected images of a white ring from the fiber optic ring illuminator of a stereoscopic zoom microscope (SMZ 1500; Nikon) were taken immediately after euthanatization, 2 hours after the last scopolamine injection. This ring light is firmly attached and surrounds the bottom of the microscope objective. Because the illumination path is nearly coincident with the optical axis of the microscope, the viewing area is evenly illuminated and nearly shadowless. The regularity of the ring light reflected off the wet cornea depends on the surface smoothness. Smoothness of the reflected rings was graded in digital images by consensus of two masked observers. The reflected ring image was divided into four quarters, of 3 clock hours each. The corneal irregularity severity score was calculated according to a 5-point scale based on the number of distorted quarters in the reflected ring: 0, no distortion; 1, distortion in one quarter of the ring (3 clock hours); 2, distortion in two quarters of the ring (6 clock hours); 3, distortion in three quarters of the ring (9 clock hours); 4, distortion in all four quarters (12 clock hours); and 5, distortion so severe that no ring could be recognized. 

Statistical Analysis

Results are presented as the mean ± SEM of at least four separate experiments. Significant differences were evaluated by two-way ANOVA, with post hoc analysis with the Bonferroni test. P ≤ 0.05 was considered significant. 

Results

Effects of Desiccating Stress on Levels of MMP and TIMP Transcripts in the Corneal Epithelium

The levels of the collagenases (MMP-1 and -13), stromelysins (MMP-3 and -10), matrilysin (MMP-7), and gelatinase (MMP-9) mRNA transcripts, their physiological inhibitors (TIMP-1, -2, -3, and -4) and the housekeeping gene GAPDH were evaluated by real-time PCR, with pooled total RNA samples of corneal epithelium (10 eyes per group) from untreated mice and mice with EDE for 5 or 10 days taken from BALB/c and C57BL/6 mice. The comparative C_T method was used to determine the relative x-fold change in expression for each gene studied, with the relative mRNA levels in untreated C57BL/6 used as a calibrator. The experiment was performed four times, to confirm the reproducibility of the results. 

There was a different response to EDE between strains (Tables 2 3) . Desiccating stress significantly increased the level of several MMP transcripts in the corneal epithelium in C57BL/6 mice. MMP-1, -3, and -10 were noted to increase by 175.84% (P < 0.01), 109.11% (P < 0.05), and 70.12% (P < 0.01), respectively, after 5 days of EDE compared with baseline levels of these transcripts in normal control C57BL/6 mice. After 10 days of desiccating stress, MMP-3 increased 111.13% (P < 0.01) and MMP-9 increased 116.11% (P < 0.05) in C57BL/6 mice compared with baseline. MMP-13 was the only MMP that decreased in the C57BL/6 corneal epithelium, and this occurred after 10 days of EDE. In contrast, MMP-13 was the only MMP transcript to increase in the BALB/c corneal epithelium, although the increase was not significant. There was no change in levels of any TIMP transcript in the cornea epithelium of either strain, except for TIMP-4 transcripts that were significantly increased at 10 days in C57BL/6 mice. 

Activity Assay for MMPs in the Mouse Corneal Epithelium

The concentrations of the collagenase MMP-1, stromelysin MMP-3, and gelatinase MMP-9 were evaluated by enzyme activity assays, with pooled protein samples of corneal epithelium (four eyes per group) from untreated mice, and mice with EDE for 5 and 10 days from both strains (BALB/c and C57BL/6). The experiment was performed three times to confirm the reproducibility of the results. 

The results of MMP activity assays are presented in Table 4 . There were no significant changes in MMP-1 and -9 concentrations compared with baseline in the corneal epithelium of either mouse strain. The MMP-3 concentration in C57BL/6 cornea increased significantly (from 0.013 ± 0.009 pg/mL in untreated control cornea to 0.146 ± 0.024 pg/mL after 4 days of EDE; P < 0.05) and to 0.327 ± 0.035 pg/mL after 10 days of EDE (P < 0.001 compared with the control). The concentrations of MMPs were significantly lower in BALB/c than in C57BL/6 mice after 2 (MMP-9), 5 (MMP-1 and -9), and 10 (MMP-3) days of EDE. 

Activity Assay for MMPs in Tear Fluid Washings

Tear fluid washings (pooled from both eyes of each mouse) were collected from six mice before and after EDE for 2, 4, 6, 8, and 10 days duration in C57BL/6 and BALB/c mice. The concentrations of MMP-1, -3, and -9 were measured in these tear washings by enzyme activity assays. The experiment was performed three times to confirm the reproducibility of the results. 

The MMP-1 concentration in C57BL/6 tears significantly increased from 30.29 ± 1.65 pg/mL in untreated control tears to 66.55 ± 1.68 pg/mL after EDE for 4 days (P < 0.001) and to 42.10 ± 0.13 pg/mL after 10 days of EDE (P < 0.01 compared with the control; Fig. 1A ). The MMP-1 concentrations in BALB/c tears significantly increased from 6.65 ± 0.77 pg/mL in control tears to 45.33 ± 1.42 pg/mL after EDE for 6 days (P < 0.001) and it returned to baseline levels at 10 days. MMP-1 tear concentrations were significantly lower in BALB/c tears than C57BL/6 tears at every time point (P < 0.01; Fig. 1B ). 

The MMP-3 concentration in C57BL/6 tears significantly increased from 94.58 ± 12.12 pg/mL in untreated control tears to 191.5 ± 12.12 pg/mL after EDE for 4 days (P < 0.05 compared with the control), and it returned to baseline levels at 10 days (Fig. 1C) . The MMP-3 concentration in BALB/c tears significantly increased from 82.47 ± 24.23 pg/mL in control tears to 239.96 ± 36.35 pg/mL after EDE for 6 days (P < 0.01), and it returned to baseline levels at 10 days. 

The MMP-9 concentration in C57BL/6 tears significantly increased from 156.70 ± 4.20 pg/mL in control tears to 358.29 ± 54.47 pg/mL after EDE for 4 days (P < 0.001) and to 262.81 ± 19.61 pg/mL after EDE for 8 days (P < 0.01) compared with the control (Fig. 1D) . The MMP-9 concentration in BALB/c tears significantly increased from 152.38 ± 11.57 pg/mL in control tears to 262.81 ± 19.60 pg/mL after EDE for 6 days (P < 0.01) and it returned to baseline levels at 10 days. MMP-9 tear concentrations were significantly lower in BALB/c tears than C57BL/6 tears after 4 days of EDE (P < 0.001). 

Immunofluorescent Staining

MMP-1, -13, -3, -10, -7, and -9 antigens were immunodetected in frozen cornea sections obtained from C57BL/6 and BALB/c mice (Figs. 2 3 ; Table 5 ). 

MMP-1 staining increased in the superficial corneal epithelial cells after 5 days of EDE in C57BL/6, whereas it was almost undetectable in the BALB/c corneal epithelium. MMP-13 staining increased in the BALB/c corneal epithelium after 2 days of EDE. MMP-3 immunoreactivity increased in the corneal epithelium in C57BL/6 mice, but it decreased in BALB/c mice. MMP-10 remained unchanged in the C57BL/6 cornea, and it was observed to increase slightly in the BALB/c corneal epithelium. MMP-7 increased in the corneal epithelium of C57BL/6 mice, and it was almost undetectable in BALB/c mice. MMP-9 increased in the corneal epithelium of both strains of mice. 

Disruption of the Corneal Epithelial Permeability Barrier

Decreased corneal epithelial barrier function is a key feature of the corneal epithelial disease in human dry eye. Corneal epithelial permeability to a high-molecular-weight fluorescent molecule OGD (70 kDa) was assessed in control mice and mice subjected to desiccating stress for 5 and 10 days. Minimal scattered punctate staining or no staining with OGD was observed in the corneas of control mice (Fig. 4) . Compared with the control corneas, uptake of OGD significantly increased (P < 0.001) after 5 and 10 days of EDE in C57BL/6 mice, and punctate and confluent dye staining, mimicking human keratitis sicca was observed (Table 6) . The OGD staining score in the BALB/c cornea was significantly lower than C57BL/6 after 5 (P < 0.001) and 10 days (P < 0.01) of EDE. 

The fluorophotometer (Fluorotron Master; Ocumetrics) has been used to measure human corneal epithelial permeability to fluorescein via both single-drop and bath emersion methods. Though used extensively in humans, the fluorophotometer has not been used to evaluate corneal epithelial permeability in a murine model. Significantly increased fluorescein permeability was noted in C57BL/6 mice on induction of EDE and it remained elevated throughout the 10 days (P < 0.001) of the experiment. The BALB/c mice showed a transient significant (P < 0.01) increase in corneal permeability at 5 days with return to baseline permeability after 10 days of EDE. The corneal fluorescence intensity in BALB/c mice was significantly lower than that in C57BL/6 after 10 days (P < 0.001) of EDE. 

The regularity of a white ring-shaped light reflected off the cornea was used to evaluate corneal epithelial smoothness (Fig. 4) . Corneal surface irregularity significantly increased after 5 and 10 days of EDE (Table 6)in both strains compared with the untreated control and was significantly lower after 10 days than after 5 days. 

Discussion

We found that desiccating ocular surface stress increased expression of several MMPs in the corneal epithelium. We used relatively quantitative real-time PCR to detect MMP RNA levels in the corneal epithelium and immunostaining to evaluate protein expression in situ. Total enzyme activity in tears and in corneal epithelial lysates was measured by commercial MMP activity assays. RNA levels for the collagenases MMP-1 and -10, the stromelysin MMP-3, and the gelatinase MMP-9 showed the greatest increase in expression in response to EDE in the C57BL/6 strain. This was accompanied by increased MMP-1 and -9 activity in the corneal epithelium and tear washings obtained from this strain. One shortcoming of our study was the inability to evaluate levels of activated enzyme. We initially attempted to measure levels of active enzyme in corneal lysates and tear washings without APMA treatment; however, levels of active enzyme were found to be below the level of detection, except in a few 5- or 12-day EDE samples. This would not permit statistical comparison between groups; therefore, total enzyme activity results are reported. In earlier studies, we have used gelatin zymography of tears and in situ zymography of the corneal epithelium to demonstrate increased expression of gelatinases in response to dry eye in the CD-1 mouse strain. ^{24 28} The purpose of the present study was to perform a comprehensive evaluation of three classes of MMPs (gelatinases, collagenases, and stromelysins) in two different mouse strains. Gelatin zymography was an option only for detection of gelatinases. In preliminary studies, we found that casein zymography lacked sufficient sensitivity to detect MMP-3 and zymographies to detect collagenases (MMP-1 and -13), stromelysins (MMP-3 and -10), and matrilysin (MMP-7) are not available. Enzyme activity assays and immunostaining permitted use of the same methods to detect all three classes of MMPs in the minute quantities of tears and corneal lysates that were obtained. Some minor discrepancies in the results of enzyme activity assays were noted between corneal epithelial lysates and tear washings. The results of the activity assays for MMP-1 and -9 were similar in tear and corneal specimens in C57BL/6 mice; however, they differed for MMP-1 in BALB/c and for MMP-3 in both strains. One possible explanation for these differences is that the tear washings contain MMPs that were released by the ocular surface epithelia (both cornea and conjunctiva) as well as inflammatory cells infiltrating the ocular surface. 

The effects of desiccating stress were compared in two well-characterized strains of mice: the C57BL/6 strain, which has a predilection for a Th1 response to proinflammatory stress, and the BALB/c strain, which manifests a Th2 response. ³⁴ BALB/c and C57BL/6 mice are widely known to express different immune responses in normal and pathologic states. These phenomena are related to interstrain differences in their genetic background. ^{40 41} For example, C57BL/6 mice are more susceptible than BALB/c mice to induction of experimental, organ-specific autoimmune diseases, such as experimental autoimmune uveitis. ^{42 44} In contrast to C57BL/6 mice, BALB/c mice display increased susceptibility to tumorigenesis. ⁴⁵  

C57BL/6 mice showed more robust stimulation of MMP production in the corneal epithelium in response to desiccating stress than did BALB/c mice. Although the immunostaining intensity of certain MMPs (MMP-9, -10, and -13) was observed to increase in the corneal epithelium of BALB/c mice, there was no significant increase in levels of transcripts for any of the MMPs in the corneal epithelium in this strain. With the tear fluid used as a measure of MMP release from the ocular surface, C57BL/6 mice showed a relatively greater and more sustained increase than the BALB/c mice. This may be related to the increased levels of proinflammatory cytokines known to stimulate MMP production (i.e., IL-1β and TNF-α) that we have detected in the tear fluid and ocular surface epithelia in C57BL/6 compared with BALB/c mice (manuscript submitted). 

Desiccating stress had no effect on levels of TIMP transcripts in the corneal epithelium in either strain, except for a significant increase in TIMP-4 transcripts on day 10 in C57BL/6 mice. This suggests that MMPs are more inducible by dry eye stress than are TIMPs. It is possible that increasing the TIMP-MMP ratio through pharmacological agents or gene therapy would blunt the increased MMP activity on the ocular surface in dry eye. 

There are multiple potential consequences of increased MMP production and activity on the ocular surface in response to dry eye. We have reported that MMP-9 plays a key role in acute disruption of corneal epithelial barrier function in response to experimental desiccating stress. ²⁸ With two different methods, C57BL/6 mice in our present study were found to have greater disruption of corneal barrier function in EDE than were BALB/c mice. Stimulated MMP production by the stressed ocular surface epithelia in patients with dry eye could be responsible for their punctate epitheliopathy and could be a factor that promotes the development of sight-threatening sterile corneal ulcers that are recognized to occur with increased frequency in dry eyes. C57BL/6 mice have been reported to be much more prone to development of corneal perforation in response to experimental pseudomonas keratitis than BALB/c mice, and higher production of MMPs by the corneal epithelium is a likely mechanism for this finding. Among patients with dry eye, the highest levels of MMP-9 in the tears were found in those with Sjögren syndrome with sterile corneal ulcers. ¹⁸ Another potential pathogenic role for MMPs in dry eye is cleavage of cell surface and tear proteins producing immunogenic peptides capable of stimulating an autoimmune response. ⁴⁶  

 Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May 2005.

 Supported by National Eye Institute Grants EY11915 (SCP) and EY016928 (CSDP), Allergan, Inc., a restricted grant from Research to Prevent Blindness, The Oshman Foundation, and The William Stamps Farish Fund.

 Submitted for publication October 24, 2005; revised March 23, 2006; accepted June, 13, 2006.

 Disclosure R. M. Corrales, Allergan, Inc. (F); M. E. Stern, Allergan, Inc. (E, F); C.S. De Paiva, Allergan, Inc. (F); J. Welch, Allergan, Inc. (F); D.-Q. Li, Allergan, Inc. (F); S. C. Pflugfelder, Allergan, Inc. (F)

 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.

 

Oligonucleotide Primers Used for Real-Time PCR

Gene Name	Symbol	Assay ID*
Glyceraldehyde-3-phosphate dehydrogenase	GAPDH	Mm99999915_g1
Matrix metalloproteinase 13	MMP-13	Mm00439491_m1
Matrix metalloproteinase 3	MMP-3	Mm00440295_m1
Matrix metalloproteinase 10	MMP-10	Mm00444630_m1
Matrix metalloproteinase 9	MMP-9	Mm00442991_m1
Matrix metalloproteinase 7	MMP-7	Mm00487724_m1
Matrix metalloproteinase 1b	MMP-1	Mm00473493_g1
Tissue inhibitor of metalloproteinase 1	TIMP-1	Mm00441818_m1
Tissue inhibitor of metalloproteinase 2	TIMP-2	Mm00441825_m1
Tissue inhibitor of metalloproteinase 3	TIMP-3	Mm00441826_m1
Tissue inhibitor of metalloproteinase 4	TIMP-4	Mm00446568_m1

*  Identification number from Applied Byosistems (www.appliedbiosystems.com).

Table 2.

 

Effect of 5 (5D) and 10 Days (10D) of Desiccating Stress on the Relative Expression of the MMPs and TIMPs in the Mouse Corneal Epithelium

	UT	EDE-5D	EDE-10D	Interstrain P
MMP-1
C57BL/6	1.00 ± 0.00	2.76 ± 0.35, **	0.91 ± 0.33, ††
BALB/c	0.96 ± 0.37	1.64 ± 0.43	0.75 ± 0.53
MMP-13
C57BL/6	1.00 ± 0.00	0.66 ± 0.12	0.32 ± 0.09, **	<0.05 in UT
BALB/c	0.45 ± 0.09	0.23 ± 0.21	0.74 ± 0.27
MMP-3
C57BL/6	1.00 ± 0.00	2.09 ± 0.45*	2.11 ± 0.19, **	<0.001 in 5D
BALB/c	0.88 ± 0.39	0.45 ± 0.15	0.47 ± 0.04	<0.001 in 10D
MMP-10
C57BL/6	1.00 ± 0.00	1.79 ± 0.36, **	1.07 ± 0.11, ††
BALB/c	1.30 ± 0.30	1.27 ± 0.25	1.51 ± 0.55
MMP-7
C57BL/6	1.00 ± 0.00	1.06 ± 0.23	0.95 ± 0.14	<0.05 in UT
BALB/c	0.33 ± 0.17	0.29 ± 0.12	0.56 ± 0.21	<0.05 in 5D
MMP-9
C57BL/6	1.00 ± 0.00	1.51 ± 0.17	2.16 ± 0.44*	<0.01 in 10D
BALB/c	1.12 ± 0.41	0.52 ± 0.18	0.85 ± 0.07
TIMP-1
C57BL/6	1.00 ± 0.00	1.02 ± 0.33	1.49 ± 0.36
BALB/c	0.78 ± 0.45	1.10 ± 0.49	1.39 ± 0.44
TIMP-2
C57BL/6	1.00 ± 0.00	1.00 ± 0.29	1.24 ± 0.48
BALB/c	0.40 ± 0.11	0.26 ± 0.24	0.55 ± 0.14
TIMP-3
C57BL/6	1.00 ± 0.00	0.80 ± 0.18	0.93 ± 0.28
BALB/c	0.25 ± 0.10	0.41 ± 0.38	0.72 ± 0.30
TIMP-4
C57BL/6	1.00 ± 0.00	0.63 ± 0.35	1.63 ± 0.30, †	<0.01 in 10D
BALB/c	0.17 ± 0.02	0.14 ± 0.09	0.26 ± 0.16

 Data were obtained from C57BL/6 and BALB/c mice by using real time PCR and are the mean (relative expression) ± SEM of results in four separate experiments. UT, untreated controls.

*   P < 0.05,

**   P < 0.01,

***   P < 0.001 vs. C57BL/6-UT or BALB/c-UT.

†   P < 0.05,

††   P < 0.01,

†††   P < 0.001 vs. C57BL/6-5D or BALB/c-5D.

Table 3.

 

Percentage Difference in Transcript Levels from the C57BL/6 Untreated Cornea

	MMP-1	MMP13	MMP3	MMP10	MMP-7	MMP-9	TIMP-1	TIMP-2	TIMP-3	TIMP-4
C57BL/6-5D	175.84	−33.77	109.11	79.12	6.00	50.64	2.31	0.36	−20.27	−36.80
C57BL/6-10D	−8.85	−68.37	111.13	7.31	−5.04	116.11	48.88	24.42	−7.06	63.05
BALB/c-UT	−3.52	−54.97	−11.66	30.12	−66.83	11.86	−22.00	−60.15	−74.95	−83.45
BALB/c-5D	64.48	−77.30	−54.74	27.16	−71.26	−48.05	9.75	−73.74	−58.68	−85.72
BALB/c-10D	−24.83	−26.18	−53.01	51.23	−44.23	−15.32	38.86	−45.28	−28.22	−74.26

 Percentages represent the difference in the number of transcripts in these samples detected by real-time PCR compared with untreated C57BL/6.

Table 4.

 

Effect of 5 (5D) and 10 Days (10D) of Desiccating Stress on MMP-1, MMP-3, and -9 Concentrations (pg/μg of total protein) in the Mouse Corneal Epithelium

	UT	EDE-2D	EDE-5D	EDE-10D	Interstrain P
MMP-1
C57BL/6	0.012 ± 0.001	0.008 ± 0.001	0.016 ± 0.002, †	0.012 ± 0.003	P < 0.05 in 5D
BALB/c	0.007 ± 0.004	0.007 ± 0.003	0.006 ± 0.001	0.005 ± 0.001
MMP-3
C57BL/6	0.013 ± 0.009	0.125 ± 0.0267	0.146 ± 0.024*	0.327 ± 0.035, *** , †† , ‡‡	P < 0.0001 in 10D
BALB/c	0.043 ± 0.010	0.030 ± 0.003	0.031 ± 0.039	0.060 ± 0.046
MMP-9
C57BL/6	0.101 ± 0.011	0.124 ± 0.007	0.128 ± 0.013	0.100 ± 0.003	P < 0.05 in 2D
BALB/c	0.065 ± 0.011	0.078 ± 0.014	0.086 ± 0.004	0.064 ± 0.011	P < 0.05 in 5D

 Data were obtained from C57BL/6 and BALB/c mice determined by enzyme activity assay and are expressed as the mean ± SD. UT, untreated control.

*   P < 0.05,

**   P < 0.01,

***   P < 0.001 vs. C57BL/6-UT.

†   P < 0.05,

††   P < 0.01 vs. C57BL/6-2D.

‡‡   P < 0.01 vs. C57BL/6-5D.

Figure 1.

 

Concentrations of MMPs in tears of C57BL/6 and BALB/c mice after 2 (2D), 4 (4D), 6 (6D), 8 (8D), and 10 (10D) days of experimentally induced dry eye, as determined by enzyme activity assay. (A) MMP-1 showing among group differences (B) MMP-1 showing between group differences (C) MMP-3 (D) MMP-9. (UT): Untreated control. Data are showed as the mean ± SEM. * P < 0.05, * * P < 0.01, ** * P < 0.01

Figure 2.

 

Immunofluorescent staining and laser scanning confocal microscopy in cornea tissue sections of C57BL/6 mice stained with polyclonal antibodies (green) to the MMPs shown. The nucleus was counterstained with PI (red), and merged images were produced. Representative staining in C57BL/6 corneal epithelium from control untreated (UT) mice and with EDE for 2 (2D), 5 (5D), or 12 (12D) days. The staining results are provided in Table 5 . Arrow: corneal epithelial layer. Scale bar, 50 μm.

Figure 3.

 

Immunofluorescent staining and laser scanning confocal microscopy in cornea tissue sections of BALB/c mice. The remaining information is as described in Figure 2 .

Table 5.

 

Quantitation of Immunofluorescent Staining Intensity for MMPs

	C57-BL6							BALB-c
		CN							CN
		UT	2D	5D	12D	UT	2D	5D	12D
MMP-1
Superficial	+	++	+++	0	+	0	0	0
Basal	0	0	0	0	0	0	+	0
MMP-13
Superficial	+	+	0	0	0	+	+	+
Basal	0	0	0	0	0	+	+	0
MMP-3
Superficial	0	+	+	+	+	0	0	+
Basal	0	+	+	+	+	0	0	+
MMP-10
Superficial	++	++	+	++	+	++	++	++
Basal	0	0	+	0	+	+	0	0
MMP-7
Superficial	+	++	+++	+	0	0	0	+
Basal	0	0	0	0	0	0	0	0
MMP-9
Superficial	+	+	+++	++	+	+	++	+
Basal	0	+	+++	++	0	+	++	+

 Levels: 0, no different than the secondary antibody control; +, slightly greater than the secondary antibody control; ++, moderate staining; +++, intense staining.

Figure 4.

 

Corneal permeability measured as the uptake of OGD (A) and corneal smoothness evaluated by reflection of white ring light (B) in C57BL/6 and BALB/c mice before (UT), after 5 days (5D; center), and after 10 days (10D; right) of EDE. Scale bar, 500 μm.

Table 6.

 

Effect of 5 (5D) and 10 Days (10D) of Desiccating Stress

	UT	EDE-5D	EDE-10D	Interstrain P
Corneal staining with OGD
C57BL/6	0.323 ± 0.119	2.598 ± 0.319, ***	2.646 ± 0.481, ***	<0.001 in 5D
BALB/c	0.778 ± 0.263	0.860 ± 0.156	0.982 ± 0.388	<0.01 in 10D
Corneal fluorescein permeability
C57BL/6	359.8 ± 24.8	634.2 ± 77.3, **	696.0 ± 83.98, ***	<0.001 in 10D
BALB/c	211.6 ± 26.87	495.0 ± 74.10, **	237.6 ± 17.89, ††
Corneal smoothness
C57BL/6	0.357 ± 0.133	2.625 ± 0.324, ***	1.250 ± 0.250* , ††
BALB/c	0.313 ± 0.162	2.438 ± 0.395, ***	1.500 ± 0.323, ** , †

 Data on corneal staining with OGD, permeability to carboxyfluorescein, and corneal smoothness (severity score) were obtained from C57BL/6 and BALB/c mice and are expressed as the mean ± SEM.

*   P < 0.05,

**   P < 0.01,

***   P < 0.001 vs. untreated (UT) mice.

†   P < 0.05,

††   P < 0.01 vs. EDE 5 days.