Neuroprotective Effect of Hepatocyte Growth Factor against Photoreceptor Degeneration in Rats

Abstract

purpose. To determine whether hepatocyte growth factor (HGF) has a neuroprotective effect against photoreceptor degeneration in rats.

methods. Eight-week-old Sprague-Dawley (SD) and 24-day-old Royal College of Surgeons (RCS) rats received an intravitreal injection of HGF in the right eyes. The left eyes were injected with vehicle and served as the control. Two days after the injections, the SD rats were exposed to fluorescent light of 3000 lux for 72 hours. Scotopic and photopic electroretinograms (ERGs) were recorded 2 weeks after the light damage and at 70 days of age in RCS rats. After the ERG recordings, the animals were killed for histologic analysis. Some RCS rats were killed at 2 weeks after HGF-treatment for TdT-dUTP terminal nick-end labeling (TUNEL) studies.

results. In both light-damaged and RCS rats, the thresholds for the scotopic and photopic b-wave were significantly lower in the HGF-treated eyes than in the control eyes (P < 0.02). The maximum b-wave amplitudes (Vb_max) of the scotopic and photopic ERGs were significantly larger in the HGF-treated eyes (P < 0.0005) with a significantly greater number of photoreceptor nuclei than in the control eyes in both animal models (P < 0.005). The vehicle-injected eyes of RCS rats had significantly larger numbers of TUNEL-positive photoreceptor nuclei than the HGF-treated eyes (P = 0.005).

conclusions. Intravitreal HGF led to the morphologic and physiological preservation of photoreceptors in rats with photoreceptor degeneration induced by phototoxicity or a gene mutation. The antiapoptotic effect may be the mechanism for the neuroprotective action of HGF.

Human hepatocyte growth factor (HGF) was first purified from the plasma of patients with fulminant hepatic failure, ^{1 2 3} and has subsequently been cloned as a highly potent mitogen of mature hepatocytes. ^{4 5} Thereafter, it was shown that HGF affects many types of cells from various organs. ^{6 7} The biological effects of HGF are not confined to cell proliferation but also to the regulation of cell development, motility, morphogenesis, migration, adhesion, and survival. ^{6 7 8 9 10}  

HGF and its receptor c-met have been identified in developing and adult mammalian brains, ¹¹ indicating that it probably plays a functional role in the central nervous system (CNS). Other studies have demonstrated that HGF promotes the survival and maturation of cultured neural cells obtained from the CNS and spinal cord motor neurons. ^{12 13} All lines of evidence suggest that HGF regulates neuronal survival. 

In the eye, HGF has been reported to promote wound healing ¹⁴ and the migration and proliferation of retinal pigment epithelial (RPE) cells. ^{15 16 17 18 19} Recently, Shibuki et al. ²⁰ demonstrated that retinal ischemia increases the expression of HGF and its receptor c-met in retinal cells in the middle and inner layers of the retina. In addition, exogenous application of HGF promoted the survival of these retinal cells in retinal ischemia–reperfusion experiments in rats, suggesting that HGF plays a role in promoting the survival of retinal cells under ischemic stress. This protection results from the suppression of apoptosis, which is known to be the mechanism of photoreceptor death in animal models of retinitis pigmentosa. ^{21 22 23}  

We hypothesized that HGF would have a neuroprotective effect on photoreceptor degeneration induced by prolonged light exposure or by a gene mutation. To test this hypothesis, we injected HGF into the vitreous cavities of rats, which were then exposed to continuous light, and also into the vitreous cavities of Royal College of Surgeon (RCS) rats. 

Materials and Methods

Animals

Nineteen 5-week-old male Sprague-Dawley (SD) rats were purchased from CLEA Japan (Tokyo, Japan). They were housed in a standard animal room illuminated with 50 lux on a 12-hour light–dark cycle for 3 weeks before the beginning of the experiments. Two pairs of mating 8-week-old RCS rats were obtained from CLEA Japan. They were of an albinotic strain and homozygous for the retinal dystrophy gene (rdy/rdy). Thirty-one RCS rats were produced by the pairs and housed in our laboratory under cyclic light of 5 lux on a 12-hour cycle. 

All the procedures conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Preparation and Intravitreal Injection of HGF

Recombinant human HGF was prepared as described previously. ²⁴ In brief, plasmid DNA containing human HGF cDNA was transfected into Chinese hamster ovarian (CHO) cells. The clones, which were highly expressed HGF, were selected by detecting the HGF in the supernatant of the cultures using an ELISA specific for human HGF. ²⁵ HGF was purified from the supernatant of the CHO cell cultures by ion-exchange chromatography (S-Sepharose; Amersham Biosciences, Tokyo, Japan). The purity of recombinant human HGF was more than 95%. HGF was dissolved in phosphate-buffered saline (PBS; Sigma-Aldrich, St. Louis, MO) with 100 IU/mL heparin, to obtain final concentrations of 0.5, 2.5, or 5 μg/μL. 

HGF was injected intravitreally in the 8-week-old SD rats. In the RCS rats (n = 5), HGF was injected at 24 days of age when photoreceptor degeneration was just beginning. ²⁶ To determine whether HGF was effective even at an advanced stage of degeneration, other RCS rats were injected with HGF when they were 30 (n = 8), 37 (n = 5), and 42 (n = 4) days old. 

All injections were performed while viewing the eye with a microscope under fluorescent lighting. The intravitreal injection of 2 μL HGF was administered to the right eye of each animal with a syringe (80001; Hamilton Co., Reno, NV) with a 30-gauge needle inserted approximately 1 mm behind the corneal limbus. The left eyes received 2 μL heparin (100 IU/mL) dissolved in PBS as the control. Animals with hemorrhage or cataract were excluded. After the intravitreal injections, all animals were returned to the animal colony room under standard cyclical lighting conditions. 

Five SD rats were not treated and were kept as non-damaged control subjects. Nine RCS rats were not injected and used as controls at 24 (n = 4) and 42 (n = 5) days of age. 

Light Damage

Two days after the intravitreal injection of HGF, the SD rats were exposed to diffuse, cool, white fluorescent light of 3000 lux for 72 hours. The animals were placed in the dark for 12 hours before the onset of the bright-light exposure at 9:00 AM. Each rat was housed in a separate, well-ventilated, transparent plastic cage so that they could not hide behind one another. During the light exposure, the temperature was kept at 23.5°C ± 1.0°C. After the light exposure, the animals were returned to the colony room and kept for 2 weeks under standard cyclical lighting. 

Recording ERGs

Scotopic and photopic ERGs were recorded in both eyes simultaneously with a Ganzfeld bowl. Rats were dark adapted overnight and prepared under dim red light. They were anesthetized by a single intramuscular injection of a mixture of ketamine (87 mg/kg) and xylazine (13 mg/kg). The pupils were maximally dilated by topical 0.5% tropicamide and 0.5% phenylephrine HCl, and the cornea was anesthetized with topical 0.4% oxybuprocaine hydrochloride. 

Gold wire (0.1 mm diameter) loops were placed on the center of the cornea and the sclera, approximately 1 mm behind the corneal limbus, as the active and reference electrodes, respectively. An aluminum sheet placed under the animals served as the ground electrode. 

The ERGs were amplified 10⁵ times and band-pass filtered from 0.5 to 1000 Hz. The intensity of the stimuli was increased from the threshold of the scotopic threshold response in 0.28-, 0.29-, or 0.43-log unit steps. The duration of the stimulus was 10 μs, and the maximum luminance was 0.84 log cd-s/m². 

Three to 25 responses were computer averaged at each stimulus intensity for the scotopic ERGs, and 40 responses were computer averaged with an interval of 1 second for the photopic ERGs. The photopic ERGs were recorded with a white rod-suppressing background of 34 cd/m², and the rats were light adapted for at least 10 minutes with the same background light before the photopic recordings. 

The ERGs were recorded 2 weeks after light-induced damage in SD rats and at 70 days of age in RCS rats. 

Histology

The day after the ERG recordings, the rats were killed with an overdose of sodium pentobarbital. Both eyes were removed and fixed in 2.5% glutaraldehyde for 2 hours and then in 5% neutral-buffered formalin overnight. Paraffin-embedded tissues were sectioned at 3 μm along the vertical meridian of the eye through the optic nerve head and stained with hematoxylin and eosin. 

The degree of retinal damage was assessed by counting the surviving photoreceptor nuclei in 100-μm lengths of the retina. For the photoreceptor nuclei, the counts were begun at approximately 200 μm from the optic nerve head and were made every 400 μm for both the superior and inferior hemispheres (a total of 20 measuring points). The width of the outer nuclear layer (ONL) and the combined thickness of the debris and rod outer segment (ROS) were measured at the 20 points. The number of photoreceptor nuclei, the thickness of the ONL, and the combined thickness of the debris zone and the ROS were averaged for analysis. 

TdT-dUTP Terminal Nick-End Labeling

Five 24-day-old RCS rats were used for the study of apoptosis. The right eyes were injected with 10 μg HGF, and the left eyes received the vehicle. Two weeks later (at 38 days of age), the rats were killed with an overdose of pentobarbital sodium, and then both eyes were immediately removed and fixed in 5% neutral-buffered formalin overnight. Paraffin-embedded tissue was sectioned at 3 μm along the vertical meridian of the eye. These sections were stained (Apop Tag; Serologicals Corp., Norcross, GA) for TUNEL assay as well as with hematoxylin. TUNEL-positive nuclei in the ONL appeared brownish in color (see Fig. 7 ). The number of TUNEL-positive nuclei was counted throughout the section three times, and the average of the numbers obtained by the three counts was used for analysis. 

Data Analyses

The amplitude of the b-wave was measured from the negative trough of the a-wave to the positive peak. When the a-wave was absent, the b-wave amplitude was measured from the baseline to the positive peak. The amplitudes of the scotopic and photopic b-waves were plotted against the stimulus intensity, and the resultant intensity–response curves were used to determine the threshold and sensitivity of the b-wave. The criterion amplitude for the threshold was set at 10 and 5 μV for the scotopic and photopic b-waves, respectively. The maximum b-wave amplitude elicited by the maximum stimulus intensity was designated as Vb_max for the scotopic and photopic ERGs. 

All intensity–response curves of the scotopic b-wave were fitted to the following equation based on the Fulton and Rushton model. ²⁷ All the fittings were performed on computer (Origin, ver. 6.1 analysis software; OriginLab Corp., Northampton, MA). The sensitivities of the rod b-waves were also determined by the following equation:  

\[V\ {=}\ V_{\mathrm{max}}\frac{I^{n}}{K^{n}\ {+}\ I^{n}}\]

where V is amplitude (in microvolts); V _max is maximum amplitude (in microvolts), defined as the amplitude at the initial plateau of the intensity–response curves and allowed to vary; I is stimulus intensity (cd-s/m²); n is the exponent, which was kept at 1; and K is the stimulus intensity necessary to obtain one-half V _max. Then, 1/K was defined as the sensitivity. A nonlinear least-squares algorithm was used for the best fit (see dashed lines in Figs. 2A 5A ). 

Statistically significant differences were determined by Student’s two-tailed t-test for paired data, with significance set at P < 0.05. 

Results

Optimum Concentration of HGF In Vivo

To determine the optimum concentration of HGF for protecting against light damage, we injected 1 (n = 5), 5 (n = 4), or 10 (n = 5) μg HGF into the vitreous cavity. The effect of the treatment was expressed as the difference in log Vb_max (log microvolts) between the treated and vehicle-injected eyes of the light-damaged rats. The difference was 0.06 ± 0.11, 1.07 ± 0.98, and 1.39 ± 0.28 log units (mean ± SD) for 1, 5, and 10 μg HGF, respectively. Whereas 1 μg HGF did not result in significant protection, 5 and 10 μg HGF provided functional preservation. Although the difference between 5 and 10 μg was not statistically significant, the protective effect of 10 μg HGF was more consistent with less variation than that of 5 μg. Therefore, 10 μg HGF was used in the experiments. 

Changes in ERGs in Light-Damaged Rats

In the normal eye, the scotopic b-wave emerged and reached the criterion voltage at −5.59 log cd-s/m² (Fig. 1) . The criterion b-wave amplitude was elicited by −4.59 log cd-sec/m² in the light-damaged rat injected with 10 μg HGF and by −2.16 log cd-s/m² in the vehicle-injected eye. Thus, the threshold was elevated by 1.0 log unit for the HGF-injected eye and by 3.43 log units for the vehicle-injected eye, compared with the normal control. At the maximum intensity, the scotopic Vb_max of the HGF-injected eye was smaller than that of the control eye, but it was much larger than that of the vehicle-injected eye with barely detectable responses. 

For the photopic ERGs, the b-wave was detectable at −0.88 for the 10 μg HGF-treated eye, and the stimulus intensity had to be increased by 1.29 log cd/m² to elicit the criterion amplitude in the vehicle-injected eye (Fig. 1) . The maximum amplitude of the photopic b-wave was reduced to approximately two thirds of normal in the HGF-injected eye, and it was more severely depressed in the vehicle-injected eye. 

The means of scotopic b-wave amplitudes were plotted against stimulus intensity to obtain the intensity–response function for the non-damaged normal eyes (n = 5), the 10 μg HGF-injected eyes (n = 5), and the vehicle-injected eyes (n = 5; Fig. 2A ). From these curves, the threshold and the sensitivity of the scotopic b-waves were obtained (Table 1) . The threshold, sensitivity, and Vb_max of the scotopic b-wave were significantly better in the 10 μg HGF-treated eyes than in the vehicle-injected eyes (Table 1 , P = 0.007 for the threshold; P = 0.01 for the sensitivity; and P = 0.0004 for the scotopic Vb_max). 

The photopic b-wave amplitudes were plotted as a function of stimulus intensity to determine the threshold of the photopic b-wave (Fig. 2B) . Although the maximum photopic b-wave amplitude was decreased and the threshold was elevated in the 10 μg HGF-treated eyes, these eyes had significantly better photopic function compared with the vehicle-injected eyes (Table 1 , photopic Vb_max: P = 0.0004; photopic b-wave threshold: P = 0.005). 

Morphologic Changes in Light-Damaged Rats

Although the ONL was thinner in the 10 μg HGF-treated eye than in the non-damaged normal eye, it was thicker with more photoreceptor nuclei and longer rod inner segments (RIS) and ROS in the 10 μg HGF-treated eye than in the vehicle-injected eye (Fig. 3A) . The number of photoreceptor nuclei in the 10 μg HGF-treated eyes was significantly greater than in the vehicle-injected eyes (Fig. 3B , P = 0.0001). In addition, the ONL was significantly thicker in the HGF-treated eyes than in the vehicle-injected eyes (Fig. 3C , P < 0.00001). 

Changes in ERGs of RCS Rats

In the scotopic ERGs, the b-wave reached criterion amplitude at an intensity of −3.88 log cd-s/m² in a 24-day-old RCS rat (baseline; Fig. 4 ). In a 70-day-old RCS rat, the b-wave appeared at −3.59 log cd-s/m² in the 10 μg HGF-treated eye and −2.16 log cd-s/m² in the vehicle-injected eye. Thus, the HGF-treated eye had a 1.43 log cd-s/m² lower threshold than did the vehicle-injected eye. 

The maximum b-wave amplitude of the HGF-treated eye was nearly the same as that of the 24-day-old rat, although the a-wave amplitude was reduced in the HGF-treated eye compared with that of the 24-day-old rat (Fig. 4) . The maximum b-wave amplitude was six times higher in the HGF-treated eye than in the vehicle-injected eye. 

For the photopic ERGs, the b-wave appeared at −0.88, −0.59, and 0.13 log cd-s/m² in the 24-day-old rat and in the HGF-treated and vehicle-injected eyes of a 70-day-old RCS rat, respectively (Fig. 4) . Thus, there was a 0.72 log unit difference in the threshold between the HGF-treated and vehicle-injected eyes. The maximum amplitude of the photopic b-wave was strongly depressed in the vehicle-injected eye. 

The mean amplitudes of the scotopic b-wave are plotted against stimulus intensity for the 24-day-old RCS rats (n = 4) and 70-day-old RCS rats (n = 5) in 10 μg HGF-treated and vehicle-injected eyes (Fig. 5A) . Although the curve of the HGF-treated eyes was shifted to the right by approximately 0.3 log units, the Vb_max was nearly the same as that in the 24-day-old rats, suggesting that the sensitivity of the b-wave was reduced in the HGF-treated eyes. The curve of the vehicle-injected eyes was greatly depressed at all intensities in contrast to that of the HGF-treated eyes. The intensity–response curve of the photopic b-wave was shifted farther to the right and to lower amplitudes in the vehicle-injected eyes than in the HGF-treated eyes (Fig. 5B) . 

As shown in Figure 5 and Table 1 , a significantly greater functional preservation was detected in the HGF-treated eyes than in the vehicle-injected eyes in the ERG parameters representing the function of rods and cones (Table 1 , P = 0.02 for scotopic b-wave threshold and P = 0.0001 for scotopic Vb_max; P = 0.0006 for photopic b-wave threshold; P = 0.0004 for the photopic Vb_max). All the functional results in the HGF-treated eyes of 70-day-old RCS rats were equivalent to those in 42-day-old RCS rats (Table 1) , indicating that HGF-treatment at 24 days of age delayed photoreceptor degeneration by approximately 28 days. 

Morphologic Results in RCS Rats

Even in a 10 μg HGF-treated eye, the ONL was reduced to two or three rows of photoreceptor nuclei, which is approximately one half that in the 24-day-old RCS rat, but the vehicle-injected eye had only a single row of nuclei in the ONL (Fig. 6A) . 

The number of surviving photoreceptor nuclei was reduced despite HGF-treatment compared with the baseline. However, a significantly larger number of photoreceptor nuclei remained in the HGF-treated eyes than in the vehicle-injected eyes (Fig. 6B , P = 0.005). The ONL was significantly thicker in the treated eyes than in the vehicle-injected eyes (Fig. 6C , P < 0.0005). There was no significant difference in the combined thickness of the debris and ROS between the HGF-treated and vehicle-injected eyes (Fig. 6D) . 

Efficacy of HGF at Different Ages of RCS Rats

RCS rats were treated with 10 μg HGF at 24 (n = 5), 30 (n = 8), 37 (n = 5), and 42 (n = 4) days of age to determine the optimum age for the preservation of the photoreceptors in RCS rats. The effect of the treatment was expressed as the difference of Vb_max between the treated and vehicle-injected eyes (Figs. 7A 7B) . The treatment efficacy sharply declined between 30 and 37 days of age, indicating that treatment is needed before this age to provide effective rescue. 

TUNEL Staining

In the vehicle-injected eye, numerous TUNEL-positive nuclei were present in the ONL, whereas in the HGF-treated eye, TUNEL-positive nuclei were sparse (Fig. 8A) . For quantitative comparison, TUNEL-positive nuclei in the ONL of the whole retinal section were counted, and the number of TUNEL-positive nuclei was significantly fewer in the HGF-treated eyes than in the vehicle-injected eyes (Fig. 8B , P = 0.005). 

Complications

The development of cataracts and retinal angiogenesis has been reported as complications in eyes treated with other growth factors. ^{28 29 30} In our preparations, significant cataracts were not detected by indirect ophthalmoscopy, and retinal neovascularization did not develop as assessed by histologic examination. 

Discussion

The results of the ERG studies showed that there was significant functional preservation in the HGF-treated eyes of the light-damaged and RCS rats. In addition, the number of photoreceptor nuclei in the HGF-treated eyes significantly outnumbered those in the control eyes in both animal models. These findings indicate that a single intravitreal injection of HGF provides functional and morphologic protection of the photoreceptors in rats with photoreceptor degeneration induced by light toxicity or by a genetic mutation. 

Mechanism of Protection

Shibuki et al. ²⁰ demonstrated that neuronal cells in the inner and middle layers of the retina, glial cells, and RPE cells were stained by anti-HGF and anti-c-met antibodies after retinal ischemia in rats. It has also been shown that the mRNA of HGF was highly expressed in the retina of RCS rats (Shuler RK, et al. IOVS 2004;45:ARVO E-Abstract 705). All evidence suggests that HGF and its receptor are expressed in the stressed retina. 

The number of TUNEL-positive cells in the ONL was significantly fewer in the HGF-treated eyes than in vehicle-injected eyes of RCS rats. This indicates that apoptosis of the photoreceptors was inhibited by the intravitreal administration of HGF. Thus, antiapoptotic factors induced by HGF may be involved in this protection. It has been reported that HGF significantly increases the expression of bcl-2, an inhibitor of apoptosis, ³¹ in endothelial cells during hypoxia or with high concentrations of d-glucose. ^{32 33} If this is the case in retinal cells, HGF could provide protection against photoreceptor death through upregulation of bcl-2 because genetically upregulated bcl-2 in photoreceptors increases photoreceptor survival in light-damaged and rd mice. ³⁴  

In contrast, it was recently reported that extracellular signal-regulated kinases (ERKs) were involved in the biological response by HGF. ^{35 36} Neurotrophic factors have been shown to induce ERK phosphorylation in glial cells in the retina. ^{37 38} Further studies are needed to investigate whether HGF functions through ERK in the retina as do other neurotrophic factors. 

Functional Measurement

In RCS rats with advanced retinal degeneration, the third-order neuronal responses including the scotopic threshold response ³⁹ and photopic negative response ⁴⁰ contribute significantly to shaping the negative deflection of the ERGs over wide range of stimuli. ⁴¹ Consequently, the third-order neuronal responses contaminate the a-wave (Machida S, et al. IOVS 2002;43:ARVO E-Abstract 3489), which makes it difficult to distinguish the photoreceptor response from the third-order neuronal response. As shown in Figure 4 , there was no substantial change in the a-wave amplitude between HGF-treated and vehicle-injected eyes of RCS rats, even though more photoreceptor nuclei survived in the HGF-treated eyes than in vehicle-injected eyes. This indicates that the a-wave may not be a good component to use in evaluating the treatment efficacy in this model of advanced photoreceptor degeneration. Therefore, we assessed the b-wave as the functional parameter throughout this study. 

Cone Preservation

Because cones are essential for visual performance in the daily life of humans, it is important to assess the protective effect of HGF against cone degeneration. The cone function was significantly preserved in both animal models. However, this evidence does not imply that HGF directly protects cones from degeneration. This protection could be a secondary effect of the rod preservation, because rods are essential for survival of cones. ⁴² Evidence on whether HGF protects cones directly could be provided by comparing cone preservation in HGF-treated and untreated eyes without surviving rods, such as in aged RCS rat eyes in which cones function exclusively (Sauve Y, et al. IOVS 2004;45:ARVO E-Abstract 5076). 

Complications

It has been shown that HGF has many biological effects, including proliferation and angiogenesis, that are unfavorable responses in the eye. One may then be concerned with retinal and choroidal neovascularization and retinal detachment. ^{15 16 17 18 19 43 44} However, in light microscopic examination, we did not find these complications in rats with a single vitreous injection of HGF in the short-term. Additional studies are warranted, using multiple injections or sustained upregulation of HGF by transfection of HGF gene into intraocular tissues. 

Comparison with LEDGF

It has been reported that lens epithelium-derived growth factor (LEDGF) protects photoreceptors against degeneration in light-damaged and RCS rats. ⁴⁵ However, there were differences in the results obtained by HGF and LEDGF. First, LEDGF-treated RCS rats showed the significant thinning of the debris zone that has been reported in other treatments. ⁴⁶ Because engorged macrophages have been observed frequently in LEDGF-treated eyes, the phagocytosis by macrophages was probably activated by the administration of LEDGF. In contrast, there was no significant difference in thickness of debris zone between HGF-treated and vehicle-injected eyes of RCS rats in the present study. 

Second, in this study we treated RCS rats earlier than in the previous report in which LEDGF provided significant protection even in the late treatment at 42 days of age in RCS rats. HGF-treatment later than 35 days of age did not show significant protection, which indicates that the treatment window is narrower for HGF than that for LEDFF in RCS rats. All these findings suggest that HGF and LEDGF have different biological effects or different mechanisms that they work through in the cells in the retina. 

Clinical Relevance

Mutation of the gene coding for the receptor tyrosine kinase gene Mertk has been identified in RCS rats, and the knockout of this gene successfully produced the RCS phenotype in mice. ^{47 48} The same gene mutation has been identified in patients with autosomal recessive retinitis pigmentosa, ⁴⁹ indicating that the RCS rat is a counterpart of human retinitis pigmentosa. Therefore, the current protective effect against photoreceptor degeneration in RCS rats suggests that HGF could be considered as a treatment for patients with some forms of retinitis pigmentosa. However, most survival-promoting factors do not work in the other genetic models. ⁵⁰ Therefore, a study should be conducted in animals with different genetic mutations before consideration of HGF for clinical use. 

It is known that HGF is involved in unfavorable biological responses in the eye, such as neovascularization in diabetic retinopathy and RPE proliferation in retinal detachment. ^{15 16 17 18 19 43 44} Thus, a strategy to reduce activity of HGF using antisense or antagonist may be used in these retinal diseases. However, these treatments would simultaneously threaten retinal cells by withdrawing the survival promoting effect of HGF. 

 Supported by grants from Japanese Retinitis Pigmentosa Society (JRPS) and Mitsubishi Pharma Corporation (SM).

 Submitted for publication April 20, 2004; revised June 25, 2004; accepted July 17, 2004.

 Disclosure: S. Machida, Mitsubishi Pharma Corp. (F); M. Tanaka, Mitsubishi Pharma Corp. (F); T. Ishii, Mitsubishi Pharma Corp. (E, F); K. Ohtaka, Mitsubishi Pharma Corp. (F); T. Takahashi, Mitsubishi Pharma Corp. (F); Y. Tazawa, Mitsubishi Pharma Corp. (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.

 

Figure 7.

 

RCS rats treated at 24 (n = 5), 30 (n = 8), 37 (n = 5), and 42 (n = 4) days of age, to determine the optimum time of treatment. The effect of the treatment was expressed as difference in the log Vb_max between the HGF-treated and vehicle-injected eyes (A, B). All ERG recordings were made at 70 days of age. Error bars, SEM.

Figure 2.

 

Mean intensity–response functions for the scotopic (A) and photopic (B) b-waves of a normal eye and a light-damaged eye treated with 10 μg HGF or vehicle. The ERG amplitudes are plotted against the stimulus intensity on a log-log scale. Criterion amplitudes were 10 μV for the scotopic b-waves and 5 μV for the photopic b-wave thresholds. Dashed lines: best fit. Error bars, SEM.

Figure 5.

 

Mean intensity–response functions for the scotopic b-wave (A) and photopic b-wave (B) of 24-day-old (baseline) and 70-day-old RCS rats. The ERG amplitudes are plotted against the stimulus intensity on a log–log scale. Criterion amplitudes were 10 μV for the scotopic b-wave and 5 μV for the photopic b-wave thresholds. The 70-day-old RCS rats were treated at 24 days of age. Dashed lines: best fitting lines. Error bars, SEM.

Figure 1.

 

Scotopic and photopic ERGs elicited by different stimulus intensities from a normal and a light-damaged rat. The right eye of the light-damaged rat was treated with 10 μg HGF while the left eye received vehicle as a control. The recordings were made 2 weeks after the light damage.

Table 1.

 

Functional Comparison between 10-μg HGF-Treated and Vehicle-Injected Eyes

	Scotopic ERG					Photopic ERG
		b-Wave Threshold* (log cd-s/m2)	Vbmax , † (log μV)	b-Wave Sensitivity(1/k) (log cd−1-s−1/m2)	b-Wave Threshold* (log cd-s/m2)	Vbmax , † (log μV)
SD rats
Non-damaged control (n = 5)	−5.64 ± 0.10	3.09 ± 0.06	3.77 ± 0.03	−0.98 ± 0.10	2.34 ± 0.07
Light damage (n = 5)
10 μg HGF-treated	−4.66 ± 0.12	2.71 ± 0.10	3.31 ± 0.19	−0.82 ± 0.07	1.91 ± 0.09
Vehicle-injected	−0.91 ± 1.53	1.32 ± 0.32	0.76 ± 1.21	0.49 ± 0.51	0.83 ± 0.20
Difference, ‡	3.75 ± 1.62	1.39 ± 0.28	2.25 ± 1.36	1.31 ± 0.52	1.08 ± 0.21
P	0.007	0.0004	0.01	0.005	0.0004
RCS rats
24 Days of age (baseline, n = 4)	−3.89 ± 0.19	2.49 ± 0.07	2.90 ± 0.11	−0.85 ± 0.03	2.18 ± 0.06
42 Days of age (n = 5)	−3.37 ± 0.19	2.27 ± 0.07	2.36 ± 0.29	−0.69 ± 0.09	1.92 ± 0.04
70 Days of age (n = 5)
10 μg HGF-treated	−3.54 ± 0.16	2.38 ± 0.04	2.45 ± 0.32	−0.56 ± 0.09	1.78 ± 0.08
Vehicle-injected	−2.03 ± 0.79	1.56 ± 0.12	1.81 ± 0.95	0.26 ± 0.20	1.01 ± 0.13
Difference, ‡	1.50 ± 0.83	0.82 ± 0.12	0.64 ± 1.15	0.82 ± 0.19	0.77 ± 0.16
P	0.02	0.0001	0.3	0.0006	0.0004

*  Criterion amplitude: 5 and 10 μV for the photopic and scotopic ERG.

†  Stimulus intensity: 0.84 log cd-s/m².

‡  difference between HGF-treated and vehicle-injected eyes.

Figure 3.

 

Light micrographs of the retina obtained from a non-damaged normal rat and a light-damaged rat injected with 10 μg HGF in the right eye and vehicle in the left eye (A). Averaged number of surviving photoreceptor nuclei per 100-μm length of the retina (B) and thickness of the ONL (C) in non-damaged control rats and light-damaged rats injected with 10 μg HGF in the right and vehicle in the left eyes. All three groups: n = 5. Error bars, SEM.

Figure 4.

 

Scotopic and photopic ERGs elicited by different stimulus intensities from a 24-day-old (baseline) and a 70-day-old RCS rat. The right eye of the 70-day-old RCS rat was treated with 10 μg HGF, and the left eye received vehicle as a control at 24 days of age.

Figure 6.

 

Light micrographs of the retinas of a 24-day-old RCS rat (baseline) and of the 10 μg HGF-treated and vehicle-injected eyes of a 70-day-old RCS rat (A). Averaged values of surviving photoreceptor nuclei per 100-μm length of the retina (B), thickness of the ONL (C), and combined thickness of debris and ROS (D) in 24-day-old baseline eyes (n = 4) and 10 μg HGF-treated (n = 5) and vehicle-injected (n = 5) eyes of 70-day-old RCS rats. Error bars, SEM.

Figure 8.

 

Retinal photographs of TUNEL staining obtained from 10 μg HGF- and vehicle-injected eyes of a 38-day-old RCS rat that was treated at 24 days of age (A). TUNEL-positive nuclei in the ONL of the whole retinal section were counted in the 10 μg HGF-treated and vehicle-injected eyes (B). Error bars, SEM.

The authors thank Paul A. Sieving (National Eye Institute, Bethesda, MD) for providing the Ganzfeld bowl system.