Corneal Endothelial Cell Survival in Organ Cultures under Acute Oxidative Stress: Effect of VIP

Abstract

purpose. Human corneal endothelium, a neural crest–derived tissue, has a very limited regenerative capacity and may depend on trophic factors for its survival throughout life, as well as after injury and during storage before transplantation. The purpose of this study was to determine whether vasoactive intestinal peptide (VIP), a 28-amino acid neurotrophic factor present in human aqueous humor, promotes the survival of corneal endothelium in corneal organ cultures, and whether VIP is produced by the corneal endothelium.

methods. Thirteen viable human donor corneas that had been received from the Central Florida Lions Eye Bank and stored in preservation medium (Optisol-GS; Chiron Vision, Irvine, CA) at 4°C for 8 to 17 days were bisected. Each half was treated with either 0 or 10 nM VIP (15 minutes) and subjected to H₂O₂ (1.4 mM, 30 minutes) treatment at 37°C. The numbers of live and dead corneal endothelial (CE) cells isolated from the corneas were then determined under fluorescence microscopy using a live–dead viability–cytotoxicity assay conducted by an observer uninformed of the treatment. The effect of VIP (10^{− 16} to 10^{− 6} M) on CE cell survival was also studied in fresh bovine corneas in situ, by using the same assay. The presence of VIP in the corneal endothelium in fresh human donor and bovine eyes was examined by immunocytochemistry, in situ hybridization, and Western blot analysis, whereas VIP in the bovine aqueous humor was assessed by radioimmunoassay.

results. VIP (10 nM) significantly increased CE survival in 10 of 13 human corneas. The mean survival of CE cells (±SEM) was 42% ± 3% in control corneas versus 59% ± 3% in VIP-treated corneas (P < 0.001). In bovine corneas, VIP at concentrations as low as 10^{− 10} M demonstrated a significant protective effect. The mean number of dead CE cells on bovine corneas was maximally decreased by 10^{− 6} M VIP from 46 ± 5 to 18 ± 3 per field. In CE cells from fresh human and bovine corneas, VIP immunoreactivity and mRNA were detected. VIP was also present in bovine aqueous humor at 40 ± 8 pM.

conclusion. VIP may be an autocrine trophic factor that protects CE cells from H₂O₂ in normal aqueous humor and possibly from other oxidative insults.

The cornea is a major site of light refraction in the eye. The transparency of the cornea is maintained by the corneal endothelium, which functions as a physical barrier to movement of fluid into the cornea and also actively pumps fluid out of the cornea. ¹ Human corneal endothelium has a very limited regenerative capacity, ² and corneal wound repair is therefore achieved primarily by cell enlargement and by redistribution of existing cells. ³ Extensive injuries do not heal well, resulting in irreversible corneal edema and corneal clouding that necessitates corneal transplantation. Donor corneas are commonly stored in the preservation medium at 4°C for several days before they are used for transplantation. ^{4 5}  

To survive throughout life and during storage, corneal endothelial (CE) cells may depend on survival-promoting factors. The corneal endothelium derives from the neural crest, ^{6 7 8 9} and expresses neuron-specific enolase. ^{10 11} Thus, neurotrophic factors, which are well known for their effects on promoting the survival of neuronal cells both in vitro and in vivo, may also play a role in the promotion of corneal endothelial cell survival. Vasoactive intestinal peptide (VIP) is a 28-amino acid neuropeptide that exhibits trophic activity on cultured cells. ^{12 13 14 15 16 17} For example, VIP has been shown to increase the survival of cultured sympathetic neuroblasts ¹³ and neurons in mixed cultures of neurons and glia ^{14 15 16} and prevents neuronal cell death in vitro induced by the external envelope protein (gp 120) of the human immunodeficiency virus, the causative agent of acquired immune deficiency syndrome (AIDS). ¹⁷ VIP attenuates cell death caused by the oxidative stress induced by glutamate in PC12 cells. ¹⁸ The glutamate-induced neurotoxicity in cultured rat retinal neurons was inhibited by VIP through a cyclic adenosine monophosphate (cAMP)/protein kinase A–dependent mechanism. ¹⁹ In isolated rat lung, VIP has been shown to protect against the nitric oxide–dependent acute injury caused by N-methyl-d-aspartate. ^{18 20}  

In addition to the neurotrophic function, VIP has also been shown to possess a wide variety of immunomodulatory activities. ^{21 22} Whereas VIP has been identified as one of the immunosuppressive factors found in the aqueous humor, ^{23 24} VIP immunoreactivity in the human aqueous humor has been reported ²⁵ and confirmed (Koh et al., unpublished data, 2000). Although the origin of VIP present in the aqueous humor has not been established, the corneal endothelium, which is one of the tissues bathed in the aqueous humor, is one possible source. Previous studies in human neuroblastoma cells have demonstrated that VIP is processed from a high-molecular-weight precursor. ^{26 27}  

Trophic factors may protect cells from acute free radical stress by increasing glutathione (GSH) synthesis and the expression of other free radical scavengers. ^{28 29} In contrast, Said et al. ¹⁸ have reported that the protective effect of VIP against the oxidative stress in PC-12 cells is not associated with the prevention of the decline of GSH synthesis. The present study showed that VIP treatment of both human and bovine corneoscleral explant organ cultures provided protection of CE cells from the acute killing effect of hydrogen peroxide, which is a normal constituent of aqueous humor. ^{30 31} VIP may thus play a role as an autocrine trophic factor of the corneal endothelium. 

Methods

Eyes and Corneoscleral Explants

Viable corneoscleral explants preserved for 8 to 17 days in preservation medium (Optisol-GS; Chiron Vision, Irvine, CA) were obtained from the Central Florida Lions Eye Bank (Table 1) . These were deemed unsuitable for transplantation because of the extended time in storage or for other reasons. Bovine eyes were obtained from the local abattoir and used within 6 hours of death. Human eyes with postmortem times of less than 24 hours that were without disease but unsuitable for cornea transplantation were obtained from the Maryland Eye Bank. The corneoscleral explants were removed from the globes using a scleral incision 2 mm posterior to the limbus and by gently teasing away the iris and ciliary body. After removal of the associated iris root using forceps, the corneoscleral explants were placed in phosphate-buffered saline (PBS) at 4°C. 

VIP Pretreatment and H₂O₂ Treatment of Corneoscleral Explants

The Optisol-GS–preserved human corneoscleral explants were removed from the preservation vials, bisected, and transferred to and incubated for 15 minutes in Eagle’s minimal essential medium plus 20 mM HEPES (EMEM-HEPES; 1 ml, pH 7.2) containing either zero or 10 nM VIP, and subsequently with 1.4-mM H₂O₂-PBS (1 ml) for 30 minutes to produce an adequate degree of CE cell injury, ³² at 37°C in 24-well tissue culture plates. 

Bovine corneoscleral explants were conditioned in EMEM-HEPES in 35-mm tissue culture dishes (3 ml) at 37°C for 60 minutes, transferred to and incubated in fresh medium containing the designated concentrations of VIP (0 and 10⁻¹⁶ to 10⁻⁶ M) for 15 minutes at 37°C, and subsequently treated with 1.4 mM H₂O₂-PBS (3 ml) for 30 minutes at 37°C. 

Viability of the Corneal Endothelium

The viable and dead corneal endothelial cells in corneoscleral explants were revealed simultaneously using fluorescence microscopy according to the procedure provided with the live–dead viability–cytotoxicity kit (Molecular Probes, Eugene, OR). The mixture of calcein acetoxymethyl ester (AM) and ethidium homodimer was added to cover and incubate the corneal endothelium in corneoscleral explants (1 and 2 ml for human and bovine corneas, respectively) for 30 minutes at 37°C. Live cells (possessing intracellular esterase activity) convert calcein AM to calcein. Calcein is retained in the cell and produces green fluorescence when excited. Dead cells (with their compromised plasma membranes) allow the entrance of ethidium homodimer which undergoes a 40-fold increase in red fluorescence after binding to nucleic acids. Therefore, the nuclei of the dead cells appear red under the fluorescence microscope. 

Human corneal endothelium sheet with areas of attached Descemet’s membrane was scraped from corneoscleral explants using a razor blade and transferred to and incubated in microfuge tubes containing 0.4 ml of 10 mM EDTA in Mg²⁺- and Ca²⁺-free PBS for 30 minutes at 37°C. After the tubes were vortexed to release Descemet’s membranes, the CE cells in the supernatants were spun down in a microfuge (12,000g, 1 minute). CE cells were resuspended in 20 μl of calcein AM-ethidium homodimer solution. From each of the tubes, duplicate 8-μl cell suspensions were placed in each of the two wells (8 mm in diameter; area, 50.2 mm²/well) of an eight-well slide (Erie Scientific, Portsmouth, NH) and each well covered by a coverslip. With identities of the corneoscleral explants withheld from the examiner and under an inverted microscope (×200 magnification) equipped with an epifluorescence attachment (Diaphot-TMD; Nikon, Tokyo, Japan), the numbers of green (live) cells and red nuclei of dead cells in the field (area, 0.33 mm²) defined by the photograph mask that was placed in the optical path of the microscope were counted. The percentage of live cells was derived by dividing the number of green cells by the combined numbers of green cells and red nuclei. Values were calculated in five masked fields in each of the two wells and were pooled to give a mean value for each hemisection. Thus, cells in a total of 10 microscopic fields (area, 10 × 0.33 mm²) in two wells (area, 2 × 50.2 mm²) were counted (3.3%). Because the two wells (2 × 8 μl) contained 80% of the total CE cell suspension (20μ l) from each cornea hemisection, the cells counted represented 2.63% of the total CE cell population from each cornea hemisection. 

The bovine corneoscleral explants were quartered and flatmounted (with the endothelium side down) on coverslips and photographed using a fluorescence microscope (at ×200 magnification), as described for human CE cells. In separate experiments, with the identities of the corneoscleral explants withheld from the examiner, the numbers of dead cells (red nuclei) in each of the fields (as for human cells) in the corneal endothelium were counted. A total of 14 fields were counted for each cornea. To determine the number of bovine CE cells that occupied one field of confluent cell layer, bovine corneoscleral explants were stained with 1% alizarin red S to reveal the cell–cell junctions, and photomicrographs of six fields were taken for cell counting. 

Statistical Analysis

Data obtained from the control and VIP-treated human donor corneoscleral explants were analyzed by Student’s t-test, whereas those in which the effects of multiple concentrations of VIP were determined were analyzed by analysis of variance (ANOVA) followed by Dunnett’s post hoc test. 

VIP Immunostaining

Corneoscleral explants dissected from fresh donor human eyes were fixed in 4% paraformaldehyde in PBS, embedded in paraffin, and sectioned (6 μm) for VIP immunostaining. Freshly dissected bovine corneoscleral explants were frozen in optimal cutting temperature compound (OCT; Miles; Elkhart, IN). Cryostat sections (8 μm) were mounted on slides (Superfrost Plus; Fisher Scientific, Pittsburgh, PA) and stored at −70°C. Slides were transferred while still frozen to 4% paraformaldehyde in PBS and immunostained for VIP using either a polyclonal rabbit anti-VIP (10 μg IgG/ml, in PBS; ICN, Costa Mesa, CA) or rabbit IgG (10 μg IgG/ml in PBS; ICN), biotinylated goat anti-rabbit IgG, extrAvidin-alkaline phosphatase conjugate (E-2636; Sigma St. Louis, MO), and a kit (Fast Red TR/Naphthol AS-MX tablet set; Sigma) to develop the red reaction product in positively stained cells. 

In Situ Hybridization

Freshly dissected human and bovine corneas were frozen in OCT. Cryostat sections (8 μm) were mounted on slides (Superfrost Plus, Fisher) and stored at −70°C. Slides were transferred while still frozen to 4% paraformaldehyde in PBS for 15 minutes, baked at 37°C for 90 minutes, then washed in PBS (three times for 5 minutes each). Slides were treated 10 minutes with 0.25% (vol/vol) acetic anhydride in 0.1 M triethanolamine (pH 8.0) at room temperature to acetylate basic residues on the slide, thus reducing nonspecific hybridization. After a brief rinse in PBS, samples were dehydrated with ascending concentrations of ethanol: 50%, 70%, 95%, and 100% (two times for 2 minutes in each concentration). Samples were then incubated for 1 to 2 hours at 60°C in a prehybridization solution of 4× SET (1× SET contains 150 mM NaCl, 1 mM EDTA, and 20 mM Tris-HCl [pH 7.8]), 1× Denhardt’s (0.2% sodium dodecyl sulfate [SDS], 100 mM dithiothreitol [DTT], 250 μg/ml transfer RNA, and 25 μg/ml each of polyA and C), and 50% formamide. Hybridization buffer consisted of prehybridization buffer (except formamide and dextran sulfate were added to 30% and 10%, respectively). To this, one-tenth volume of freshly-labeled mouse cDNA probe ³⁴ was added so that the final probe concentration was 50 to 100 × 10³ cpm/μl. An aliquot of 20 μl of this was added to each tissue section, and the sections were then sealed under a silanized glass coverslip and incubated for 16 hour at 60°C in a humid chamber. Slides were immersed the next day briefly in 4× SSC/1 mM DTT (20× SSC containing 175.3 g/l NaCl and 88.2 g/l Na citrate [pH 7.0]) to remove coverslips, in 2× SSC/1 mM DTT for 1 hour at room temperature, in wash buffer (500 mM NaCl, 10 mM Tris [pH 7.6], and 1 mM EDTA) containing 0.8 μg/ml RNase A for 30 minutes at 37°C, in wash buffer containing 1 mM DTT for 15 minutes at 37°C, in 2× SSC/1 mM DTT for 30 minutes at 37°C, in 0.1× SSC for 20 minutes (two times) at 60°C, then in 1× SSC 15 minutes at room temperature. Slides were dehydrated in ascending alcohols containing 300 mM ammonium acetate with a final dehydration in 100% ethanol. For autoradiography, slides were dipped in emulsion (1:1 with distilled water; NTB-2; Eastman Kodak, Rochester, NY) at 45°C. After developing, slides were counterstained with hematoxylin. 

SDS-PAGE and Western Blot Analysis

CE cells with some areas of attached Descemet’s membrane were scraped from the corneas, homogenized in the RIPA buffer (25 mM Tris[ pH 7.2], 0.1% SDS, 1% Triton X-100, 1% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA), followed by centrifugation at 12,000g (10 minutes) to remove the insoluble materials including the Descemet’s membrane and to obtain the extract fractions. Cell extracts were diluted (1:1) with a sample buffer containing 5%β -mercaptoethanol for sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) using preformed Tris-glycine 8% to 16% polyacrylamide gels (Novex; San Diego, CA). The electrophoresed proteins were electrophoretically transferred to a nitrocellulose membrane for Western blot analysis using an enhanced chemiluminescence kit (ECL; Amersham Pharmacia, Piscataway, NJ). A rabbit polyclonal anti-VIP antibody (ICN Biochemicals, Aurora, OH) was used at 1:1250 dilution in 1% bovine serum albumin (BSA)-PBS. To block the specific anti-VIP antibody reactivity, it was preincubated with the synthetic VIP (molecular weight, 3326; Sigma) at 1.7 × 10⁻⁷ M at 4°C for 20 hours. 

Radioimmunoassay (RIA) for VIP in Bovine Aqueous Humor

VIP-immunoreactivity in bovine aqueous humor was assayed using a radioimmunoassay kit (Research and Diagnostic Antibodies, Berkeley, CA) according to the instruction provided by the manufacturer. 

Results

Increased CE Cell Survival in H₂O₂-Treated Human Corneoscleral Explants by VIP Pretreatment

Corneas were bisected and each half treated for 15 minutes with either 0 or 10 nM VIP. After a subsequent 30-minute incubation with 1.4 mM H₂O₂, the percentage of live CE cells was determined in each hemisection in 10 microscopic fields containing 2.63% of the total CE cell population (see the Materials and Methods section). As shown in Figure 1 , the corneoscleral explant hemisections treated with VIP showed a higher mean percentage of live CE cells after H₂O₂ treatment than untreated hemisections from the same corneoscleral explants. An increase in CE cell survival in VIP-treated portions was observed in 12 of the 13 corneoscleral explants. Ten of these showed a statistically significant increase. The maximal increase in percentage of live CE cells in a given cornea was approximately threefold, from 23% ± 2% to 70% ± 2% (mean ± SEM; donor 1). The percentages of live CE cells in control or VIP-treated corneoscleral explants did not correlate with the age of the donor, or the number of days in the preservation medium (given in Table 1 ; data not shown). The overall mean survival (± SEM, n = 13) was 42% ± 3% in control versus 59% ± 3% in VIP-treated explants (P < 0.001). VIP treatment of corneoscleral explants significantly increased the numbers of live CE cells after the subsequent H₂O₂ treatment. The mean numbers of live cells per field (± SEM) was 9.4 ± 2.3 in control versus 22.7 ± 7.6 in VIP-treated explants (P = 0.037). In contrast, although VIP also increased the total number of (live plus dead) cells, the effect of VIP was not statistically significant. The mean numbers of live plus dead cells per field (± SEM) was 23.5 ± 4.9 in control versus 37.9 ± 11.5 in VIP-treated explants (P = 0.12). 

VIP Concentration Dependency of CE Cell Survival in H₂O₂-Treated Bovine Corneoscleral Explants

Using the live–dead viability–cytotoxicity kit, the nuclei of the dead cells appeared as red dots on a carpet of green (live) CE cells in flatmounted corneas under the fluorescence microscope. As shown in Figure 2 , VIP pretreatment of the cornea helped CE cells survive the subsequent H₂O₂ treatment in a VIP concentration-dependent manner. For each cornea, the numbers of red nuclei (of the dead CE cells) were counted in 14 fields, each with an area of 0.33 mm² in which 592 ± 31 (mean ± SEM, n = 6) CE cells formed a confluent cell layer. As shown in Figure 3 , the numbers of red nuclei decreased in a VIP concentration–dependent manner. The averaged numbers of red nuclei per field were 46, 38, 35, 23, 22, 23, 24, 30, and 18 in the bovine corneoscleral explants treated with 0, 10⁻¹⁶, 10⁻¹⁴, 10⁻¹², 10⁻¹⁰, 10⁻⁹, 10⁻⁸, 10⁻⁷, and 10⁻⁶ M VIP, respectively (P = 0.0001, ANOVA; Fig. 3 ). The Dunnett’s post hoc test showed that VIP at concentrations of 10⁻¹² M and higher (with the exception of 10⁻⁷ M) produced significant effects (P < 0.05) on reducing the number of red nuclei in the bovine corneal endothelium. The decrease in cell survival seen at 10⁻⁷ M VIP was surprising but was observed in four of six experiments. 

VIP Immunostaining

Figure 4 shows the presence of endogenous VIP immunoreactivity in the CE cells in corneoscleral explants dissected from fresh donor human eyes (without storage in the preservation medium; Fig. 4A ) and those from fresh bovine eyes (Fig. 4C) . VIP immunoreactivity was detected in the corneal endothelium in all donor eyes (ages 4, 82, and 92). Results shown here were those from the 82-year-old donor. VIP immunoreactivity was observed in the corneal endothelium sectioned from either frozen or paraffin-embedded corneoscleral explants. 

In Situ Hybridization of VIP mRNA

In situ hybridization revealed VIP gene transcripts in the CE cells in corneoscleral explants of all eyes examined (two fresh human donor eyes and one bovine eye). Figure 5 shows corneal cryosections from a 72-year-old human donor and a bovine eye, hybridized to VIP mRNA riboprobes. Specific hybridization signals were observed with antisense riboprobe (Figs. 5A 5B 5C) , whereas no significant signals were observed with sense probe (Figs. 5D 5E 5F) . 

Western Blot Analysis of VIP Immunoreactive Molecules

Western blot analysis showed that both human and bovine CE cell extracts contained a anti-VIP immunoreactive molecule with molecular weight of 40 kDa and a doublet with approximate molecular weight of 20 kDa (Fig. 6) . Human retina also expressed these VIP-immunoreactivities (Fig. 6) . The immunoreactivity of the 20-kDa molecule in human and bovine CE cell extract was blocked by preincubating the anti-VIP antibody with the synthetic VIP (Figs. 7A 7B ). The 40-kDa immunoreactive species in the human CE extract was not blocked by preincubation (Fig. 7A) , and that in the bovine CE cell extract was partially blocked (Fig. 7B) , indicating that the 40-kDa band probably was nonspecific. 

VIP in Bovine Aqueous Humor

VIP immunoreactivity was present at 40 ± 8 pM in the bovine aqueous humor (n = 4 eyes). 

Discussion

The aqueous humor has been postulated to contain immunosuppressive factors, including VIP. ^{23 24} Although the origin and nature of many factors present in the aqueous humor has not been established, the corneal endothelium, is one possible source. The current studies are the first to demonstrate VIP immunoreactivity in the corneal endothelium (Fig. 4) . That this represents authentic VIP is supported by the finding that VIP gene expression was clearly detected in the corneal endothelium of freshly dissected human and bovine eyes by in situ hybridization (Fig. 5) . On the contrary, Denis et al. ³³ reported that VIP gene transcripts could not be detected in the cornea of rat eyes. Although this could represent a species difference, the ability to detect VIP gene transcripts in fresh human and bovine corneas dissected from whole globes may be because the assay was more sensitive in current studies or because an upregulation of VIP gene expression occurred after enucleation and storage (e.g., injury response). The high specificity and sensitivity of the assay was demonstrated in previous studies that demonstrated VIP gene expression at very early stages of embryonic brain development. ³⁴ Moreover, the present studies demonstrated two specific anti-VIP immunoreactive molecules with molecular weight close to that of the prepro-VIP and pro-VIP ^{26 27} (Figs. 6 7) in cell extracts of both human and bovine corneal endothelium. Other investigators have shown that VIP and VIP mRNA are present in other nonneuronal cells, such as polymorphonuclear leukocytes, mast cells, ³⁵ T and B lymphocytes, ³⁶ non–small-cell lung cancer cells, ³⁷ and endothelial cells of human umbilical blood vessels. ³⁸ Although the primary in vitro translation product of the mRNA encoding VIP is the 20-kDa prepro-VIP, a 17.5-kDa pro-VIP has been isolated from the human neuroblastoma cells. ^{26 27}  

The current studies demonstrated for the first time that VIP immunoreactivity, VIP mRNA, and two anti-VIP immunoreactive molecules with approximate molecular weight of 20 kDa are expressed in the corneal endothelium. Whereas the VIP immunoreactivity in human retina has been demonstrated by immunocytochemical methods, ^{39 40} the present study also demonstrated in the human retina the presence of anti-VIP immunoreactive molecules with molecular weight close to that of known VIP precursors (Fig. 6) . 

The present studies showed that VIP pretreatment of the corneas increased CE cell survival after subsequent hydrogen peroxide challenge (Figs. 1 2 3) . Hydrogen peroxide is a normal constituent of the aqueous humor and has been hypothesized to play some role in the loss of corneal endothelial cells during normal aging. VIP immunoreactivity has been detected in the aqueous humor from rabbits ²⁴ and humans ²⁵ and now in bovine aqueous humor. We have also confirmed the presence of VIP immunoreactivity in the aqueous humor of patients undergoing cataract surgery (Koh et al., unpublished data, 2000). The protective effect of VIP on CE cells against hydrogen peroxide may be critical in eyes after corneal transplantation, trabeculectomy, or cataract surgery, in which the level of hydrogen peroxide present in the aqueous humor is likely elevated because of the presence of invading macrophages and other immune cells ⁴¹ that produce hydrogen peroxide. ⁴² Whether a brief VIP treatment of the donor corneas before transplantation may have a beneficial effect on the survival of the CE cells in the recipient eyes remains to be investigated. 

In summary, the present study indicates that VIP plays a role in promoting the survival of corneal endothelium under acute oxidative stress. The expression and release of VIP, a putative immunomodulator, by the CE cells may help to maintain an immune-privileged anterior chamber in the eye. 

 Supported in part by Grants EY11607, HD04612, HD06576, and HD34475 from the National Institutes of Health; the Pangborn award from University of Maryland, and a development grant from Research to Prevent Blindness.

 Submitted for publication February 11, 2000; revised August 14, 2000; accepted September 6, 2000.

 Commercial relationships policy: N.

 

Table 1.

 

Donor Human Corneas Subjected to Acute Oxidative Stress

Donor	Donor Age	Race/Gender	Cause of Death	Days Preserved
1	72	W/M	Lung cancer	8
2*	68	W/F	Cardiac crisis	16
3	75	W/M	Stroke	17
4*	68	W/F	Cardiac crisis	16
5	68	W/M	Lung cancer	15
6, †	18	W/F	Auto accident	12
7, †	18	W/F	Auto accident	12
8	74	W/F	Cardiac crisis	13
9	69	W/M	Lung cancer	12
10, ‡	45	W/F	Cardiac crisis	14
11	40	W/M	Auto accident	17
12, ‡	45	W/F	Cardiac crisis	14
13	66	W/F	Lung cancer	13

 Donor corneas were numbered in the order of increasing percentage of live cells found in the control hemisection of the corneas (see Fig. 1 ).

*  ,

†  ,

‡  : Denote corneas from the same donors.

Figure 1.

 

The effect of VIP pretreatment on CE cell survival in H₂O₂-treated human corneoscleral explant organ cultures. Bisected explant halves were treated with either 0 (hatched bar) or 10 nM VIP (filled bar) before H₂O₂ treatment. After isolation of CE cells and determination of live and dead cell numbers, the percentage of live cells was derived by dividing the number of live cells by the combined numbers of live cells and red nuclei (of the dead cells). The number assignments for the donor eyes (corneoscleral explants) are in order of increasing percentage of live cells found in the control halves of the explants, from the lowest (donor 1) to the highest (donor 13). Significant difference between the control and VIP-treated halves of the explant: *P < 0.05; **P < 0.005.

Figure 2.

 

Fluorescent photomicrographs of bovine corneal endothelium in flatmounted corneas. Explants were pretreated with varying concentrations of VIP (15 minutes), treated with H₂O₂ (1.4 mM; 30 minutes), and subsequently reacted with reagents from the live–dead viability–cytotoxicity kit. In this black-and-white picture, the red fluorescent nuclei from dead cells appear as dots on a carpet of live cells. VIP concentrations: 0 (A), 1 × 10⁻¹⁴ (B), 1 × 10⁻¹² (C), 5 × 10⁻¹¹ (D), 1 × 10⁻¹⁰ (E), and 1 × 10⁻⁸ (F) M.

Figure 3.

 

VIP-concentration dependency of corneal endothelium survival in VIP-H₂O₂–treated bovine corneoscleral explants. Pretreatment of explants with VIP (15 minutes) promoted corneal endothelial cell survival during subsequent H₂O₂ treatment (1.4 mM; 30 minutes). The differences among the nine groups were significant at P = 0.0005 (ANOVA). The Dunnett’s post hoc test showed that VIP at concentrations of 10^{− 12} M and higher (with the exception of 10^{− 7} M) produced significant effects (P < 0.05). The results presented are the averaged data from six experiments. The number of corneas used for each of the VIP concentrations: 0 M (11), 1 × 10^{− 16} M (4), 1 × 10^{− 14} (6), 1 × 10^{− 12} M (10), 1 × 10^{− 10} (11), 1 × 10^{− 9} (10), 1 × 10^{− 8} (9), 1 × 10^{− 7} (8), and 1 × 10^{− 6} M (9).

Figure 4.

 

VIP immunoreactivity in corneal endothelium of an 82-year-old donor human eye (A, B) and a bovine eye (C, D). (A, C) anti-VIP; (B, D) rabbit IgG. (A, B) Counterstained with hematoxylin.

Figure 5.

 

VIP mRNA in human and bovine corneal endothelium. Photomicrographs of cryosections of a human cornea from a 72-year-old donor (A, B, D, and E) and that of a bovine cornea (C, F) processed for in situ hybridization using antisense (A, B, and C) and sense (D, E, and F) VIP cDNA. Dark-field images demonstrate autographic signals in the human corneal endothelium (A, D). Bright-field images demonstrate autographic signals over human corneal endothelium (B, E); bovine corneal endothelium (C, F).

Figure 6.

 

Western blot analysis of the anti-VIP immunoreactive molecules in cell extracts of corneal endothelium (human and bovine) and the retina (human). Human corneal endothelium of fresh human eyes (both OD and OS) of three donors (ages: 34, 46, and 65 years), human retina from both eyes of a 67-year-old donor, and corneal endothelium from fresh bovine eyes were extracted. Each lane of an 8% to 16% polyacrylamide gel contained approximately 45 μg protein. Three separate experiments were conducted with the same results.

Figure 7.

 

Anti-VIP antibody immunoreactivity in cell extracts of corneal endothelium was blocked by VIP. Preincubation of the anti-VIP antibody solution with VIP blocked the immunoreactivity of two molecules (molecular weight ∼20 kDa) in cell extracts of corneal endothelium from fresh human donor (A) and bovine (B) eyes. Human corneal endothelium from both eyes of three donors (age: 34, 46, and 65 years) and corneal endothelium from fresh bovine eyes were extracted. Each lane contained 27 μg protein. The experiment was repeated three times.

The authors thank Timothy Coll, Robert Casillas, and William Rodriguez for technical assistance and the Central Florida Lions Eye Bank and the Maryland Eye Bank for providing human donor corneas and eyes.