WO2000024359A2

WO2000024359A2 - Use of gndf for treating retinal degeneration

Info

Publication number: WO2000024359A2
Application number: PCT/FR1999/002586
Authority: WO
Inventors: José SAHEL; Maria Frasson; Serge Picaud; David Hicks; Saddek Mohand Said; Thierry Leveillard
Original assignee: Universite Louis Pasteur
Priority date: 1998-10-26
Filing date: 1999-10-25
Publication date: 2000-05-04
Also published as: FR2784898A1; WO2000024359A3; AU6345699A

Abstract

The invention concerns the use of the neurotrophic factor derived from glyal cells (GNDF), or a substance of the family comprising said factor, as active protective agent for treating retinal degeneration, more particularly pigmentary retinopathy (RP) in human beings.

Description

Use of GDNF for the treatment of retinal degeneration

The present invention relates to the field of treatments for certain retinal diseases, more particularly agents capable of acting in the context of these treatments, and relates to the use of the neurotrophic factor derived from glial cells for this purpose. Pigmentary retinitis (RP) corresponds to a group of hereditary retinal diseases inducing the progressive loss of photoreceptors and an extreme visual handicap.

To date, there is no known treatment for these diseases, which can be caused by numerous mutations in a variety of photoreceptor specific genes (rhodopsin, cGMP-phosphodiesterase).

Several possible treatment solutions are currently being studied to limit or prevent the loss of photoreceptors associated with this disease, namely, gene therapy (Bennett et al., Nature Médicine 1996, 2: 649-654), pharmacological treatments (ie application of growth factors: Steinberg, Curr. Opinion in Neurobiology 1994; 4: 515-524) and retinal transplantation (Del Cerro, J. Neural Transplantation 1989; 1: 1-10 / Mohand-Saïd et al, Opthal. Res. 1997; 29, 290).

Among these approaches, the use of growth factors has been considered since the aforementioned publication by Steinberg et al., Who have shown that intraocular injections of basic fibroblastic growth factor (FGF-2) retard the degeneration of photoreceptors in a model. animal of RP, the rat RCS (Royal College of Surgeons) (Faktorovich et al., Nature 1990; 347, 83). The same authors have published data showing that several growth factors, including FGF-2 and ciliary neurotrophic factor (CNTF) (Lavail et al., Invest.Ophtalmol.Vis.Sci., 1998; 39: 592-602 ), exerted beneficial effects on the survival of photoreceptors in rats in the degeneration of photoreceptors (Faktorovich et al., J. Neuroscience 1992; 12, 3554 / Steinberg, Curr. Opinion Neurobiology 1994; 4, 515).

Regarding GDNF, its messenger RNA is produced in the retina. The retinal cells express these receptors and it has a stimulating activity on the survival of in vitro photoreceptors isolated from newborn rats (Jing et al., Cell 1996; 85, 1113). However, these results do not allow any valid deduction on the effects that GDNF could have on the degeneration of photoreceptors in vivo on pathological models which are not mentioned in this article, and in particular in the mouse, a species carrying mutations similar to those encountered in human pathology.

This is all the more true since it has been found that the CNTF has opposite effects on chicken photoreceptors and on rat photoreceptors (Kirsch et al ;, Neuroreport 1996; 7, 697) and that FGF-2 protected photoreceptors in hereditary degeneration in the retina of rats but not in the retina of mice (Faktorovich et al, Nature 1996; 381, 789).

In addition, the abovementioned results (Jing et al., 1996) were observed in vitro in a monolayer culture of immature cells, and do not allow any deduction in terms of effects under the conditions resulting from the complex environment of the whole retina. mature in animals. Finally, all of these publications refer only to rod type photoreceptors and none mentions any effect at the level of cone type photoreceptors which are however essential for color vision and visual acuity, nor any influence on visual function. . Furthermore, the results obtained in vitro with GDNF do not seem to have been judged sufficiently conclusive for this neurotrophic factor to be taken into account in recent studies and systematic tests relating to the effects of a large number of photoreceptor survival factors. in retinal degeneration (La Vail et al., Invertigative Ophtalmology & Visual Science, March 1998, vol 39, n ° 3, 592-602).

However, the authors of the present invention have observed and been able to observe that the neurotrophic factor derived from glial cells (GDNF) exhibits, in vivo, a protective effect against the degeneration of photoreceptors on a mutant animal, model of human pigmentary retinopathy, this protective effect applying as much to rods as to cones and indicating a preservation of the visual function, at least partial.

Thus, the main object of the present invention consists in the use of the neurotrophic factor derived from glial cells (GDNF) or a substance of the GDNF family, as an active protective agent both at the structural level and at the level functional, for the treatment of retinal degenerations, more particularly retinitis pigmentosa (RP) in humans. Obtaining an active substance which protects against the effects of retinal degeneration is not limited to GDNF alone, but can be extended, according to the research carried out by the inventors, at least to some of the substances belonging to the GDNF family. , such as, for example, neurturin or TGFB.

According to a first embodiment of the protective active agent according to the invention, the neurotrophic factor derived from glial cells is administered by subretinal intraocular injection and consists of recombinant exogenous GDNF. In accordance with a second embodiment of the invention, the

GDNF can be released in situ for example by fibroblasts implanted by grafting, after transfection.

In addition, it should be noted that GDNF is a particularly interesting neuroprotective molecule with regard to its tolerance and side effects because, unlike FGF-2, it is not angiogenic and does not normally cause neovascular complications.

The in vivo effects of GDNF mentioned above were verified using the tests described below, in relation to Figures 1 to 6 of the accompanying drawings. These figures show more particularly:

Figure 1 (Figures 1A and 1B): Evolution of ERG (electroretinograms) of wild type and rd / rd mice in response to mesopic stimulation.

Figure 2: Persistence of ERG at a detectable level in rd / rd mice aged 21 and 24 days having received an injection of GDNF and their disappearance in identical mice having received an injection of PBS. (Figure

2A) Histograms showing the amplitudes of the ER and A waves observed on separate animals. (Figure 2B) Recording ERG on eyes given an injection of GDNF. Figure 3: Low magnification view of whole rd / rd retinas laid out flat having received an injection of PBS (Figure 3A) and GDNF (Figure 3B), immunolabeled with rho-4D2. The rod-type photoreceptors surviving following treatment with GDNF are much more numerous. This example is representative of the mouse which presents the largest numerical difference between the eye treated with GDNF and the contralateral eye having received an injection of PBS. Scale bar = 50 μm. Figure 4: Illustration of the systematic random sampling approach used (Figure 4A) and high magnification view of the immunolabelled rd / rd retina showing the appearance of the rods and the superimposed counting frame (Figure 4B). Scale bar (Figure 4B) = 10 μm. Figure 5 (Figures 5A and 5B): Counts of immunostained rods in rd / rd mice treated with GDNF, PBS and untreated. (Figure 5 A) Total number of surviving rods for each mouse taken individually having received an injection of GDNF in one eye (white bars) and of the PBS vehicle in the other (black bars). The reproducible beneficial effect of growth factor therapy compared to vehicle injected eyes can be observed. The mice which enabled ERG to be successfully registered were numbered 3, 4, 7 and 9. The total numbers of surviving rods for non-operated rd / rd retinas taken from another group of 5 mice are presented in columns 11 to 15 (black bars). It is noted that the surgical intervention itself led to a variation in the number of rods compared to the non-operated eyes, and that the injection of PBS generated a slight protection of the rods. (Figure 5B) Average numbers (mean +/- standard deviation) of the sticks calculated from (Figure 5A). The increase in the numbers of rods following treatment with GDNF compared to untreated controls and having received treatment with PBS is clearly visible.

Figure 6: The expression of GDNF mRNA is not altered during the degeneration of photoreceptors in rd / rd mice. The expression of GDNF mRNA was followed by semi-quantitative RT-PCR. (Figure 6A) The expression of GDNF was compared between retinas with the rest of the eye. Total RNA was isolated from rd / rd retinas (lanes 1, 3) and from the rest of the eye (lanes 2, 4) and analyzed with specific primers of GDNF (1, 2) and G6PDH ( 3, 4). (Figure 6B) Total RNA isolated from whole eyes taken from rd / rd mice aged 12, (lanes 1, 12, 23); 22 (tracks 2, 3, 13, 14, 23, 24); 28 (tracks 4, 5, 15, 16, 26, 27); 35 (tracks 6, 7, 17, 18, 28, 29) and 120 days post-natal (tracks 8, 9, 19, 20, 30, 31), and from normal mice (C57B16) aged 35 days post -natal (lanes 10, 11, 21, 22, 23, 32, 33) were analyzed with primers specific for the rhodopsin gene (RHO-1- 11), GDNF (12 - 22) and the gene for interview coding for glucose-6-phosphate dehydrogenase (G6PDH, 23 - 33). The reduction in the levels of mRNA coding for rhodopsin, and the uniform profile of the levels of constant expression mRNA coding for GDNF and G6PDH, can be observed. 1) Materials and Methods used a) Animals: C3H / He / J mice homozygous for the retinal degeneration (rd) gene (rd / rd mice) and C57 / BL6 / J control mice aged 12, 15, 17, 19 and 25 days were obtained from the Janvier establishment for the supply of laboratory animals (Le Genest, St. Isle, France), kept in transparent plastic cages and subjected to standard 12-hour day cycles. All protocols complied with the ARVO Resolution on the use and care of animals for research purposes. b) In vivo application of GDNF to rd / rd mice: GDNF injections were carried out in C3H mice aged 16 and 20 days. The mice (n = 10) were anesthetized by intraperitoneal injection (16.7 μl / g body weight) of etomidate (0.5 mg / ml) and midazolane (0.5 mg / ml). The pupils were dilated by applying 0.5% tropicamide, and the corneas were anesthetized with 0.5% propacaine for topical application. A μl of GDNF (recombinant human GDNF) from the company Promega, Madison, IL., EUA (330 ng / μl) in 0.1 M sterile phosphate buffered saline (PBS) was injected using a syringe of the type Hamilton fitted with a 30-gauge needle at a distance of 1 mm from the corneal limbus in the subretinal space. The contralateral eye received an identical volume injection of a PBS solution only.

Bilateral eye surgery was necessary to distinguish the protective effect due to GDNF itself from that due to the experimental procedure. In rare cases, a partial reflux phenomenon has been observed. After injection, an examination of the back of the eye was carried out to confirm the induction of a local detachment of the retina. c Electroretinographic recordings (ERG) in rd mice: ERGs were performed using a cotton wick connected to an Ag: AgCl electrode by applying physiological water to the apex of the cornea using '' a micro-manipulator. A stainless steel reference electrode was inserted under the skin at the top of the head. The responses were amplified by a factor of 10,000 and filtered using filters with resolution thresholds low of 0.1 Hz and high of 1000 Hz, respectively present on the amplifier of the company Universal Gould Amplifier. The responses were digitized using an interface card of the type known under the designation labmaster (Scientific Solutions, USA), connected to a PC-type microcomputer. Experimental data was collected and analyzed using software packages of the type known as the Patchit and Tack designations. Following a 24 hour dark acclimation period, the mice were prepared under low intensity red lighting. They were anesthetized as described above and placed in a Faraday cage, the head being supported by a U-shaped support. The upper and lower eyelids were brought back to keep the eye open in a state of exophthalmosis . The light stimulus was obtained using a quartz-xenon lamp (from the company Muller Instruments, RFA). The light beam was focused at infinity through a heat-absorbing filter towards a hole made on the Faraday cage. The light intensity of the beam was 2.9 log cd / m ² as measured by a luxmeter at eye level. A computer-controlled shutter was used to deliver lightning with a duration of 300 ms. For the mice having received an injection of GDNF, recordings were obtained successively from each eye without the experimenter knowing which of the two eyes had received an injection of GDNF. d) Immunohistochemical marking of rods:

The mice were sacrificed by overdosing them with anesthetic following ERG recordings at 25 days. Unoperated eyes were also removed from five other rd / rd mice. The optical cups were removed and fixed in 4% paraformaldehyde. The retinas have been immunostained with an anti-rhodopsin antibody (rho-4D2) (Hicks and Molday, Exp. Eye Research 1986; 42, 55-71) which specifically stains rod-shaped photoreceptors. The retinas were washed three times with PBS, permeabilized in PBS containing 0.1% Triton X-100, incubated with the rho-4D2 antibody (10 μg / MI) for one hour, washed and incubated then with a goat anti-mouse igG coupled to Texas Red® (Jackson Laboratories, West Grove, PA) for one hour. The retinas were washed and flattened in PBS / glycerol (50/50 vol./vol.) Orienting the photoreceptor layer upwards, e) Cell counts: The quantification technique used consisted of a stereological approach allowing unbiased sampling (Gundersen et al., APMIS 1988; 96, 379). The immunolabelled cells were observed with a Nikon Optiphot 2 microscope under epifluorescent lighting using a 40/070 DIC objective (160 / 0.17). 120 fields of 8000 μm ² were selected over the entire retinal surface (approximately 13 mm ² ) using an encoder mounted on a plate and a systematic random procedure (Gundersen et al., 1988). The fields were viewed with a color graphic screen camera type Sony ® trinitron and digitized using Automator for Windows ® software from Biocom.

In each of the 120 fields, the cells were counted in two unbiased counting frames of 900 μm ² of surface generated by the software. The total number of rod-shaped cells over the entire retinal surface was estimated by dividing the values obtained by the total retinal surface measured by an Optiscan type image analysis system (Macintosh LC II Ci) so as to obtain standardized values. f) Analysis of the GDNF mRNA by a polymerase chain reaction coupled with reverse transcription (RT-PCR):

Total RNA was isolated with an RNA / DNA extraction kit from the company Quiagen (Valencia, CA, EUA) using a standard protocol. The RNA samples (1 μg) were brought into contact with a random hexameric primer and incubated for 2 hours at 37 ° C. with reverse transcriptase from the Moloney Murine Leukemia Virus. 1/20 (2 μl) of the cDNA was amplified by carrying out 35 cycles with pairs of specific primers. The sequences of the primers read from 5 ′ to 3 ′ were: rhodopsin, AAGCCGATGAGCAACTTCC,

TCATCTCCCAGTGGATTCTT; GDNF, ACCAGATAAACAAGCGGCAG, TCAGATACATCCA CACCGTTTAG; glucose-6-phosphate dehydrogenase (MG6PDH), GCAGTCACCAAGAACATTCAAG, CCCAAATTCATCAAAATAGCCC. 2) Results a) Electroretino grams: Figure 1 illustrates the parallel evolution of the ERG of wild type (C57) and rd / rd mice. In wild type mice, the amplitude of the a and b waves increases regularly from day 12 to day 24 postnatal. In the rd rd mice, the ERG was similar to that of the wild type on day 12 post-natal but thereafter the waves a and b saw their amplitude gradually decrease from day 15 post-natal, to the point of becoming undetectable after 24 days (Figure 1).

In order to determine whether GDNF is capable of improving visual functions in rd / rd mice, injections of this trophic factor were performed in the subretinal space on days 16 and 20, in conjunction with injections of PBS as a vehicle in the contralateral eye as witnesses. The animals underwent ERG recordings on days 21 and 24 postnatal at the time when the ERG signals had completely disappeared in the rd / rd mice. None of the eyes that received PBS had detectable ERG (n = 10). On the other hand, ERGs were detectable in 4 eyes having received a GDNF infection out of 10. (Figure 2A, numbers 3, 4, 7 and 9). The mean amplitudes of waves a and b for the mice injected with GDNF were 8.5 and 15.5 μV, respectively. A representative model of ERG recorded on an eye having received an injection of GDNF is presented in Figure 2B. b) Immunohistochemical and morphometric analyzes:

Following an ERG recording, the animals which received injections of GDNF and PBS (n = 10), and a second group of control mice of identical age (n = 5), were sacrificed and the rods were were marked and counted in flat retina montages. Figure 3 shows a low magnification view of rd / rd integral retina assemblies injected with PBS (Figure 3 A) and GDNF (Figure 3B) immuno-labeled with rho-4D2. There are obviously more rod-shaped photoreceptors that survive GDNF treatment. The images obtained at higher magnification showed rods marked in the form of small rounded cells, often having a short external segment (Figure 4). The stereological counting framework, as well as the principle of the systematic random sampling procedures employed, are likewise presented in Figure 4.

The counts of rod-shaped cells for each animal subject are presented in FIG. 4A [matched rd / rd mice having received an injection of PBS and GDNF (1-10), and rd / rd mice having not undergone d 'surgery (11-15)]. The numbers of opsin immunostained rods between different mice are obviously variable, but in each case treatment with GDNF resulted in higher cell counts.

There was a certain link between the intensity of protective effect and the functional state, it being specified that 3 mice out of 4 in which an ERG was detectable (Figure 5 A, numbers 3, 4 and 7) showed the greatest relative differences in terms of number of sticks [the average rate of increase in the number of sticks between animals given an injection of GDNF and PBS was 70% for the ERG-positive group (n = 4) and 34% for the ERG-negative group (n = 6)]. The average number of rod-shaped cells in the eyes given a GDNF injection (mean +/- standard deviation = 89232 +/- 8033) had increased by 46% compared to what was observed for eyes given an injection of PBS (61165 +/- 3259), and 98% compared to the control eyes that did not receive an injection (44958 +/- 1378) (Figure 5B). These gains were highly significant statistically speaking. It should also be noted that the injection of PBS led to an increase in the number of rods (+ 36%) compared to the untreated eyes. c) Expression of GDNF mRNA in rd / rd mice: In order to study the possible role of GDNF as an endogenous trophic factor in rods, the expression of endogenous GDNF mRNA in the retinas of mice was analyzed ( Figure 6A).

Lane 1 shows the lasting expression in larger amounts of GDNF mRNA in the retinas in comparison with the rest of the eye (6/4 ratio). 12-day-old rd / rd mice expressed rod-specific rhodopsin mRNA at rates comparable to those of 35-day-old C57 mice (lanes 1 and 10-11). Marked decreases in the level of rhodopsin mRNA expression were observed in rd / rd mice between days 12 and

22 (compare track 1 to tracks 2, 3) and between days 22 and 120 (compare track 4 to track 9). Throughout the study interval, the expression levels of GDNF mRNA and those of the maintenance enzyme G6PDH remained unchanged (lanes

12 - 20), and were comparable to the respective expression levels in the eyes of C57 mice at day 35 (lanes 21, 22).

3) Discussion The above results demonstrate in particular a stimulating effect of intraocular injections of the neurotrophic factor GDNF on the survival rate of rod-type photoreceptors in rd / rd mice. This effect has been demonstrated both by carrying out immunohistochemical and morphometric analyzes and by measuring photo-induced responses. The general correspondence between the degree of protection and the functional state suggests that even slight increases in the number of photoreceptors can be of considerable functional importance.

This observation makes it possible to formally report that the application of a neurotrophic growth factor and more particularly of GDNF contributes to improving the retinal function in an animal model of retinal degeneration close to human RP.

In the present study, the injection of PBS alone induced a slight protective effect, but did not lead to any detectable functional improvement.

The above results support the hypothesis that GDNF constitutes one of the few neurotrophic factors identified affecting the survival of rod-type photoreceptors in hereditary retinal degenerations in which the defective gene is located in the rod. It should be noted that criteria such as the chronology and the choice of ocular injection site are essential and that the methodological approach used in this study is probably just as decisive. The stereological counting method, originally developed to assess the number of neurons in the brain areas (Gundersen et al., 1988; 96, 379 / Coggleshall and Lekan, Journal of Comparative Neurology 1996; 364: 6-15), allows precise, unbiased estimation of total cell populations, greatly reducing the variability induced by local differences (upper hemisphere as opposed to lower hemisphere, injection site). With this method, it is possible to reliably screen the effects of various therapeutic compounds on the survival of photoreceptors, especially in rd / rd mice.

The tests described above and illustrated by the accompanying drawings show that there is a decrease in amplitude of ERG in the rd / rd mouse as described in several previous studies. The detection of an ERG signal in the eyes having received an injection of GDNF indicates that the visual function has been preserved in these retinas. This physiological observation is correlated in most cases with very large relative increases in the number of rods.

Furthermore, it was previously demonstrated that the transplant of rod-type photoreceptors has protective effects on the survival of the cones in the rd / rd mouse model and suggested that this effect was mediated by a diffusible factor released by the rods (Mohand -Saïd et al., Ophtal Research, 1997; 29, 290). Additional evidence for the existence of such diffusible signals produced by normal retinas and influencing the survival of cones was obtained using co-culture models (Mohand-Saïd et al., Proc. Natl. Acad. Sci. 1998 , Vol. 95 (14), 8357-8362).

This suggests that the observed survival of the rods after administration of GDNF should normally lead to the survival of cones necessary for daytime vision. Although exogenous GDNF protects rod-like photoreceptors, it is unlikely to represent this endogenous signal for several reasons. Expression of GDNF persists in the retina of rd / rd mice even after complete loss of rods, which would indicate that GDNF does not express itself in these cells (although its expression can be modulated positively in internal retinal cells during degeneration of photoreceptors as has been demonstrated in the case of retinal damage in the case of FGF-2 and CNTF). As it has been shown that the survival promoting effect cones is abolished following the loss of photoreceptors in the rd / rd retina (Mohand-Saïd et al., Ophthalmia Res. 1997; 29: 290-297), the absence of GDNF is therefore not due origin of degeneration of cones secondary to the loss of rods. Endogenous GDNF is not sufficient to reverse the apoptotic process in mutant rods, and may have functions other than the survival of photoreceptors in the retina of mice. Although its expression is not sufficient to prevent the loss of photoreceptors, its administration using a gene delivery system as successfully attempted in a model of Parkinson's disease in rats (Clarkson, 1995) could be useful to protect from retinal degeneration.

The description above mentions the invention more particularly in relation to the GDNF used for the treatment of pigmentary retinitis but, as already indicated previously, the invention can be extended at least to certain other members composed of the GDNF family for treatment of other types of retinal degeneration.

It should be noted that the invention is not limited to the embodiment, in particular to the mode of application of the molecule, described. Modifications remain possible, in particular from the point of view of the constitution of the various elements or by substitution of technical equivalents, without thereby departing from the scope of protection of the invention.

Claims

R E V E N D I C A T I O N S

1) Neurotrophic factor derived from glial cells (GDNF), or substance of the family comprising said factor, for use in obtaining a protective active agent for the treatment of retinal degenerations, more particularly pigmentary retinitis (RP) in the man. 2) Active protective agent according to claim 1, characterized in that the neurotrophic factor derived from glial cells consists of recombinant exogenous GDNF in a form which can be administered by injection.

3) Active protective agent according to claim 1, characterized in that the GDNF is released in situ by fibroblasts implanted by grafting, after transfection.