Final Report Summary - RDCVF (Rod-derived Cone Viability Factor)
Executive Summary:
The discovery of the RdCVF protein (Rod-derived Cone Viability Factor) has provided a clue to understanding the loss of cone photoreceptors and therewith of the central and light-adapted vision. In most cases of rod-cone degenerations, like retinitis pigmentosa (RP) the loss of the cone photoreceptors is a secondary effect caused by the degeneration of rod photoreceptors due to mutations expressed only in rods. It has been clearly shown that RdCVF protein protects cone viability in experimental models of retinal degenerative disorders.
Moreover, the RdCVF protein not only rescues cones but also significantly preserves their function, offering major therapeutic potential for human retinal neurodegenerative diseases.
In order to produce RdCVF under GMP conditions in mammalian cell lines, analytical methods development have been carry out and selective and specific polyclonal antibodies have been produced. Cellular expression of RdCVF was obtained at a very low efficiency of secretion which is not compatible with protein purification at a large scale. However we could assess the functionality of the protein on a range of assays and Chicken cone viability in vitro functional assay has been standardized.
Major advances have been also achieved in in vivo functional testing assay:
- Cone cell semi‐automated quantification has been standardized and compared to manual counting.
- rd1 mice showed a reproducible photoreceptor degeneration as measured by cone cell quantification.
- P23H rats showed a reproducible photoreceptor degeneration as measured by cone cell quantification, photopic ERG, cone tip area quantification.
During the last period of the project we have also observed that the secretion of RdCVF can be enhanced by treating the cells with 50 µM H2O2, only in the presence of RdCVFL. Such observation fits with a mechanistic model in which oxidative stress (H202) is acting on RdCVF expression and secretion mechanisms through RdCVFL, the thioredoxin enzyme.
Since there are currently still no data on protein therapy based on protein injection in retinal dystrophies, information on toxicity tests, relying on various animal models have been carried out. As the protein RdCVF was not available, we investigated the fate of two different proteins after intravitreal injection. As model proteins, we investigated bevacizumab and peroxidase. Peroxidase is a marker protein which can be very well displayed with the electron microscope. Bevacizumab was injected intravitreally into cynomolgus monkeys for this purpose. Our results are showing that after intravitreal injection both bevacizumab and the marker protein peroxidase are transported to and into the photoreceptors within a week by active transport mechanisms. We can now say that after intravitreal injection the proteins are transported into the blood vessels of the choroid were they induced alterations of the plasma composition and can interact with other plasma proteins. This interaction can induce fibrin formation and thrombocyte activation. The therapeutic proteins also interact with red blood cells. Within 24 hours after intravitreal injection the proteins were detected within the glomeruli where they persisted within the observation period of 1 week which is shown here for the first time. It is essential that these mechanisms are further investigated and understood for therapeutical experiments in order to understand whether this is clinically significant or not.
Although the RdCVF protein was not available we work on a delivery system based on nanoparticles, which should replace intravitreal injections, providing a stable release of the protein and also reducing the concentration of the injected protein, we could develop new methods for the synthesis of polymers, published in reference journals of the area, and we are sure that these results will be very useful for the preparation of polymers to biological applications.
Project Context and Objectives:
RdCVF is the abbreviation for a protein called Rod-derived Cone Viability Factor and at the same time the name of an European research network, which aimed at developing a therapeutic strategy for preserving the function of cone photoreceptors in patients suffering of retinal degenerations like retinitis pigmentosa (RP). This research network is coordinated by the Institut National de la Santé et de la Recherche Médicale (Inserm) through Prof. José Alain Sahel (Director of the Vision Institute, Paris, France). The consortium consists of 5 participants from 3 different European countries: France, Germany and Portugal
Retinitis Pigmentosa (RP) is the most common form of inherited retinal diseases that affects altogether 1,500,000 people worldwide. It is a genetically heterogeneous group of disorders having common characteristics that lead to irreversible loss of vision. Initially, in affected people, rod photoreceptors that are responsible for night vision degenerate, followed by cone celles, responsible for central and light)adapted vision. RP is still an untreatable disorder and preventing cone cell death is a very promising therapeutic approach, since even when 95% of the cones have been lost in patients the vision remains substantial. Hence we decided to target more specifically the secondary degeneration of cones, delivering and injecting a well-defined protein for preserving central vision. The protein in question is the “Rod derived Cone Viability Factor” (RdCVF), expressed and secreted by the rod photoreceptor, protecting cone photoreceptors from cell death.
To reach an appropriate clinical stage to apply a RdCVF protein therapy in inherited retinal degenerations patients we had to achieve the following goals:
1) The production of a preclinical Good Laboratory Practices (GLP*) batch which will be performed in a Good Manufacturing Practices (GMP) compliant facility and finally its functionality will be validated in in vitro and in vivo assays
2) The determination of the dosage of the protein by pharmacokinetic and pharmacodynamic studies
3) The determination of the toxicology of the different dosages of this protein in normal and mutant mice, in rats and in monkeys
4) Optimization of the therapeutic effect, productivity and purificability of the RdCVF protein through the modification of its aminoacid sequence
5) The development of a polymeric device capable of an efficient delivery of the RdCVF protein in the retina
However the RdCVF consortium had to face to a major unexpected technical issue, namely that the RdCVF protein is hydrophobic and insoluble. This, as far, unknown nature of the functional RdCVF protein made impossible, using the tested methods, its production in a quantity necessary for a therapy. Through this unexpected result, the consortium wanted to refocus its work plan proposing 2 alternatives:
1) Design and synthesis of molecular agonists of the protein
2) A gene-based, instead of a protein therapy.
It was estimated by EU-appointed experts that these alternative strategies could not bear fruit during the span of time of the funded project.
Project Results:
What the RdCVF consortium experienced, during the last 24 months, was a very new situation and this was very difficult to manage for everybody. During the whole duration of the project, unfortunately, not really the lack of the RdCVF protein, but much more the uncertainty to be able to produce it in the necessary quantity requested for therapeutics aims and especially well timed to at least achieve some of the foreseen objectives, overshadowed and influenced our work plan. Nevertheles we have been able to work on standardization and optimization of methods, ameliorating the procedures (WP1 and WP3) and introducing new methods for the synthesis of polymers relevant to biological applications (WP5). In the meantime P1 (WP1 and WP4) could acquaint very new knowledge on the complexity and functionality of the RdCVF protein.
In order to produce RdCVF under GMP conditions in mammalian cell lines, analytical methods development have been carry out and selective and specific polyclonal antibodies have been produced. Cellular expression of RdCVF was obtained at a very low efficiency of secretion which is not compatible with protein purification at a large scale. However we could assess the functionality of the protein on a range of assays and Chicken cone viability in vitro functional assay has been standardized.
The Chick embryo cone enriched culture system was described in Léveillard et al., (2004) and is adapted from the high content screening that led to the identification of RdCVF. Retinal precursor cells are isolated from chicken embryo at day 6 of embryogenesis and seeded at two different densities 1 and 2x105 cells/cm2. The cells are then incubated for a period of seven days. After this period, the viability of the cells is measured using Live/Dead assay using a cell counting platform composed of an inverted microscope and a CCD camera driven by an in house application base on the software Metamorph. Only the viable cells are taken into consideration for scoring the activity. The assay is based on the evaluation of the viability activity as average of four wells of culture for each seeding density.
Kinetics of cone viability has been established at four time points (D+1, 3, 5 and 7) during the process of cell degeneration in vitro. The viability of the cells in these cultures is scored using a viability assay that relies on the use of two fluorescent probes (Live/Dead assay). The calcein is metabolized to a fluorescent molecule emitting in green only if the cell is alive, while the dimer of ethidium is penetrating the nucleus where it binds to the DNA enhancing its red fluorescence only if the cell is dead. These cultures contain 70-80% of cone cells as seen after labelling with anti-visinin antibodies, a cone marker. The cell viability increases between D1 and D3 which corresponds to a period of proliferation of the retinal precursors when isolated from the chicken embryo at Day 6 of in ovo development (Stage 29 of chicken development). Thereafter, the cells die between D3 to D7. For further tests, trophic activity of hRdCVF109 was tested at D7.
Tests of the different batches of hRdCVF proteinsQuantification of viability activity of different batches of hRdCVF protein produced in mammalian cells and in produced in E. coli was assayed in cone enriched culture assay.
Major advances have been also achieved in in vivo functional testing assay:
- Cone cell semi‐automated quantification has been standardized and compared to manual counting.
- rd1 mice showed a reproducible photoreceptor degeneration as measured by cone cell quantification.
- P23H rats showed a reproducible photoreceptor degeneration as measured by cone cell quantification, photopic ERG, cone tip area quantification.
Cone cell semi‐automated quantification has been standardized and compared to manual counting.
Standardization of the cone counting of rd1 retina: Partner 1 has developed an automated cone counting system for scoring the trophic activity of the hRdCVF109 proteins rd1 mouse cones labelled with PNA. The system involves an inverted microscope driven by a software that allows the scanning of the retinal surface, the focussing onto 300 to 400 zones that represent the entire retinal surface, the acquisition of a pile of 9 images and the treatment of these piles of images with a deconvolution step in order to calculate the cone density of the retina.
In order to validate this cone counting platform, a comparison of densities of cones of the rd1 retina during the period of the secondary degeneration of cones to those obtained with a semi-automated system that relies on the visual observation of the labelled images as used by Mohand-Saïd et al. (1998). The two methods giving similar results the automated system was validated.
Because the two methods also differ in their coverage of the retinal surface, and because the automated method is not a stereological approach but a global counting, an additional automated method has been developed that is more directly related to the stereological counting of cones, using an unbiased sampling approach and a counting of these samples in order to provide an estimation of the cone density over the entire retina. This novel method was validated by measuring the density of cones during the period of their degeneration in the rd1 mouse
The automated global method was used to evaluate the trophic activity of the hRdCVF109 proteins ex vivo on retinal explants (data not shown), while the automated stereological approach was use after in vivo sub-retinal injection in the rd1 mouse (task 1.10 of the WP1 in Annex I)
During the last period of the project we have also observed that the secretion of RdCVF can be enhanced by treating the cells with 50 µM H2O2, only in the presence of RdCVFL. Such observation fits with a mechanistic model in which oxidative stress (H202) is acting on RdCVF expression and secretion mechanisms through RdCVFL, the thioredoxin enzyme.
As reviewed previously, we were not able to achieve scalable production of hRdCVF protein using CHO‐S cells line as well as HEK293 cell lines, but also COS‐1 as well as in E. coli bacteria. During this last period, we have developed an alternative approach aimed at enhancing the secretion of hRdCVF from HEK293 cells using an oxidative agent in order to achieve a level of the protein in the conditioned media that would resolve the problem of its purificability.
The importance of oxidation in this protein function and production is emphasized by evolutionary science facts. The inverse correlation between the rate of oxygen consumption and the life-span of mammals has directed the attention of aging research onto oxygen metabolism. It was discovered that reactive oxygen species (ROS) which are too reactive to exist in biological systems, are formed in situ in response to radiation and oxygen poisoning and are responsible for the associated toxicity. Free radicals are continuously produced in the cell as a product of aerobic life. In order to avoid the damage of macromolecules by ROS proper redox conditions must be maintained within the intracellular environment. Therefore, aerobic organisms have developed several antioxidant systems including superoxide dismutase, catalase and thioredoxin (TXN) systems. The prototype of the thioredoxin proteins, TXN, is a 12 kDa protein with a redox active disulfide/dithiol group within the conserved active-site sequence CGPC. Reduced TXN catalyses the reduction of disulfide bounds in many proteins, and oxidized TXN1 is reversibly reduced by the action of the thioredoxin reductase and NADPH (Holmgren et al, 2005). Secretion of TXN by a hitherto unknown mechanism that is not dependent on a signal peptide has been observed under conditions of oxidative stress (Kondo et al, 2004; Rubartelli et al, 1992).
We have shown that the Nxnl1 participates in a physiological signal involved in the maintenance of photoreceptors during aging and exposure to oxidative stress (Cronin et al, 2010). It is tempting to speculate that the two isoforms of the Nxnl1 gene, RdCVF and RdCVFL work in the same regulatory system: the RdCVFL enzyme would be the sensor for oxidative conditions, coupling to the trophic protein RdCVF for an environmentally adapted response (Leveillard & Sahel, 2010). Indeed, while we have identified a functional interaction between RdCVFVL and TAU metabolism (Fridlich et al, 2009), we are still investigating a mechanistic link between the two products of the Nxnl1 gene
Since there are currently still no data on protein therapy based on protein injection in retinal dystrophies, information on toxicity tests, relying on various animal models have been carried out. As the protein RdCVF was not available, we investigated the fate of two different proteins after intravitreal injection. As model proteins, we investigated bevacizumab and peroxidase. Peroxidase is a marker protein which can be very well displayed with the electron microscope. Bevacizumab was injected intravitreally into cynomolgus monkeys for this purpose. Our results are showing that after intravitreal injection both bevacizumab and the marker protein peroxidase are transported to and into the photoreceptors within a week by active transport mechanisms. We can now say that after intravitreal injection the proteins are transported into the blood vessels of the choroid were they induced alterations of the plasma composition and can interact with other plasma proteins. This interaction can induce fibrin formation and thrombocyte activation. The therapeutic proteins also interact with red blood cells. Within 24 hours after intravitreal injection the proteins were detected within the glomeruli where they persisted within the observation period of 1 week which is shown here for the first time. It is essential that these mechanisms are further investigated and understood for therapeutical experiments in order to understand whether this is clinically significant or not.
For this project proposed rat strain P23H was evaluated for classification and characterization of retinal development and development of disease progression. As known from the literature the P23H rat was characterized up to day 150, we were able to detect remaining retinal function over an observation period up to 350 days. The retinal function assessed by fullfield electroretinography using more differentiated ERG protocols can evoke clear responses to light up to day 350. Especially retinal illumination with a 5, 10 and 15 Hz flicker stimulus could evoke retinal response in a really late stage of the P23H rat. Overall the development of specific ERG protocols in case of animal models is strongly recommended.
All these results can be used later for validation of GMP grade therapy on this degeneration model.
Publication is foreseen by end of 2013
Although the RdCVF protein was not available we work on a delivery system based on nanoparticles, which should replace intravitreal injections, providing a stable release of the protein and also reducing the concentration of the injected protein, we could develop new methods for the synthesis of polymers, published in reference journals of the area, and we are sure that these results will be very useful for the preparation of polymers to biological applications.
Potential Impact:
Our consortium held a unique competitive position as it comprises the discoverers of RdCVF and the SME that holds the rights to develop this compound. The participation of other EU partners had to ensure that the validation and toxicological studies will gain wide acceptance.
The clinical trials that should follow this program cannot be conducted at a national level as the main database for affected patients has been developed at a EU level. No single country could succeed in performing efficient clinical phase II or III trials on this orphan diseases group. The US Foundation Fighting Blindness has recognized the strong potential of RdCVF (Trustees Award to the discoveries in 2005) and will support the extension of EU results to US groups.
There is little doubt that in this orphan condition, preserving central vision in patients affected with a wide range of mutations, would have a strong impact. There is accumulating evidence that the transfer of RdCVF into the clinical setting would still have a tremendous impact. Clearly, the production of a clinical grade protein represents a major hurdle. The pharmacological, pharmacokinetics, toxicological studies proposed herein should still pave the way to a protein therapy. The development of innovative delivery approaches will enhance the potential and range of applications.
However due to unexpected characteristics of RdCVF, it was impossible for the consortium, during the project period, to produce the GMP-grade protein in a quantity necessary for therapy. During this time the participants investigated on the complexity of this protein and on the improvement of the validation and toxicological methods, understanding better transport mechanisms of protein in the blood vessels. In addition new methods for the synthesis of polymers for protein delivery, which will be published in reference journals of the area, have been also developed.
Due to the lack of the planned deliverables and results, no main dissemination activities and the exploitation of results have been implemented. Most of the dissemination activies have been initiated during the first period of the project: leaflets and posters have been printed and distributed, partners participated to meetings and confencerences to present the potentiality of the project and a website has been set up.
Nevertheless there is a large awareness in the patient and scientific community regarding the potential of RdCVF (http://www.ffb.ca/patient_resources/factsheets/cone_cells.html) a protein that keeps raising important expectations.
The consortium during the the last months of the project discussed the eventuality of 2 alternatives work plans, which have been presented to external experts, who evaluated these 2 new proposals as valuable, but not realistic for the remaining short timeframe.
List of Websites:
www.rdcfv.eu
José Alain Sahel (scientific coordinator)
Institut de la Vision
17 rue Moreau
75012 Paris - France
j.sahel@gmail.com
Emanuela De Luca (project manager)
Fondation de Coopération Scientifique
Voir & Entendre
Institut de la Vision
17 rue Moreau
75012 Paris - France
emanuela.de-luca@institut-vision.org
