Expression and Neuroprotective Effect of Hepatocyte Growth Factor in Retinal Ischemia–Reperfusion Injury

Abstract

purpose. To investigate the expression and possible neuroprotective effects of hepatocyte growth factor (HGF) in a rat model of retinal ischemia–reperfusion injury.

methods. Retinal ischemia was induced in adult male Sprague-Dawley rats by raising the intraocular pressure to 110 mm Hg for 45 minutes. To study expression of HGF and its receptor c-Met, reverse transcription–polymerase chain reaction (RT-PCR), Western blot analysis, and immunohistochemical staining were performed on eyes enucleated at 6, 12, 24, 48, and 96 hours after reperfusion. To examine the neuroprotective effects of HGF, recombinant human (rh)HGF (1, 6, and 12 μg in 2 μL PBS) or vehicle was administered intravitreally 1 minute after reperfusion, and the eyes were enucleated at 6, 12, 24, 48, and 96 hours and 28 days after reperfusion. The retinal damage was assessed by electroretinogram (ERG) recordings, by measuring the inner retinal thickness, and by counting the number of TUNEL-positive cells in each retinal layer.

results. RT-PCR and Western blot analyses showed upregulation of HGF and c-Met–HGF receptor mRNA at 6, 12, 24, and 48 hours after reperfusion, compared with the normal rat retina. Immunohistochemically, expression of HGF was found in the retinal pigment epithelial cells at 6 hours after reperfusion and in some cells in the ganglion cell layer and inner nuclear layer at 24 hours after reperfusion. The amplitudes of the ERG b-wave and oscillatory potentials were significantly larger in the eyes treated with 6 and 12 μg rhHGF than in those of vehicle-treated control rats (P < 0.01). On day 28, the thicknesses of the inner retina of vehicle-treated rats and that of 6-μg rhHGF-treated rats were 54.4 ± 6.12 (mean ± SD, n = 9) and 71.5 ± 9.81 μm (n= 8), respectively (P < 0.01). The number of TUNEL-positive cells at 6, 12, 24, and 48 hours after reperfusion was decreased significantly by treatment with 6 μg rhHGF, compared with those in the control rats (P < 0.01).

conclusions. Upregulation of HGF in the retina may play a role in retinal ischemia–reperfusion injury. Intravitreal injection of rhHGF is neuroprotective against the injury.

Hepatocyte growth factor (HGF) was originally identified and cloned as a potent mitogen for hepatocytes. ^{1 2 3} This growth factor acts as a mitogen and a morphogen for a variety of epithelial cells. ^{4 5 6 7} Recent studies have suggested that HGF has various other activities and plays an important role as an organotrophic factor responsible for regeneration of the liver, kidney, and lung. ^{3 8 9 10} Application of HGF or HGF gene therapy induces potent therapeutic effects on various types of injuries and diseases in experimental animals. ^{11 12}  

Expression of HGF is not confined to the liver; it is also expressed in the central nervous system, and functional coupling between HGF and the c-Met–HGF receptor is known to enhance the survival of hippocampal neurons in primary culture. HGF has also been shown to induce neurite outgrowth during neuronal development in vitro. ^{13 14 15 16} HGF is as potent a survival factor for motor neurons as other survival factors described to date, ¹⁷ such as brain-derived neurotrophic factor (BDNF) and ciliary neurotrophic factor (CNTF). ^{18 19}  

Recently, we showed that most of the neuronal cell death in retinal ischemia–reperfusion injury is due to apoptosis, as determined by analysis of the ultrastructure, by the TdT-dUTP terminal nick-end labeling (TUNEL) method, and by DNA ladder formation. ²⁰ Although not completely understood, glutamate toxicity ^{21 22} and production of free radicals, including superoxide and nitric oxide, ^{23 24} play an important role in the pathogenesis of retinal neuronal death. Accordingly, recent studies have shown that some materials inhibiting production of glutamate and reactive oxygen intermediates have neuroprotective effects. For example, MK-801, an N-methyl-d-aspartate (NMDA) receptor inhibitor ^{25 26} ; catalase or thioredoxin, free radical scavengers ^{27 28} ; and nitric oxide synthase inhibitors ^{29 30} have been shown to have neuroprotective effects against retinal ischemia–reperfusion injury. Also, neurotrophic and growth factors, such as BDNF, CNTF, basic fibroblast growth factor (bFGF), and nerve growth factor (NGF) have neuroprotective effects against retinal ischemia-reperfusion. ^{31 32 33 34}  

In this study, we investigated the expression of HGF and c-Met–HGF receptor using a rat model of retinal ischemia–reperfusion injury. Reverse transcription–polymerase chain reaction (RT-PCR), Western blot analysis, and immunohistochemical studies were used to study the expression of HGF and c-Met–HGF receptor. We also examined the possible neuroprotective effects of exogenous HGF against such injury, by using electroretinogram (ERG), by measuring the inner retinal thickness, and by counting the number of cells labeled by the TUNEL method in the ganglion cell layer (GCL), inner nuclear layer (INL), and outer nuclear layer (ONL). 

Materials and Methods

Animals

A total of 195 adult male Sprague-Dawley rats weighing 250 to 300 g were used in the study. All studies were conducted in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Institutional Animal Care and Use Committee of Shinshu University School of Medicine. 

Rats were anesthetized by intraperitoneal injections of pentobarbital (60 mg/kg) and the pupils dilated with topical phenylephrine hydrochloride and tropicamide. The anterior chamber of the left eye was cannulated with a 27-gauge infusion needle connected to a reservoir containing normal saline. The intraocular pressure (IOP) was raised to 110 mm Hg for 45 minutes by elevating the saline reservoir. ²⁸ Retinal ischemia was confirmed by the whitening of the iris and fundus. Sham-treated control right eyes underwent a similar procedure, but without the elevation of the saline bag, so that normal ocular tension was maintained. The 45-minute duration of ischemia was chosen on the basis of previous studies. ^{35 36}  

To investigate the neuroprotective effects of rhHGF, the rats were treated intravitreally with 1, 6, and 12 μg rhHGF in 2 μL phosphate-buffered saline (PBS) or 2 μL PBS without rhHGF 1 minute after reperfusion. In six sham-treated control right eyes, 2 μL PBS without rhHGF was injected intravitreally 1 minute after the sham operation. rhHGF was prepared as described previously. ^{3 14} The purity of rhHGF exceeds 98%, as determined by SDS-PAGE and after protein staining. 

Expression of HGF and c-Met–HGF Receptor by RT-PCR

To collect retinal tissues for RT-PCR, eyes were enucleated at 6, 12, 24, 48, and 96 hours after a 45-minute ischemic insult (n= 5 for each time point), and the retina was removed from the eye immediately. PolyA⁺ RNA was extracted from the experimental and normal rat retinas. A total of 0.1 μg RNA was used to make the cDNA with a first-strand cDNA synthesis kit (Pharmacia Biotech, Uppsala, Sweden), and PCR was performed. The following conditions were used: denaturation at 94°C for 45 seconds, annealing at 55°C for 45 seconds, and extension at 72°C for 90 seconds, for 30 cycles. The DNA thermal cycler and Taq DNA polymerase were obtained from Perkin Elmer (Foster City, CA). The primers used for HGF were 5′-CCATGAATTTGACCTCTATG-3′ (sense) and 5′-ACTGACGAATGTCACAGACT-3′ (antisense). The primers used for c-Met–HGF receptor were 5′-AGATGAACGTGAACATGAAG-3′ (sense) and 5′-CTGATGAGCTGGTCGTCATAG-3′ (antisense). Expression of β-actin was used as the internal standard. PCR products were electrophoresed on a 3% agarose gel and visualized with ethidium bromide. Semiquantitative analysis was performed by using the digital photograph (Gel Plotting Macros; NIH Image, ver. 1.62; provided in the public domain by the National Institutes of Health, Bethesda, MD, and available at http://rsb.info.nih.gov/nih-image/). PCR products were subcloned into pCR II plasmid vector (Invitrogen, San Diego, CA). Nucleotide sequencing of the cloned DNA was performed by the dideoxynucleotide chain termination method. ³⁷ PCR products were run on a gene analyzer (ABI Prism 310; Perkin Elmer) to examine the sequences of HGF and c-Met–HGF receptor. The sequence data of the PCR product were identical with HGF and c-Met–HGF receptor sequences found in GenBank (provided in the public domain by the National Center for Biotechnology Information, Bethesda, MD, and available at http://www.ncbi.nlm.nih.gov/genbank/). 

Western Blot Analysis

For Western blot analysis, eyes were enucleated at 6, 12, 24, 48, and 96 hours after reperfusion (n = 4 for each time point) and the sensory retina was removed immediately. Retinas were also obtained from four normal rats. These samples were homogenized in 200 μL of 4% sodium dodecyl sulfate (SDS) and were then subjected to SDS-PAGE with 8% polyacrylamide gel and electrophoretically transferred to a nitrocellulose membrane. ³⁸ The membrane was incubated with goat polyclonal anti-HGF antibody (Santa Cruz Biotechnology, Santa Cruz, CA) or rabbit polyclonal anti-c-Met antibody (Santa Cruz Biotechnology) at a dilution of 1:1000 with 1% BSA in Tris-buffered saline. Alkaline phosphatase-conjugated anti-goat IgG was used as the second antibody at a dilution of 1:5000. 

Immunohistochemical Study and PI Staining

At 6 and 24 hours after reperfusion, rats were killed with an overdose of sodium pentobarbital and the eyes immediately enucleated and fixed in 4% paraformaldehyde in phosphate buffer (n = 4 for each time point). Frozen sections were used for immunofluorescence staining of HGF and c-Met–HGF receptor. The sections were incubated with 2% normal goat or rabbit serum for 30 minutes at room temperature. After rinsing, the sections were incubated overnight at 4°C with either 1.5 μg/mL goat polyclonal anti-HGF antibody or 2μ g/mL rabbit polyclonal anti-c-Met antibody. The working concentrations of the antibodies were determined after testing different concentrations. Double staining of the retinal sections with anti-HGF antibody and propidium iodide (PI) or anti-c-Met–HGF receptor antibody and PI was performed as previously described. ^{39 40} Anti-HGF antibody and anti-c-Met–HGF receptor antibody were used as the primary antibodies, and FITC-conjugated secondary antibodies were used to obtain a green fluorescence. The nuclei were then counterstained with PI (20 μg/mL), which has been used previously as a marker of cell death. ^{40 41 42} Double staining of HGF and glial fibrillary acidic protein (GFAP) or c-Met–HGF receptor and GFAP was also performed to study cell types that expressed HGF and c-Met–HGF receptor. Mouse monoclonal anti-GFAP protein (Chemicon, Temecula, CA) was used at a concentration of 2 μg/mL and rhodamine-labeled anti-mouse IgG was used as the second antibody to depict red fluorescence. All specimens were examined with a scanning laser confocal microscope (model LSM410; Carl Zeiss, Oberkochen, Germany) in the fluorescence mode. 

Neuroprotective Effects of Recombinant Human HGF

Electroretinograms.

The rats were anesthetized by intramuscular injections of ketamine hydrochloride (70 mg/kg) and xylazine (10 mg/kg) and the pupils dilated with phenylephrine hydrochloride and tropicamide. Rats were dark adapted for at least 60 minutes before ERGs were recorded. The temperature of the rats was measured by a rectal sensor and maintained at 37°C with a heated blanket during anesthesia. ERGs were recorded 1 day before ischemia and on days 1, 2, 4, 7, 14, and 28 after reperfusion, as previously described. ²⁸ The amplitudes and the implicit times of the a-waves, b-waves, and oscillatory potentials (OPs) for three ERGs were measured and the results averaged. 

Morphometric Measurements.

On day 28 after reperfusion, the rats were killed with an overdose of sodium pentobarbital. The eyes were immediately enucleated and fixed in 2.5% glutaraldehyde in phosphate buffer for morphometric measurements. The specimens fixed in glutaraldehyde were osmicated, dehydrated, and embedded in epoxy resin. Sections of 1-μm thickness, cut along the vertical meridian of the eye and passed through the optic nerve head, were stained with toluidine blue. The ischemic changes were evaluated by measuring the inner retinal thickness (thickness between the inner limiting membrane and the boundary of the ONL and the OPL) at 10 points for each eye, according to previously described methods. ^{28 43} Data are presented as the mean of data from seven eyes with an average of 10 measurements made on each retina. 

DNA Nick-End Labeling by the TUNEL Method.

At 6, 12, 24, 48, and 96 hours after reperfusion, rats were killed with an overdose of sodium pentobarbital. The eyes were immediately enucleated and fixed in 4% paraformaldehyde in PBS for the TUNEL studies. The specimens were then dehydrated and embedded in paraffin and 4-μm sections were cut. These sections were stained by the TUNEL method using 3,3′-diaminobenzidine as the substrate. ⁴⁴ The number of TUNEL-positive cells was counted in 20 areas of approximately 6000 μm² in the GCL, INL, and ONL of each section. Data are the mean positive cells per square millimeter. 

The measurements of the inner retinal thickness and count of TUNEL-positive cells were digitized by a computer-controlled display on a computer screen using the scanning laser confocal microscope with an area measure function (LSM410; Zeiss). 

Statistical Analysis

The amplitude and implicit times of the a-wave, b-wave, and OPs were analyzed by repeated measures of analysis of variance (ANOVA) followed by Scheffé post hoc test. Data from the measurement of the inner retinal thickness were analyzed by one-way ANOVA followed by the Scheffé post hoc test and data on the number of TUNEL-positive cells were analyzed by two-way ANOVA followed by the Scheffé post hoc test. P < 0.05 was considered statistically significant. 

Results

Expression of HGF and c-Met–HGF Receptor by RT-PCR

Only very weak expression of HGF and c-Met–HGF receptor mRNA was detected in the normal rat retina. However, mRNA expression of HGF and c-Met–HGF receptor was upregulated in the experimental retinas at 6, 12, 24, and 48 hours after reperfusion. Particularly at 6 and 24 hours after reperfusion, the expression levels of HGF and c-Met mRNA were approximately three times higher than those of the normal control rat retina as determined by the NIH Image program. Thereafter, the mRNA expression decreased until 96 hours after reperfusion (Fig. 1) . 

Western Blot Analysis

In the control retinas, expression of HGF and c-Met–HGF receptor protein was hardly detectable. However, anti-HGF and anti-c-Met antibodies detected a 69-kDa band and a 145-kDa band, respectively, in the retina obtained from 6 to 96 hours after reperfusion (Fig. 2) . The data agree with those from the RT-PCR experiments. The apparent molecular weights of the bands (69 kDa for HGF and 145 kDa for c-Met) agree with those of the active form of HGF α-chain and c-Metβ -chain. 

Immunohistochemical Study and PI Staining

Specific immunostaining in the retina after ischemic insult was obtained by using anti-HGF and anti-c-Met antibodies (Fig. 3) . Very weak immunostaining with anti-HGF and anti-c-Met antibodies was observed in the retinal pigment epithelial cells of normal rat eyes (Figs. 3A 3D) and also in the sham-treated right eyes (not shown). In contrast, at 6 hours after reperfusion, specific immunostaining with anti-HGF was visible in the retinal pigment epithelial cells (Fig. 3B) . The anti-c-Met antibodies showed positive staining in the retinal pigment epithelial cells and cells in the retinal GCL at 6 hours after reperfusion (Fig. 3E , arrowheads). At 24 hours after reperfusion, specific staining with the anti-HGF and anti-c-Met antibodies was found in the GCL, the IPL, and the INL (Figs. 3C 3F , arrowheads). Also, at 24 hours the number of anti-c-Met–positive cells in the INL and the IPL was higher than that of anti-HGF–positive cells. From the morphology and location of the cells, both neuronal and glial cells seemed to be HGF- and c-Met–positive. To further characterize such cell types, double staining of HGF and GFAP or c-Met–HGF receptor and GFAP was performed. At 24 hours after reperfusion, the number of the cells in the retina double-stained with anti-HGF and anti-GFAP antibodies or with the anti-c-Met–HGF receptor and anti-GFAP antibodies (total cells in the GCL, IPL, and INL) were 256 ± 31 and 272 ± 36 cells/mm², respectively. However, the number of cells stained with the anti-HGF antibody only or the anti-c-Met–HGF receptor antibody were 116 ± 18 and 240 ± 18 cells/mm², respectively. The data show that glial cells constituted a major population of cells that expressed HGF, but a smaller population of neuronal cells were also HGF-positive. Both glial and neuronal cells seemed to express c-Met–HGF receptor. 

There were no PI-positive dying cells in the retina of normal rat eyes. At 6 hours after reperfusion, PI-positive dying cells were observed in the GCL (Figs. 3B 3E , arrows), and at 24 hours after reperfusion, PI-positive dying cells were mainly observed in the INL (Figs. 3C 3F , arrows). It is of interest that there was no colocalization of PI signal and immunostaining for anti-HGF and anti-c-Met antibodies. 

Neuroprotective Effects of rhHGF

Electroretinograms.

An ischemic insult of 45 minutes decreased the amplitudes of the ERG a- and b-waves and the OPs of the vehicle-treated rats throughout the follow-up period (Fig. 4) . In these rats, all the components of the ERG showed partial recovery on day 28, with the amplitude of the a-wave in both the vehicle-treated and 6-μg rhHGF-treated rats showing approximately 90% recovery on day 28. The difference between the two groups was not statistically significant (Fig. 4A) . The differences in the implicit times of the a-waves, b-waves, and OPs were not statistically significant between the vehicle-treated rats and rhHGF-treated rats (data not shown). 

A different pattern was found in the amplitudes of the ERG b-wave and OPs. Except on day 1, the amplitudes of the b-waves were significantly larger than those of the vehicle-treated rats on days 2, 4, 7, 14, and 28 (P < 0.01; Scheffé post hoc test; Fig. 4B ). Similarly, the amplitudes of the OPs were significantly larger than those of the vehicle-treated rats on days 2 (P < 0.05), 4, 7, 14, and 28 (P < 0.01; Scheffé post hoc test; Fig. 4C ). With 1 μg rhHGF, no statistically significant changes were observed in the amplitudes of the b-wave on day 28 after reperfusion; however, with 12 μg rhHGF, the amplitude of the b-waves and OPs were significantly larger on day 28 (P < 0.01; one-way ANOVA, Scheffé post hoc test; Table 1 ). A statistically significant change was not found in the sham-treated right eyes, demonstrating that the rhHGF, per se, did not change the amplitude of the a-wave, b-wave, or OPs (data not shown). 

Histologic Findings.

Typical histologic changes on day 28 after ischemia–reperfusion injury are shown in Figure 5 . The inner retinal layer of the 6-μg rhHGF-treated rats (Fig. 5B) showed relatively well-maintained structures compared with those of vehicle-treated control animals (Fig. 5C) . On day 28 after the ischemia-reperfusion, no proliferative changes, including neovascularization, were observed in the rhHGF-treated eyes. No histologic changes were detected in the sham-treated right eyes. 

On day 28 after reperfusion, the inner retinal thicknesses of the 1-, 6-, and 12-μg rhHGF-treated rats were 59.8 ± 7.06 (mean ± SD; n = 6), 71.5 ± 9.81 (n = 8), and 80.2 ± 8.33 μm (n = 5), respectively. In the control rats, the thickness was 54.4 ± 6.12 μm (n = 9). The variations in the 10 measurements on each retinal specimen were 4% to 6% of the mean value. The retinal thickness of the 6- and 12-μg rhHGF-treated rats was significantly thicker than that of the control rats (P < 0.01; one-way ANOVA, Scheffé post hoc test; Fig. 6 ). 

In contrast, the inner retinal thicknesses of normal rat eyes and sham-treated right eyes were 107.5 ± 8.36 (n = 6) and 110.2 ± 12.07 μm (n = 5), respectively. A minor change was observed in the outer retinal layer, but the thickness of the ONL in the rats of all experimental groups was not significantly different from that of normal rats (data not shown). 

The number of TUNEL-positive cells was lower in the 6-μg rhHGF–treated eyes than in the vehicle-treated eyes (Fig. 7) . There are no TUNEL-positive cells in the sham-treated right eyes. At 6 hours after reperfusion, TUNEL-positive cells were mainly observed in the GCL (Figs. 7A 7D) ; at 24 hours, such cells were mainly in the INL (Figs. 7B 7E) ; and at 48 hours, TUNEL-positive cells were found in the INL and the ONL (Figs. 7C 7F) . The number of TUNEL-positive cells in the GCL of both 6-μg rhHGF-treated eyes and vehicle-treated control eyes at 6 hours after reperfusion were 332 ± 40 (n = 6) and 459 ± 38 cells/mm² (n = 6), respectively. This difference was significant (P < 0.01; two-way ANOVA, Scheffé post hoc test; Fig. 8A ). The number of TUNEL-positive cells in the INL of both 6-μg rhHGF-treated eyes and vehicle-treated control eyes reached a peak at 24 hours after reperfusion and were 2150 ± 249 (n = 6) and 2706 ± 273 cells/mm² (n = 7), respectively. At 12, 24, and 48 hours after reperfusion, the number of TUNEL-positive cells in the INL of 6-μg rhHGF-treated eyes was significantly lower than those of the vehicle-treated control eyes (P < 0.01; two-way ANOVA, Scheffé post hoc test; Fig. 8B ). The number of TUNEL-positive cells in the ONL of both 6-μg rhHGF-treated eyes and vehicle-treated control eyes reached a peak at 48 hours after reperfusion: 872 ± 77 (n = 6) and 910 ± 65 cells/mm² (n = 7), respectively, but the difference was not significant (Fig. 8C) . 

Discussion

The expression of HGF in the central nervous system has been reported during the development of the cerebral cortex and in the hippocampus at the time of cerebral ischemia. ^{13 14 45 46} The present study has clearly shown the expression of HGF and c-Met–HGF receptor in the retina of a rat retinal ischemia–reperfusion injury model. From the results of RT-PCR, Western blot analysis, and immunohistochemical staining of the ischemic retina, it may be concluded that HGF and c-Met–HGF receptor are expressed mainly in the retinal pigment epithelial cells at 6 hours after reperfusion and mainly in cells of the GCL, IPL, and INL at 24 hours after reperfusion. The time course of expression may agree with previous reports that argue that the retinal pigment epithelial cells are affected by ischemia–reperfusion injury at the early phase of reperfusion. ⁴⁷ Of particular interest, at 24 hours after reperfusion when the number of dying cells in the retina reached its peak, most of the HGF- and c-Met–positive cells did not show condensed nuclei, as judged by PI staining. Thus, expression of HGF and c-Met–HGF receptor may be induced by neuronal cell death but from the present study, it is not clear whether the expression of HGF exerts a protective effect or inflicts injury on retinal neurons. However, because the larger population of the cells expressing HGF are glial cells including Müller cells, it is likely that HGF acts in an autocrine fashion, although paracrine action also plays a role. In the brain, both HGF and c-Met–HGF receptors have been reported to be expressed by various cell types, including astrocytes, microglia, and neurons. ^{13 14 15 45 46 48 49}  

To study the possible neuroprotective effects of HGF and particularly to examine the protective effects on neuronal apoptosis caused by ischemia–reperfusion injury, exogenous rhHGF was administered, and retinal function and morphology were studied. HGF is known to suppress the delayed cell death of hippocampal neurons after cerebral ischemia. ⁵⁰ Our results also demonstrated the neuroprotective effects of HGF on cells in the GCL and INL. Several mechanisms can be considered for HGF’s role in preventing neuronal apoptosis after ischemia–reperfusion injury. First, HGF can act directly on neuronal cells as a neurotrophic factor through c-Met–HGF receptor, because the expression of c-Met–HGF receptor was found on intraretinal cells. Second, HGF may act on the retinal glial cells to induce the expression of Bcl-2, which is well known to have a neuroprotective effect. ⁵¹ Third, HGF could increase the activity of the other neurotrophic factors, such as bFGF, which are expressed in the ischemic retina and thereby have neuroprotective and tissue-regenerating effects in cooperation with the other neurotrophic factors: neuronal growth factor (NGF), CNTF, and bFGF. ^{52 53}  

With the experimental model used in this study, rhHGF was injected only once into the vitreous body, immediately after reperfusion. Because neuronal cell death in the INL reaches a peak 24 hours after reperfusion, the neuroprotective effect of rhHGF may be increased by administering additional rhHGF or by changing the administration schedules. However, there is a possibility that the continuous administration of rhHGF over a long period would promote neovascularization and exacerbate the pathophysiology. With the experimental model used in this study, side effects, including neovascularization of rhHGF-treated retina 28 days after reperfusion, were not found histologically. Although only a few reports are available regarding HGF in the retina, it has been shown that the serum and vitreous HGF concentrations are significantly higher in patients with proliferative diabetic retinopathy than in normal subjects. ^{54 55} Similar to VEGF, HGF may play a role in the progression of diabetic retinopathy. 

Because apoptosis of neuronal cells represents the basic pathophysiology in many retinal diseases, such as retinitis pigmentosa and diabetic retinopathy, the use of rhHGF to suppress apoptosis may be a new treatment modality for many retinal diseases. However, HGF treatment for retinal disease accompanied by neovascularization and proliferation should be carefully planned, because of possible side effects of accelerating neovascularization and proliferation. Establishing a more effective method of administration may lead to the clinical application of HGF in eye diseases, as well. 

 Supported in part by a Grant-in-Aid for Scientific Research (13470364) from the Ministry of Education, Culture, Sports, Science and Technology of the Japanese Government.

 Submitted for publication January 29, 2001; revised July 12, 2001; accepted August 8, 2001.

 Commercial relationships policy: P (KM, TN); N (all others).

 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.

 

RT-PCR analysis of HGF (A) and c-Met–HGF receptor (B) expression in the retina after an ischemic insult. A single band was detected in the samples of normal rat retina and those examined 6 to 96 hours after reperfusion. Expression of HGF and c-Met–HGF receptor mRNA was upregulated in the retina at 6 to 48 hours after reperfusion, compared with that of normal control rat retina. Such mRNA expression was transient and decreased at 96 hours after reperfusion. The expression of HGF and c-Met–HGF receptor mRNA at 96 hours after reperfusion was similar to that in normal rat retinas. The sequence data of the PCR products were identical with those of HGF and c-Met–HGF receptor.

Figure 2.

 

Western blot analysis of HGF (A) and c-Met–HGF receptor (B). Immunoreactivity for HGF and c-Met–HGF receptor was not detected in the samples of normal rat retinas. However, the anti-HGF antibody detected a 69-kDa band and the anti-c-Met antibody a 145-kDa band after an ischemic insult. Expression of HGF and c-Met–HGF receptor at 6 and 24 hours after reperfusion was upregulated compared with that of normal control rat retina.

Figure 3.

 

Immunohistochemical studies of HGF (A, B, C) and c-Met–HGF receptor (D, E, F). Double staining with anti-HGF or anti-c-Met antibody and PI of normal rat retina (A, D) and of retinas at 6 (B, E) and 24 hours (C, F) after reperfusion: Green, HGF immunostaining; red, PI staining. Very weak immunostaining with anti-HGF and anti-c-Met antibodies and no PI-positive dying cells was observed in the normal rat retina (A, D). At 6 hours after reperfusion, however, HGF- (B) and c-Met–positive cells (E) were observed in the retinal pigment epithelium (RPE), and there are PI-positive dying cells and c-Met–positive cells (arrows and arrowbeads) in the GCL. In the GCL, the IPL, and in the INL, specific immunostaining with the anti-HGF (C) and anti-c-Met antibody (F) was seen at 24 hours after reperfusion (arrowheads). At 24 hours after reperfusion, PI-positive dying cells were mainly observed in the INL (arrows; C, F). Few cells were colabeled with anti-HGF or anti-c-Met antibody, and immunopositive cells did not show condensed PI staining. Bar, 50 μm.

Figure 4.

 

Changes in the amplitudes of the a-wave (A), b-wave (B), and OPs (C) after intravitreal injection of 6 μg rhHGF or vehicle. Rats were initially subjected to 45 minutes of retinal ischemia. There was no statistically significant difference in the amplitude of the a-wave between vehicle-treated control rats and rhHGF-treated animals (A). The amplitudes of both the b-wave and OPs from the rhHGF-treated rats showed statistically significant changes, compared with those in vehicle-treated control rats (B, C; P < 0.01; repeated-measures ANOVA). Statistically significant difference (*P < 0.05; **P < 0.01; Scheffé post hoc test) between rhHGF- and vehicle-treated eyes. Results are mean ± SD.

Table 1.

 

Effects of rhHGF on the ERG Amplitude on Day 28 after 45 Minutes of Ischemia

Treatment	n	a-Wave	b-Wave	OPs
Vehicle	10	166.8 ± 20.74	221.5 ± 21.26	181.6 ± 38.81
1 μg rhHGF	6	182.5 ± 27.62	254.8 ± 28.19	210.3 ± 30.52
6 μg rhHGF	8	179.5 ± 28.39	285.6 ± 41.39*	258.9 ± 32.95*
12 μg rhHGF	6	176.2 ± 30.87	331.4 ± 38.85*	274.0 ± 35.46*

 There was no statistically significant difference between vehicle control rats and all doses (1, 6, 12 μg) in rhHGF-treated rats in the amplitudes of the a-wave on day 28 after reperfusion. The amplitudes of the b-wave and OPs from 6- and 12-μg rhHGF-treated rats showed statistically significant changes compared with those in the vehicle-treated control rats. Results are mean microvolts ± SD.

*  Statistically significant difference (p < 0.01 by one-way ANOVA followed by the Scheffé post hoc test) between rhHGF- and vehicle-treated eyes.

Figure 5.

 

Light micrographs of the retina. (A) Sham-operated rat retina; (B) retina from a rat in the ischemic-insult and 6-μg rhHGF-treated group; and (C) retina from the ischemic-insult and vehicle-treated group. The inner retinal layers (IRL) of the 6-μg rhHGF-treated retinas were relatively well maintained, compared with those of the vehicle-treated control retinas. Bar, 50 μm.

Figure 6.

 

Measurement of the inner retinal thickness on day 28 after 45 minutes of ischemia, with and without rhHGF treatment. There was no statistically significant difference between vehicle-treated control rats and 1 μg rhHGF-treated rats in the thickness of the inner retinal layer 28 days after reperfusion. However, the inner retinal thickness of 6- and 12-μg rhHGF-treated rats was significantly preserved, compared with that of vehicle-treated control rats. Statistically significant difference (*P < 0.01; one-way ANOVA followed by Scheffé post hoc test) between rhHGF- and vehicle-treated eyes. Results are mean ± SD; n= 5–9.

Figure 7.

 

In situ labeling of retina by the TUNEL method. Rat retina treated with 6 μg rhHGF (A, B, C) or vehicle (D, E, F) at 6, 24, and 48 hours after 45 minutes of ischemic insult. At 6, 24, and 48 hours after reperfusion, TUNEL-positive cells were mainly observed in the GCL (A, D), INL (B, E), and INL and ONL (C, F). The number of TUNEL-positive cells was decreased by the treatment, compared with that of vehicle-treated control rats at 6 and 24 hours after reperfusion. At 48 hours after reperfusion, the TUNEL-positive cells in the INL of rhHGF-treated rat retina were fewer than those in vehicle-treated rat retina; however, there were no obvious differences in the number of the TUNEL-positive cells in the ONL between the rhHGF- and vehicle-treated rat retinas. Bar, 50 μm.

Figure 8.

 

Time course of TUNEL-positive cells after 45 minutes of ischemia–reperfusion injury. Number of TUNEL-positive cells in the GCL (A), INL (B), and ONL (C). There was a statistically significant difference between vehicle-treated control rats and 6-μg rhHGF-treated rats in the number of TUNEL-positive cells in the GCL (A) and INL (B); however, there was no statistically significant difference in the ONL (C). Statistically significant difference (*P < 0.01, two-way ANOVA followed by Scheffé post hoc test) between rhHGF- and vehicle-treated eyes. Results are mean ± SD; n = 6 or 7 for each time point.