Gene therapy for the treatment of a disease of retinal cone cells

Abstract

The present invention relates to a polynucleotide configured for the treatment of a disease of retinal cone cells, such as achromatopsia, a nucleic acid vector comprising said polynucleotide, a pharmaceutical composition comprising said nucleic acid vector, a kit comprising said polynucleotide or said nucleic acid vector, a method of making said nucleic acid vector, and a method for treating a disease of the retinal cone cells.

Claims

1. A polynucleotide, comprising a transgene expression cassette, comprising (a) a nucleic acid encoding the promoter of human retinal arrestin 3 gene (hArr3); (b) a nucleic acid encoding the human cone cyclic nucleotide-gated channel alpha 3 subunit (hCNGA3) or fragments thereof exhibiting hCNGA3 activity, and (c) a nucleic acid encoding regulatory elements necessary for effective expression of hCNGA3.

2. The polynucleotide of claim 1, wherein said regulatory elements comprise (c1) a woodchuck stomatitis virus posttranscriptional regulatory element (WPRE).

3. The polynucleotide of claim 2, wherein said WPRE is a mutated WPRE (WPREm), said WPREm comprising non-expressible woodchuck hepatitis virus X protein (WHX) open reading frame (WHX OR).

4. The polynucleotide of any of claim 1, wherein said regulatory elements comprise (c2) a polyadenylation signal (pA).

5. The polynucleotide of claim 4, wherein said pA is a bovine growth hormone pA (BGH pA).

6. The polynucleotide of claim 1, wherein it further comprises a nucleic acid encoding inverted terminal repeats (ITRs) flanking said transgene expression cassette, wherein at least one ITR is adjacent to said hArr3 promoter (L-ITR) and at least one ITR is adjacent to said pA (R-ITR).

7. The polynucleotide of claim 6, wherein said ITRs are derived from AAV serotype 2 (ITR AAV2).

8. The polynucleotide of claim 1 comprising an arrangement order selected from the following group: (a)-(b)-(c), (a)-(b)-(c1)-(c2), and (L-ITR)-(a)-(b)-(c1)-(c2)-(R-ITR).

9. The polynucleotide of claim 1, wherein said nucleic acid encoding the promoter of hArr3 comprises the nucleotide sequence of SEQ ID No. 1.

10. The polynucleotide of claim 1, wherein said nucleic acid encoding hCNGA3 comprises the nucleotide sequence of SEQ ID No. 2 or a nucleotide sequence encoding the amino acid sequence of SEQ ID No. 3.

11. The polynucleotide of claim 3, wherein said nucleic acid encoding WPREm comprises the nucleotide sequence of SEQ ID No. 4.

12. The polynucleotide of claim 5, wherein said nucleic acid encoding BGH pA comprises the nucleotide sequence of SEQ ID No. 5.

13. The polynucleotide of claim 6, wherein said nucleic acid encoding L-ITR comprises the nucleotide sequence of SEQ ID No. 6 and/or said nucleic acid encoding R-ITR comprises the nucleotide sequence of SEQ ID No. 7.

14. A nucleic acid vector comprising a polynucleotide comprising a transgene expression cassette, comprising (a) a nucleic acid encoding the promoter of human retinal arrestin 3 gene (hArr3); (b) a nucleic acid encoding the human cone cyclic nucleotide-gated channel alpha 3 subunit (hCNGA3) or fragments thereof exhibiting hCNGA3 activity, and (c) a nucleic acid encoding regulatory elements necessary for effective expression of hCNGA3.

15. The nucleic acid vector of claim 14, which is a circular plasmid further comprising a backbone having a length of5,000 bp or 5,500 bp.

16. The nucleic acid vector of claim 15, wherein said backbone comprises 0 to 5 open reading frames (ORFs).

17. The nucleic acid vector of claim 15, wherein said backbone comprises a selection marker selected from the group consisting of: an antibiotic resistance encoding nucleic acid and a kanamycin resistance encoding nucleic acid (KanR).

18. The nucleic acid vector of claim 17, wherein said selection marker is at its 5 and 3 termini remotely spaced apart from the polynucleotide by 1,900 bp.

19. The nucleic acid vector of claim 15, wherein said backbone comprises 0 to 10 restriction enzyme recognition sites (RERSs).

20. The nucleic acid vector of claim 15, wherein said backbone comprises 0 to 5 promoters.

21. The nucleic acid vector of claim 15, wherein said backbone further comprises a pUC18 origin of replication (ORI) ORI.

22. The nucleic acid vector of claim 15, wherein said backbone comprises the nucleotide sequence of SEQ ID No. 8.

23. The nucleic acid vector of claim 15, wherein the vector is an adeno-associated viral (AAV) vector.

24. The nucleic acid vector of claim 23, wherein the serotype of the AAV capsid sequence of said AAV vector is selected from the group consisting of: AAV2, AAV5, AAV8, and combinations thereof.

25. A pharmaceutical preparation comprising the nucleic acid vector of claim 15, and a pharmaceutically acceptable carrier.

26. The pharmaceutical preparation of claim 25, wherein said pharmaceutically acceptable carrier is selected from the group consisting of: saline solution, balanced sterile saline solution, surfactant, and micronized poloxamer.

27. A method of making a recombinant adeno-associated viral (rAAV) vector comprising inserting into an adeno-associated viral vector the polynucleotide of claim 1.

28. The method of claim 27, wherein said recombinant adeno-associated viral vector is a nucleic acid vector comprising a polynucleotide, comprising a transgene expression cassette, comprising (a) a nucleic acid encoding the promoter of human retinal arrestin 3 gene (hArr3); (b) a nucleic acid encoding the human cone cyclic nucleotide-gated channel alpha 3 subunit (hCNGA3) or fragments thereof exhibiting hCNGA3 activity, and (c) a nucleic acid encoding regulatory elements necessary for effective expression of hCNGA3.

29. A method for treating a disease associated with a genetic mutation that affects retinal cone cells, wherein the method comprises administering to a subject in need of such treatment the nucleic acid vector of claim 14, thereby treating the subject.

30. The method of claim 29, wherein the disease is achromatopsia (ACHM).

31. The method of claim 29, wherein the disease is achromatopsia type 2 (ACHM2).

32. The method of claim 29, wherein the vector is administered subretinally or intravitreally.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows the structure of the rAAV.hCNGA3 vector genome;

(2) FIGS. 2A-2B show two embodiments of the phArr3.hCNGA3.WPREm cis vector plasmid map;

(3) FIG. 3 shows representative ERG measurements from CNGA3-deficient mice treated on one eye with the vector according to the invention; and

(4) FIG. 4 depicts representative confocal images from immunohistological stainings of hCNGA3 in CNGA3-deficient mice (A3KO) treated with rAAV.hCNGA3 vector.

(5) FIGS. 5A-5B show the longitudinal course of best corrected visual acuity (BCVA) after treatment with AAV8.hCNGA3. Upper graph (FIG. 5A) shows data from the treated study eyes, lower graph (FIG. 5B) from the untreated fellow control eyes. Each line represents one patient. BCVA was determined by means of the standardized EDTRS charts under constant illumination. The absolute numbers of letters correctly read by each patient relative to his/her pre-injection baseline (0) are plotted against time (days) after injection. Immediately after injection BCVA drops in most patients but after 180 day every single patient's BCVA has improved in comparison to baseline, indicating a clear tendency of improvement not observed in the control eyes.

(6) FIGS. 6A-6B depict the change in metabolic peak activity in the visual cortex after treatment with AAV8.hCNGA3 (example). Before treatment (6A), this achromatic patient responded to a luminance contrast pattern (left column), but not to an isoluminant chromatic contrast pattern (right column=0). After treatment, the isoluminant chromatic signal (6B, right column) reached similar levels as after luminance contrast stimulation.

(7) FIGS. 6C-6D depict fMRI signals from the same patient, resolved over time after stimulus presentation. Grey lines present signal amplitude in response to contrast stimuli; blue lines in response to isoluminant chromatic contrast stimuli. Before treatment (FIG. 6C) chromatic stimuli do not produce increase in fMRI signal amplitude, after treatment (FIG. 6D) signal amplitude is clearly increased.

EXAMPLES

(8) 1. Nucleic Acid Vector of the Invention

(9) In this exemplary embodiment the rAAV.hCNGA3 vector is a hybrid AAV-based vector carrying the cDNA of the human CNGA3 subunit of the cone photoreceptor cyclic nucleotide-gated (CNG) cation channel. The hCNGA3 cDNA expression is under the control of the cone-specific human arrestin 3 (hArr3) promoter and is enhanced using a mutated woodchuck stomatitis virus posttranscriptional regulatory element (WPRE) sequence. The expression cassette is flanked by the AAV serotype 2 inverted terminal repeats (ITRs) and the recombinant genome is packaged in the AAV serotype 8 capsid, resulting in an AAV2/8 hybrid vector. The expression cassette comprises the following elements: Promoter of the human arrestin 3 (hArr3) gene: 0.4 Kb cDNA of the human CNGA3 subunit of the cone photoreceptor cyclic nucleotide-gated cation channel: 2 Kb Woodchuck stomatitis virus posttranscriptional regulatory element (WPRE) with a point mutation in the ATG codon of the WHV-X open reading frame: 0.54 Kb Polyadenylation signal of the Bovine Growth Hormone (BGH): 0.2 Kb AAV serotype 2 inverted terminal repeats (ITRs): 0.13 Kb The structure of the rAAV.hCNGA3 vector genome is depicted in FIG. 1.
2. pGL2.hArr3.hCNGA3.WPREm Cis Vector Plasmid

(10) In one exemplary embodiment the pGL2.hArr3.hCNGA3.WPREm cis vector-plasmid backbone is used that contains an expression cassette comprising a 405 bp cone photoreceptor-specific human cone arrestin (hArr3) promoter [see Li et al, Retinoic acid upregulates cone arrestin expression in retinoblastoma cells through a Cis element in the distal promoter region, Investigative ophthalmology & visual science, 43 (2002) 1375-1383, and Carvalho et al., Long-term and age-dependent restoration of visual function in a mouse model of CNGB3-associated achromatopsia following gene therapy, Human molecular genetics, 20 (2011) 3161-3175] and the full-length (2085 bp) human CNGA3 cDNA [see Wissinger et al., Cloning, chromosomal localization and functional expression of the gene encoding the alpha-subunit of the cGMP-gated channel in human cone photoreceptors]. The expression cassette also contains a 543 bp woodchuck hepatitis virus post-transcriptional regulatory element (WPRE) with mutated WXF-open reading frame [Zanta-Boussif et al., Validation of a mutated PRE sequence allowing high and sustained transgene expression while abrogating WHV-X protein synthesis: application to the gene therapy of WAS, Gene therapy, 16 (2009), 605-619] and a 207 bp bovine growth hormone polyadenylation signal (BGHpA). The 5591 bp vector backbone with the nucleotide sequence depicted in SEQ ID No. 8 containing a kanamycin resistance (KanR) positioned 1943 bp from the L-ITR and 2853 bp from the R-ITR and 2024 bp from a pUC18 ori.

(11) The rAAV.hCNGA3 vector is produced using transient double-transfection of the cis vector plasmid and a trans pDP8-KanR helper plasmid in the human embryonic kidney 293 cells (HEK293). The cell lysate is clarified by a low-speed centrifugation and the vector is then purified by 2 consecutive rounds of cesium chloride gradients ultracentrifugation followed by a tangential flow filtration step for concentration and buffer exchange. The resulting rAAV.hCNGA3 vector suspension is then sterile-filtered and vialed as drug product.

(12) Two embodiments of the phArr3.hCNGA3.WPREm vector plasmid map are shown in FIG. 2A,B.

(13) 3. Biological Activity and Transgene Expression Conferred by the rAAV.hCNGA3 Vector

(14) To verify biological activity and transgene expression the inventors delivered the rAAV.hCNGA3 vector into the subretinal space of 2-week-old Cnga3-deficient mice [Biel et al., Selective loss of cone function in mice lacking the cyclic nucleotide-gated channel CNG3, Proc Natl Acad Sci USA, 96(13):7553-7557 (1999). The delivery procedure was similar to the one described for the mouse-specific vector [Michalakis et al., Restoration of cone vision in the CNGA3/ mouse model of congenital complete lack of cone photoreceptor function, Molecular therapy: The Journal of the American Society of Gene Therapy, 18 2057-2063 (2010)]. The mice received a subretinal injection in the treated eye (TE), whereas the other, untreated eye (UE) served as control. The vector efficacy was evaluated at 8 weeks following the injection by means of electroretinography (ERG), an objective functional in vivo assay. Cnga3-deficient mice lack any cone-mediated vision. Therefore, ERG protocols specifically testing for cone function are suitable as an indirect measure for CNGA3 function and for the assessment of biological activity (BAA) of the rAAV.hCNGA3 vector.

(15) For representative results see FIG. 3: Representative ERG measurements from CNGA3-deficient mice treated on one eye (treated eye, black traces) with rAAV.hCNGA3 vector. The traces from the untreated control eye are shown in grey. The biological activity conferred by the rAAV.hCNGA3 vector-mediated expression of hCNGA3 is clearly evident as elevation of specific ERG components (7 Hz scotopic flicker, 5 Hz photopic flicker and photopic flash) that are mediated by cone photoreceptors and are missing in CNGA3-deficient mice.

(16) The rAAV.hCNGA3 vector treatment resulted in a clear therapeutic effect in the treated eye reflected by elevation of specific ERG components. After completion of the ERG measurements mice were sacrificed, the eyes enucleated and processed for immunohistological analysis of hCNGA3 transgene expression (transgene expression assay, TEA). For this, the tissue was fixed and cryoembedded. Vertical cryosections were stained with a rat monoclonal antibody that binds to mouse and human CNGA3 protein. The immunosignal was detected with a Cy3 tagged donkey anti-rat IgG secondary antibody. Confocal images from the immunostained cryosections were collected using a Leica SP8 SMD confocal laser scanning microscope. The anti-CNGA3 antibody also detects mouse Cnga3 protein and gives a specific signal in cone photoreceptor outer segments of wildtype mouse retina and no signal in Cnga3-deficient retina. After treatment with rAAV.hCNGA3 vector a clear and specific signal for CNGA3 was observed in the cone photoreceptor outer segments in the treated eye, which was absent in the untreated eye.

(17) For representative results see FIG. 4: Representative confocal images from immunohistological stainings of hCNGA3 in CNGA3-deficient mice (A3KO) treated with the rAAV.hCNGA3 vector. The anti-CNGA3 antibody (working dilution in all experiments 1:50) also detects mouse Cnga3 protein and gives a specific signal in cone photoreceptor outer segments (rod-shaped structures in the upper part of the image) of wildtype mouse retina (left two panels) and no specific signal in the retina of untreated CNGA3-deficient mice (right two panels). The specific signal for the hCNGA3 protein encoded by the rAAV.hCNGA3 vector is shown in the third panel, which is absent in the untreated A3KO shown in the fourth panel. Panel two shows a secondary antibody only control staining. The lower panels show an overlay of the CNGA3 signal with the cone photoreceptor-specific marker peanut agglutinin and the nuclear dye Hoechst 4322.

(18) In conclusion, the rAAV.hCNGA3 vector expresses the hCNGA3 transgene efficiently and specifically in cone photoreceptors of CNGA3-deficient mice and confers cone-mediated vision to these mice that lack cone function from birth (biological activity).

(19) 4. AAV8 Biodistribution and Shedding after Subretinal Injection in Non-Human Primates

(20) In another study the virus distribution and shedding was analysed after a single subretinal administration of clinical grade recombinant adeno-associated virus (rAAV) in non-human primates. This is important for an environmental risk assessment of the gene therapeutic method according to the invention.

(21) 18 non-human primates (Macacca fascicularis) underwent 23G pars plana vitrectomy and subretinal injection in three cohorts (high dose: 110.sup.12 vector genomes [vg], low dose: 110.sup.11 vg, or vehicle only). Four additional animals received intravitreal injections to mimic via falsa biodistribution. Tissues samples were harvested at necropsy (day 91) from the treated eye, draining lymph nodes, salivary gland and spleen, optic nerve, brain and spinal cord, heart, lung, liver, adrenal glands and gonads. Blood, urine, lacrimal and nasal swabs were harvested from each animal prior to dosing and 1, 2 and 3 days and 1, 4 and 13 weeks after application of the vector for DNA extraction and quantification of vector genomes by qPCR.

(22) Dose dependent rAAV DNA was found in the treated retina and optic nerve. Quantifiable levels of rAAV DNA were also detected in optic chiasm of 2 animals of the high dose group. Transient shedding was found in all bio fluids. The highest concentrations were found in lacrimal fluid of the high dose group. DNA was not detected in the germ line tissues and apart from sporadic signals detected in a small number of animals in the lymph nodes and spleen, all remaining tissues were negative. Blood samples showed quantifiable levels of rAAV vector DNA at 24 and 72 hours after treatment, but were negative at all other time points tested.

(23) These data are relevant for the clinical implementation of the invention, where trial subjects, investigators and regulators alike are interested to identify environmental risks associated with application of genetically modified organisms. While shedding into biofluids seems to occur in a dose dependent manner, transduction of off-target organs seems minimal.

(24) 5. Humoral Immune Response to Subretinal AAV8 in Non-Human Primates

(25) Knowledge of the humoral immune response to single subretinal administration of clinical grade recombinant adeno-associated virus (rAAV) in non-human primates is a key factor for the development of safe and efficient clinical trial protocols for the retinal gene therapy according to the invention. For this reason the inventors explored anti-drug-antibody (ADA) titres in non-human primates (Macacca fascicularis) after single subretinal administration of a rAAV8-pseudotyped virus.

(26) 18 monkeys received subretinal injections in three cohorts (high dose: 110.sup.12 vector genomes [vg], low dose: 110.sup.11 vg, or vehicle only) and concomitant immunosuppressive therapy equivalent to a clinical trial scenario. Four additional animals received intravitreal injections to mimic biodistribution e.g. after surgical complications. Baseline samples were compared to those taken 1, 2 and 3 days and 1, 4 and 13 weeks after application of the vector.

(27) The anti-drug-antibody (ADA) titres in all animals of the low dose group stayed constant over the 13 week observation period. The subretinal high dose group showed greatest variability over time, but no clear pattern of humoral immune response.

(28) This study provides data relevant for a clinical application of the invention, where rAAV8 might be used for subretinal delivery of the hCNGA3 transgene. When mimicking the clinical scenario with clinical grade vector, surgery and concomitant immunosuppression, no induction of anti-drug-antibodies occurred in non-human primates.

(29) 6. Successful Delivery of rAAV8.CNGA3 in a Patient with CNGA3 Based Achromatopsia

(30) The aim of this clinical interventional study (NCT02610582) was to test safety aspects of the AAV8 based supplementation gene therapy according to the invention in patients with CNGA3 based achromatopsia.

(31) After extensive safety testing in a dose escalation study in 34 non-human primates (NHP) the inventors selected a dosing range of 110.sup.10, 510.sup.10, and 110.sup.11 vector genomes (vg) for an exploratory, dose-escalation clinical phase I/II trial. A total of 9 patients with homozygous or compound heterozygous mutations in CNGA3 received a single subretinal injection of either 110.sup.10 vg (n=3), 510.sup.10 vg (n=3), or 110.sup.11 vg (n=3) each in 0.2 ml balanced salt solution. Concomitant steroid treatment (Prednisolone 1 mg/kg/d) was initiated 1 day prior surgery. The primary endpointsafety of applicationwas assessed by clinical examination and best corrected visual acuity (BCVA).

(32) NHP safety data showed no persisting test item-related changes after application of 110.sup.12 vg 90 days after dosing. In the clinical trial, all patients received the respective dose (110.sup.10-110.sup.11 vg) safely and without surgical or post-surgical complications such as retinal detachment, hemorrhage or inflammation unresponsive to treatment. BCVA reached baseline levels as soon as 14 days post treatment. Structural changes at the level of the retinal pigment epithelium and inner/outer photoreceptor segments were attributed to the surgical procedure (see above).

(33) The NHP safety study showed that 110.sup.12 vg can be applied without relevant sequelae. This was the first clinical application of AAV8 mediated subretinal gene therapy in the eye. The application was well tolerated and did not lead to clinically apparent inflammation under concomitant Prednisolone treatment. Even though the application involved macular detachment, visual acuity reached baseline levels within 14 days.

(34) 7. Safety after Subretinal Delivery of AAV8.hCNGA3 in Patients with Achromatopsia

(35) Safety as the primary endpoint was assessed by clinical examination of ocular inflammation (slit lamp, fundus biomicroscopy, angiography, perimetry or electrophysiology). At the current stage (15 months into the trial) with all patients having been treated and followed up for minimum of three months, not a single serious adverse event had to be documented. Additionally, there was not a single ocular adverse event, which required additional action and no non-ocular adverse events, which were not resolved without sequelae. Generally, this reflects the excellent safety profile already seen in the pre-clinical toxicology study in NHPs.

(36) 8. Efficacy after Subretinal Delivery of AAV8.hCNGA3 in Patients with Achromatopsia

(37) Although not representing a main goal of this safety study, explorative efficacy endpoints were chosen to screen for their suitability in future efficacy studies. These included best corrected visual acuity, patient reported outcome measures and others.

(38) One of the most relevant endpoints in ocular clinical trials is the best corrected visual acuity. In this endpoint, all available data at this time-point of submission of this document indicate no sustained and/or substantial deleterious effect of the treatment. While the surgery can lead to transient reduction of visual acuity (as expected), all patients with a follow up of at least 6 months show improvement in visual acuity and all patients with a follow up of 12 months continue to show also improvement in visual acuity. This is illustrated by the graphs depicted in FIG. 5, top (A) for the treated eye, bottom (B) for the untreated eye.

(39) Patient reported outcome measures gain importance in trial protocols as they typically reflect parameters important for our patients' quality of life. Interim results of the ongoing trial (NCT02610582) demonstrate that the vast majority of the study patients reported a fast and relevant improvement of their key symptom glare after subretinal injection of rAAV.hCNGA3. The majority also reported an improvement in recognition of letters and numbers and in their fixation ability. These preliminary results were found distributed quite homogenously in all three dosage groups.

(40) Ganzfeld stimulation and functional magnetic resonance (fMRI) imaging was used to quantify localized metabolic activity in the visual cortex dependent on stimuli originating from the treated area and the cone photoreceptor system. Three groups of three patients each were treated with low-, medium- or high-dose of AAV8.hCNGA3 gene therapy in the study eye to restore local cone function. fMRI was performed before and at three months after the treatment. This allowed us to assess the (re)organization of the visual cortex as well the whole brain network at these time-points, and to compare with the corresponding responses we collected from normal-trichromatic subjects. In each fMRI session the subjects performed three visual stimulation experiments under mesopic light conditions: a) retinotopic mapping, b) isoluminant color contrast (not resolved by achromatic retinae) vs luminance contrast, and c) spatial frequency gratings (0.3, 1, 5 cycles per degree) at low (3%) and high (50%) contrast. For all patients, the baseline response before the treatment was, as expected, much higher for luminance contrast (grey columns on the left in FIG. 6A) in comparison to isoluminant chromatic stimuli (on the right in FIG. 6A) that gave very weak to no response. In comparison with the control group all responses were lower and slower. Three months after, a general increase of the signal was observed in all cases. Importantly, in some subjects from the medium-, and high-dose groups (blue column on the right of FIG. 6B), the amplitude of the MRI signals reached levels similar to those observed in response to luminance contrast stimulation (grey column on the left of FIG. 6B). FIG. 6C (before treatment) and 6D (after treatment) show similar data as 6A and 6B, respectively; however, here the fMRI signals are resolved over time (in s) after presenting the stimuli (contrast stimulus in grey, isoluminant colour stimuli in blue). The peak of the signal is reached after approximately 10 s. patients (representative case from the intermediate dose group).

(41) An increase of the responses to intermediate and higher spatial frequencies was also observed (not shown). These results are an indication of brain activation with isoluminant chromatic stimuli not resolved without cone function and also point to brain plasticity after AAV8.hCNGA3 gene therapy and provide the first evidence of successfully activating cone-related brain pathways in these patients (representative case from the intermediate dose group).

(42) 9. Nucleic Acid Sequences

(43) The following nucleotide and amino acid sequences are identified in the sequence listing. hArr3 promoter nucleotide sequence: SEQ ID No. 1 hCNGA3 nucleotide sequence: SEQ ID No. 2 hCNGA3 amino acid sequence: SEQ ID No. 3 WPREm nucleotide sequence: SEQ ID No. 4 BGH pA nucleotide sequence: SEQ ID No. 5 L-ITR nucleotide sequence: SEQ ID No. 6 R-ITR nucleotide sequence: SEQ ID No. 7