Chimeric promoter for targeted expression in AII amacrine cells

Abstract

The present invention relates to the use of a chimeric promoter having a promoter activity in AII amacrine cells. The present invention also relates to expression cassettes or vectors comprising said promoter operably linked to a coding sequence of interest as well as viral particles or host cells comprising said expression cassette or vector. The present invention also relates to the use of said expression cassettes, vectors, viral particles or cells in the treatment of ocular disease, in particular ocular disease associated with photoreceptor cell degeneration.

Description

[0001] CHIMERIC PROMOTER FOR TARGETED EXPRESSION IN All AMACRINE CELLS

[0002] FIELD OF THE INVENTION

[0003] The present invention relates to the field of medicine, in particular to the treatment and prevention of ocular diseases associated with photoreceptor cell degeneration.

[0004] BACKGROUND OF THE INVENTION

[0005] Blindness affects 45 million people worldwide. In many cases of retinal degeneration photoreceptors are lost, while retinal ganglion cells (that provide visual signals to the brain) as well as many interneurons, such as amacrine cells, are maintained. This opens the possibility to stimulate the remaining retinal ganglion or amacrine cells directly to restore visual function.

[0006] Retinal prostheses are a promising solution and have been found to restore some useful perception in blind patients. However, the acuity of the existing devices remains very low, below the level of legal blindness (Lorach et al. 2012, da Cruz et al. 2016). Patients also report that percepts evoked by electrical stimulation of retinal neurons are not easily interpretable as visual stimuli (Beyleler et al. 2019) and therefore are often not sufficient to identify objects or to navigate in complex environments.

[0007] Optogenetic therapies provide a promising alternative to restore vision with a higher resolution and specificity that can better mimic the natural output of the retina (Ferrari et al. 2020). In this strategy, a light sensitive protein is expressed in specific neural populations in a blind retina.

[0008] Expressing light sensitive proteins in retinal ganglion cells can be an efficient way to restore vision through the stimulation of these newly light-sensitive cells with patterned light to evoke visual perception (Bi et al. 2006, Caporale et al. 2011, Sengupta et al. 2016, Chaffiol et al. 2017, Berry et al. 2019), although the first results show that the acuity is still low with this strategy (Sahel et al. 2021). How to optimize visual acuity and perceptual performance when restoring vision using optogenetics is an active area of investigation.

[0009] In a healthy retina, the ganglion cell population can be divided into about 20 to 40 cell types that each perform a different computation on the visual scene (Sanes & Masland 2015, Baden et al. 2016). Each cell type is classically assumed to be selective of a specific feature of the visual scene and therefore convey a corresponding feature map to the brain (Deny et al. 2017). Altering specifically one of these cell type populations can lead to specific impairments in visual perception and motor output, including specific defects in perceiving moving objects and eye movement control (Merigan et al. 1991, Yonehara et al. 2016, Hillier et al. 2017). In particular, ganglion cells usually respond either to light increase (ON ganglion cells) or light decrease (OFF ganglion cells). Inactivating ON ganglion cells leads to a reduced ability to detect increase of luminance at the perceptual level while ability to detect decrease of luminance is not affected (Schiller et al. 1986). Optogenetic strategies targeting ganglion cells will not restore the computations performed in the normal retina. In particular, making ganglion cells light-sensitive will result in a retina where all ganglion cells become de facto ON cells (only responding to light increase). It is unclear how this synthetic visual signal will affect the physiological processing performed by downstream areas in the brain and what will be the resulting restored perception, but this loss of retinal computations could severely impair perceptual performance.

[0010] To restore some of the response diversity found in normal retinas with optogenetic therapy, other cells such as “dormant” cones (Busskamp et al. 2010) or bipolar cells (Mace et al. 2015, Gaub et al. 2015, van Wyk et al. 2015, Cehajic-Kapetanovic et al. 2015) have been targeted specifically. However, In many retinal dystrophy patients, photoreceptors are not present in late stages, and bipolar cells can be partially degenerate (Francis et al. 2013). Indeed, retinal ganglion cells and All amacrine cells have been shown to be the most robust neuronal cell types during retinal degeneration. All amacrine cells that are stable over a longer time (Strettoi et al. 2002), can therefore be used at the most advanced stages of degeneration.

[0011] The expression of Channelrhodopsin2-GFP using AAV2 under a ubiquitous promoter was shown both in short term and long term studies for up to 18 months (Bi et al., 2006). In these studies expression of channelrhodopsin2-GFP was observed predominantly in retinal ganglion as well as in displaced amacrine cells of the ganglion cell layer. Although these studies showed that it is possible to obtain weak expression in All amacrine cells, finding a promoter to allow higher level expression in this target cell population is still needed to obtain a reactivation strategy based on All amacrine cells.

[0012] SUMMARY OF THE INVENTION

[0013] In a first aspect, the present invention relates to an expression cassette comprising a nucleic acid having a promoter activity in All amacrine cells operably linked to a coding sequence of interest, for use in the treatment of an ocular disease, wherein said nucleic acid having a promoter activity comprises an enhancer element of an inter-photoreceptor retinoid-binding protein (IRBP) promoter that is fused to a promoter nucleotide sequence of a cone transducin alpha-subunit (GNAT2) gene. The present invention also relates to the use of an expression cassette comprising a nucleic acid having a promoter activity in All amacrine cells operably linked to a coding sequence of interest, for the manufacture of a medicament for the treatment of an ocular disease, wherein said nucleic acid having a promoter activity comprises an enhancer element of an inter-photoreceptor retinoid- binding protein (IRBP) promoter that is fused to a promoter nucleotide sequence of a cone transducin alpha-subunit (GNAT2) gene. The present invention further relates to a method for treating an ocular disease comprising administering a therapeutically efficient amount of an expression cassette comprising a nucleic acid having a promoter activity in All amacrine cells operably linked to a coding sequence of interest, to a subject in need thereof, wherein said nucleic acid having a promoter activity comprises an enhancer element of an inter-photoreceptor retinoid-binding protein (IRBP) promoter that is fused to a promoter nucleotide sequence of a cone transducin alpha-subunit (GNAT2) gene.

[0014] Preferably, the treatment involves the intravitreal administration of said expression cassette to the subject being treated.

[0015] The enhancer element of an IRBP promoter may comprise, or consist of, the nucleotide sequence of SEQ ID NO: 2, or a functional variant thereof having at least 80% identity to SEQ ID NO:2 or comprising at least 150 consecutive nucleotides of SEQ ID The promoter nucleotide sequence of a GNAT2 gene may comprise, or consist of, the nucleotide sequence of SEQ ID NO: 3, or a functional variant thereof having at least 80% identity to SEQ ID NO:3 or comprising at least 200 consecutive nucleotides of SEQ ID NO:3.

[0016] Preferably, the enhancer element of an IRBP promoter is positioned upstream of the promoter nucleotide sequence of a GNAT2 gene.

[0017] In particular, the nucleic acid having a promoter activity in All amacrine cells may comprise, or consist of, the sequence of SEQ ID NO: 4 or a functional variant thereof exhibiting a promoter activity in All amacrine cells and having at least 80% identity to

[0018] The expression cassette may further comprise a splice donor / splice acceptor element, preferably comprised in an intronic sequence known to increase splicing efficiency and expression of the coding sequence. Preferably, the splice donor / splice acceptor element is position downstream of the nucleic acid having a promoter activity in All amacrine cells and upstream of the coding sequence of interest. In particular, the splice donor / splice acceptor element may be comprised in an intronic sequence comprising, or consisting of, sequence of SEQ ID NO: 10 or a functional variant thereof having at least 80% identity to SEQ ID NO: 10.

[0019] The expression cassette may be provided in a vector, preferably a viral vector, more preferably an adeno-associated viral (AAV) vector. In particular, the expression cassette may be provided in AAV particle comprising an AAV vector. Preferably the AAV particle comprises an AAV-derived capsid selected from the group consisting of AAV2 and AAV2-7m8. More preferably, the AAV particle comprises an AAV2-7m8 capsid.

[0020] The expression cassette may be provided in a host cell transformed or transfected with said expression cassette, or with a vector or viral particle comprising said expression cassette. Preferably, the host cell is a human host cell, more preferably an All amacrine cell, and even more preferably a human All amacrine cell.

[0021] The coding sequence of interest may be a therapeutic protein, an optogenetic protein or a reporter protein. Preferably, the coding sequence of interest is an optogenetic protein. Preferably, the ocular disease is a disease associated with photoreceptor cell degeneration, more preferably is selected from the group consisting of retinitis pigmentosa, age-related macular degeneration, cone-rod dystrophy, Leber congenital amaurosis, Stargardt’s disease, diabetic retinopathy, Best’s disease, choroideremia or a tapetoretinal degeneration, even more preferably is retinitis pigmentosa.

[0022] In another aspect, the present invention also relates to

[0023] - an expression cassette comprising a nucleic acid having a promoter activity in All amacrine cells operably linked to a sequence encoding an optogenetic protein, wherein said nucleic acid having a promoter activity comprises an enhancer element of an inter-photoreceptor retinoid-binding protein (IRBP) promoter that is fused to a promoter nucleotide sequence of a cone transducin alpha-subunit (GNAT2) gene,

[0024] - a vector comprising said expression cassette, preferably an adeno-associated viral (AAV) vector,

[0025] - an AAV particle comprising said a vector, preferably comprising an AAV- derived capsid selected from the group consisting of AAV2 and AAV2-7m8, more preferably an AAV2-7m8 capsid, and

[0026] - an isolated host cell transformed or transfected with said expression cassette, said vector, or said AAV particle, preferably an All amacrine cell, more preferably a human All amacrine cell.

[0027] In a further aspect, the present invention relates to a pharmaceutical composition comprising an expression cassette, a vector, an AAV particle or a cell of the invention, and a pharmaceutically acceptable excipient.

[0028] In another aspect, the present invention also relates to a method for expressing a coding sequence of interest in an All amacrine cell, said method comprising introducing into said cell an expression cassette comprising the coding sequence of interest operably linked to a nucleic acid having a promoter activity in All amacrine cells, said nucleic acid having a promoter activity comprising comprises an enhancer element of an interphotoreceptor retinoid-binding protein (IRBP) promoter that is fused to a promoter nucleotide sequence of a cone transducin alpha-subunit (GNAT2) gene. It also relates to the use of a nucleic acid having a promoter activity in All amacrine cells for expressing a coding sequence of interest operably linked to said promoter in an All amacrine cell, said nucleic acid having a promoter activity comprising comprises an enhancer element of an inter-photoreceptor retinoid-binding protein (IRBP) promoter that is fused to a promoter nucleotide sequence of a cone transducin alpha-subunit (GNAT2) gene.

[0029] BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1: Investigation of expression pattern upon intravitreal delivery of AAV2- 7m8-IRBP / GNAT2-eGFP. A. Representative eye fundus image of a wild type mouse eye treated with AAV2-7m8-IRBP / GNAT2-GFP. B, B’. Post mortem retinal flatmount with endogenous eGFP fluorescence and immuno staining against GAD to visualize GABAergic cells. Scale bar is 50um.

[0031] Figure 2: AAV-mediated gene delivery to All amacrine cells from the vitreous A. Whole-mount view showing eGFP expression in the plane of All somas (INL). B. Retinal cryosections showing eGFP expression and labeling of starburst amacrine cells with a ChAT antibody. C,C’,C” . Retinal cryosections showing eGFP expression under the control of our promoter and Proxl immunostaining. D, D’, D”. Retinal cryosections showing ReachR-eYFP expression under the control of our promoter, Proxl immunostaining. For all panels, retinal layers are shown on the left: inner nuclear layer (INL), ganglion cell layer (GCL). DAPI labels all nuclei, while Proxl labels bipolar and All amacrine cells. Co-labeled All amacrine cells are indicated with white arrows. Colabeled retinal ganglion cells are indicated by white arrowheads. Scale bars are 50um.

[0032] Figure 3: ON and OFF RGC responses elicited upon light stimulation of ReaChR- expressing All amacrine cells A. All amacrine cells connect to the ON pathway through gap junctions, and to the OFF pathway through glycinergic inhibitory connections. Stimulation of Alls hence produces responses of opposite polarity on ON and OFF retinal ganglion cells. We target Alls through optogenetic stimulation consisting of a series of full field flashes, and record the responses of the retinal ganglion cells with a multielectrode array. Pharmacology (ACET and L-AP4) blocks synaptic transmission from photoreceptors. B. Responses of representative ON (left column) and OFF (right column) retinal ganglion cells to photoreceptor stimulation with white full-field flashes. Top: spiking activity across different trials. Bottom: mean response. The time intervals of the flashes are shown in gray. C. Responses of the same ON (left column) and OFF (right column) RGCs shown in B to optogenetic stimulation, after blocking photoreceptor synaptic transmission. Top: spiking activity across different trials. Bottom: mean response. D. Percentage of retinal ganglion cells responding to optogenetic stimulation. Left: percentage of ON RGCs responding to the flash respectively at onset, offset, or never, for both low and high luminance. Center and Right: same plot for OFF and ON- OFF RGCs, respectively. E. Percentage of retinal ganglion cells responding to optogenetic stimulation for a control population with no opsin expressed. Same plots as in D.

[0033] Figure 4: Off RGC Responses of inverted polarity are due to off-target opsin expression A. Control protocol showing direct ganglion cell activation due to off-target expression of ReaChr: the application of CNQX and CPP disrupts all the excitatory synaptic connections. Responses induced by visual stimulations under this condition are only due to direct activation of the retinal ganglion cells expressing the opsin. B. Mean responses of three representative RGCs (one ON and two OFF) to simple photoreceptor stimulation. The stimulation time interval is depicted in gray. C. Mean responses of the same three RGCs to optogenetic stimulation, after blocking the photoreceptor transmission. D. Mean responses of the same RGCs after blocking all the excitatory synaptic connections. Responses of RGC 3 can only be explained by off-target expression of the opsin in ganglion cells. E. Percentage of ganglion cells responding to direct optogenetic stimulation, due to off-target expression. Left: percentage of ON RGCs responding respectively to the stimulus onset or offset (or not responding), at both low and high luminance. Center and Right: same plot for OFF and ON-OFF RGCs, respectively. F. Percentage of ganglion cells responding to direct optogenetic stimulation for a control population with no opsin expressed. Same plots as in E.

[0034] Figure 5: The optogenetic activation of RGCs (mediated by ReaChR-expressing All amacrine cells) varies with the stimulus light intensity. A. Activation of ON RGCs with optogenetic stimulation (mediated by ReaChR-expressing All amacrine cells) at different light intensities. The X axis (logarithmic scale) represents the light intensity of the stimulus (a white flash). For each of seven luminance levels we show the percentage of RGCs that responded to the stimulation with ON, OFF, and no responses. The two green dashed lines represent the two light levels - low intensity and high intensity, respectively - chosen for our study. B. Activation of OFF RGCs with optogenetic stimulation (mediated by ReaChR-expressing All amacrine cells) at different light intensities. Same plot as in A. C. Activation of ON-OFF RGCs with optogenetic stimulation (mediated by ReaChR-expressing All amacrine cells) at different light intensities. Same plot as in A.

[0035] Figure 6: The diversity of RGC responses is really due to All activation, and not to photoreceptor transmission or off-target expression A. Examples of responses to photoreceptor stimulation for a representative ON (left column, red) and OFF (right column, blue) retinal ganglion cells. Top: raster plot of RGC responses across trials. Bottom: mean responses. The stimulation period is represented by the gray regions. B. same as A for optogenetic stimulation, after blocking photoreceptor transmission: this time All amacrine cells express an inhibitory opsin, gtACR. C. Percentage of RGCs responding to optogenetic stimulation (with the inhibitory opsin gtACR). Top: percentage of ON RGCs responding respectively at stimulus onset, stimulus offset, or not responding, for both low and high luminance. Bottom and Right: same plot for OFF and respectively.

[0036] Figure 7: All activation generates diverse RGC responses A. On top in black: the temporal profile of the full-filed chirp stimulus (luminance over time). Below: mean responses of four representative retinal ganglion cells to the chirp stimulus. In the top rows: responses mediated by photoreceptors (no optogenetics involved). In the bottom rows: responses mediated by the optogenetic activation of Alls (with blocked photoreceptor transmission) B. Comparison between the Sustained-Transient index computed on photoreceptor and optogenetic responses for 48 selected RGCs. Each dot shows the index for a different RGCC, for both photoreceptor responses (X axis) and optogenetic responses (Y axis). Values close to one mean the response is predominantly sustained; values close to zero mean the response is predominantly transient. C. Distribution of the comparative sustained-transient index AST for the ON and OFF RGC populations (same 48 RGCs shown in B). An index close to 1 means that the cell responds more transiently when activated optogenetically with respect to its normal photoreceptor responses. Conversely, an index close to -1 means that the optogenetic responses are more sustained than the normal photoreceptor responses. D. Response correlations across retinal ganglion cell pairs from a population of 40 RGCs, for both photoreceptor and optogenetic stimulations. Pearson correlation coefficient computed on the responses of pairs of RGCs to photoreceptor stimulation (X axis) and to optogenetic stimulation (Y axis). ON RGCs dots represent ON to ON response correlations: OFF RGCs dots represent OFF to OFF response correlations. E. Principal Component Analysis of the mean RGC responses to the chirp stimulus for a selected population of 40 cells (same shown in D). Each curve shows the number of principal components needed (X axis) to explain a given percentage of variance in the ganglion cell responses (Y axis). We show the curves for both photoreceptor responses and optogenetic responses. Light dashed curves represent analysis conducted on the individual experiments. Dark, continuous lines represent the average across all experiments.

[0037] Figure 8: Optogenetic Activation of Alls produces ON and OFF RGC responses also in dystrophic retinas A,B,C. Cross section of a dystrophic mouse retina (rdl) showing ReaChR-eYFP expression (panels A and C) under HKamac promoter; co-localization with Proxl antibody (panels B and C) indicates ReaChR expression in All amacrine cells. Expression under control of DAPI is shown in blue (panel C). All amacrine cells are indicated with white arrows. Retinal layers are shown on the left: outer plexiform layer (OPL), inner nuclear layer (INL), inner plexiform layer (IPL), ganglion cell layer (GCL). D. Optogenetic responses of 10 representative retinal ganglion cells from dystrophic retinas to a series of flashes. Left: raster plots of the responses; each row represents a different cell; the stimulation period is highlighted in gray. Right: mean responses over trials. E. Activation of RGCs in dystrophic retina due to optogenetic stimulation. For each of five luminance levels we plot the percentage of retinal ganglion cells that responded to the optogenetic stimulation with pure ON, pure OFF, ON-OFF or no responses.

[0038] Figure 9: Promoter and intronic sequences preceding transgene (eGFP or ReaChR-eYFP) sequences. IRPB enhancer, GNAT2 enhancer and SV40 SD / SA sequences are framed. Restriction enzymes sites inserted for cloning purposes are underlined.

[0039] DETAILED DESCRIPTION OF THE INVENTION

[0040] A valuable strategy to restore visual function in patients suffering from photoreceptor degeneration would be to target other cell types in the intermediate layers of the retina that are not damaged. All amacrine cells are an ideal target because they connect with both ON and OFF bipolar cells with different types of synapses (Famiglietti & Kolb 1975, Vaney 1985, Kolb et al. 1992, Wassle et al. 1995, Xin & Bloomfield 1999, Guadagni et al. 2019). They form gap junctions with most ON bipolar cell types (Tsukamoto & Omi 2017), and can therefore excite them when they are activated. At the same time, they form glycinergic inhibitory synapses with most OFF bipolar cell types (Kolb & Nelson 1983, Strettoi et al. 1992, Tsukamoto & Omi 2017).

[0041] For the first time, the inventors herein demonstrated that All amacrine cells can be targeted thanks to a chimeric promoter comprising an enhancer element of the interphotoreceptor retinoid-binding protein combined with a minimal human transducin alpha- subunit promoter. This promoter was previously described as being able to direct weak cone-specific expression in mice and dogs (Dyka et al., 2014; Ye et al., 2016, US 9,375,491). The inventors herein found that this promoter is able to target specifically All amacrine cells and to provide high level expression of a transgene in these cells.

[0042] Using this chimeric promoter, they obtained high levels of microbial opsin expression in this cell type rendering them light sensitive. This strategy has the advantage of being mutation-independent, and can potentially be used for different genotypes of retinal dystrophies or other retinal diseases associated with photoreceptor cell degeneration. Compared to ganglion cell targeting, which is currently in clinical trials, this approach has the advantage of restoring a significant part of the retinal computations performed by a normal retina. In particular, the inventors have shown that this strategy allows restoration of both ON and OFF ganglion cell responses. They also observed diverse responses, presumably corresponding to the activation of several different cell types and pathways in the retinal circuit, similar to what happens in the normal retina.

[0043] Definitions

[0044] In the present description, the terms “nucleic acid”, “nucleic sequence”, “polynucleotide” , “oligonucleotide” , “nucleotide sequence” are used interchangeably and refer refers to a polymeric form of nucleotides of any length, either ribonucleotides or deoxyribonucleotides. Thus, this term includes, but is not limited to, single-, double- or multi- stranded DNA or RNA, genomic DNA, cDNA, DNA-RNA hybrids, or a polymer comprising purine and pyrimidine bases, or other natural, chemically or biochemically modified, non-natural, or derivatized nucleotide bases. The backbone of the polynucleotide can comprise sugars and phosphate groups (as may typically be found in RNA or DNA), or modified or substituted sugar or phosphate groups. Alternatively, the backbone of the polynucleotide can comprise a polymer of synthetic subunits such as phosphoramidates and thus can be an oligodeoxynucleoside phosphoramidate (P-NH2) or a mixed phosphoramidate- phosphodiester oligomer. Methods for preparing nucleic acids are well known to one skilled in the art and include chemical synthesis, recombination, and mutagenesis. Unless otherwise specified, a nucleic acid disclosed herein is preferably a DNA molecule.

[0045] As used herein, the term "promoter” refers to a regulatory element that has the ability to initiate transcription of a nucleic acid to which it is operably linked. A promoter can regulate both rate and efficiency of transcription of an operably linked nucleic acid. The activity of such a promoter can be evaluated by any method known to the skilled person, and in particular by any method for quantifying the mRNA or protein obtained from the gene controlled by the promoter. By way of illustration, the activity can be measured by evaluating the amount of mRNA, for example by Northern blot or RT-PCR, or by evaluating the amount of translated protein, for example by Western blot, ELISA, colorimetric tests, enzyme activity tests or by using a reporter gene system.

[0046] As used herein, the term “operably linked” refers to nucleotide sequences that are combined / arranged in such a way that the function of one acts on the other. For example, a promoter and a nucleotide sequence, such as a coding sequence, are operably linked when the expression of the nucleotide sequence is controlled by said promoter.

[0047] As used herein, the term “expression cassette” refers to a nucleic acid construct comprising a coding sequence and one or more control sequences required for expression of said coding sequence. In particular, one of these control sequence is a IRBP / GNAT2 promoter as defined below. Generally, the expression cassette comprises a coding sequence and regulatory sequences preceding (5' non-coding sequences) and following (3' non-coding sequences) the coding sequence that are required or improve the expression of the selected gene product. Thus, an expression cassette typically comprises a promoter sequence, a coding sequence and a 3' untranslated region that usually contains a polyadenylation site and / or transcription terminator. The expression cassette may also comprise additional regulatory elements such as, for example, enhancer sequences, a polylinker sequence facilitating the insertion of a DNA fragment within a vector and / or splicing signal sequences. The expression cassette is usually included within a vector, to facilitate cloning and transformation.

[0048] The term “ortholog” or “orthologous gene” as used herein refers a functional counterpart (i.e. retaining the same function) of a gene in another species. Orthologous genes are similar to each other because they originated from a common ancestor. Sequence differences between the orthologs are thus the result of speciation. Methods for identification of orthologous gene are well known in the art.

[0049] The terms "polypeptide” and "protein” are used interchangeably to refer to a polymer of amino acid residues, and are not limited to a minimum length. Such polymers of amino acid residues may contain natural or non-natural amino acid residues, and include, but are not limited to, peptides, oligopeptides, dimers, trimers, and multimers of amino acid residues. Both full-length proteins and fragments thereof are encompassed by the definition. The terms also include post-expression modifications of the polypeptide, for example, glycosylation, sialylation, acetylation, phosphorylation, and the like. Furthermore, for purposes of the present invention, a "polypeptide" may refer to a protein which includes modifications, such as deletions, additions, and substitutions (generally conservative in nature), to the native sequence, as long as the protein maintains the desired activity. These modifications may be deliberate, as through site-directed mutagenesis, or may be accidental, such as through mutations of hosts which produce the proteins or errors due to PCR amplification.

[0050] The term “% sequence identity” or “% identity”, as used herein, refers to the percentage of identical nucleotides in an alignment of two polynucleotides. This percentage can be easily calculated by the skilled person using a sequence comparison computer program. Sequence identity is determined by comparing the sequences after they have been aligned by maximizing overlaps and identities while minimizing spaces. In particular, sequence identity can be determined using one of the global or local mathematical alignment algorithms available. Sequences of similar size are preferably aligned using a global alignment algorithm (e.g. Needleman and Wunsch, 1970) that optimally aligns sequences along their entire length. Sequences of substantially different lengths are preferably aligned using a local alignment algorithm (e.g. Smith and Waterman, 1981; Altschul et al., 1997; Altschul et al., 2005). Alignments to determine percent identity can be performed by any method known to the skilled person, in particular using computer programs available on the internet (for example, http: / / blast.ncbi.nlm.nih.gov / or http: / / www.ebi.ac.uk / Tools / emboss / ). Preferably, the value of the percent identity refers here to a value obtained with the EMBOSS Needle program which uses the Needleman-Wunsch global alignment algorithm with the default parameters, namely: Scoring matrix = BLOSUM62, Gap open = 10, Gap extend = 0.5, End gap penalty = false, End gap open = 10 and End gap extend = 0.5.

[0051] As used herein, the term “variant" refers to a nucleotide or protein sequence that differs from the parent sequence but that retains its essential properties. The sequence of the variant may differ from the parent sequence by one or more substitutions, deletions and / or insertions. The variant may have the same length of the parent sequence, or may be shorter or longer.

[0052] The term "subject" or “patient” refers to an animal having retina, preferably to a mammal, even more preferably to a human, including adult, child and human at the prenatal stage.

[0053] IRBP / GNAT2 promoter

[0054] Amacrine cells represent the most diverse class of retinal neuron, comprising dozens of distinct cell types. Each type exhibits a unique morphology and generates specific visual computations through its synapses with a subset of excitatory interneurons (bipolar cells), other amacrine cells, and output neurons (ganglion cells).

[0055] The All (or A2) amacrine cells are arrow-field, bistratified, axon-less interneurons with a dendritic morphology. These cells connect rod and cone photoreceptor pathways, forming an essential link in the circuit for rod-mediated (scotopic) vision. All amacrine cells connect with both ON and OFF bipolar cells with different types of synapses (Famiglietti & Kolb 1975, Vaney 1985, Kolb et al. 1992, Wassle et al. 1995, Xin &Bloomfield 1999, Guadagni et al. 2019). They form gap junctions with most ON bipolar cell types (Tsukamoto & Omi 2017), and can therefore excite them when they are activated. At the same time, they form glycinergic inhibitory synapses with most OFF bipolar cell types (Kolb &Nelson 1983, Strettoi et al. 1992, Tsukamoto & Omi 2017). The All amacrine cells have been found in all mammalian retinas and, to the extent investigated, the details of their cellular and network properties are remarkably conserved among different species (reviewed by Wassle and Boycott, 1991). As demonstrated in the experimental section, the inventors identified a promoter that can drive expression in All amacrine cells. This promoter is able to target All amacrine cells and to provide high level expression of a transgene in these cells.

[0056] The nucleic acid having a promoter activity in All amacrine cells used in the present invention, termed “IRBP / GNAT2 promoter” in the following, comprises an enhancer element of an inter-photoreceptor retinoid-binding protein (IRBP) promoter that is fused to a promoter nucleotide sequence of a cone transducin alpha-subunit (GNAT2) gene.

[0057] Interphotoreceptor retinoid-binding protein (IRBP), also known as retinol binding protein 3 (RBP3), is a large glycoprotein known to bind retinoids and found primarily in the interphotoreceptor matrix of the retina between the retinal pigment epithelium and the photoreceptor cells. It is thought to transport retinoids between the retinal pigment epithelium and the photoreceptors, a critical role in the visual process. The chromosomal location of the human IRBP gene is lOql 1.22 and the NCBI Gene ID is 5949.

[0058] In the IRBP / GNAT2 promoter, the enhancer element of an IRBP promoter, termed “IRBP enhancer” in the following, may be (i) an enhancer element of the human IRBP gene or a functional variant thereof or (ii) an enhancer element of an orthologous gene of the human IRBP gene, or a functional variant thereof. In particular, orthologous genes may be from any non-human mammal such as mouse, rat, dog and primate.

[0059] In preferred embodiments, the IRBP enhancer is an enhancer element of the human IRBP gene or a functional variant thereof.

[0060] In particular, the IRBP enhancer may comprise, or consist of, sequence -1650 to - 1350 of the human IRBP gene (SEQ ID NO: 6), or a functional variant thereof. More particularly, the IRBP enhancer may comprise, or consist of, sequence from -1650, -1640, -1630 or -1620 to -1350, -1360, -1370, -1380, -1390, -1400 or -1410 of the human IRBP gene, or a functional variant thereof. The functional variant may have a sequence comprising at least 150, at least 160, at least 170, at least 180, at least 185, at least 190 or at least 200, consecutive nucleotides of SEQ ID NO:6, or a sequence having at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:6. Preferably, the functional variant has a sequence comprising at least 180, at least 185, at least 190 or at least 200, consecutive nucleotides of SEQ ID NO:6, or a sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:6. In particular, the sequence of the variant may differ from the sequence SEQ ID NO: 6 by a maximum of 15, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 or 15, substitutions, insertions and / or deletions, preferably by a maximum of 10, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10, substitutions, insertions and / or deletions, more preferably by a maximum of 5, i.e. 1, 2, 3, 4 or 5, substitutions, insertions and / or deletions.

[0061] Preferably, the IRBP enhancer may comprise, or consist of, sequence -1650 to - 1411 of the human IRBP gene (SEQ ID NO: 7), or a functional variant thereof. The functional variant may have a sequence comprising at least 150, at least 160, at least 170, at least 180, at least 185, at least 190 or at least 200, consecutive nucleotides of SEQ ID NO:7, or a sequence having at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:7. Preferably, the functional variant has a sequence comprising at least 180, at least 185, at least 190 or at least 200, consecutive nucleotides of SEQ ID NO:7, or a sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:7. In particular, the sequence of the variant may differ from the sequence SEQ ID NO: 7 by a maximum of 15, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 or 15, substitutions, insertions and / or deletions, preferably by a maximum of 10, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10, substitutions, insertions and / or deletions, more preferably by a maximum of 5, i.e. 1, 2, 3, 4 or 5, substitutions, insertions and / or deletions.

[0062] More preferably, the IRBP enhancer comprises, or consists of, sequence -1619 to -1411 of the human IRBP gene (SEQ ID NO: 2), or a functional variant thereof. The functional variant may have a sequence comprising at least 150, at least 160, at least 170, at least 180, at least 185, at least 190 or at least 200, consecutive nucleotides of SEQ ID NO:2, or a sequence having at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:2. Preferably, the functional variant has a sequence comprising at least 180, at least 185, at least 190 or at least 200, consecutive nucleotides of SEQ ID NO:2, or a sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:2. In particular, the sequence of the variant may differ from the sequence SEQ ID NO: 2 by a maximum of 15, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 or 15, substitutions, insertions and / or deletions, preferably by a maximum of 10, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10, substitutions, insertions and / or deletions, more preferably by a maximum of 5, i.e. 1, 2, 3, 4 or 5, substitutions, insertions and / or deletions.

[0063] The term "functional variant of an enhancer element of an IRBP promoter” refers to a variant that retains the activity of the parent sequence, i.e. that retains the activity of the enhancer element of an IRBP promoter in the IRBP / GNAT2 promoter. The skilled person can easily assess this activity by replacing the sequence of the IRBP enhancer in SEQ ID NO: 1 (see Figure 9, SEQ ID NO: 2) by the sequence of the variant and repeating the experiments described in the experimental section. Briefly, AAV2-7m8 vectors comprising an expression cassette comprising the IRBP / GNAT2 promoter (including the variant) operably linked to a reporter gene, e.g. a GFP gene, are delivered intravitreally to 4- week-old C57BL / 6J wild-type mice. 120-140 days after injection, retinas are harvested, fixed, and embedded in tissue freezing medium for analysis. The presence of the reporter signal in All amacrine cells indicates that the promoter exhibits a promoter activity in All amacrine cells and therefore that the variant is a functional variant. Preferably, the promoter activity is specific of All amacrine cells.

[0064] Transducin is a 3-subunit guanine nucleotide-binding protein (G protein) which stimulates the coupling of rhodopsin and cGMP-phoshodiesterase during visual impulses. The transducin alpha subunits in rods and cones are encoded by separate genes. The GNAT2 gene encodes the alpha subunit in cones. The chromosomal location of the human GNAT2 gene is lp 13.3 and the NCBI Gene ID is 2780. In the IRBP / GNAT2 promoter, the promoter nucleotide sequence of a cone transducin alpha-subunit (GNAT2) gene, termed GNAT2 promoter in the following may be (i) a promoter nucleotide sequence of the human GNAT2 gene or a functional variant thereof or (ii) a promoter nucleotide sequence of an orthologous gene of the human GNAT2 gene, or a functional variant thereof.

[0065] In preferred embodiments, the GNAT2 promoter is a promoter nucleotide sequence of the human GNAT2 gene or a functional variant thereof.

[0066] In particular, the GNAT2 promoter may comprise, or consist of, sequence -200 to +200 of the human GNAT2 gene (SEQ ID NO: 8), or a functional variant thereof. More particularly, the GNAT2 promoter may comprise, or consist of, sequence from -200, - 190, -180, -170 or -160 to +130, +140, +150, +160, +170, +180, +190 or +200 of the human IRBP gene, or a functional variant thereof. The functional variant may have a sequence comprising at least 200, at least 210, at least 220, at least 230, at least 240, at least 250 or at least 260, consecutive nucleotides of SEQ ID NO:8, or a sequence having at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:8. Preferably, the functional variant has a sequence comprising at least 230, at least 240, at least 250 or at least 260, consecutive nucleotides of SEQ ID NO:8, or a sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:8. In particular, the sequence of the variant may differ from the sequence SEQ ID NO: 8 by a maximum of 15, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 or 15, substitutions, insertions and / or deletions, preferably by a maximum of 10, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10, substitutions, insertions and / or deletions, more preferably by a maximum of 5, i.e. 1, 2, 3, 4 or 5, substitutions, insertions and / or deletions.

[0067] Preferably, the GNAT2 promoter may comprise, or consist of, sequence -151 to +200 of the human GNAT2 gene (SEQ ID NO: 9), or a functional variant thereof. The functional variant may have a sequence comprising at least 200, at least 210, at least 220, at least 230, at least 240, at least 250 or at least 260, consecutive nucleotides of SEQ ID NO:9 or a sequence having at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:9. Preferably, the functional variant has a sequence comprising at least 230, at least 240, at least 250 or at least 260, consecutive nucleotides of SEQ ID NO:9, or a sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:9. In particular, the sequence of the variant may differ from the sequence SEQ ID NO: 9 by a maximum of 15, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 or 15, substitutions, insertions and / or deletions, preferably by a maximum of 10, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10, substitutions, insertions and / or deletions, more preferably by a maximum of 5, i.e. 1, 2, 3, 4 or 5, substitutions, insertions and / or deletions.

[0068] More preferably, the GNAT2 promoter may comprise, or consist of, sequence - 151 to +126 of the human GNAT2 gene (SEQ ID NOG), or a functional variant thereof. The functional variant may have a sequence comprising at least 200, at least 210, at least 220, at least 230, at least 240, at least 250 or at least 260, consecutive nucleotides of SEQ ID NOG, or a sequence having at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NOG. Preferably, the functional variant has a sequence comprising at least 230, at least 240, at least 250 or at least 260, consecutive nucleotides of SEQ ID NOG, or a sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NOG. In particular, the sequence of the variant may differ from the sequence SEQ ID NO: 3 by a maximum of 15, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 or 15, substitutions, insertions and / or deletions, preferably by a maximum of 10, i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10, substitutions, insertions and / or deletions, more preferably by a maximum of 5, i.e. 1, 2, 3, 4 or 5, substitutions, insertions and / or deletions.

[0069] The term "functional variant of a promoter nucleotide sequence of a GNAT2 gene” refers to a variant that retains the activity of the parent sequence, i.e. that retains the activity of the promoter nucleotide sequence of a GNAT2 gene in the IRBP / GNAT2 promoter. The skilled person can easily assess this activity by replacing the sequence of the GNAT2 promoter in SEQ ID NO:1 (see Figure 9, SEQ ID NO: 3) by the sequence of the variant and repeating the experiments described in the experimental section. Briefly, AAV2-7m8 vectors comprising an expression cassette comprising the IRBP / GNAT2 promoter (including the variant) operably linked to a reporter gene, e.g. a GFP gene, are delivered intravitreally to 4- week-old C57BL / 6J wild-type mice. 120-140 days after injection, retinas are harvested, fixed, and embedded in tissue freezing medium for analysis. The presence of the reporter signal in All amacrine cells indicates that the promoter exhibits a promoter activity in All amacrine cells and therefore that the variant is a functional variant. Preferably, the promoter activity is specific of All amacrine cells.

[0070] The IRBP / GNAT2 promoter used in the present invention comprises an IRBP enhancer as defined above that is fused to a GNAT2 promoter as defined above. The IRBP enhancer may be positioned upstream or downstream of the GNAT2 promoter. Preferably, the IRBP enhancer is preferably positioned upstream of the GNAT2 promoter.

[0071] As used herein, the term “upstream” refers to a relative position which is toward the 5' end of the coding strand for the gene operably linked to the promoter of the invention and the term “downstream” refers to a relative position which is toward the 3' end of the coding strand for said gene.

[0072] The IRBP enhancer and the GNAT2 promoter may be directly adjacent or may be separated by an intervening sequence. When present, the intervening sequence may have a length of 1 to 200 nucleotides, preferably 5 to 50, and more preferably 10 to 20 nucleotides.

[0073] In a particular embodiment, the IRBP / GNAT2 promoter comprises an intervening sequence between the IRBP enhancer and the GNAT2 promoter, said intervening sequence comprising, or consisting of, ACTAGTGTTGCTAGC (SEQ ID NO: 5), or a variant thereof that differs from the sequence of SEQ ID NO: 5 by a maximum of 5, i.e. 1, 2, 3, 4, 5 substitutions, insertions and / or deletions.

[0074] In a more particular embodiment, the IRBP / GNAT2 promoter comprises, or consists of, the sequence of SEQ ID NO: 4 or a functional variant thereof exhibiting a promoter activity in All amacrine cells, preferably a promoter activity which is specific of All amacrine cells. Said variants may have a sequence having at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:4, preferably at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO:4. Preferably, the IRBP / GNAT2 promoter comprises, or consists of, the sequence of SEQ ID NO: 4.

[0075] The IRBP / GNAT2 promoter exhibits a promoter activity in All amacrine cells, i.e. when introduced in All amacrine cells, it can initiate transcription of a nucleic acid to which it is operably linked. Preferably, the promoter activity is specific of All amacrine cells. The term “specific of All amacrine cells” shall be understood to mean a promoter mainly active in All amacrine cells. It shall be understood that a residual expression, generally lower, in other tissues or cells cannot be entirely excluded.

[0076] Expression cassettes, vectors and viral particles

[0077] The present invention relates to an expression cassette comprising a IRBP / GNAT2 promoter as defined above operably linked to a coding sequence of interest. All the embodiments of the IRBP / GNAT2 promoter as defined above are also contemplated in this aspect.

[0078] The coding sequence of interest may encode a polypeptide of interest or a nucleic acid of interest.

[0079] In some embodiments, the coding sequence of interest encodes a polypeptide of interest. The polypeptide of interest may be any polypeptide of which expression in All amacrine cells is desired. In particular, the polypeptide of interest may be a therapeutic polypeptide, reporter protein or optogenetic protein.

[0080] In an embodiment, the coding sequence of interest is a therapeutic gene, i.e. encodes a therapeutic polypeptide.

[0081] As used herein, the term "therapeutic gene" refers to a gene encoding a therapeutic protein which is useful in the treatment of a pathological condition. The therapeutic gene, when expressed, confers a beneficial effect on the cell or tissue in which it is present, or on a patient in which the gene is expressed. Examples of beneficial effects include amelioration of a sign or symptom of a condition or disease, prevention or inhibition of a condition or disease, or conferral of a desired characteristic. Therapeutic genes include genes that partially or wholly correct a genetic deficiency in the patient. In particular, the therapeutic gene may be, without limitation, a nucleic acid sequence encoding a protein useful in gene therapy to relieve deficiencies caused by missing, defective or sub-optimal levels of said protein in a cell or tissue of a subject. The therapeutic polypeptide may, e.g., supply a polypeptide and / or enzymatic activity that is absent, defective or present at a sub-optimal level in All amacrine cells, supply a polypeptide and / or enzymatic activity that indirectly counteracts an imbalance in All amacrine cells. The therapeutic polypeptide may also be used to reduce the activity of a polypeptide by acting, e.g., as a dominant-negative polypeptide. Preferably, the therapeutic polypeptide supplies a polypeptide and / or enzymatic activity that is absent, defective or present at a sub-optimal level in All amacrine cells, more preferably a polypeptide and / or enzymatic activity that is absent or defective in All amacrine cells.

[0082] Additional signal peptide may be added to therapeutic proteins, in particular in order to import them inside certain organelles (such as mitochondria), to secrete them from the cell, or to insert them into cellular membrane.

[0083] In preferred embodiments, the polypeptide of interest is an optogenetic protein.

[0084] As used herein, the term "optogenetic protein " refers to a photochemically reactive polypeptide that uses vitamin A or isoforms thereof as its chromophore. An optogenetic protein is a light-gated ion pump or channel that absorbs light and is activated by light or a G protein coupled vertebrate opsin. The optogenetic protein may be from a prokaryotic organism or a eukaryotic organism. In particular, it may be a microbial opsin or a vertebrate opsin. The optogenetic protein may be an optogenetic activator (depolarizing optogenetic protein) or an optogenetic inhibitor (hyperpolarizing optogenetic protein).

[0085] An optogenetic activator causes a cell to depolarize upon exposure to light. When a cell depolarizes, the negative internal charge of the cell becomes positive for a brief period. The shift from a negative to a positive internal cellular environment allows the transmission of electrical impulses both within a cell and, optionally, between cells. Examples of optogenetic activators include, but are not limited to, rhodopsins, photopsins, melanopsins, pinopsins, parapinopsins, VA opsins, peropsins, neuropsins, encephalopsins, retinochromes, RGR opsins, microbial opsins, in particular microbial opsins with red-shifted spectral properties such as ReaChR, Chrimson or ChrimsonR, vertebrate opsins that can recruit Gi / Osignalling such as short wavelength vertebrate opsin or long wavelength vertebrate opsin, channelrhodopsins such as Channelrhodopsin-1 (ChRl ), Channelrhodopsin-2 (ChR2), and Volvox Channelrhodopsin (VChRl), and variants thereof. Numerous variants (e.g. codon optimized variants, mutants, chimeras) of channelrhodopsin are being generated to improve certain features of these proteins. Examples of these variants include, but are not limited to, hChR2 (L132C), ChR2 (H134R), ChETA (E123T), C1V1 (E122T), C1V1 (E162T), C1V1 (E122 / 162T), hChR2 hCatch (T159S), hChief, hChR2 (C128S / D156A), hChR2 (T159C), hChR2 (E123T / T159C), hChR2c (C128T), ChR2c (C128T), ChR2e (Q117C) and SwitChR (for review see Prakash et al. Nat Methods. 2012 Dec;9(12): 1171-9).

[0086] In a particular embodiment, the optogenetic protein is an optogenetic activator, preferably selected from the group consisting of microbial opsins, in particular with red- shifted spectral properties such as ReaChR, Chrimson or ChrimsonR, preferably ReaChR.

[0087] An optogenetic inhibitor causes a cell to hyperpolarize upon exposure to light. When a cell hyperpolarizes, the negative internal charge of the cell becomes more negative for a brief period. The shift to more negative inhibits action potentials by increasing the stimulus required to move the membrane potential to the action potential threshold. In particular, an optogenetic inhibitor may be a light-gated ion pump that upon absorption of a photon transports chloride ions inward and / or transports cations outward. Any suitable light-gated, retinal-dependent, ion pump that transports chloride ions inward or cations outward upon absorption of a photon may be used as an optogenetic inhibitor. Examples of optogenetic inhibitors include, but are not limited to, halorhodopsins such as halorhodopsin (NpHR), enhanced halorhodopsins (eNpHR2.0 and eNpHR3.0) and the red-shifted halorhodopsin Halo57, archaerhodopsin-3 (AR-3), archaerhodopsin (Arch), bacteriorhodopsins such as enhanced bacteriorhodopsin (eBR), proteorhodopsins, xanthorhodopsins, Leptosphaeria maculans fungal opsins (Mac), the cruxhalorhodopsin Jaws, GtACRl and 2 (Govorunova et al. “Science 349.6248 (2015), pp. 647-650), and variants thereof. In a particular embodiment, the optogenetic protein is an optogenetic inhibitor, preferably selected from the group consisting of the cruxhalorhodopsin Jaws and GtACRl and 2, and variants thereof. .

[0088] In a further embodiment, the coding sequence of interest encodes a reporter protein. Preferably, the reporter protein is detectable in living All amacrine cells. The expression of a reporter protein under the control of the IRBP / GNAT2 promoter as defined above allows specifically detecting or identifying All amacrine cells. The reporter protein may be, for example, a fluorescent protein (e.g., GFP), calcium indicator (e.g. GCamP), luciferase, alkaline phosphatase, beta-galactosidase, beta-lactamase, horseradish peroxidase, and variants thereof. In a particular embodiment, the reporter protein is selected from the group consisting of fluorescent proteins, calcium indicators, alkaline phosphatases, beta-galactosidases, beta-lactamases, horseradish peroxidase, and variants thereof.

[0089] In some other embodiments, the coding sequence of interest encodes a nucleic acid of interest.

[0090] The nucleic acid of interest may be any nucleic acid of which expression in All amacrine cells is desired. In particular, the nucleic acid of interest may be a therapeutic nucleic acid.

[0091] The nucleic acid may be, for example, an siRNA, an shRNA an RNAi, an miRNA, an antisense RNA, a ribozyme or a DNAzyme.

[0092] In a particular embodiment, the nucleic acid encodes an RNA that when transcribed from the coding sequence of interest can treat or prevent an ocular disease by interfering with translation or transcription of an abnormal or excess protein associated with said disorder. For example, the nucleic acid of interest may encode for an RNA, which treats the disease by highly specific elimination or reduction of mRNA encoding the abnormal and / or excess proteins.

[0093] The expression cassette of the invention may comprise one or more coding sequences of interest operably linked to the IRBP / GNAT2 promoter. For example, the promoter may be operably linked to a sequences of interest coding an optogenetic protein and a sequences of interest coding a reporter protein. Preferably, at least one of the coding sequences of interest operably linked to the IRBP / GNAT2 promoter is a sequence coding an optogenetic protein.

[0094] Preferably, in the expression cassette of the invention, the IRBP / GNAT2 promoter as defined above is not operably linked to a sequence encoding a cone cyclic nucleotide- gated channel (CNG) protein, such as CNGB3 and CNGA3, and / or to a sequence encoding a reporter protein and / or to a sequence encoding a guide RNA that targets a GUCY2D gene (the gene encoding retinal guanylate cyclase-1) and / or to a sequence encoding a RNA-guided endonuclease.

[0095] The expression cassette may further comprise regulatory sequences preceding (5' non-coding sequences) and following (3' non-coding sequences) the coding sequence that are required or improve the expression of the selected gene product. Thus, in addition to the IRBP / GNAT2 promoter and at least one coding sequence of interest, the expression cassette may further comprise splicing signal sequences and / or a 3' untranslated region that usually contains a polyadenylation site and / or transcription terminator.

[0096] In particular, the expression cassette of the invention may comprise a splice donor / splice acceptor element, preferably comprised in an intronic sequence known to increase splicing efficiency and expression of the coding sequence. Such intronic sequences are well known and easily chosen by the skilled person. Preferably, the intronic sequence comprises, or consists of, SV40 intron of the sequence of SEQ ID NO: 10, or a functional variant thereof having at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO: 10, preferably at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99%, identity to SEQ ID NO: 10. More preferably, the intronic sequence is SV40 intron of the sequence of SEQ ID NO: 10.

[0097] The splice donor / splice acceptor element or intronic sequence can be positioned downstream or upstream of the coding sequence of interest. Preferably, splice donor / splice acceptor element or intronic sequence is position downstream of the promoter and upstream of the coding sequence of interest. The present invention also relates to a vector comprising the expression cassette of the invention.

[0098] As used herein, the term "vector" refers to a nucleic acid molecule used as a vehicle to transfer genetic material, and in particular to deliver a nucleic acid into a host cell, either in vitro or in vivo. Vectors include, but are not limited to, plasmids, phasmids, cosmids, transposable elements, viruses, and artificial chromosomes (e.g., YACs).

[0099] Preferably, the vector of the invention is a vector suitable for use in gene or cell therapy, and in particular is suitable to target All amacrine cells.

[0100] The vector of the invention is preferably a viral genome vector including any element required to establish the expression of the coding sequence of interest in a host cell such as, for example, a promoter, e.g., the IRBP / GNAT2 promoter as defined above, an ITR, a ribosome binding element, terminator, enhancer, selection marker, intron, polyA signal, and / or origin of replication.

[0101] In some embodiments, the vector is a viral vector, such as vectors derived from Moloney murine leukemia virus vectors (MoMLV), MSCV, SFFV, MPSV or SNV, lentiviral vectors (e.g. derived from human immunodeficiency virus (HIV), simian immunodeficiency virus (SIV), feline immunodeficiency virus (FIV), bovine immunodeficiency virus (BIV) or equine infectious anemia virus (EIAV)), adenoviral (Ad) vectors, adeno-associated viral (AAV) vectors, simian virus 40 (SV-40) vectors, bovine papilloma virus vectors, Epstein-Barr virus, herpes virus vectors, vaccinia virus vectors, Harvey murine sarcoma virus vectors, murine mammary tumor virus vectors, Rous sarcoma virus vectors.

[0102] In particular embodiments, the vector is a retroviral vector, preferably a lentiviral vector or a non-pathogenic parvovirus.

[0103] As is known in the art, depending on the specific viral vector considered for use, suitable sequences should be introduced in the vector of the invention for obtaining a functional viral vector, such as AAV ITRs for an AAV vector, or LTRs for lentiviral vectors.

[0104] In preferred embodiments, the vector is an AAV vector. As used herein, the term "AAV vector" refers to a polynucleotide vector comprising one or more heterologous sequences (i.e., nucleic acid sequence not of AAV origin) that are flanked by at least one AAV inverted terminal repeat sequence (ITR), preferably two ITRs. Such AAV vectors can be replicated and packaged into infectious viral particles when present in a host cell that has been infected with a suitable helper virus (or that is expressing suitable helper functions) and that is expressing AAV rep and cap gene products (i.e. AAV Rep and Cap proteins).

[0105] An "inverted terminal repeat" or "ITR" sequence is a term well understood in the art and refers to relatively short sequences found at the termini of viral genomes which are in opposite orientation. An "AAV inverted terminal repeat (ITR)" sequence is an approximately 145-nucleotide sequence that is present at both termini of the native singlestranded AAV genome. The outermost 125 nucleotides of the ITR can be present in either of two alternative orientations, leading to heterogeneity between different AAV genomes and between the two ends of a single AAV genome. The outermost 125 nucleotides also contains several shorter regions of self-complementarity (designated A, A', B, B', C, C and D regions), allowing intra-strand base-pairing to occur within this portion of the ITR. AAV ITRs for use in the vectors of the invention may have a wild-type nucleotide sequence or may be altered by the insertion, deletion or substitution. The serotype of the inverted terminal repeats (ITRs) of the AAV vector may be selected from any known human or nonhuman AAV serotype.

[0106] When an AAV vector is incorporated into a larger polynucleotide (e.g., in a chromosome or in another vector such as a plasmid used for cloning or transfection), then the AAV vector may be referred to as a "pro-vector" which can be "rescued" by replication and encapsidation in the presence of AAV packaging functions and suitable helper functions. The AAV vector of the invention can be in any of a number of forms, including, but not limited to, plasmids, linear artificial chromosomes, complexed with lipids, encapsulated within liposomes, and encapsidated in a viral particle, e.g., an AAV particle.

[0107] The expression cassette of the invention may be introduced into the vector by any method known by the skilled person. The vector may further comprise one or more nucleic acid sequences encoding selectable marker such as auxotrophic markers (e.g., LEU2, URA3, TRP 1 or HIS3), detectable labels such as fluorescent or luminescent proteins (e.g., GFP, eGFP, DsRed, CFP), or protein conferring resistance to a chemical / toxic compound (e.g., MGMT gene conferring resistance to temozolomide). These markers can be used to select or detect host cells comprising the vector and can be easily chosen by the skilled person according to the host cell.

[0108] All the embodiments of the IRBP / GNAT2 promoter as defined above and the expression cassette of the invention are also contemplated in this aspect.

[0109] The vector of the invention may be packaged into a virus capsid to generate a "viral particle". Thus, in a further aspect, the present invention also relates to a viral particle comprising a vector of the invention.

[0110] In a particular embodiment, the vector is an AAV vector and is packaged into an AAV-derived capsid to generate an "adeno-associated viral particle" or "AAV particle". Thus, used herein, the term "AAV particle" refers to a viral particle composed of at least one AAV capsid protein and an encapsidated AAV vector genome.

[0111] The capsid serotype determines the tropism range of the AAV particle.

[0112] Multiple serotypes of adeno-associated virus (AAV), including 12 human serotypes and more than 100 serotypes from nonhuman primates have now been identified (Howarth al., 2010, Cell Biol Toxicol 26: 1-10). Among these serotypes, human serotype 2 was the first AAV developed as a gene transfer vector. Other currently used AAV serotypes include, but are not limited to, AAV1, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAVrh8, AAV9, AAV10, AAVrhlO, AAV11, AAV12, AAVrh74 and AAVdj, etc.. In addition, non-natural engineered variants and chimeric AAV can also be useful. In particular, the capsid proteins may be variants comprising one or more amino acid substitutions enhancing transduction efficiency.

[0113] Different AAV serotypes are used to optimize transduction of particular target cells or to target specific cell types within a particular target tissue (e.g., All amacrine cells). An AAV particle can comprise viral proteins and viral nucleic acids of the same serotype or any natural or artificial sequence variant of AAV. For example, the AAV particle may comprise AAV2 capsid proteins and at least one, preferably two, AAV2 ITR. Any combination of AAV serotypes for production of an AAV particle is provided herein as if each combination had been expressly stated herein.

[0114] In a particular embodiment, the AAV particle comprises an AAV-derived capsid selected from the group consisting of AAV2, AAV-5, AAV9 and AAV8 capsid, and variants thereof.

[0115] Preferably, the AAV particle comprises an AAV-derived capsid selected from the group consisting of AAV2 and variants thereof. An example of AAV2 variants is AAV2- 7m8 (Dalkara et al. Sci Transl Med (2013), 5, 189ra76).

[0116] In preferred embodiments, the AAV particle comprises an AAV-derived capsid selected from the group consisting of AAV2 and AAV2-7m8, preferably AAV2-7m8.

[0117] AAV viruses may be engineered using conventional molecular biology techniques, making it possible to optimize these particles for cell specific delivery of nucleic acid sequences, for minimizing immunogenicity, for tuning stability and particle lifetime, for efficient degradation, for accurate delivery to the nucleus.

[0118] Alternatively to using AAV natural serotypes, artificial AAV serotypes may be used in the context of the present invention, including, without limitation, AAV with a non-naturally occurring capsid protein. Such an artificial capsid may be generated by any suitable technique, using a selected AAV sequence (e.g., a fragment of a VP1 capsid protein) in combination with heterologous sequences which may be obtained from a different selected AAV serotype, non-contiguous portions of the same AAV serotype, from a non-AAV viral source, or from a non-viral source. An artificial AAV serotype may be, without limitation, a chimeric AAV capsid or a mutated AAV capsid.

[0119] A chimeric capsid comprises VP capsid proteins derived from at least two different AAV serotypes or comprises at least one chimeric VP protein combining VP protein regions or domains derived from at least two AAV serotypes.

[0120] Capsid proteins may also be mutated, in particular to enhance transduction efficiency. Mutated AAV capsids may be obtained from capsid modifications inserted by error prone PCR and / or peptide insertion or by including one or several amino acids substitutions. In particular, mutations may be made in any one or more of tyrosine residues of natural or non-natural capsid proteins (e.g. VP1, VP2, or VP3). Preferably, mutated residues are surface exposed tyrosine residues. Exemplary mutations include, but are not limited to tyrosine-to-phenylalanine substitutions such as Y252F, Y272F, Y444F, Y500F, Y700F, Y704F, Y730F, Y275F, Y281F, Y508F, Y576F, Y612G, Y673F and Y720F as well as other mutations having an impact on Heparan Sulfate Binding by AAV (see Boye et al., J Virol. 2016 Mar 28;90(8):4215-4231. ).

[0121] In addition, the genome vector (i.e. a vector of the invention) of the AAV particle may either be a single stranded or self- complementary double-stranded genome. Self- complementary double- stranded AAV vectors are generated by deleting the terminal resolution site (trs) from one of the AAV terminal repeats. These modified vectors, whose replicating genome is half the length of the wild type AAV genome have the tendency to package DNA dimers. In preferred embodiments, the AAV particle implemented in the practice of the present invention has a single stranded genome.

[0122] Numerous methods are known in the art for production of viral particles, and in particular AAV particles, including transfection, stable cell line production, and infectious hybrid virus production systems which include adenovirus-AAV hybrids, herpesvirus- AAV hybrids and baculovirus-AAV hybrids.

[0123] AAV production cultures for the production of AAV virus particles all require; 1) suitable host cells, including, for example, human-derived cell lines such as HeEa, A549, or 293 cells, or insect-derived cell lines such as SF-9, in the case of baculovirus production systems; 2) suitable helper virus function, provided by wild- type or mutant adenovirus (such as temperature sensitive adenovirus), herpes virus, baculovirus, or a plasmid construct providing helper functions; 3) AAV rep and cap genes and gene products; 4) a nucleic acid of interest flanked by at least one AAV ITR sequences, e.g., a vector of the invention; and 5) suitable media and media components to support AAV production that are well-known in the art.

[0124] In practicing the invention, host cells for producing AAV particles include mammalian cells, insect cells, plant cells, microorganisms and yeast. Host cells can also be packaging cells in which the AAV rep and cap genes are stably maintained in the host cell or producer cells in which the AAV vector genome is stably maintained. Exemplary packaging and producer cells are derived from 293, A549 or HeEa cells. AAV particles are then purified and formulated using standard techniques known in the art. All the embodiments of the IRBP / GNAT2 promoter as defined above and the expression cassette and vector of the invention are also contemplated in this aspect.

[0125] Host cell

[0126] In another aspect, the present invention also relates to an isolated host cell transformed or transfected with an expression cassette, vector or viral particle of the invention.

[0127] The host cell may be any animal cell, plant cell, bacterium cell or yeast. Preferably, the host cell is a mammalian cell or an insect cell. More preferably, the host cell is a human cell.

[0128] In preferred embodiments, the host cell is an All amacrine cell, in particular a human All amacrine cell.

[0129] The expression cassette or vector of the invention may be transferred into host cells using any known technique including, but being not limited to, calcium phosphate- DNA precipitation, DEAE-Dextran transfection, electroporation, microinjection, biolistic, lipofection, or viral infection, and may be maintained in the host cell in an ectopic form or may be integrated into the genome.

[0130] In preferred embodiments, the expression cassette or vector of the invention is transferred into the host cell by viral infection, preferably using a viral particle of the invention, more preferably using an AAV particle of the invention.

[0131] All the embodiments of the IRBP / GNAT2 promoter as defined above and the expression cassette, vector and viral particle of the invention are also contemplated in this aspect.

[0132] Pharmaceutical composition

[0133] The present invention also relates to a pharmaceutical composition comprising an expression cassette, vector, viral particle or cell of the invention.

[0134] Such compositions comprise a therapeutically effective amount of the therapeutic agent (an expression cassette, vector, viral particle or cell of the invention), and a pharmaceutically acceptable excipient. As used herein, the term "pharmaceutically acceptable" means approved by a regulatory agency or recognized pharmacopeia such as European Pharmacopeia, for use in animals and / or humans. The term "excipient" refers to a diluent, adjuvant, carrier, or vehicle with which the therapeutic agent is administered.

[0135] As is well known in the art, pharmaceutically acceptable excipients are relatively inert substances that facilitate administration of a pharmacologically effective substance and can be supplied as liquid solutions or suspensions, as emulsions, or as solid forms suitable for dissolution or suspension in liquid prior to use. For example, an excipient can give form or consistency, or act as a diluent. Suitable excipients include but are not limited to stabilizing agents, wetting and emulsifying agents, salts for varying osmolality, encapsulating agents, pH buffering substances, and buffers. Such excipients include any pharmaceutical agent suitable for direct delivery to the eye which may be administered without undue toxicity. Pharmaceutically acceptable excipients include, but are not limited to, sorbitol, any of the various tween compounds, and liquids such as water, saline, glycerol and ethanol. Pharmaceutically acceptable salts can be included therein, for example, mineral acid salts such as hydrochlorides, hydrobromides, phosphates, sulfates, and the like; and the salts of organic acids such as acetates, propionates, malonates, benzoates, and the like. A thorough discussion of pharmaceutically acceptable excipients is available in Remington's Pharmaceutical Sciences, 15th Edition.

[0136] Preferably, the composition is formulated to be administered to the eye, in particular by intraocular injection, e.g., by subretinal and / or intravitreal administration, preferably by intravitreal administration. Accordingly, the composition can be combined with pharmaceutically acceptable excipient such as saline, Ringer's balanced salt solution (pH 7.4), and the like.

[0137] The pharmaceutical compositions described herein can be packaged in single unit dosages or in multidosage forms.

[0138] In an embodiment, the pharmaceutical composition comprises a vector or viral particle of the invention, more preferably an AAV vector or particle.

[0139] In another embodiment, the pharmaceutical composition comprises host cells of the invention, preferably human host cell of the invention, i.e. transformed or transfected with an expression cassette, vector or viral particle of the invention, preferably with an AAV particle. Optionally, the composition comprising host cells may be frozen for storage at any temperature appropriate for storage of the cells. For example, the cells may be frozen at about -20°C, -80°C or any other appropriate temperature. Cryogenically frozen cells may be stored in appropriate containers and prepared for storage to reduce rick of cell damage and maximize the likelihood that the cells will survive thawing. Alternatively, the cells may also be maintained at room temperature of refrigerated, e.g. at about 4°C.

[0140] The amount of pharmaceutical composition to be administered may be determined by standard procedure well known by those of ordinary skill in the art. Physiological data of the patient (e.g. age, size, and weight) and type and severity of the disease being treated have to be taken into account to determine the appropriate dosage.

[0141] The pharmaceutical composition of the invention may be administered as a single dose or in multiple doses.

[0142] In a particular embodiment, the composition comprises viral particles of the invention and each unit dosage comprises from 108to 1013viral particles, preferably from 109to 1012particles.

[0143] The pharmaceutical composition may further comprise one or several additional active compounds such as corticosteroids, antibiotics, analgesics, immunosuppressants, trophic factors, or any combinations thereof.

[0144] All the embodiments of the IRBP / GNAT2 promoter as defined above and the expression cassette, vector, viral particle and host cell of the invention are also contemplated in this aspect.

[0145] Methods and uses

[0146] In a further aspect, the present invention also relates to a method for expressing a coding sequence of interest in an All amacrine cell, said method comprising introducing into said cell an expression cassette, vector or viral particle of the invention comprising the coding sequence of interest operably linked to a IRBP / GNAT2 promoter as defined above. The method may be an in vitro or in vivo method.

[0147] It also relates to the use of a IRBP / GNAT2 promoter as defined above for expressing a coding sequence of interest operably linked to said promoter in an All amacrine cell. The use may be an in vitro or in vivo use. All amacrine cells connect with both ON and OFF bipolar cells with different types of synapses. They form gap junctions with most ON bipolar cell types and can therefore excite them when they are activated. At the same time, they form glycinergic inhibitory synapses with most OFF bipolar cell types. Expressing optogenetic proteins in this cell type rendering them light sensitive and can be used for the treatment of retinal diseases associated with photoreceptor cell degeneration. Compared to ganglion cell targeting, which is currently in clinical trials, this approach has the advantage of restoring a significant part of the retinal computations performed by a normal retina. In particular, the inventors have shown that this strategy allows restoration of both ON and OFF ganglion cell responses and thus allows restoration of retinal function and visual perception in patients suffering from retinal degeneration.

[0148] Thus, the present invention also relates to

[0149] - an expression cassette, vector, viral particle, host cell or a pharmaceutical composition of the invention for use in the treatment of an ocular disease,

[0150] - the use of an expression cassette, vector, viral particle or host cell of the invention for the manufacture of a medicament for the treatment of an ocular disease, and

[0151] - a method for treating an ocular disease comprising administering a therapeutically efficient amount of an expression cassette, vector, viral particle, host cell or a pharmaceutical composition of the invention to a subject in need thereof.

[0152] Preferably, the ocular disease is a disease associated with photoreceptor cell degeneration.

[0153] In particular, the disease associated with photoreceptor cell degeneration may be selected from the group consisting of retinitis pigmentosa, age-related macular degeneration, cone-rod dystrophy, Leber congenital amaurosis, Stargardt’s disease, diabetic retinopathy, Best’s disease, choroideremia and a tapetoretinal degeneration.

[0154] Preferably, the disease associated with photoreceptor cell degeneration is retinitis pigmentosa.

[0155] In the aspects of the invention relating to therapeutic applications, the expression of the coding sequence of interest operably linked to the IRBP / GNAT2 promoter in All amacrine cells provides a treatment of the ocular disease. Preferably, the treatment includes restoration of a significant part of the retinal computations performed by a normal retina, in particular includes restoration of both ON and OFF ganglion cell responses.

[0156] In a preferred embodiment, the ocular disease is a disease associated with photoreceptor cell degeneration and the coding sequence of interest operably linked to the IRBP / GNAT2 promoter encodes a optogenetic protein.

[0157] As used herein, the term “treatment”, “treat” or “treating” refers to any act intended to ameliorate the health status of patients such as therapy, prevention, prophylaxis and retardation of the disease. In certain embodiments, such term refers to the amelioration or eradication of a disease or symptoms associated with a disease. In other embodiments, this term refers to minimizing the spread or worsening of the disease resulting from the administration of one or more therapeutic agents to a subject with such a disease.

[0158] In particular, the term “treatment of an ocular disease” may refer to a treatment to provide enhanced vision, to prevent progression of the disease to total blindness, to prevent spread of damage to uninjured ocular cells, to improve damage in injured ocular cells, to prevent the occurrence of retinal damage or to rescue eyes having mild or advanced disease. In some embodiments, this term refers to a treatment to prevent, reduce or stop retinal cell degeneration by providing a therapeutic protein correcting a genetic deficiency of the patient. In preferred embodiments, this term refers to a treatment to reanimate the retina, reactivate retinal computations, and in particular ON-OFF selectivity, and / or restore vision, using optogenetics.

[0159] By a “therapeutically efficient amount” is intended an amount of pharmaceutical composition of the invention administered to a subject that is sufficient to constitute a treatment as defined above of ocular disease.

[0160] The pharmaceutical composition may be administered as a single dose or in multiple doses.

[0161] In a particular embodiment, the pharmaceutical composition comprises viral particles of the invention and each unit dose comprises from 108to 1013viral particles, preferably from 109to 1012particles. In the method for treating ocular disease of the invention, the pharmaceutical composition of the invention is preferably administered intraocularly, more preferably by intravitreal administration.

[0162] The method of the invention may also further comprise administering at least one additional therapeutic agent to the subject. In particular, said therapeutic agent may be selected from the group consisting of a corticosteroid, an antibiotic, an analgesic, an immunosuppressant, or a trophic factor, or any combinations thereof.

[0163] The composition of the invention may be administered before or after the disease becomes symptomatic, e.g., before or after partial or complete photoreceptor cell degeneration and / or before or after partial or complete loss of vision.

[0164] All the embodiments of the IRBP / GNAT2 promoter as defined above and the expression cassette, vector, viral particle, host cell and pharmaceutical composition of the invention are also contemplated in this aspect.

[0165] All patents, patent applications, provisional applications, and publications referred to or cited herein are incorporated by reference in their entirety, including all figures and tables, to the extent they are not inconsistent with the explicit teachings of this specification.

[0166] Further aspects and advantages of the present invention will be described in the following examples, which should be regarded as illustrative and not limiting.

[0167] EXAMPLES

[0168] Material and methods

[0169] AAV Productions

[0170] Recombinant AAVs were produced by the plasmid cotransfection method, and the resulting lysates were purified via iodixanol gradient ultracentrifugation as previously described (Dalkara et al. 2013). Briefly, 40% iodixanol fraction was concentrated and buffer exchanged using Amicon Centrifugal Filter Units (Millipore, Molsheim, France). Vector stocks were then titrated for DNase-resistant vector genomes with ITR primers by real-time qPCR with SybrGreen (ThermoFischer Scientific) relative to a standard.

[0171] Animals And Intravitreal Injections

[0172] All experiments were done in accordance with Directive 2010 / 63 / EU of the European Parliament. The protocol was approved by the Local Animal Ethics Committee of Paris 5 (CEEA 34). All mice used in this study were C3H / HeN (rdl mice) or C57B16J mice (wild type) from Janvier Laboratories (Le Genest Saint Isle, France). For injections, mice were anesthetized with isoflurane (5% induction, 2% during the procedure). Pupils were dilated, and an ultrafine 30-gauge disposable needle was passed through the sclera, at the equator and next to the limbus, into the vitreous cavity. Injection of 1.5 pl stock containing 3e9 vg / eye of AAV-GFP or ReaChR-eYFP was made with direct observation of the needle in the center of the vitreous cavity.

[0173] Immunohistochemistry

[0174] Mice were sacrificed in accordance with all animal facility protocols at the Institut de la Vision by CO2 inhalation and cervical dislocation. Eyes were removed and fixed 2 hours in 4% paraformaldehyde solution at room temperature. Eyes for flatmounts were dissected and retinal whole mounts were prepared for imaging. Eyes for sectioning were cryopreserved in 30% sucrose prior to embedding in OCT (Thermofisher, Waltham MA) and cut into 12 pm sections using a cryostat. Slides were washed for lOmin, blocked for Ihour in 6% NDS / 1% BSA / 0.5% triton in PBS IX, and incubated overnight in 50% block solution diluted in PBS with anti-Proxl 1:500 (Biolegend, San Diego, CA; Rb), anti- ChAT 1:1000 (Chemicon, Gt), washed 3x5 min in PBS IX, incubated with Alexa Fluor conjugated secondary antibody 1:1000 and DAPI 1:2000 for l-2hours at room temperature, washed 3x5 min, and coverslipped in mounting medium (Vectashield). Images were then acquired using a confocal microscope and analyzed using Fiji by Z- stacking.

[0175] Multi-Electrode Array

[0176] MEA recordings were obtained from ex-vivo isolated flat mounted retinas of wild type mice and rdl mice aged from 132 to 324 days. Mice were sacrificed by quick cervical dislocation, and eyeballs were removed and placed in Ames medium (Sigma- Aldrich, St Louis, MO; A1420) bubbled with 95% 02 and 5% C02 at room temperature. Isolated retinas were placed on a cellulose membrane and gently pressed against a MEA (MEA256 iR-ITO; Multi-Channel Systems, Reutlingen, Germany), with the RGCs facing the electrodes. For wild type and rdl recordings MEA with respectively 3 pm and 60 pm electrode spacing were used. Pharmacology was used to block photoreceptor to bipolar cell transmission with 5 pm L-AP4 and 1 pm ACET, followed by 200 pm CNQX and 10 pm CPP in some experiments to block all transmission to RGCs (Tocris, Bristol UK). All of the multi-electrode array recordings were processed with the software spyking-circus (Yger et al. 2018) to sort the recorded spikes and obtain templates of individual retinal ganglion cell responses.

[0177] Light Stimulation

[0178] To study the responses of RGCs to optogenetic stimulation we used a flickering stimulus (referred to as flicker in the following) consisting of a series of white full-field flashes of one second duration, interleaved with one second intervals of darkness. Flashes were played both at low (-2.5.1013photons ern es’1) and high (-7.1014photons, ern es’1) light intensities (see (Sengupta et al. 2016) for a similar intensity tuning). To study the diversity of the cell responses, we used a chirp stimulus. This is a full field stimulus, lasting 25 seconds, designed to test the reaction of ganglion cells to changes in light intensity at different regimes of contrast and frequency. After a 1 s flash, the light intensity varies at constant speed and increasing contrast for 10 s, and with constant contrast and increasing frequency for other 10 s.

[0179] Detection of Ganglion Cell Responses

[0180] To check if a stimulus s evoked a response in a given ganglion cell, we did the following test. First, we considered a control window of 300 ms right before the stimulus onset. For a given cell c, we calculated the average spontaneous firing rate rctric sand its standard deviation cctric sacross repetitions in this time interval. We defined an activation threshold Tc sas: With h equal to lOhz. Then, we looked at a response window of 300 ms after the presentation of the stimulus (specifically, right after the stimulus onset for ON responses, and after the stimulus offset for OFF responses), and computed the peri-stimulus time histogram psthc sof these responses (time bin equal to 50 ms). We considered that there was a response if the mean response psthc sexceeded the activation threshold Tc s.

[0181] Classification Of ON And OFF Ganglion Cells

[0182] We used the criterion above to classify ON and OFF retinal ganglion cells. We labeled as ON or OFF all the cells that responded respectively to the onset or to the offset of the flicker. Cells responding at both onset and offset were labeled as ON-OFF. Cells that were not responding at all were not labeled, and have not been considered in further analyses. For the excitatory opsin protocol (ReaChR), we pooled data from three different experiments, obtaining a total population of 173 ON, 65 OFF and 44 ON-OFF ganglion cells. For the control protocol, we recorded from two different retinas, and identified a total of 113 ON, 94 OFF and 51 ON-OFF ganglion cells. For the inhibitory opsin protocol (gtACR), we collected data from a single experiment, and found 36 ON, 125 OFF and 9 ON-OFF ganglion cells.

[0183] Analysis Of RGC Responses To Ontogenetic Stimulation

[0184] To study the responses of RGC to the optogenetic stimulation, we displayed the flicker at low (~0.1 pW cm'2) and high (~2.8 pW cm'2) light intensities. Then, for each luminance level, we calculated the percentage of ON and OFF cells with detectable responses at both stimulus onset and offset. We did the same analysis to test the impact of both the excitatory and inhibitory opsins (respectively ReaChR and gtACRl, fig. 3.D, 4.E, 6.C), and for all the control protocols with L-AP4, ACET and CNQX and CPP (fig. 3.E, 4.F). For the control experiments on rdl mice, since the photoreceptor responses were not available, we could not preliminarily classify the ganglion cells as ON or OFF. As a consequence, we computed the percentages of RGCs activated on the entire population of cells, without any subdivision based on polarity (fig 8.E)

[0185] Off-Target Expression Of Retinal Ganglion Cells To identify the retinal ganglion cells directly expressing ReaChR due to the expression leakage, we looked at the optogenetic responses to the flicker after application of CNQX and CPP. We labeled as affected by the leakage all those cells for which a detectable response (either at stimulus onset and / or offset) was detectable. We did not consider these RGCs for the sustained-transient analysis nor for the complexity analysis described below.

[0186] Sustained-Transient Index

[0187] We defined a sustained-transient index STC, to assess which component is prevailing (transient or sustained) in the responses of a given ganglion cell c to a certain stimulus s. We considered two response windows subsequent to the presentation of the stimulus: a first one capturing transient responses (from 0 ms to 300 ms after stimulus onset for ON cells, and after stimulus offset for OFF cells) and a second one for sustained responses (from 300 ms to 600 ms after onset for ON cells and after offset for OFF cells). We computed the peri-stimulus time histogram of the responses (time bin equal to 50 ms) on both windows: psthtransc srepresenting the transient component, and psthSUstc srepresenting the sustained component. We then computed the sustained-transient index STC Sas the ratio:

[0188] An index STC Sclose to 1 entails that the cell c produces a sustained response to the stimulus s. Conversely, an index close to 0 means the response is predominantly transient.

[0189] Sustained-Transient Analysis

[0190] We used the ST index defined above to compare the transience of RGC responses to photoreceptors and optogenetic stimulations respectively. For this analysis, we only considered cells with detectable responses to both the photoreceptor and optogenetic responses. We also excluded all the cells for which optogenetic responses showed a different polarity with respect to the photoreceptor responses, and all those cells affected by the expression leakage, leaving us with a total of 48 good cells. We then calculated the sustained-transient index described above for all cells in both optogenetic and photoreceptor responses, and compared the distribution of the index under the two conditions using the Wilcoxon rank sum test. We found that, for the population of ON ganglion cells, these two distributions are significantly different, with the responses to optogenetic stimulation being more sustained (the test rejects the null hypothesis that the two distributions have equal medians with p-val < le-5). For the population of OFF cells, we did not observe a significant difference among the two distributions (fig 7.B). To consolidate this result at single cell level, we computed for each cell a comparative index AST, defined as the difference of the ST indices computed respectively on the photoreceptor and optogenetic responses:

[0191] This index has value close to 0 if the photoreceptor and optogenetic responses are similar in transience, value close to 1 if the cell has an optogenetic response more transient with respect to its normal photoreceptor responses, and value close to -1 if its optogenetic responses are more sustained. We looked at the distribution of this index for both ON and OFF ganglion cells (fig 7.C), and observed that the vast majority of ON retinal ganglion cells (31 cells out of 34) has a negative relative index, indicating that ON ganglion cells tend to lose their transient component when activated optogenetically through the All pathway.

[0192] Analysis Of Complexity Of The Responses

[0193] To assess the complexity of RGC responses we relied on principal component analysis (PCA). We computed the principal components of the ganglion cell responses to the chirp stimulus for both photoreceptor and optogenetic stimulation, and compared the number of components needed to explain different percentages of variance. We applied the selection criteria used for the sustained-transient analysis described above: we only kept cells consistently responding to both photoreceptor and optogenetic stimulation, and excluded all the cells featuring off-target expression, for a total of 40 good cells. As we did not want to account for inter-experimental variability, we ran the analysis independently for each of the three experiments. To make the results comparable, we needed to keep the population size constant across the different experiments. As a consequence, we ran the principal component analysis on each experiment on a sampled subpopulation of fixed size (30 cells). We repeated this procedure 100 times, and for each experiment we computed the average curve showing the variance explained by each number of principal components under both normal and optogenetic conditions (fig. 7.E). We then obtained the final results by averaging the curves across all three experiments (opaque lines in figure fig. 7.E).

[0194] Results

[0195] Targeting of All Amacrine Cells

[0196] To identify a promoter driving expression in All amacrine cells, we generated plasmid constructs with several known retinal specific promoters, encoding eGFP expression. Then, we produced AAV vectors with the corresponding expression cassettes and tested the resulting expression pattern, via intravitreal injection route with AAV2- 7m8, a genetic variant of AAV2 (Dalkara et al. 2013, Khabou et al. 2018) specifically selected for enhanced retinal transduction properties when delivered intravitreally. We incidentally found a sequence driving specific expression in All amacrine cells. This sequence (SEQ ID NO: 1, Fig. 9), that we refer to as HKamac in the following, has been derived from the IRBP enhancer and the human GNAT2 promoter, previously described for weak targeting of photoreceptors (Ying et al. 1998, Yeh CY, et al. 2013, Ye et al. 2016), and followed by an intronic SV40 sequence. C57BL / 6J wild-type mice were injected at 4 weeks of age using AAV2-7m8-IRBP / GNAT2-eGFP. 6 weeks after injection, eye fundus showed high expression levels (Fig. 1). Retinas were then harvested, fixed, and embedded in tissue freezing medium for histology and immunohistochemistry analysis 120-140 days postinjection.

[0197] Wild-type retinal flat mounts showed that a homogeneous population of cells with large somas expressed GFP in the INL (fig. 1 and 2.A). In cross-sections, the labeled cells showed dendritic stratification in both ON and OFF layers, a pattern reminiscent of All morphology (Helmstaedter et al. 2013). To determine precisely the subtype of amacrine cell, we first showed that GFP did not colocalize with GABAergic (GAD staining, see fig. l.B,B’) or with Starburst amacrine cell marker (ChAT staining) (fig. 2.B). To determine whether they were glycinergic All amacrine cells, we performed cryosections labelings with a Proxl antibody (Perez de Sevilla Muller L, 2017), labeling both glycinergic bipolar cells and All amacrine cells, and found clear co-localization of GFP with Proxl, thereby confirming GFP positive cells are glycinergic All amacrine cells (Fig. 2.C,C’ ,C” ). We then replaced GFP by ReaChR-eYFP in the plasmid construct while keeping the same promoter. ReachR was then delivered in wild-type mice using the same AAV2-7m8 capsid variant. ReaChR-eYFP was successfully expressed in All amacrine cells’ membranes, although eYFP was detected in some RGCs as well (fig. 2.D,D’, D”). The off-target expression pattern can depend on the viral dose and the injection route (Khabou et al, 2018). ReaChR is also very efficiently expressed which skews expression towards a broader range of cell types.

[0198] Optogenetic Stimulation Of Alls Produces On And Off Responses

[0199] All amacrine cells excite ON bipolar cells through gap junctions, and inhibit OFF bipolar cells through glycinergic, inhibitory synapses. We tested if stimulation of All amacrine cells with optogenetics could evoke ON and OFF responses in retinal ganglion cells (RGCs). For this we recorded RGC spiking activity from wild-type retinas expressing ReaChR under the HKamac promoter on a multielectrode array (fig. 3. A). To test if All stimulation could activate similar circuits to photoreceptor stimulation, we first measured the responses of ganglion cells to stimuli at low light intensity, which only activated photoreceptors, and were not sufficiently strong to activate ReaChR (termed photoreceptor stimulation in the following). We observed both ON and OFF responses (fig. 3.B). We then blocked the synaptic transmission from photoreceptors to the ON and OFF bipolar cells using pharmacology (L-AP4 to block the transmission from photoreceptors to ON bipolar cells, and ACET to block transmission from photoreceptors to OFF bipolar cells). At the same light intensity, responses disappeared. This is expected since the impact of photoreceptor activation on the rest of the retinal circuit has been blocked, and the light intensity is too low to activate ReaChR. We then increased light intensity to reach the activation of ReaChR and observed both ON and OFF responses to light stimulation (fig. 3.C and D) for a large portion of ganglion cells. This activation (termed optogenetic stimulation in the following) is due to the stimulation of All since stimulation in untreated retinas at similar intensity with the same concentration of blockers, did not show any response (fig. 3.E). Activation of All with optogenetic stimulation is thus able to evoke both ON and OFF responses in the retina.

[0200] To understand if our strategy allows reactivation of the same computations performed in the normal retina, we categorized the cells as ON, OFF or ON-OFF depending on their responses to photoreceptor stimulation. ON ganglion cells were defined as responding to the onset of the photoreceptor stimulation, OFF cells as responding to the offset and ON-OFF as responding to both. We then asked if ON cells and OFF cells responded to both the onset and offset of the optogenetic stimulation. If the responses were consistent for photoreceptor and optogenetic stimulation, this would suggest that All stimulation is able to reactivate some of the circuits that are active during photoreceptor stimulation. For example, ON ganglion cells receive their inputs from ON bipolar cells. At the onset of All stimulation, ON bipolar cells should be activated, and should therefore stimulate ON ganglion cells. Conversely, OFF bipolar cells, which provide the main excitatory input to OFF ganglion cells, should be inhibited during All stimulation, and uninhibited at the offset of All stimulation. As a result, they should be able to excite OFF ganglion cells at the offset of the All stimulation. If this hypothesis were correct, ON ganglion cells should be activated at the onset of All stimulation, and OFF ganglion cells at the offset.

[0201] We identified a total population of 173 ON and 65 OFF (and 44 ON-OFF) ganglion cells across 3 different experiments. The large majority of ON cells responded to the onset of the optogenetic stimulation (74%) while their responses to the offset were almost not present (only 0.5%), which is consistent with our hypothesis. A significant portion of OFF cells showed OFF responses to the optogenetic stimulation (25%). However, we also observed that a fraction of cells classified as OFF based on response to photoreceptor stimulation, turned ON during optogenetic stimulation (36%). We hypothesized that this could be due to off-target expression of the ReachR protein in ganglion cells, observed in our histology experiments.

[0202] Off-target Expression Explains Changes In ON-OFF Selectivity

[0203] If the observed responses at the onset for OFF cells were due to direct expression of ReaChR in ganglion cells, these responses would still be present when fully blocking glutamatergic synaptic transmission. To test this, we performed additional experiments on the same cell populations where we fully blocked this transmission using a pharmacological cocktail composed of L-AP4, ACET, CNQX and CPP (fig. 4.A). The responses at the stimulation onset in OFF ganglion cells were still present after the application of this cocktail (fig. 4.B, C, D): 41% of the OFF cell population responded to the stimulus onset at bright luminance, while responses to the stimulus offset completely disappeared (fig. 4.E). In control retinas where no optogenetic protein is expressed, light responses were almost completely abolished (99% or the OFF population did not respond to the stimulus, fig. 4.F). This confirms that these onset responses in OFF cells are due to off-target expression in ganglion cells.

[0204] This off-target expression had two consequences. First, when varying the stimulation light intensity, there were more and more ganglion cells responding to light onset when increasing light intensity (fig. 5). Our results suggest that this is due to a weaker off target expression in ganglion cells, which require a higher intensity to reach the spike threshold. When increasing light intensity, more ganglion cells have a light response due to off-target expression. Second, ON-OFF cells, characterized with photoreceptor stimulation, usually respond to light onset for optogenetic stimulation. This might be due to off-target expression in ganglion cells: to respond to light onset, a ganglion cell can either receive ON bipolar cell output, or express ReaChR due to off- target expression. To respond to light offset, a cell has to receive OFF bipolar cell output. This asymmetry might explain why we observed less responses to light offset than to light onset in ON-OFF ganglion cells.

[0205] Ganglion Cell ON And OFF Responses Can Be Evoked By All Stimulation

[0206] To further demonstrate that the observed responses were mostly due to All activation and not off-target expression in ganglion cells, nor to a failure of the pharmacological blocking of photoreceptor transmission, we performed additional experiments. We reasoned that if we use an inhibitory opsin, which will hyperpolarize the cells upon light stimulation, it will inactivate ganglion cells. As a consequence, off- target expression will not allow any spiking response. On the contrary, if All are hyperpolarized upon light stimulation, they should still evoke responses, except that they should be inverted: ON ganglion cells should respond at light offset, and OFF ganglion cells at light onset. If the responses are due to a failure of the pharmacological blocking, we should not see this inversion. We injected the same construct but replaced ReaChR with gtACRl (Govorunova et al. 2015). We performed the same protocols (2 retinas, 36 ON, 125 OFF and 9 ON-OFF RGCs) and found that most ganglion cells for which a response was detected showed the predicted inversion of polarity (fig. 6. A, B, C). Observed responses are thus due to All modulation, and not to off-target expression, nor to a failure of the pharmacological block. These results confirm that our approach allows stimulation of All to modulate differentially ON and OFF ganglion cells.

[0207] Diversity Of Ganglion Cell Responses To All Stimulation

[0208] All stimulation can restore ON and OFF responses, but can it restore more features of the normal retinal responses? In particular, beyond the ON and OFF classification, previous works on normal retinas have shown that retinal responses to more complex stimuli like “chirp” stimuli uncover a large diversity of responses, corresponding to the different types of ganglion cells. Is this diversity still present in retinas reactivated by our All stimulation strategy? To test this, we displayed the chirp stimulus, previously used to classify different types of ganglion cells (Baden et al. 2016), at high light intensity so that it activates the ReaChR protein expressed in All amacrine cells. We found a large diversity of responses to this stimulus (fig. 7.A). Beyond responses to onset and offset, some cells responded to different parts of this stimulus, showing different tunings to temporal frequencies.

[0209] A few cells responded transiently while most of them had more sustained responses. To quantify this, we first measured an index of how transient or sustained ganglion cells responses were. We computed this index both on the responses to optogenetic stimulation and to normal photoreceptor responses: we found that a large majority of cells lost their transient component when reactivated through the All pathway (fig. 7.B). We also defined a comparative sustained-transient index, that measures how transient is the response of a ganglion cell when activated optogenetically, with respect to its normal photoreceptor response (fig. 7.C). An analysis of the distribution of these indices showed that ON ganglion cells are consistently more sustained in their optogenetic responses (tested with the Wilcoxon rank-sum test, p-value below le-5), while the result is less clear for OFF cells. We then checked if the optogenetic stimulation of Alls preserved or disrupted the functional organization of retinal ganglion cells. Retinal ganglion cells can be classified into several types: cells belonging to the same type are spatially arranged to cover uniformly the visual field, and produce similar responses to the same stimuli. To check if this organization was still present, we looked at correlations of the responses of pairs of retinal ganglion cells for both photoreceptor and optogenetic stimulations (fig. 7.D). If the functional arrangement is preserved in optogenetically reactivated retinas, these correlations should not vary significantly across these two conditions. We used the Pearson Correlation Coefficient to estimate the correlation of two cells responses. We looked at the correlations of all cell pairs with the same polarity, for both photoreceptor and optogenetics responses. We observed that for most of the ganglion cell pairs considered, these correlations were still present during optogenetic stimulation (fig. 7.D).

[0210] As a further estimation of the diversity of the responses, we calculated the dimensionality of the space of possible responses. For this we performed a PCA on the ensemble of all the ganglion cell average responses to the chirp stimulus. If all the cells responded the same way to the stimulus, the first principle component would explain all the variance in these responses. On the contrary, if all the responses were very different, it would take a lot of components to explain most of the variance. We found that, for both normal and reactivated retinas, we needed more than 6 components to explain more than 95% of the total variance in the response (fig. 7.E). This shows that our strategy is able to restore a large part of the diversity in the visual responses.

[0211] All stimulation in degenerated retinas restore ON and OFF responses

[0212] So far we worked with wild-type retinas where we could compare the same ganglion cells responding to photoreceptor stimulation and optogenetic stimulation of All amacrine cells. However, in retinal dystrophies, the retinal network is rewired following the degeneration (Marc & Jones 2003). Is All stimulation still able to evoke responses after the rewiring imposed by degeneration? To test this, we performed the same experiments on rdl mice. We obtained a similar expression pattern in rdl mice as in wild type mice (fig. 8. A, B, C). We recorded the retinas at the age of 3 months when there are no measurable responses to light due to photoreceptor degeneration. We collected data accounting for a total population of 160 RGCs, measured their responses to light flashes, and consistently found both ON and OFF responses (fig. 8.D, E). This demonstrates that the rewiring of the network following degeneration does not affect the ability of our strategy to restore diverse responses in ganglion cells.

[0213] References

[0214] Baden et al. 2016 - Tom Baden et al. ImNature 529.7586 (2016), pp. 345-350.

[0215] Berry et al. 2019 - Michael H. Berry et al. In : Nature Communications 10.1 (2019), p. 1221.

[0216] Beyleler et al. 2019 - Michael Beyeler et al. In: Scientific Reports 9.1 (2019), pp. 1-16.

[0217] Bi et al. 2006 - Anding Bi et al. In: Neuron 50.1 (2006), pp. 23-33

[0218] Busskamp et al. 2010 - Volker Busskamp et al. In : science 329.5990 (2010), pp. 413- 417.

[0219] Caporale et al. 2011 - Natalia Caporale et al. In: Molecular Therapy 19.7 (2011), pp. 1212-1219.

[0220] Cehajic-Kapetanovic et al. 2015 - Jasmina Cehajic-Kapetanovic et al. In: Current Biology 25.16 (2015), pp. 2111-2122.

[0221] Chaffiol et al. 2017 - Antoine Chaffiol et al. In : Molecular Therapy 25.11 (2017), pp. 2546-2560. da Cruz et al. 2016 - Lyndon da Cruz et al. In: Ophthalmology 123.10 (2016), pp. 2248- 2254.

[0222] Dalkara et al. 2013 - Deniz Dalkara et al. In: Science translational medicine 5.189 (2013), 189ra76-189ra76.

[0223] Deny et al. 2017 - Stephane Deny et al. In: Nature Communications 8.1 (Dec. 2017), p. 1964.

[0224] Dyka et al. Adv Exp Med Biol. 2014;801:695-701 -

[0225] Famiglietti & Kolb 1975 - Famiglietti, E V Jr, and H Kolb. Brain research vol. 84,2 (1975) : 293-300. Ferrari et al. 2020 - Ulisse Ferrari et al. In:PloS computational biology 16.7 (2020), el007857.

[0226] Francis et al. 2013 - Peter J. Francis, Brian Mansfield, and Stephen Rose. In: Translational Vision Science And Technology 2.7 (2013), p. 4.

[0227] Gaub et al. 2015 - Benjamin M. Gaub et al. In: Molecular Therapy 23.10 (2015), pp. 1562-1571.

[0228] Govorunova et al. 2015 - Elena G Govorunova et al. In : Science 349.6248 (2015), pp. 647-650.

[0229] Guadagni et al. 2019 - Guadagni, Viviana et al. FASEB journal : official publication of the Federation of American Societies for Experimental Biology vol. 33,9 (2019): 10177- 10192.

[0230] Helmstaedter et al. 2013 - Moritz Helmstaedter et al. In: Nature 500.7461 (2013), pp. 168-174.

[0231] Hillier et al. 2017 - Daniel Hillier et al. In : Nature Neuroscience 20.7 (2017), pp. 960- 968.

[0232] Khabou et al. 2018 - Hanen Khabou et al. In: JCI insight 3.2 (2018).

[0233] Kolb & Nelson 1983 - Kolb, H, and R Nelson. Vision research vol. 23,4 (1983): 301-12.

[0234] Kolb et al. 1992 - Kolb, H et al. The Journal of comparative neurology vol. 318,2 (1992): 147-87.

[0235] Lorach et al. 2012 - Henri Lorach et al. In: Journal of Neural Engineering 9.6 (2012).

[0236] Mace et al. 2015 - Emilie Mace et al. In: Molecular Therapy 23.1 (2015), pp. 7-16.

[0237] Marc & Jones 2003 - Robert E. Marc and Bryan W. Jones. In: Molecular Neurobiology 28.2 (2003), pp. 139-147.

[0238] Merigan et al. 1991 - William H Merigan, Carey E Byrne, and John H R Maunsell. In : Journal of Neuroscience l l.November (1991), pp. 3422-3429.

[0239] Perez de Sevilla Muller L, Front Neuroanat. 2017 May 5; 11:39.

[0240] Sahel et al. 2021 - Jose Alain Sahel et al. In: Nature Medicine 27.7 (2021), pp. 1223- 1229. Sanes & Masland 2015 - Joshua R. Sanes and Richard H. Masland. In: Annual Review of Neuroscience 38.1 (2015), pp. 221-246.

[0241] Schiller et al. 1986 - Peter H. Schiller, Julie H. Sandell, and John H. R. Maunsell. In: Progress in Retinal and Eye Research 322 (1986), pp.824-825.

[0242] Sengupta et al. 2016 - Abhishek Sengupta et al. In: EMBO Molecular Medicine 8.11 (2016), pp. 1248-1264.

[0243] Strettoi et al. 1992 - Strettoi, E et al. The Journal of comparative neurology vol. 325,2 (1992) : 152-68.

[0244] Strettoi et al. 2002 - Enrica Strettoi et al. In: Journal of Neuroscience 22.13 (2002), pp. 5492-5504

[0245] Tsukamoto and Omi 2017 - Yoshihiko Tsukamoto and Naoko Omi. In: Frontiers in neuroanatomy 11 (2017), p. 92. van Wyk et al. 2015 - Michiel van Wyk et al. In: PloS biology 13.5 (2015), el002143.

[0246] Vaney 1985 - Vaney, D I. Proceedings of the Royal Society of London. Series B, Biological sciences vol. 224,1237 (1985):475-88.

[0247] Wassle et al. 1995 - Wassle, H et al. The Journal of comparative neurology vol. 361,3 (1995): 537-51.

[0248] Wassle & Boycott 1991 - Physiol. Rev. 71, 1991, 447-480

[0249] Xin & Bloomfield 1999 - Xin, D, and S A Bloomfield. Visual neuroscience vol. 16,4 (1999) : 653-65.

[0250] Ye et al 2016 - Ye et al Hum Gene Ther 2016 Jan;27(l):72-82.

[0251] Yeh et al 2013 - Yeh CY, et al. In: ARVO 2013 Annual Meeting (Seattle, WA), 2013; p. abstract 1937.30

[0252] Ying et al 1998 - Ying et al, Curr Eye Res 1998 Aug;17(8):777-82.

[0253] Yger et al. 2018 - Pierre Yger et al. In: Elife 7 (2018), e34518.

[0254] Yonehara et al. 2016 - Keisuke Yonehara et al. In: Neuron 89.1 (2016), pp. 177-193.

Claims

CLAIMS1. An expression cassette comprising a nucleic acid having a promoter activity in All amacrine cells operably linked to a coding sequence of interest, for use in the treatment of an ocular disease, wherein said nucleic acid having a promoter activity comprises an enhancer element of an inter-photoreceptor retinoid-binding protein (IRBP) promoter that is fused to a promoter nucleotide sequence of a cone transducin alphasubunit (GNAT2) gene, and wherein the treatment involves the intravitreal administration of said expression cassette to the subject being treated.

2. The expression cassette for use according to claim 1, wherein the enhancer element of an IRBP promoter comprises, or consists of, the nucleotide sequence of SEQ ID NO: 2, or a functional variant thereof having at least 80% identity to SEQ ID NO:2 or comprising at least 150 consecutive nucleotides of SEQ ID NO:2.

3. The expression cassette for use according to 1 or 2, wherein the promoter nucleotide sequence of a GNAT2 gene comprises, or consists of, the nucleotide sequence of SEQ ID NO: 3, or a functional variant thereof having at least 80% identity to SEQ ID NOG or comprising at least 200 consecutive nucleotides of SEQ ID NOG.

4. The expression cassette for use according to any one of claims 1 to 3, wherein the enhancer element of an IRBP promoter is positioned upstream of the promoter nucleotide sequence of a GNAT2 gene.

5. The expression cassette for use according to any one of claims 1 to 4, wherein the nucleic acid having a promoter activity in All amacrine cells comprises, or consists of, the sequence of SEQ ID NO: 4 or a functional variant thereof exhibiting a promoter activity in All amacrine cells and having at least 80% identity to SEQ ID NOG.

6. The expression cassette for use according to any one of claims 1 to 5, wherein the expression cassette further comprises a splice donor / splice acceptor element, preferably comprised in an intronic sequence known to increase splicing efficiency and expression of the coding sequence.

7. The expression cassette for use according to claim 6, wherein the splice donor / splice acceptor element is position downstream of the nucleic acid having a promoter activity in All amacrine cells and upstream of the coding sequence of interest.

8. The expression cassette for use according to claim 6 or 7, wherein the splice donor / splice acceptor element is comprised in an intronic sequence comprising, or consisting of, sequence of SEQ ID NO: 10 or a functional variant thereof having at least 80% identity to SEQ ID NO: 10.

9. The expression cassette for use according to any one of claims 1 to 8, wherein the expression cassette is provided in a vector, preferably a viral vector.

10. The expression cassette for use according to claim 9, wherein the vector is an adeno-associated viral (AAV) vector.

11. The expression cassette for use according to any one of claims 1 to 10, wherein the expression cassette is provided in AAV particle comprising an AAV vector.

12. The expression cassette for use according to claim 11, wherein the AAV particle comprises an AAV-derived capsid selected from the group consisting of AAV2 and AAV2-7m8.

13. The expression cassette for use according to claim 11 or 12, wherein the AAV particle comprises an AAV2-7m8 capsid.

14. The expression cassette for use according to any one of claims 1 to 13, wherein the expression cassette is provided in a host cell transformed or transfected with said expression cassette, or with a vector or viral particle comprising said expression cassette.

15. The expression cassette for use according to claim 14, wherein the host cell is a human host cell.

16. The expression cassette for use according to claim 14 or 15, wherein the host cell is an All amacrine cell, preferably a human All amacrine cell.

17. The expression cassette for use according to any one of claims 1 to 16, wherein the coding sequence of interest is a therapeutic protein, an optogenetic protein or a reporter protein.

18. The expression cassette for use according to any one of claims 1 to 16, wherein the coding sequence of interest is a therapeutic protein.

19. The expression cassette for use according to any one of claims 1 to 17, wherein the coding sequence of interest is an optogenetic protein.

20. The expression cassette for use according to claim 19, wherein the optogenetic protein is an optogenetic activator.

21. The expression cassette for use according to claim 19, wherein the optogenetic protein is an optogenetic inhibitor.

22. Use an expression cassette as defined in any of claims 1 to 21 for the manufacture of a medicament for the treatment of an ocular disease.

23. A method for treating an ocular disease comprising administering a therapeutically efficient amount of an expression cassette as defined in any of claims 1 to 21 to a subject in need thereof.

24. The expression cassette for use according to any one of claims 1 to 21, the use of claim 22 or the method of claim 23, wherein the ocular disease is a disease associated with photoreceptor cell degeneration.

25. The expression cassette for use according to any one of claims 1 to 21 and 24, the use of claim 22 or 24 or the method of claim 23 or 24, wherein the ocular disease is selected from the group consisting of retinitis pigmentosa, age-related macular degeneration, cone-rod dystrophy, Leber congenital amaurosis, Stargardt’s disease, diabetic retinopathy, Best’s disease, choroideremia or a tapetoretinal degeneration.

26. The expression cassette for use according to any one of claims 1 to 21 and 24, the use of claim 22 or 24 or the method of claim 23 or 24, wherein the ocular disease is retinitis pigmentosa.

27. An expression cassette comprising a nucleic acid having a promoter activity in All amacrine cells operably linked to a sequence encoding an optogenetic protein, wherein said nucleic acid having a promoter activity comprises an enhancer element of an inter-photoreceptor retinoid-binding protein (IRBP) promoter that is fused to a promoter nucleotide sequence of a cone transducin alpha-subunit (GNAT2) gene.

28. A vector comprising an expression cassette of claim 27.

29. The vector of claim 28, wherein said vector is an adeno-associated viral (AAV) vector.

30. An AAV particle comprising a vector of claim 29.

31. The AAV particle of claim 30, comprising an AAV-derived capsid selected from the group consisting of AAV2 and AAV2-7m8.

32. The AAV particle of claim 30, comprising an AAV2-7m8 capsid.

33. A isolated host cell transformed or transfected with an expression cassette of claim 27, a vector of claim 28 or 29, or an AAV particle of any of claims 30 to 32.

34. The cell of claim 33, which is an All amacrine cell, preferably a human All amacrine cell.

35. A pharmaceutical composition comprising an expression cassette of claim 27, a vector of claim 28 or 29, an AAV particle of any of claims 30 to 32, or a cell of claim 33 or 34, and a pharmaceutically acceptable excipient.

36. A method for expressing a coding sequence of interest in an All amacrine cell, said method comprising introducing into said cell an expression cassette comprising the coding sequence of interest operably linked to a nucleic acid having a promoter activity in All amacrine cells, said nucleic acid having a promoter activity comprising comprises an enhancer element of an inter-photoreceptor retinoid-binding protein (IRBP) promoter that is fused to a promoter nucleotide sequence of a cone transducin alpha- subunit (GNAT2) gene.

37. Use of a nucleic acid having a promoter activity in All amacrine cells for expressing a coding sequence of interest operably linked to said promoter in an All amacrine cell, said nucleic acid having a promoter activity comprising comprises an enhancer element of an inter-photoreceptor retinoid-binding protein (IRBP) promoter that is fused to a promoter nucleotide sequence of a cone transducin alpha-subunit (GNAT2) gene.

Images