Compositions useful in the treatment of CDKL5 deficiency (CDD)

Abstract

Recombinant adeno-associated viruses (rAAVs) are provided that have an AAV capsid and a vector genome that includes a nucleic acid sequence encoding a functional CDKL5 (hCDKLK5). Also provided are production systems useful for producing the rAAVs, pharmaceutical compositions that include the rAAVs, and methods for treating a subject with CDD, or alleviating the symptoms of CDD, or slowing the progression of CDD, via administering an effective amount of the rAAV to a subject in need thereof.

Description

[Technical field]

[0001] CDKL5 deficiency (CDD) is a severe neurodevelopmental disorder that affects young children. The underlying cause is lack of CDKL5 protein expression due to mutations in the X-linked cyclin-dependent kinase-like 5 gene, CDKL5 (Mendelian Inheritance in Man, MIM:300203, formerly known as STK9), which has been shown to cause EIEE2 (MIM:300672), a form of early infantile epileptic encephalopathy [Bahi-Buisson, N. et al. Key clinical features to identify girls with CDKL5 mutations. Brain 131, 2647-2661, doi:10.1093 / brain / awn197(2008)], and infantile spasms [Fehr, S. et al. Eur J Hum Genet 21, 266-273, doi:10.1038 / ejhg.2012.156(2013); Kalscheuer, VM et al., American Journal of Human Genetics, 72, 1401-1411, doi:10.1086 / 375538(2003); Tao, J. et al. American Journal of Human Genetics 75, 1149-1154, doi:10.1086 / 426460(2004); Weaving, LSet al. American Journal of Human Genetics 75, 1079-1093, doi:10.1086 / 426462(2004)]. In addition to the characteristic early onset of seizures, the phenotype may also include several other features such as stereotypical hand movements, severe psychomotor retardation and general hypotonia. The early postnatal onset of symptoms indicates that CDKL5 plays an important role in brain development. CDKL5 is also expressed within the mature adult nervous system. CDKL5 is expressed throughout the cell, including the nucleus, as well as the cytoplasm of the cell body and dendrites.

[0002] CDKL5 gene mutations are responsible for most cases of CDD, a progressive neurodevelopmental disorder and one of the most common causes of cognitive impairment in women. Males with the gene mutation that causes CDD are devastated; most of them die before birth or in early infancy. See, e.g., ninds.nih.gov / Disorders / Patient-Caregiver-Education / Fact-Sheets / Rett-Syndrome-Fact-Sheet and omim.org / entry / 312750.

[0003] Currently, there is no cure for CDD, and treatments focus on alleviating disease symptoms. Because seizures are often poorly controlled, there is an urgent medical need to find novel therapeutic approaches. Summary of the Invention

[0004] Provided herein is a recombinant adeno-associated virus (rAAV) useful for treating CDKL5 deficiency (CDD) in a subject in need of such treatment. The rAAV has a vector genome that includes an inverted terminal repeat (ITR) and a novel nucleic acid sequence that encodes a functional human CDKL5 protein under the control of a regulatory sequence that directs hCDKL5 expression in a target cell.

[0005] In certain embodiments, a recombinant adeno-associated virus (rAAV) useful for treating CDD is provided. The rAAV comprises: (a) an AAVhu68 or AAVrh91 capsid; and (b) a vector genome in the AAV capsid of (a), the vector genome comprising a 5′AAV inverted terminal repeat (ITR), an expression cassette comprising the human CDKL5 sequence of nucleotides 1-2883 of SEQ ID NO:22 operably linked to a regulatory sequence directing its expression and further comprising four tandem miR183 targeting sequences, and a 3′AAV ITR. In certain embodiments, the regulatory sequence further comprises a UbC promoter or an hSyn promoter. In certain embodiments, the UbC promoter has the sequence of SEQ ID NO:52. In certain embodiments, the expression cassette comprises the nucleic acid sequence of nucleotides 220 to 4609 of SEQ ID NO:49 (or SEQ ID NO:50), the nucleic acid sequence of nucleotides 226 to 4608 of SEQ ID NO:29 (or SEQ ID NO:59), or the nucleic acid sequence of nt 224 to 4191 of SEQ ID NO:31 (or SEQ ID NO:60). In certain embodiments, the AAV capsid is an AAVhu68 capsid. In certain embodiments, the vector genome comprises an AAV 5'ITR, a UbC promoter, a Kozak sequence, a hCDKL5 coding sequence, four miR183 targeting sequences in the 3'UTR of the hCDKL5 coding sequence, a rabbit globin polyA signal, and an AAV 3'ITR. In certain embodiments, at least one miR183 targeting sequence has the sequence AGTGAATTCTACCAGTGCCATA (miR183, SEQ ID NO:11). In certain embodiments, two, three, or four of the miR183 targeting sequences have SEQ ID NO: 11. In certain embodiments, the four miR183 targeting sequences are located in tandem in the 3'UTR and separated by a spacer sequence.

[0006] In certain embodiments, a composition is provided that comprises a stock of rAAV described herein and an aqueous suspension medium.

[0007] In certain embodiments, a method of treating CDD is provided comprising administering to a subject in need thereof an effective amount of a rAAV described herein.

[0008] In certain embodiments, an rAAV production system useful for producing the vectors described herein is provided.

[0009] In a further aspect, provided herein is a composition comprising a rAAV or vector described herein and an aqueous suspension medium.

[0010] In another aspect, a method of treating a subject with CDD, or alleviating symptoms of CDD, or delaying the progression of CDD is provided. The method comprises administering to a subject in need thereof an effective amount of a rAAV or vector described herein. In certain embodiments, the vector or rAAV can be administered to the patient via intracisternal injection (ICM).

[0011] These and other aspects of the invention will become apparent from the following detailed description of the invention. [Brief description of the drawings]

[0012] [Figure 1A] 1 shows the AAV vector design of the AAV CDKL5 vector genome, including the 5' AAV inverted terminal repeat (ITR), an expression cassette containing a human synapsin neuronal promoter, an engineered human CDKL5 DNA coding sequence, a woodchuck hepatitis virus post-transcriptional regulatory element (WPRE), and polyA, and the AAV 3' ITR. [Figure 1B] 1 shows the AAV vector design of the AAV CDKL5 vector genome, which includes a 5' AAV ITR, an expression cassette containing the human ubiquitin C (UbC) promoter, an engineered human CDKL5 DNA coding sequence, a drg-miRNA to reduce drg expression, polyA, and the AAV 3' ITR. [Figure 1C]1 shows the AAV vector design of the AAV CDKL5 vector genome, including the 5' AAV ITR, an expression cassette containing the chicken beta-actin hybrid promoter (CBh), an engineered human CDKL5 DNA coding sequence, miRNA sequence, polyA, and the AAV 3' ITR. [Figure 2A] Analysis of mouse hippocampus assessed with anti-CDKL5 antibody (S957D, University of Dundee, UK). Mice were treated with 5x1010GC AAV-hSyn-CDKL5-1co.WPRE by neonatal intracerebroventricular injection. CDKL5 expression by Western blotting charting CDKL5 / tubulin levels in PBS-injected wild-type mice, PBS-injected KO mice, and treated KO mice. [Figure 2B] Analysis of mouse hippocampus assessed with anti-CDKL5 antibody (S957D, University of Dundee, UK). Mice were treated with 5x1010GC AAV-hSyn-CDKL5-1co.WPRE by neonatal intracerebroventricular injection. CDKL5 activity as determined using pS222EB2 (Baltussen et al,2018 Chemical genetic identification of CDKL5 substrates reveals its role in neuronal microtubule dynamics,EMBO J,37:e99763) levels in PBS-injected wild-type mice, PBS-injected KO mice, and treated KO mice are shown. [Diagram 3] This provides a graph of mouse maturation and survival after injection of AAV-hSyn-hCDKL5-1co.WPRE in a CDD mouse model. All injected mice survived and gained weight after treatment. The mice showed no obvious signs of adverse outcomes. [Figure 4]Results from behavioral assessment in CDD mice receiving AAV-hSyn-CDKL5-1co.WPRE are provided. The figure shows the results of the Elevated Zero Maze, which assesses balance between risk taking, curiosity, and anxiety. The first bar represents wt mice receiving PBS. Wt mice are curious but cautious and spend limited time in the open. The middle bar shows Cdkl5-ko mice receiving PBS only, which shows reduced anxiety and spends more time in the open; these mice entered the open zone more frequently. After AAV-CDKL5 treatment, the behavior of Cdkl5-ko mice reverts towards that of wt (time in the open zone and for entries from the closed zone to the open zone). [Figure 5A] Figure 1 provides results from behavioral assessment in CDD mice receiving AAV-hSyn-CDKL5-1co.WPRE. Shows the exploratory activity of mice in an open field arena plotted as beam breaks / bin versus time (min). Dotted line indicates wt mice, which are curious but explore the arena within 10 min and settle down. Line with long dashes indicates Cdkl5-ko spend more time exploring but eventually settle down. Dot-dash line indicates that after AAV-CDKL5 treatment, activity of Cdkl5-ko mice is reduced and their overall activity level is similar to wt mice. [Figure 5B] Figure 5A provides results from behavioral assessment in CDD mice receiving AAV-hSyn-CDKL5-1co.WPRE. Cumulative activity data of total beam breaks acquired by the mice are shown, confirming the results in Figure 5A. [Figure 6]Figure 1 shows graphs of measured latency to fall (seconds) in locomotor activity and agility assessment (rotarod) in mice over three consecutive days. Wild type mice are observed to increase their performance while learning over time. Cdkl5-ko mice show improved performance compared to wt mice, likely due to the initial hyperactivity previously observed. After AAV-hSyn-CDKL5-1co.WPRE treatment, the behavior of Cdkl5-ko mice approaches that of WT mice and matches the performance of WT mice after two days of learning. [Figure 7A] Figure 1 shows the results of hippocampal learning and memory (Y-maze). The percentage of spontaneous changes of the test group and the two control groups are shown. [Figure 7B] Results of hippocampal learning and memory (Y-maze). Distance traveled (m) for the test and two control groups is shown. wt mice show a strong tendency to explore maze arms that have not been recently visited (spontaneous change behavior), whereas Cdkl5-ko mice have a lower tendency for this memory-dependent behavior. After gene therapy, performance shows a trend toward improvement. [Figure 8A] 1 shows CDKL5 expression or activity levels of AAV.CDLK5 vector constructs for expressed isoform 1, isoform 2, isoform 3, or isoform 4. Quantified expression levels of CDKL5 isoforms 1, 2, 3, and 4 in knockout mice injected with AAV vector (5x1010 GC, neonatal ICV) compared to vehicle-injected wild-type mice and vehicle-injected knockout mice. [Figure 8B] 1 shows CDKL5 expression or activity levels of AAV.CDLK5 vector constructs for expression isoform 1, isoform 2, isoform 3, or isoform 4. Shown is CDKL5 activity as determined from quantified signal of pS222EB2 from Western blot analysis of tissues from treated wild type mice (injected with vehicle (PBS)), knockout mice (injected with vehicle or AAV.CDKL5-1co). [Figure 8C]Shown are CDKL5 expression or activity levels of AAV.CDLK5 vector constructs for expression isoform 1, isoform 2, isoform 3, or isoform 4. Shown are CDKL5 activity as determined from the quantified signal of pS222EB2 from Western blot analysis of tissues from treated wild type mice (injected with vehicle (PBS)), knockout mice (injected with vehicle, or AAV.CDKL5-isoform 1, 2, 3, or 4 (from FIG. 8A)). [Figure 8D] Shown are CDKL5 expression or activity levels of AAV.CDLK5 vector constructs for expressed isoform 1, isoform 2, isoform 3, or isoform 4. Shown are quantified CDKL5 expression levels of isoform 1 in KO mice (from FIG. 8B). [Figure 9A] 1 shows the therapeutic efficacy of AAV.CDKL5 gene therapy in a mouse study comparing different vector doses (5×10 GC, 2.5×10 GC, 1×10 GC, and 6×10 GC) in knockout and wild-type mice. Weight gain (g) over 10 weeks in mice treated with AAV.CDKL5 at a dose of 5×10 GC or PBS is shown. [Figure 9B] 1 shows the therapeutic efficacy of AAV.CDKL5 gene therapy in a mouse study comparing different vector doses (5×10 GC, 2.5×10 GC, 1×10 GC, and 6×10 GC) in knockout and wild-type mice. Weight gain (g) over 10 weeks in mice treated with AAV.CDKL5 at a dose of 2.5×10 GC or PBS is shown. [Figure 9C] 1 shows the therapeutic efficacy of AAV.CDKL5 gene therapy in a mouse study comparing different vector doses (5×10 GC, 2.5×10 GC, 1×10 GC, and 6×10 GC) in knockout and wild-type mice. Dose-dependent results of the hindlimb clasping test of the AAV.CDKL5-treated group at a dose of 5×10 GC compared to untreated Cdkl5-ko mice are shown. [Figure 9D]1 shows the therapeutic efficacy of AAV.CDKL5 gene therapy in a mouse study comparing different vector doses (5×10 GC, 2.5×10 GC, 1×10 GC, and 6×10 GC) in knockout and wild-type mice. Dose-dependent results of the hindlimb clasping test of the AAV.CDKL5-treated group at a dose of 2.5×10 GC compared to untreated Cdkl5-ko mice are shown. [Figure 9E] 1 shows the therapeutic efficacy of AAV.CDKL5 gene therapy in a mouse study comparing different vector doses (5×10 GC, 2.5×10 GC, 1×10 GC, and 6×10 GC) in knockout and wild-type mice. Dose-dependent results of the hindlimb clasping test of the AAV.CDKL5-treated group at a dose of 1×10 GC compared to untreated Cdkl5-ko mice are shown. [Figure 9F] Figure 1 shows the therapeutic efficacy of AAV.CDKL5 gene therapy in a mouse study comparing different vector doses (5x1010GC, 2.5x1010GC, 1x1010GC, and 6x109GC) in knockout and wild-type mice. Figure 2 shows dose-dependent results of hindlimb clasping assay of AAV.CDKL5 treated group at a dose of 6x109GC compared to untreated Cdkl5-ko mice. WT mice showed no clasping and KO mice showed significant clasping. After treatment, KO mice showed significantly reduced clasping among clasping. Injected WT mice were unaffected. [Figure 10A] 1 shows the therapeutic efficacy of AAV.CDL5 gene therapy in CDD ko mice study. Results are shown for nest construction (nesting quality / score) in AAV.CDKL5 treated groups at a dose of 5×10 10 GC compared to untreated Cdkl5-ko mice. [Figure 10B] 1 shows the therapeutic efficacy of AAV.CDL5 gene therapy in a CDD ko mouse study. Results from the glass ball burying task are shown with a trend towards normalization in the AAV.CDLK5 treatment group at a dose of 5×10 GC compared to WT and Cdkl5-ko mice. [Figure 10C]1 shows the therapeutic efficacy of AAV.CDL5 gene therapy in CDD ko mice study. Results are shown for nest construction (nesting quality / score) in AAV.CDKL5 treated groups at a dose of 2.5×10 10 GC compared to untreated Cdkl5-ko mice. [Figure 10D] 1 shows the therapeutic efficacy of AAV.CDL5 gene therapy in a CDD ko mouse study. Results from the glass ball burying task are shown with a trend towards normalization in the AAV.CDLK5 treatment group at a dose of 2.5×10 10 GC compared to WT and Cdkl5-ko mice. [Figure 10E] Figure 10 shows the therapeutic efficacy of AAV.CDL5 gene therapy in CDD ko mice study. Results in nest construction (nest quality / score) of AAV.CDKL5 treatment group at a dose of 1x1010GC compared to untreated Cdkl5-ko mice are shown. Figure 10F shows the results of nest construction test in male Cdkl5KO / Y mice and female Cdkl5KO / X mice after ICV administration of AAV vector expressing human CDKL5, plotted as a percentage of intact original nestlet weight. [Figure 10F] 1 shows the results of nest building assays in male Cdkl5KO / Y and female Cdkl5KO / X mice following ICV administration of an AAV vector expressing human CDKL5, plotted directly as a percentage of intact original nestlet weight. [Figure 11A] Figure 1 shows correction of hyperactivity in AAV.CDKL5-treated ko mice as assessed by open field activity test. Figure 2 shows ambulatory activity / bin vs. time (5 min intervals for 30 min) in AAV.CDKL5-treated ko mice at a dose of 5x1010 GC. [Figure 11B] 1 shows correction of hyperactivity in AAV.CDKL5 treated ko mice as assessed by open field activity test. Total activity in AAV.CDKL5 treated ko mice at a dose of 5×10 10 GC is shown. [Figure 11C]Figure 1 shows correction of hyperactivity in AAV.CDKL5-treated ko mice as assessed by open field activity test. Figure 2 shows ambulatory activity / bin vs. time (5 min intervals for 30 min) in AAV.CDKL5-treated ko mice at a dose of 2.5x1010 GC. [Figure 11D] 1 shows correction of hyperactivity in AAV.CDKL5 treated ko mice as assessed by open field activity test. Total activity in AAV.CDKL5 treated ko mice at a dose of 2.5×10 10 GC is shown. [Figure 11E] Figure 1 shows correction of hyperactivity in AAV.CDKL5-treated ko mice as assessed by open field activity test. Figure 2 shows ambulatory activity / bin vs. time (5 min intervals for 30 min) in AAV.CDKL5-treated ko mice at a dose of 6x109 GC. [Figure 11F] Figure 1 shows correction of hyperactivity in AAV.CDKL5 treated ko mice as assessed by open field activity test. Figure 2 shows total activity in AAV.CDKL5 treated ko mice at a dose of 6x109GC. Normalization of increased risk taking is observed in the elevated zero maze and normalization of hippocampal learning deficits is seen in the Y maze in AAV.CDKL5 treated ko mice. [Figure 12] We show that expression of CDKL5 isoforms 2-4 provides a significant correction of the hindlimb clasping phenotype when assessed at a dose of 5x1010 GC in ko mice. [Figure 13A] Shows a strong trend towards correction in KO mice treated with AAV.CDKL5-isoform 1. Shows elevated activity in KO mice treated with AAV.CDKL5-isoform 1 at a dose of 5x10 10 GC. [Figure 13B] 1 shows a strong trend towards correction in KO mice treated with AAV.CDKL5-isoform 1. FIG. 2 shows elevated activity in KO mice treated with AAV.CDKL5-isoform 1 at a dose of 2.5×10 10 GC. [Figure 13C]A strong trend towards correction in KO mice treated with AAV.CDKL5-isoform 1. Y-maze activity in KO mice treated with AAV.CDKL5-isoform 1 at a dose of 5x1010 GC. [Figure 13D] A strong trend towards correction in KO mice treated with AAV.CDKL5-isoform 1. Y-maze activity in KO mice treated with AAV.CDKL5-isoform 1 at a dose of 2.5x1010 GC. [Figure 14A] 1 shows gender-specific results in hindlimb clasping following treatment of knockout mice with AAV.CDKL5-isoform 1. FIG. 2 shows hindlimb clasping following treatment of male knockout mice with AAV.CDKL5-isoform 1. [Figure 14B] Figure 1 shows gender-specific results in hindlimb clasping after treatment of knockout mice with AAV.CDKL5-isoform 1. Figure 2 shows hindlimb clasping after treatment of female knockout mice with AAV.CDKL5-isoform 1. In Cdkl5-ko mice, both hemizygous males and heterozygous females showed hindlimb clasping, which was significantly reduced after treatment. None of the WT groups showed clasping. [Figure 14C] Figure 1 shows gender-specific results in hindlimb clasping following treatment of knockout mice with AAV.CDKL5-isoform 1. Locomotor activity at the high dose (5x1010GC, neonatal ICV) is shown, with a marked improvement in female heterozygous mice. [Figure 15A] 1 shows the results of the open field test in male Cdkl5KO / Y mice and female Cdkl5KO / X mice after ICV administration of an AAV vector expressing human CDKL5. 2 shows the results of the horizontal activity open field test in males plotted as X / Y axis beam breaks. [Figure 15B]1 shows the results of the open field test in male Cdkl5KO / Y mice and female Cdkl5KO / X mice after ICV administration of an AAV vector expressing human CDKL5. 2 shows the results of the horizontal activity open field test in females plotted as X / Y axis beam breaks. [Figure 15C] 1 shows the results of the open field test in male Cdkl5KO / Y mice and female Cdkl5KO / X mice following ICV administration of an AAV vector expressing human CDKL5. 2 shows the results of the rearing open field test in males plotted as Z-axis beam break. [Figure 15D] 1 shows the results of the open field test in male Cdkl5KO / Y mice and female Cdkl5KO / X mice after ICV administration of an AAV vector expressing human CDKL5. 2 shows the results of the rearing open field test in females plotted as Z-axis beam break. [Figure 15E] Figure 1 shows the results of the open field test in male Cdkl5KO / Y and female Cdkl5KO / X mice following ICV administration of an AAV vector expressing human CDKL5. Figure 2 shows the results of the central activity open field test in males plotted as central beam break percentage. [Figure 15F] Figure 1 shows the results of the open field test in male Cdkl5KO / Y and female Cdkl5KO / X mice following ICV administration of an AAV vector expressing human CDKL5. Figure 2 shows the results of the central activity open field test in females plotted as central beam break percentage. [Figure 16A] Figure 1 shows gender differences in ko mice treated with AAV.CDKL5-isoform1 vectors. Figure 1 shows the results of open field-locomotor activity in the elevated zero maze assessment in male (KO) mice treated with AAV.CDKL5-isoform1 plotted as time spent in the open zone (sec). [Figure 16B]Gender differences in ko mice treated with AAV.CDKL5-isoform1 vectors. Open field-locomotor activity results in the elevated zero maze assessment in female (ht) mice treated with AAV.CDKL5-isoform1 plotted as time spent in the open zone (sec). Risk-prone behavior was modified, with a more pronounced size effect in males. [Figure 17] Vector distribution in various tissue samples from a NHP study is provided (representative of a 1x1014 GC dose). The graph provides rAAV.CDKL5 in the gc / diploid genome of various non-neural tissues, spinal tracts, and brain tissue. Strong transduction of dorsal root ganglia (DRG) is observed. Moderate to low transduction of brain tissue is observed, with some leakage into non-neural tissues. [Figure 18] 4 provides the results of quantification of hCDKL5 expression (measured by RT-qPCR) in NHP studies shown in the cerebellum, frontal cortex, occipital cortex, parietal cortex, and temporal cortex. [Figure 19A] 1 shows the results of a dose escalation study measuring behavioral changes following administration of CDKL5 gene therapy to WT mice, showing no significant change in hindlimb clasp severity scores in WT mice injected with 7.5x1010 GC and 1x1011 GC AAV compared to PBS-treated control mice. [Figure 19B] 1 shows the results of a dose escalation study measuring behavioral changes following administration of CDKL5 gene therapy to WT mice, showing no significant changes in locomotor activity in WT mice injected with 7.5×10 10 GC and 1×10 11 GC AAV compared to PBS-treated control mice. [Figure 20] Shown is CDKL5 expression, as determined qualitatively by Western blot, 14 days after administration of either AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183 via neonatal ICV at a dose of 3x1010GC. [Figure 21A]Mice were administered AAVrh91.UbC.CDKL5-1co.miR183 via neonatal ICV at doses of 3x1010, 6x1010GC and CDKL5 expression as measured qualitatively by Western blotting at 4 months of age. [Figure 21B] Mice were administered AAVrh91.CBh.CDKL5-1co.miR183 via neonatal ICV at doses of 3x1010, 1x1010GC, and CDKL5 expression as measured qualitatively by Western blotting at 4 months of age is shown. [Figure 21C] Shown is CDKL5 expression quantified from Western blot analysis, charted as CDKL5 / tubulin levels, in wild-type and knockout mice administered AAVrh91.UbC.CDKL5-1co.miR183 at doses of 3x1010, 6x1010GC via neonatal ICV compared to AAVhu68.hSyn-CDKL5 at a dose of 5x1010GC. [Figure 22A] Representative images from immunofluorescence microscopy analysis of CDKL5 expression following administration of AAVrh91.UbC.CDKL5-1co.miR183 at a dose of 3x1010 GCs via neonatal ICV are shown. [Figure 22B] Representative images from immunofluorescence microscopy analysis of CDKL5 expression following administration of AAVrh91.CBh.CDKL5-1co.miR183 at a dose of 3x1010 GCs via neonatal ICV are shown. [Figure 23A] Representative images (enlarged) from immunofluorescence microscopy analysis of CDKL5 expression (samples were also probed for NeuN, a neuronal marker) following administration of AAVrh91.UbC.CDKL5-1co.miR183 at a dose of 3 x 1010 GCs via neonatal ICV are shown. [Figure 23B] Representative images (enlarged) from immunofluorescence microscopy analysis of CDKL5 expression (samples were also probed for NeuN, a neuronal marker) following administration of AAVrh91.CBh.CDKL5-1co.miR183 at a dose of 3 x 10 GCs via neonatal ICV are shown. [Figure 24]Quantification of CDKL5 expressing neurons (plotted as percent CDKL5 in positively identified neurons above background levels) following administration of AAVhu68.hSyn.CDKL5 compared to previous results is shown. [Figure 25A] Figure 1 shows the results of a survival study using enumeration survival rates at postnatal day 16 (PND16) for mice administered AAVrh91.CBh.CDKL5-1co.miR183 at doses of 1x1010, 3x1010, and 6x1010 GCs via neonatal ICV. Figure 1 shows the results of a survival study using enumeration survival rates at postnatal day 16 (PND16) for mice administered AAVrh91.CBh.CDKL5-1co.miR183 at doses of 1x1010 GCs via neonatal ICV. [Figure 25B] Figure 1 shows the results of a survival study using enumeration survival rates at postnatal day 16 (PND16) for mice administered AAVrh91.CBh.CDKL5-1co.miR183 at doses of 1x1010, 3x1010, and 6x1010 GCs via neonatal ICV. Figure 1 shows the results of a survival study using enumeration survival rates at postnatal day 16 (PND16) for mice administered AAVrh91.CBh.CDKL5-1co.miR183 at doses of 3x1010 GCs via neonatal ICV. [Figure 25C] Figure 1 shows the results of a survival study using enumeration survival rates at postnatal day 16 (PND16) for mice administered AAVrh91.CBh.CDKL5-1co.miR183 at doses of 1x1010, 3x1010, and 6x1010 GCs via neonatal ICV. Figure 1 shows the results of a survival study using enumeration survival rates at postnatal day 16 (PND16) for mice administered AAVrh91.CBh.CDKL5-1co.miR183 at doses of 6x1010 GCs via neonatal ICV. [Figure 26A]Figure 1 shows the severity scores observed in DRG neurons from collected tissues of the cervical, thoracic, and lumbar regions from NHPs treated with AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183 vectors at a dose of 3x1010 GCs via the ICM route. When scored by a board-certified veterinary pathologist, a score of 0 indicates no signs of toxicity and a score of 5 indicates severe toxicity. Scores below 0.5 are considered background based on similar evaluation of normal tissues. [Figure 26B] 1 shows the severity scores observed in spinal cord neurons from cervical, thoracic, and lumbar tissue collected from NHPs treated with AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183 vectors at a dose of 3×10 10 GCs via the ICM route. [Figure 26C] Shown are the severity scores observed in the sural nerve from proximal and distal harvested tissues from NHPs treated with AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183 vectors at a dose of 3x1010 GCs via the ICM route. [Figure 27] 1 shows the results of vector copy number plotted as GC / diploid genome in various tissues of NHPs after administration of AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183 vectors. [Figure 28] Relative expression of CDKL5 plotted per 100 ng of cDNA in various CNS tissues of NHPs (motor cortex, somatosensory (som.Sens) cortex, parietal cortex, hippocampus, thalamus) compared to the results observed in mouse brain. [Figure 29A]Shown is CDKL5 expression quantified from Western blotting analysis charted as CDKL5 / tubulin levels in knockout mice administered AAVrh91.UbC.CDKL5-1co.miR183 at a dose of 3x1010 GCs compared to WT and knockout mice treated with PBS (control group). [Figure 29B] Kinase activity quantified from Western blotting analysis charted as pEB2pS222 / total EB2 levels in knockout mice administered AAVrh91.UbC.CDKL5-1co.miR183 at a dose of 3x1010 GC compared to WT and knockout mice treated with PBS (control group). [Diagram 30] Kinase activity, as measured qualitatively by Western blotting (using pEB-S222 antibody, Baltussen at al., Chemical genetic identification of CDKL5 substrates reveals its role in neuronal microtubule dynamics, 2018, EMBO J, 37:e99763), is shown in knockout mice administered AAVrh91.UbC.CDKL5-1co.miR183 at a dose of 3x1010 GC compared to WT and knockout mice (control group) treated with PBS. [Figure 31A] Shown are the results of percent neurons with CDKL5 protein expression in mouse cortical and hippocampal tissues following neonatal ICV administration of AAVrh91.UbC.CDKL5-1co.miR183 at doses of 1x1010, 3x1010, and 6x1010 GCs compared to PBS-treated WT. [Figure 31B] Representative microscopy images from immunofluorescence staining with DAPI (nuclei), CDKL5, and NeuN (neuronal marker) of cortical section tissue following neonatal ICV administration of AAVrh91.UbC.CDKL5-1co.miR183 at a dose of 3x1010 GCs are shown. [Diagram 32]Analysis of measured body weights of wild type and CDKL5-ko upon administration of PBS or AAV.UbC.CDKL5-1co.miR183 at doses of 1x1010, 3x1010, and 6x1010 GC is shown. [Figure 33A] Figure 1 shows the results of the hind paw clasping test for the AAV.UbC.CDKL5-1co.miR183 treatment group at a dose of 3x1010 GC compared to the untreated group in Cdkl5-ko and WT mice. Statistically significant improvement from ko is shown (*p<0.05, **p<0.01). [Figure 33B] Figure 1 shows dose-dependent effects on hyperactivity measured in the open field activity test and plotted as locomotor activity (beam breaks) in Cdkl5-ko and WT mice following administration of AAV.UbC.CDKL5-1co.miR183 at doses of 1x1010, 3x1010, and 6x1010 GCs. Statistically significant improvement from ko is shown (*p<0.05, **p<0.01). [Figure 34A] Figure 1 shows binned locomotor activity results for groups of WT and Cdkl5-ko mice administered AAV.UbC.CDKL5-1co.miR183 at a low dose of 1x1010 GC. Statistically significant improvement from ko is shown (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001). [Figure 34B] Binned locomotor activity results are shown for groups of WT and Cdkl5-ko mice administered AAV.UbC.CDKL5-1co.miR183 at a medium dose of 3x1010 GC. [Figure 34C] Binned locomotor activity results are shown for groups of WT and Cdkl5-ko mice administered AAV.UbC.CDKL5-1co.miR183 at a high dose of 6x1010 GCs. [Diagram 35] Shown are nest construction (nesting quality / score) results for AAV.UbC.CDKL5-1co.miR183-treated WT and Cdkl5-ko mice with doses of 1x1010, 3x1010, and 6x1010 GC. [Figure 36A]A schematic diagram of the intracisternal catheter (ICM) administration technique is shown. [Figure 36B] A more detailed overview of ICM administration as a fluorescence-guided procedure is provided. [Figure 37A] FIG. 13 shows analysis of brain transduction as measured by vector genome copies via qPCR of DNA / RNA extracted from different brain regions of NHPs following administration of AAVrh91.UbC.CDKL5-1co.miR183. [Figure 37B] Relative CDKL5 transgene expression (mRNA) measured via qPCR of RNA extracted from different NHP brain regions following administration of AAVrh91.UbC.CDKL5-1co.miR183 (compared to expression in mouse brain when administered at a dose of 3x1010 GC). [Figure 38A] FIG. 1 shows the results of a molecular analysis of CDKL5 gene therapy outcome on a single neuron basis plotted as the percentage of transduced neurons as measured by vector genome copies. [Figure 38B] Shown are CDKL5 transgene expression levels measured from detectable CDKL5 transgene mRNA in single neurons plotted as percentage of transgene-expressing neurons. [Figure 39A] FIG. 1 shows the results of the elevated zero maze test in male Cdkl5KO / Y mice following ICV administration of an AAV vector expressing human CDKL5, plotted as time in the open zone (seconds). [Figure 39B] FIG. 1 shows the results of the elevated zero maze test in female Cdkl5KO / X mice following ICV administration of an AAV vector expressing human CDKL5, plotted as time in the open zone (seconds). [Figure 40A] FIG. 1 shows the results of the elevated zero maze test in male Cdkl5KO / Y mice following ICV administration of an AAV vector expressing human CDKL5, plotted as open zone entries. [Figure 40B]FIG. 1 shows the results of the elevated zero maze test in female Cdkl5KO / X mice following ICV administration of an AAV vector expressing human CDKL5, plotted as open zone entries. [Figure 41A] Shown are the results of the elevated zero maze test in male Cdkl5KO / Y mice following ICV administration of an AAV vector expressing human CDKL5, plotted as total distance traveled. [Figure 41B] FIG. 1 shows the results of the elevated zero maze test in female Cdkl5KO / X mice following ICV administration of an AAV vector expressing human CDKL5, plotted as total distance traveled. [Figure 42A] 1 shows the results of the Y-maze test in male Cdkl5KO / Y mice following ICV administration of an AAV vector expressing human CDKL5 plotted as percent spontaneous change. [Figure 42B] 1 shows the results of the Y-maze test in female Cdkl5KO / X mice following ICV administration of an AAV vector expressing human CDKL5 plotted as percent spontaneous change. [Figure 43A] FIG. 1 shows the results of a contextual fear conditioning test in male Cdkl5KO / Y mice following ICV administration of an AAV vector expressing human CDKL5 plotted as percent freezing. [Figure 43B] FIG. 1 shows the results of a contextual fear conditioning test in female Cdkl5KO / X mice following ICV administration of an AAV vector expressing human CDKL5 plotted as percent freezing. [Figure 44A] 1 shows the results of transgene product expression in male Cdkl5KO / Y mice and female Cdkl5KO / X mice following ICV administration of an AAV vector expressing human CDKL5 (CDKL5 / tubulin). [Figure 44B] Shown are activity results in male Cdkl5KO / Y and female Cdkl5KO / X mice following ICV administration of an AAV vector expressing human CDKL5 (pS222 / total EB2). [Figure 45A]13 shows vector biodistribution results in adult rhesus macaques following ICM administration of AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 at a low dose. [Figure 45B] 13 shows vector biodistribution results in adult rhesus macaques following ICM administration of AAVhu68.UbC.hCDKL5-1co.SV40 at low doses. [Figure 45C] 1 shows vector biodistribution results in adult rhesus macaques following ICM administration of AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 at a medium dose. [Figure 45D] 13 shows vector biodistribution results in adult rhesus macaques following ICM administration of AAVhu68.UbC.hCDKL5-1co.SV40 at a medium dose. [Figure 45E] 1 shows vector biodistribution results in adult rhesus macaques following ICM administration of AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 at a high dose. [Fig.45F] 13 shows vector biodistribution results in adult rhesus macaques following ICM administration of high dose AAVhu68.UbC.hCDKL5-1co.SV40. [Figure 46] Figure 1 shows the results of transgene product expression in adult rhesus macaque brains following ICM administration of an AAV vector expressing human CDKL5. Adult (3-10 years old) male and female rhesus macaques received a single ICM administration of either AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 or AAVhu68.UbC.hCDKL5-1co.SV40 at low dose (3.0x1012 GC), medium dose (1.0x1013 GC), or high dose (3.0x1013 GC) (N=1 animal per vector per dose). [Figure 47A] FIG. 1 shows a flow diagram of the upstream manufacturing process for the drug substance. [Figure 47B] FIG. 1 shows a flow diagram of the downstream manufacturing process for the drug substance. [Figure 48] 1 shows an overview of the manufacturing process flow diagram for the drug substance. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Compositions and methods for treating CDD are provided herein: An effective amount of a recombinant adeno-associated virus (rAAV) having an AAV capsid (e.g., AAVhu68) and having a vector genome packaged therein encoding a functional human cyclin-dependent kinase-like 5 (hCDKL5) is delivered to a subject in need thereof.

[0014] I. Human CDKL5 The cyclin-dependent kinase-like 5 (CDKL5, also known as CFAP247, serine / threonine kinase 9, STK9; Uniprot#076039) gene is naturally located on the short (p) arm of the X chromosome at position 22.13. The N-terminus of the CDLK5 protein acts as a kinase, an enzyme that alters the activity of other proteins. Several direct substrates of CDKL5 have been identified (Baltussen et al., 2018; Munoz et al., 2018). The C-terminus of CDKL5 is of unknown function.

[0015] As used herein, functional hCDKL5 protein refers to an isoform, natural variant, variant, polymorph, or truncation of CKDL5 protein that is not associated with CDD and / or whose delivery or expression may alleviate symptoms or delay progression of CDD in animal models or patients. See OMIM#300203, each webpage is incorporated herein by reference in its entirety. In certain embodiments, functional hCDKL5 has an amino acid sequence of SEQ ID NO:2 (isoform 1), or an amino acid sequence at least about 90% (e.g., at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 99.9%) identical thereto. In certain embodiments, a functional hCDKL5 protein has an amino acid sequence of SEQ ID NO: 19 (isoform 2), or an amino acid sequence at least about 90% (e.g., at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 99.9%) identical thereto. In certain embodiments, a functional hCDKL5 protein has an amino acid sequence of SEQ ID NO: 20 (isoform 3), or an amino acid sequence at least about 90% (e.g., at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 99.9%) identical thereto. In certain embodiments, the functional hCDKL5 protein has an amino acid sequence of SEQ ID NO:21 (isoform 4), or an amino acid sequence at least about 90% (e.g., at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 99.9%) identical thereto. In certain embodiments, the functional hCDKL5 is a truncated hCDKL5 that includes a methyl-CpG binding domain (MBD) and an NCoR / SMRT interaction domain (NID) having the sequence:

[0016] In certain embodiments, functional hCDKL5 protein ameliorates or slows the progression of CDD symptoms in an animal model. One exemplary animal model is the CDKL5-ko mouse. Other suitable models are described herein.

[0017] Symptoms or progression of CDD may be assessed using a variety of assays / methods, including, but not limited to, survival plots (e.g., Kaplan-Meier survival plots), monitoring of body weight, and observation of behavioral changes (e.g., hindlimb clasping, open field assay (motor function), elevated zone maze (anxiety / risk vs. exploration), Y-maze (learning and memory / hippocampus), glass ball burying assay (innate behavior and motor), nest building (innate social behavior), and rotarod assay (motor function, integrity). In certain embodiments, administration or expression of functional hCDKL5 protein in animal models is associated with increased phenotype in corresponding wild-type animals. leads to an alleviation of CDD symptoms or a delay in CDD progression as indicated by an assay result that is at least about 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, 100%, or greater than 100% of that obtained. In certain embodiments, administration or expression of a functional hCDKL5 protein in a CDD animal model leads to an alleviation of CDD symptoms or a delay in CDD progression as indicated by improved assay results that are at least about 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, 100%, or greater than 100% of that obtained from a corresponding untreated CDD animal.

[0018] Provided herein is a nucleic acid sequence encoding a functional hCDKL5 protein, referred to herein as the hCDKL5 coding sequence or CDKL5 coding sequence. In certain embodiments, the hCDKL5 coding sequence is SEQ ID NO:3, or a sequence that is at least about 95% identical to SEQ ID NO:3. In certain embodiments, the hCDKL5 coding sequence is SEQ ID NO:2 (referred to as CDKL5 or CDKL5co or CDKL5-1 or CDKL5-1co), or the NCBI reference sequence NM_001037343.1 (referred to as CDKL5 or CDKL5e1, SEQ ID NO:16), which encodes the amino acid sequence NP_001310218.1 (SEQ ID NO:19), No. 20), and NM_003159.2 (SEQ ID NO: 18) which encodes the amino acid sequence NP_003150.1 (SEQ ID NO: 21), or a nucleic acid sequence at least about 70% (e.g., at least about 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 99.9%) identical thereto. Each of the NCBI reference sequences is incorporated herein by reference in its entirety. In certain embodiments, the hCDKL5 coding sequence is modified or engineered (hCDKL5 or hCDKL5co or CDKL5-1 or CDKL5-1co). The modified or engineered sequences share less than about 70% (e.g., about 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 99.9%) identity to the NCBI reference sequence.

[0019] In certain embodiments, the hCDKL5 coding sequence is SEQ ID NO:22, or a nucleic acid sequence at least about 70% identical thereto (e.g., at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, or at least about 99.9%). In certain embodiments, the hCDKL5 coding sequence is SEQ ID NO:24, or a nucleic acid sequence at least about 70% identical thereto (e.g., at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, or at least about 99.9%). In certain embodiments, the hCDKL5 coding sequence is SEQ ID NO:25, or a nucleic acid sequence at least about 70% identical thereto (e.g., at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, or at least about 99.9%). In certain embodiments, the hCDKL5 coding sequence is SEQ ID NO:26, or a nucleic acid sequence at least about 70% identical thereto (e.g., at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, or at least about 99.9%).

[0020] In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:37, or a sequence at least about 95% identical to SEQ ID NO:37. In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:38, or a sequence at least about 95% identical to SEQ ID NO:38. In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:39, or a sequence at least about 95% identical to SEQ ID NO:39. In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:40, or a sequence at least about 95% identical to SEQ ID NO:40. In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:41, or a sequence at least about 95% identical to SEQ ID NO:41. In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:42, or a sequence at least about 95% identical to SEQ ID NO:42. In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:43, or a sequence at least about 95% identical to SEQ ID NO:43. In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:44, or a sequence that is at least about 95% identical to SEQ ID NO:44. In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:45, or a sequence that is at least about 95% identical to SEQ ID NO:45. In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:46, or a sequence that is at least about 95% identical to SEQ ID NO:46. In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:47, or a sequence that is at least about 95% identical to SEQ ID NO:47. In certain embodiments, the hCDKL5 coding sequence is an engineered sequence of SEQ ID NO:48, or a sequence that is at least about 95% identical to SEQ ID NO:48.

[0021] A "nucleic acid" as described herein may be RNA, DNA, or modifications thereof, may be single-stranded or double-stranded, and may be selected from the group including, for example, nucleic acids encoding proteins of interest, oligonucleotides, nucleic acid analogs, such as peptide nucleic acids (PNAs), pseudocomplementary PNAs (pc-PNAs), locked nucleic acids (LNAs), etc. Such nucleic acid sequences include, but are not limited to, nucleic acid sequences encoding, for example, proteins, that act as, for example, transcription repressors, antisense molecules, ribozymes, hypo-inhibitory nucleic acid sequences, such as, but not limited to, RNAi, shRNAi, siRNA, microRNAi (mRNAi), antisense oligonucleotides, etc.

[0022] The terms "percent (%) identity," "sequence identity," "percent sequence identity," or "percent identical" in the context of nucleic acid sequences refer to residues in two sequences that are the same when aligned for correspondence. The length of sequence identity comparison may be over the entire length of a genome, the entire length of a gene coding sequence, or a fragment of at least about 500-5000 nucleotides, which is preferred. However, identity between smaller fragments, e.g., of at least about 9 nucleotides, usually at least about 20-24 nucleotides, at least about 28-32 nucleotides, at least about 36 or more nucleotides, may also be desired.

[0023] Percent identity can be readily determined for amino acid sequences spanning the entire length of a protein, polypeptide, about 32 amino acids, about 330 amino acids, or peptide fragments thereof, or the corresponding nucleic acid sequence encoding the sequence. Suitable amino acid fragments can be at least about 8 amino acids in length and can be up to about 700. Generally, when referring to "identity", "homology", or "similarity" between two different sequences, the "identity", "homology", or "similarity" is determined with reference to an "aligned" sequence. An "aligned" sequence or "alignment" refers to multiple nucleic acid or protein (amino acid) sequences, often including modifications for missing or additional bases or amino acids, as compared to a reference sequence.

[0024] Alignment is performed using a variety of multiple sequence alignment programs, either public or commercially available. Sequence alignment programs are available for amino acid sequences, including, for example, "Clustal X", "Clustal Omega", "MAP", "PIMA", "MSA", "BLOCKMAKER", "MEME", and "Match-Box" programs. Generally, any of these programs are used with default settings, but those skilled in the art may change these settings as needed. Alternatively, those skilled in the art may utilize other algorithms or computer programs, providing at least the level of identity or alignment as provided by the referenced algorithms and programs. See, for example, JD Thomson et al, Nucl. Acids. Res., "A comprehensive comparison of multiple sequence alignments", 27(13):2682-2690 (1999).

[0025] Multiple sequence alignment programs are also available for nucleic acid sequences. Examples of such programs include "Clustal W", "Clustal Omega", "CAP Sequence Assembly", "BLAST", "MAP" and "MEME" and are accessible through web servers on the Internet. Other sources of such programs are known to those skilled in the art. Alternatively, the Vector NTI utility is also used. There are also several algorithms known in the art that can be used to measure nucleotide sequence identity, including those included in the programs described above. As another example, polynucleotide sequences can be compared using Fasta™, a program in GCG version 6.1. Fasta™ provides alignment and percent sequence identity of the best overlapping regions between the query sequence and the search sequence. For example, percent sequence identity between nucleic acid sequences can be determined using Fasta™ with its default parameters (word size 6 and NOPAM factor for the scoring matrix) as provided in GCG version 6.1 (herein incorporated by reference).

[0026] II. Expression Cassettes Provided herein is a nucleic acid sequence comprising a hCDKL5 coding sequence under the control of regulatory sequences directing hCDKL5 expression in a target cell, also referred to as an expression cassette. As used herein, "expression cassette" refers to a nucleic acid molecule comprising a coding sequence (e.g., CDKL5 coding sequence) and a regulatory sequence operably linked thereto. In certain embodiments, a vector genome comprises two or more expression cassettes. The term "transgene" refers to a DNA sequence from an exogenous source that is inserted into a target cell, typically a transgene encoding a product (e.g., CDKL5). Typically, such an expression cassette packaged into a viral vector comprises a coding sequence for a gene product as described herein adjacent to a packaging signal of the viral genome, and other expression control sequences such as those described herein. The necessary regulatory sequences are operably linked to the hCDKL5 coding sequence in a manner that permits its transcription, translation and / or expression in the target cell. As used herein, "operably linked" sequences include sequences that regulate transcription, translation, and / or expression contiguous with the hCDKL5 coding sequence, as well as regulatory sequences that act in trans or at a distance to control the hCDKL5 coding sequence. The expression cassette may contain, among other elements, regulatory sequences upstream (5') of the gene sequence, such as one or more of a promoter, enhancer, intron, etc., and enhancer, or regulatory sequences downstream (3') of the gene sequence, such as one or more of a 3' untranslated region (3'UTR) including a polyadenylation site. Such regulatory sequences typically include, for example, one or more of a promoter, enhancer, intron, Kozak sequence, polyadenylation sequence, and TATA signal. In certain embodiments, the promoter is a tissue-specific promoter, such as a CNS-specific or neuron-specific promoter. In certain embodiments, the promoter is the human synapsin promoter (SEQ ID NO: 23).In certain embodiments, additional or alternative neuron-specific promoter sequences may be selected from the neuron-specific enolase (NSE) promoter (Andersen et al., (1993) Cell. Mol. Neurobiol., 13:503 15), the neurofilament light chain gene promoter (Piccioli et al., (1991) Proc. Natl. Acad. Sci. USA, 88:5611 5), the neuron-specific vgf gene promoter (Piccioli et al., (1995) Neuron, 15:373 84), and / or others.

[0027] In certain embodiments, the human synapsin promoter has the sequence (eg, nt 213 to nt 678) of SEQ ID NO: 1, 3, 5, 7, 9, or SEQ ID NO: 23 (also referred to herein as hSyn or Syn).

[0028] In other embodiments, the promoter is a constitutive promoter, such as chicken β-actin promoter with cytomegalovirus enhancer (CB7) promoter, human elongation initiation factor 1 alpha promoter (EF1a) promoter, human ubiquitin C (UbC) promoter. In certain embodiments, the regulatory sequence directs the expression of hCDKL5 in central nervous system (CNS) cells. In certain embodiments, the UbC promoter comprises the nucleic acid sequence of SEQ ID NO:52.

[0029] In certain embodiments, the target cell may be a central nervous system cell. In certain embodiments, the target cell is one or more of an excitatory neuron, an inhibitory neuron, a glial cell, a cortical cell, a frontal cortical cell, a cerebral cortical cell, and a spinal cord cell. In certain embodiments, the target cell is a peripheral nervous system (PNS) cell, such as a retinal cell. Also, cells other than cells from the nervous system may be selected as target cells, such as monocytes, B lymphocytes, T lymphocytes, NK cells, lymph node cells, tonsillar cells, bone marrow mesenchymal cells, stem cells, bone marrow stem cells, cardiac cells, epithelial cells, esophageal cells, stomach cells, fetal transection cells, colon cells, rectal cells, liver cells, kidney cells, lung cells, salivary gland cells, thyroid cells, adrenal cells, breast cells, pancreatic cells, islet cells of Langerhans, gallbladder cells, prostate cells, bladder cells, skin cells, uterine cells, cervical cells, testicular cells, or any other cells that express functional CDKL5 protein in subjects without CDD.

[0030] In certain embodiments, additional or alternative promoter sequences may be included as part of the expression control sequence (regulatory sequence), for example, located between the selected 5'ITR sequence and the coding sequence. Constitutive promoters, regulatable promoters [see, e.g., WO2011 / 126808 and WO2013 / 04943], tissue-specific promoters, or promoters responsive to physiological cues may be utilized in the vectors described herein. The promoter(s) may be selected from different sources, for example, human cytomegalovirus (CMV) immediate early enhancer / promoter, SV40 immediate early enhancer / promoter, JC polymomavirus promoter, myelin basic protein (MBP) or glial fibrillary acidic protein (GFAP) promoter, herpes simplex virus (HSV-1) latency-associated promoter (LAP), Rous sarcoma virus (RSV) long terminal repeat (LTR) promoter, neuron-specific promoter (NSE), platelet-derived growth factor (PDGF) promoter, hSYN, melanin-concentrating hormone (MCH) promoter, CBA, matrix metalloprotein promoter (MPP), and chicken beta-actin promoter.

[0031] In addition to the promoter, the vector may include one or more other suitable transcription initiation sequences, transcription termination sequences, enhancer sequences, efficient RNA processing signals such as splicing and polyadenylation (polyA) signals; sequences that stabilize cytoplasmic mRNA (e.g., WPRE); sequences that enhance translation efficiency (i.e., Kozak consensus sequences); sequences that enhance protein stability; and, optionally, sequences that enhance secretion of the encoded product. One example of a suitable enhancer is the CMV enhancer. Other suitable enhancers include those suitable for the desired target tissue indication. In one embodiment, the regulatory sequence includes one or more expression enhancers. In one embodiment, the regulatory sequence contains two or more expression enhancers. These enhancers may be the same or different from each other. For example, the enhancer may include a CMV immediate early enhancer. This enhancer may be present in two copies located adjacent to each other. Alternatively, the double copy of the enhancer is separated by one or more sequences. In yet another embodiment, the expression cassette further contains an intron, for example, a chicken beta-actin intron. In certain embodiments, the intron is a chimeric intron (CI), which is a hybrid intron consisting of a human β-globin splice donor and an immunoglobulin G (IgG) splice acceptor element. Other suitable introns include those known in the art, such as those described in WO2011 / 126808. Examples of suitable polyA sequences include, for example, rabbit globin polyA, SV40, SV50, bovine growth hormone (bGH), human growth hormone, and synthetic polyA. In certain embodiments, the polyA sequence is an SV40 polyA sequence. In certain embodiments, the polyA sequence is a rabbit beta globin (RBG or rbg or rBG) polyA sequence. In certain embodiments, the polyA is a rabbit beta globin polyA comprising the nucleic acid sequence of SEQ ID NO:53. Optionally, one or more sequences may be selected to stabilize the mRNA.The provided expression cassettes may include one or more expression enhancers, such as post-transcriptional regulatory elements from woodchuck (WPRE), human (HPRE), ground squirrel (GPRE), or arctic ground squirrel (AGSPRE) hepatitis virus, or synthetic post-transcriptional regulatory elements. These expression enhancing elements are particularly advantageous when placed in the 3'UTR and can significantly increase mRNA stability and / or protein yield. In certain embodiments, the provided expression cassettes include a regulatory sequence that is a woodchuck hepatitis virus post-transcriptional regulatory element (WPRE) or a variant thereof. Suitable WPRE sequences are provided in the vector genomes described herein and are known in the art (e.g., as described in U.S. Pat. Nos. 6,136,597, 6,287,814, and 7,419,829, which are incorporated by reference). In certain embodiments, the WPRE is a variant mutated to eliminate expression of the woodchuck hepatitis B virus X (WHX) protein, including, for example, a mutation in the start codon of the WHX gene (see Zanta-Boussif et al., Gene Ther. 2009 May;16(5):605-19, incorporated by reference). In other embodiments, the enhancer is selected from a non-viral source. In certain embodiments, the WPRE sequence is absent.

[0032] In certain embodiments, the expression cassette refers to a nucleic acid molecule having a sequence of nt 213 to 4439 of SEQ ID NO: 1, which encodes the amino acid sequence of hCDKL5 (isoform 1, SEQ ID NO: 2). In certain embodiments, the expression cassette refers to a nucleic acid molecule having a sequence of nt 213 to 4562 of SEQ ID NO: 5, which encodes the amino acid sequence of hCDKL5 (isoform 2 or 2GS, SEQ ID NO: 6). In certain embodiments, the expression cassette refers to a nucleic acid molecule having a sequence of nt 213 to 4388 of SEQ ID NO: 7, which encodes the amino acid sequence of hCDKL5 (isoform 3 or 3GS, SEQ ID NO: 8). In certain embodiments, the expression cassette refers to a nucleic acid molecule having a sequence of nt 213 to 4511 of SEQ ID NO: 9, which encodes the amino acid sequence of hCDKL5 (isoform 4 or 4GS, SEQ ID NO: 10). In certain embodiments, the expression cassette comprises an engineered nucleic acid sequence selected from SEQ ID NO: 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, or 48, and encoding the amino acid sequence of hCDKL5 (isoform 1, SEQ ID NO: 2). In certain embodiments, the expression cassette refers to a nucleic acid molecule having a sequence of nt 213 to 4555 of SEQ ID NO: 3, encoding the amino acid sequence of hCDKL5 (isoform 1, SEQ ID NO: 4) and consisting of miRNA183 (SEQ ID NO: 11). In certain embodiments, these expression cassettes further comprise one, two, three, four, or more miRNA sequences for reducing drg expression. In certain embodiments, the expression cassette refers to a nucleic acid molecule having a sequence of nt 226 to 4608 of SEQ ID NO: 29 (or SEQ ID NO: 59), which encodes the amino acid sequence of hCDKL5 (isoform 1, SEQ ID NO: 30) and is composed of four tandem repeats of miRNA183 (SEQ ID NO: 11). In certain embodiments, the expression cassette refers to a nucleic acid molecule having a sequence of nt 220 to 4609 of SEQ ID NO: 49 (or SEQ ID NO: 50), which encodes the amino acid sequence of hCDKL5 (isoform 1) and is composed of four tandem repeats of miRNA183 (SEQ ID NO: 11).In certain embodiments, the expression cassette refers to a nucleic acid molecule having a sequence of nt 224 to 4191 of SEQ ID NO: 31 (or SEQ ID NO: 60), encoding the amino acid sequence of hCDKL5 (isoform 1, SEQ ID NO: 32) and composed of four tandem repeats of miRNA183 (SEQ ID NO: 11). See, e.g., PCT / US19 / 67872, filed December 20, 2019, now published as WO2020 / 132455.

[0033] In certain embodiments, the expression cassette comprises four copies of the miR183 expression cassette. In certain embodiments, the expression cassette comprises a miR-183 target sequence comprising AGTGAATTCTACAGTGCCATA (SEQ ID NO: 11), with the sequence complementary to the miR-183 seed sequence underlined. In certain embodiments, the expression cassette comprises two or more copies (e.g., two or three copies) of a sequence that is 100% complementary to the miR-183 seed sequence. In certain embodiments, the miR-183 target sequence is from about 7 nucleotides to about 28 nucleotides in length and comprises at least one region that is at least 100% complementary to the miR-183 seed sequence. In certain embodiments, the miR-183 target sequence comprises a sequence that is partially complementary to SEQ ID NO: 11, and thus has one or more mismatches when aligned to SEQ ID NO: 11. In certain embodiments, the miR-183 target sequence comprises a sequence having at least 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 mismatches when aligned with SEQ ID NO:11, and the mismatches can be non-contiguous. In certain embodiments, the miR-183 target sequence comprises a region of 100% complementarity and comprises at least 30% of the length of the miR-183 target sequence. In certain embodiments, the region of 100% complementarity comprises a sequence having 100% complementarity with the miR-183 seed sequence. In certain embodiments, the remainder of the miR-183 target sequence has at least about 80% to about 99% complementarity with miR-183. In certain embodiments, the expression cassette comprises a truncated SEQ ID NO:11, i.e., a sequence lacking at least 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 nucleotides at either or both of the 5' or 3' ends of SEQ ID NO:11. In certain embodiments, the expression cassette comprises a transgene and one miR-183 target sequence. In yet other embodiments, the expression cassette comprises at least two, three, or four miR-183 target sequences. In certain embodiments, the inclusion of two, three, or four miR-183 target sequences in the expression cassette increases the level of transgene expression in a target tissue, such as the heart.

[0034] In one embodiment, the expression cassette comprises a UbC promoter, a hCDKL5-1 coding sequence, four copies of a miR183 targeting sequence, and a polyA sequence. In another embodiment, the expression cassette comprises a hSyn promoter, a hCDKL5-1 coding sequence, four copies of a miR183 targeting sequence, and a polyA sequence. In another embodiment, the expression cassette comprises a CBh promoter, a hCDKL5-1 coding sequence, four copies of a miR183 targeting sequence, and a polyA sequence. In certain embodiments, the expression cassette further comprises at least one intron and / or at least one enhancer sequence. In certain embodiments, the enhancer is a mutant WPRE element lacking the ability to express woodchuck hepatitis B virus X (WHX) protein.

[0035] In certain embodiments, the vector genome comprises a 5'-AAV ITR sequence, a spacer sequence, an expression cassette as described herein, a spacer sequence, and a 3'-AAV ITR. Suitably, there may be a non-coding spacer sequence between the 5'ITR sequence and the 5' end of the expression cassette, and there may be a non-coding spacer sequence between the 3' end of the ITR sequence and the 3'ITR.

[0036] In certain embodiments, the expression cassette comprises the nucleic acid sequence from nt 220 to 4609 of SEQ ID NO: 49 (or SEQ ID NO: 50). In certain embodiments, the expression cassette comprises the nucleic acid sequence from nt 226 to 4608 of SEQ ID NO: 29 (or SEQ ID NO: 59). In certain embodiments, the expression cassette comprises the nucleic acid sequence from nt 224 to 4191 of SEQ ID NO: 31 (or SEQ ID NO: 60).

[0037] III. rAAV Provided herein is a recombinant adeno-associated virus (rAAV) useful for treating CDD. The rAAV comprises (a) an AAV capsid, and (b) a vector genome packaged in the AAV capsid of (a). Suitably, the selected AAV capsid targets the cells to be treated. In certain embodiments, the capsid is from clade F. However, in certain embodiments, another AAV capsid source may be selected. The vector genome comprises an inverted terminal repeat (ITR) and a nucleic acid sequence encoding a functional human cyclin-dependent kinase-like 5 (hCDKL5) under the control of a regulatory sequence directing hCDKL5 expression. In certain embodiments, CDKL5 may refer to CDKL5 or hCDKL5, CDKL5-2GS or hCDKL5-2GS, CDKL5-3GS or hCDKL5-3GS, and CDKL5-4GS or hCDKL5-4GS. In certain embodiments, the hCDKL5 coding sequence is at least about 95% identical to SEQ ID NO:22 (encoding the amino acid sequence of CDKL5-1 or hCDKL5-1, SEQ ID NO:2). In certain embodiments, the hCDKL5 coding sequence is at least about 95% identical to SEQ ID NO:24 (encoding the amino acid sequence of CDKL5-2GS or hCDKL5-2GS, SEQ ID NO:6). In certain embodiments, the hCDKL5 coding sequence is at least about 95% identical to SEQ ID NO:25 (encoding the amino acid sequence of CDKL5-3GS or hCDKL5-3GS, SEQ ID NO:8). In certain embodiments, the hCDKL5 coding sequence is at least about 95% identical to SEQ ID NO:26 (encoding the amino acid sequence of CDKL5-4GS or hCDKL5-4GS, SEQ ID NO:10). In certain embodiments, the hCDKL5 coding sequence is less than 80% identical to any one of hCDKL5 transcript variants 1-3 (NM_001037343.1 having SEQ ID NO: 16, which encodes the amino acid sequence NP_001032420.1 having SEQ ID NO: 19; NM_001323289.2 having SEQ ID NO: 17, which encodes the amino acid sequence NP_001310218.1 having SEQ ID NO: 20; NM_003159.2 having SEQ ID NO: 18, which encodes the amino acid sequence NP_003150.1 having SEQ ID NO: 21).In certain embodiments, the hCDKL5 coding sequence is, or is at least about 95% identical to, SEQ ID NO: 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, and 47 (encoding the amino acid sequence of CDKL5-1 or hCDKL5-1, SEQ ID NO: 2). In certain embodiments, the functional hCDKL5 has the amino acid sequence of SEQ ID NO: 2 (CDKL5-1 or hCDKL5-1). In certain embodiments, the functional hCDKL5 has the amino acid sequence of SEQ ID NO: 6 (CDKL5-2GS or hCDKL5-2GS). In certain embodiments, the functional hCDKL5 has the amino acid sequence of SEQ ID NO: 8 (CDKL5-3GS or hCDKL5-3GS). In certain embodiments, the functional hCDKL5 has the amino acid sequence of SEQ ID NO: 10 (CDKL5-4GS or hCDKL5-4GS). In certain embodiments, the regulatory sequence induces hCDKL5 expression in central nervous system cells. In certain embodiments, the regulatory sequence comprises a human synapsin promoter (hSyn) or a CB7 promoter. In certain embodiments, the regulatory sequence comprises a human ubiquitin C (hUbC or UbC) promoter. In certain embodiments, the regulatory element comprises one or more of a Kozak sequence, a polyadenylation sequence, an intron, an enhancer, and a TATA signal. In certain embodiments, the vector genome further comprises at least two tandem repeats of a dorsal root ganglion (drg)-specific miRNA target sequence, the at least two tandem repeats comprising at least a first miRNA target sequence and at least a second miRNA target sequence, which may be the same or different. In certain embodiments, the vector genome is a nucleic acid sequence that is at least about 70% (e.g., at least about 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 99.9%) identical to nt1 through nt4634 of SEQ ID NO:1, or nt1 through nt4750 of SEQ ID NO:3, or nt1 through nt4757 of SEQ ID NO:5, or nt1 through nt4583 of SEQ ID NO:7, or nt1 through nt4706 of SEQ ID NO:9.

[0038] In certain embodiments, vector genome refers to a nucleic acid molecule comprising SEQ ID NO:1, which encodes the amino acid sequence of hCDKL5 (isoform 1, SEQ ID NO:2). In certain embodiments, vector genome refers to a nucleic acid molecule comprising SEQ ID NO:5, which encodes the amino acid sequence of hCDKL5 (isoform 2 or 2GS, SEQ ID NO:6). In certain embodiments, vector genome refers to a nucleic acid molecule comprising SEQ ID NO:7, which encodes the amino acid sequence of hCDKL5 (isoform 3 or 3GS, SEQ ID NO:8). In certain embodiments, vector genome refers to a nucleic acid molecule comprising SEQ ID NO:9, which encodes the amino acid sequence of hCDKL5 (isoform 4 or 4GS, SEQ ID NO:10). In certain embodiments, vector genome refers to a nucleic acid molecule comprising SEQ ID NO:3, which encodes the amino acid sequence of hCDKL5 (isoform 1, SEQ ID NO:4) and which comprises miRNA183 (SEQ ID NO:11). In certain embodiments, vector genome refers to a nucleic acid molecule comprising SEQ ID NO: 29, encoding the amino acid sequence of hCDKL5 (isoform 1, SEQ ID NO: 30) and comprising tandem repeats of miRNA183 (SEQ ID NO: 11). In certain embodiments, vector genome refers to a nucleic acid molecule comprising SEQ ID NO: 31, encoding the amino acid sequence of hCDKL5 (isoform 1, SEQ ID NO: 32) and comprising tandem repeats of miRNA183 (SEQ ID NO: 11).

[0039] In certain embodiments, in addition to the hCDKL5 coding sequence, another non-AAV coding sequence may be included, such as a peptide, polypeptide, protein of interest, a functional RNA molecule (e.g., miRNA, miRNA inhibitor), or other gene product. Useful gene products may include miRNAs. miRNAs and other small interfering nucleic acids regulate gene expression through cleavage / degradation of target RNA transcripts or translational repression of target messenger RNAs (mRNAs). miRNAs are typically naturally expressed as final 19-25 untranslated RNA products. miRNAs exhibit their activity through sequence-specific interactions with the 3' untranslated region (UTR) of target mRNAs. These endogenously expressed miRNAs form hairpin precursors, which are then processed into miRNA duplexes and further processed into "mature" single-stranded miRNA molecules. This mature miRNA directs the multiprotein complex miRISC, which identifies target sites (e.g., in the 3'UTR region) of target mRNAs based on complementarity to the mature miRNA.

[0040] As used herein, a "miRNA target sequence" is a sequence located on the DNA plus strand (5' to 3') that is at least partially complementary to a miRNA sequence, including a miRNA seed sequence. The miRNA target sequence is exogenous to the untranslated region of the encoded transgene product and is designed to be specifically targeted by the miRNA in cells where suppression of transgene expression is desired. The term "miR183 cluster target sequence" refers to a target sequence that responds to one or more members of the miR183 cluster (alternatively referred to as family), including miR-183, -96, and -182 (described by Dambal, S. et al. Nucleic Acids Res 43:7173-7188, 2015, which is incorporated herein by reference).

[0041] Typically, the miRNA target sequence is at least 7 nucleotides to about 28 nucleotides in length, at least 8 nucleotides to about 28 nucleotides in length, 7 nucleotides to 28 nucleotides, 8 nucleotides to 18 nucleotides, 12 nucleotides to 28 nucleotides in length, about 20 nucleotides to about 26 nucleotides, about 22 nucleotides, about 24 nucleotides, or about 26 nucleotides, and contains at least one contiguous region (e.g., 7 or 8 nucleotides) that is complementary to the miRNA seed sequence. In certain embodiments, the target sequence comprises a sequence that has exact complementarity (100%) with the miRNA seed sequence, or a sequence that has partial complementarity with some mismatches with the miRNA seed sequence. In certain embodiments, the target sequence comprises at least 7 to 8 nucleotides that are 100% complementary to the miRNA seed sequence. In certain embodiments, the target sequence consists of a sequence that is 100% complementary to the miRNA seed sequence. In certain embodiments, the target sequence comprises multiple copies (e.g., 2 or 3 copies) of a sequence that is 100% complementary to the seed sequence. In certain embodiments, the region of 100% complementarity comprises at least 30% of the length of the target sequence. In certain embodiments, the remainder of the target sequence has at least about 80% to about 99% complementarity to the miRNA. In certain embodiments, in an expression cassette comprising a DNA positive strand, the miRNA target sequence is the reverse complement of the miRNA.

[0042] As used herein, a "miRNA target sequence" is a sequence located on the DNA plus strand (5' to 3') that is at least partially complementary to a miRNA sequence, including a miRNA seed sequence. The miRNA target sequence is exogenous to the untranslated region of the encoded transgene product and is designed to be specifically targeted by the miRNA in cells where suppression of transgene expression is desired. The term "miR183 cluster target sequence" refers to a target sequence that responds to one or more members of the miR183 cluster (alternatively referred to as family), including miR-183, -96 and -182 (described by Dambal, S. et al. Nucleic Acids Res 43:7173-7188, 2015, which is incorporated herein by reference). Without intending to be bound by theory, the messenger RNA (mRNA) of the transgene (which encodes a gene product) is present in the cell type into which the expression cassette containing the miRNA is delivered, such that specific binding of the miRNA to its target sequence in the 3'UTR miRNA results in silencing and cleavage of the mRNA, thereby reducing or eliminating expression of the transgene only in cells that express the miRNA.

[0043] Typically, the miRNA target sequence is at least 7 nucleotides to about 28 nucleotides in length, at least 8 nucleotides to about 28 nucleotides in length, 7 nucleotides to 28 nucleotides, 8 nucleotides to 18 nucleotides, 12 nucleotides to 28 nucleotides in length, about 20 nucleotides to about 26 nucleotides, about 22 nucleotides, about 24 nucleotides, or about 26 nucleotides, and contains at least one contiguous region (e.g., 7 or 8 nucleotides) that is complementary to the miRNA seed sequence. In certain embodiments, the target sequence comprises a sequence that has exact complementarity (100%) with the miRNA seed sequence, or a sequence that has partial complementarity with some mismatches with the miRNA seed sequence. In certain embodiments, the target sequence comprises at least 7 to 8 nucleotides that are 100% complementary to the miRNA seed sequence. In certain embodiments, the target sequence consists of a sequence that is 100% complementary to the miRNA seed sequence. In certain embodiments, the target sequence comprises multiple copies (e.g., 2 or 3 copies) of a sequence that is 100% complementary to the seed sequence. In certain embodiments, the region of 100% complementarity comprises at least 30% of the length of the target sequence. In certain embodiments, the remainder of the target sequence has at least about 80% to about 99% complementarity to the miRNA. In certain embodiments, in an expression cassette comprising a DNA positive strand, the miRNA target sequence is the reverse complement of the miRNA.

[0044] In certain embodiments, the miRNA target sequence of at least the first and / or at least the second miRNA target sequence of the expression cassette mRNA or DNA positive strand is selected from (i) AGTGAATTCTAACCAGTGCCATA (miR183, SEQ ID NO: 11), or (ii) AGTGTGAGTTCTAACCATTGCCAAA (miR182, SEQ ID NO: 13). In other embodiments, AGGGATTCCTGGGAAAACTGGAC (SEQ ID NO: 14) is selected.

[0045] In certain embodiments, the vector genome or expression cassette comprises at least one miRNA target sequence that is a target sequence of miR-183. In certain embodiments, the vector genome or expression cassette comprises a miR-183 target sequence comprising AGTGAATTCTACAGTGCCATA (SEQ ID NO: 11), where the sequence complementary to the miR-183 seed sequence is GTGCCAT). In certain embodiments, the vector genome or expression cassette comprises two or more copies (e.g., two or three copies) of a sequence that is 100% complementary to the miR-183 seed sequence. In certain embodiments, the miR-183 target sequence is about 7 nucleotides to about 28 nucleotides in length and comprises at least one region that is at least 100% complementary to the miR-183 seed sequence. In certain embodiments, the miR-183 target sequence comprises a sequence that is partially complementary to SEQ ID NO: 11, and thus has one or more mismatches when aligned to SEQ ID NO: 11. In certain embodiments, the miR-183 target sequence comprises a sequence having at least 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 mismatches when aligned with SEQ ID NO: 11, and the mismatches can be non-contiguous. In certain embodiments, the miR-183 target sequence comprises a region of 100% complementarity and comprises at least 30% of the length of the miR-183 target sequence. In certain embodiments, the region of 100% complementarity comprises a sequence having 100% complementarity with the miR-183 seed sequence. In certain embodiments, the remainder of the miR-183 target sequence has at least about 80% to about 99% complementarity with miR-183. In certain embodiments, the expression cassette or vector genome comprises a miR-183 target sequence comprising a truncated SEQ ID NO:11 (i.e., a sequence lacking at least 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 nucleotides at either or both the 5' or 3' end of SEQ ID NO:11). In certain embodiments, the expression cassette or vector genome comprises a transgene and one miR-183 target sequence. In yet other embodiments, the expression cassette or vector genome comprises at least two, three, or four miR-183 target sequences: (i) AGTGAATTCTACAGTGCCATA (miR183, SEQ ID NO:11).

[0046] In certain embodiments, the vector genome or expression cassette comprises at least one miRNA target sequence that is a miR-182 target sequence. In certain embodiments, the vector genome or expression cassette comprises a miR-182 target sequence comprising AGTGTGAGTTCTACCATTGCCAAA (SEQ ID NO: 13). In certain embodiments, the vector genome or expression cassette comprises two or more copies (e.g., two or three copies) of a sequence that is 100% complementary to the miR-182 seed sequence. In certain embodiments, the miR-182 target sequence is about 7 nucleotides to about 28 nucleotides in length and comprises at least one region that is at least 100% complementary to the miR-182 seed sequence. In certain embodiments, the miR-182 target sequence comprises a sequence that is partially complementary to SEQ ID NO: 13, and thus has one or more mismatches when aligned with SEQ ID NO: 13. In certain embodiments, the miR-183 target sequence comprises a sequence having at least 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 mismatches when aligned with SEQ ID NO: 13, and the mismatches can be non-contiguous. In certain embodiments, the miR-182 target sequence comprises a region of 100% complementarity, which also comprises at least 30% of the length of the miR-182 target sequence. In certain embodiments, the region of 100% complementarity comprises a sequence having 100% complementarity with the miR-182 seed sequence. In certain embodiments, the remainder of the miR-182 target sequence has at least about 80% to about 99% complementarity with miR-182. In certain embodiments, the expression cassette or vector genome comprises a miR-182 target sequence comprising a truncated SEQ ID NO: 13 (i.e., a sequence lacking at least 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 nucleotides at either or both the 5' or 3' end of SEQ ID NO: 13). In certain embodiments, the expression cassette or vector genome comprises a transgene and one miR-182 target sequence. In yet other embodiments, the expression cassette or vector genome comprises at least two, three, or four miR-182 target sequences.

[0047] The term "tandem repeat" is used herein to refer to the presence of two or more consecutive miRNA target sequences. These miRNA target sequences can be consecutive, i.e., directly following each other, with the 3' end of one immediately upstream of the 5' end of the next sequence, without any intervening sequence, or vice versa. In another embodiment, two or more of the miRNA target sequences are separated by a short spacer sequence.

[0048] As used herein, a "spacer" is any selected nucleic acid sequence, e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 nucleotides in length, located between two or more consecutive miRNA target sequences. In certain embodiments, the spacer is 1-8 nucleotides in length, 2-7 nucleotides in length, 3-6 nucleotides in length, 4 nucleotides in length, 4-9 nucleotides in length, 3-7 nucleotides in length, or longer. Suitably, the spacer is a non-coding sequence. In certain embodiments, the spacer may be four (4) nucleotides. In certain embodiments, the spacer is GGAT. In certain embodiments, the spacer is six (6) nucleotides. In certain embodiments, the spacer is CACGTG or GCATGC.

[0049] In certain embodiments, the tandem repeat comprises two, three, four or more of the same miRNA target sequences. In certain embodiments, the tandem repeat comprises at least two different miRNA target sequences, at least three different miRNA target sequences, or at least four different miRNA target sequences, etc. In certain embodiments, the tandem repeat may comprise two or three of the same miRNA target sequences and a different fourth miRNA target sequence.

[0050] In certain embodiments, there may be at least two different sets of tandem repeats in the expression cassette. For example, the 3'UTR may include a tandem repeat immediately downstream of the transgene, a tandem repeat immediately downstream of the UTR sequence, and two or more tandem repeats near the 3' end of the UTR. In another example, the 5'UTR may contain one, two or more miRNA target sequences. In another example, the 3' may include a tandem repeat and the 5'UTR may include at least one miRNA target sequence.

[0051] In certain embodiments, the expression cassette contains two, three, four or more tandem repeats that begin within about 0-20 nucleotides of the transgene's stop codon, hi other embodiments, the expression cassette includes miRNA tandem repeats at least 100 to about 4000 nucleotides from the transgene's stop codon.

[0052] In certain embodiments, the spacers between the miRNA target sequences are the same. As used herein, CDKL5 or hCDKL5 refers to isoform 1 unless otherwise specified. Isoforms 2-4 can be designated as CDKL5-2GS or hCDKL5-2GS, CDKL5-3GS or hCDKL5-3GS, and CDKL5-4GS or hCDKL5-4GS. Expression cassettes and vector genomes carrying these isoforms can be constructed as described for isoform 1.

[0053] See PCT / US19 / 67872 (now WO2020 / 132455), filed December 20, 2019, and U.S. Provisional Patent Application No. 63 / 023,593, filed May 12, 2020, U.S. Provisional Patent Application No. 63 / 038,488, filed June 12, 2020, U.S. Provisional Patent Application No. 63 / 043,562, filed June 24, 2020, and U.S. Provisional Patent Application No. 63 / 079,299, filed September 16, 2020, and U.S. Provisional Patent Application No. 63 / 152,042, filed February 22, 2011, which are incorporated by reference herein.

[0054] In certain embodiments, the clade F AAV capsid is selected from an AAVhu68 capsid, an AAV9 capsid, an AAVhu31 capsid, an AAVhu32 capsid, or an engineered variant of one of these capsids. A nucleic acid sequence encoding an AAVhu68 capsid protein is utilized in the following examples for the production of an AAV.hCDKL5 recombinant AAV (rAAV) having a vector genome. Additional details related to AAVhu68 are provided in WO2018 / 160582 and US2015 / 0079038, each of which is incorporated herein by reference in its entirety. The clade F vectors described herein are highly suitable for delivering vector genomes containing hCDKL5 coding sequences to cells within the central nervous system, including the brain, hippocampus, motor cortex, cerebellum, and motor neurons. These vectors can be used to target other cells within the central nervous system (CNS), as well as specific other tissues and cells outside the CNS.

[0055] In certain embodiments, the AAV capsid for the compositions and methods described herein is selected based on target cells.In certain embodiments, the AAV capsid transduces CNS cells and / or PNS cells.In certain embodiments, the AAV capsid is selected from cy02 capsid, rh43 capsid, AAV8 capsid, rh01 capsid, AAV9 capsid, rh8 capsid, rh10 capsid, bb01 capsid, hu37 capsid, rh02 capsid, rh20 capsid, rh39 capsid, rh64 capsid, AAV6 capsid, AAV1 capsid, hu44 capsid, hu48 capsid, cy05 capsid, hu11 capsid, hu32 capsid, pi2 capsid, or variations thereof. In certain embodiments, the AAV capsid is a clade F capsid, such as an AAV9 capsid, an AAVhu68 capsid, a hu31 capsid, a hu32 capsid, or a variation thereof. See, for example, WO2005 / 033321, published April 14, 2015, WO2018 / 160582, and US2015 / 0079038, each of which is incorporated herein by reference in its entirety. In certain embodiments, the AAV capsid is a non-clade F capsid, such as a clade A, B, C, D, or E capsid. In certain embodiments, the non-clade F capsid is AAV1 or a variation thereof. In certain embodiments, the AAV capsid transduces a target cell other than a nervous system cell. In certain embodiments, the AAV capsid is a clade A capsid (e.g., AAV1, AAV6, AAVrh91), a clade B capsid (e.g., AAV2), a clade C capsid (e.g., hu53), a clade D capsid (e.g., AAV7), or a clade E capsid (e.g., rh10). In certain embodiments, the AAV capsid is a clade A capsid, such as an AAVrh91 capsid (the nucleic acid sequences of SEQ ID NOs: 33 and 35).See PCT / US20 / 030266, filed April 29, 2020, now published WO2020 / 223231 (incorporated herein by reference), and U.S. Provisional Patent Application No. 63 / 065,616, filed April 29, 2019 (incorporated herein by reference). See also U.S. Provisional Patent Application No. 63 / 065,616, filed August 14, 2020, and U.S. Provisional Patent Application No. 63 / 109,734, filed November 4, 2020, and International Application No. PCT / US21 / 45945, filed August 13, 2021 (incorporated herein by reference). Nevertheless, other AAV capsids may be selected.

[0056] In certain embodiments, the AAV capsid is an AAVhu68 capsid or an AAVrh91 capsid. In certain embodiments, the AAVhu68 capsid is produced from a nucleic acid sequence encoding the amino acid sequence of SEQ ID NO:61. In certain embodiments, the AAVhu68 capsid comprises (i) an AAVhu68 vp1 protein, an AAVhu68 vp2 protein, and an AAVhu68 vp3 protein produced from a nucleic acid sequence encoding SEQ ID NO:61; or (ii) a heterogeneous population of AAVhu68 vp1, AAVhu68 vp2, and AAVhu68 vp3 proteins, wherein the AAVhu68 vp1, AAVhu68 vp2, and AAVhu68 vp3 proteins contain at least 50% to 100% deamidated asparagine (N) at each of the asparagine-glycine pairs at positions 57, 329, 452, 512, respectively, compared to the amino acid of SEQ ID NO:61, and wherein the deamidated asparagine is deamidated to aspartic acid, isoaspartic acid, an interconverted aspartic acid / isoaspartic acid pair, or a combination thereof, as determined using mass spectrometry. In certain embodiments, the nucleic acid sequence encoding the AAVhu68 vp1 protein is SEQ ID NO:57, or a sequence that is at least 80% to at least 99% identical to SEQ ID NO:57 that encodes the amino acid sequence of SEQ ID NO:61. Optionally, the nucleic acid sequence is at least 80% to 97% identical to SEQ ID NO:57. See, e.g., WO2018 / 160582 and WO2019 / 169004, which are incorporated by reference herein in their entireties.

[0057] As used herein, the term "clade" in reference to a group of AAVs refers to a group of AAVs that are phylogenetically related to each other, as determined based on an alignment of AAV vp1 amino acid sequences using a Neighbor-Joining algorithm with a bootstrap value of at least 75% (out of at least 1000 replicates) and a Poisson-corrected distance measure of 0.05 or less. Neighbor-joining algorithms have been described in the literature. See, e.g., M. Nei and S. Kumar, Molecular Evolution and Phylogenetics (Oxford University Press, New York (2000). Computer programs are available that can be used to implement this algorithm. For example, the MEGA v2.1 program implements a modified Nei-Gojobori method. Using these techniques and computer programs, and the sequence of the AAV vp1 capsid protein, one of skill in the art can readily determine whether a selected AAV falls within one clade identified herein, another clade, or is outside these clades. See, e.g., G Gao, et al, J Virol 2004 Jun;78(10):6381-6388, which identifies clades A, B, C, D, E, and F, and provides the nucleic acid sequences of novel AAVs (GenBank accession numbers AY530553-AY530629). See also WO2005 / 033321.

[0058] rAAV is composed of an AAV capsid and a vector genome. The AAV capsid is a collection of a heterogeneous population of vp1, a heterogeneous population of vp2, and a heterogeneous population of vp3 proteins. As used herein, when used to refer to vp capsid proteins, the term "heterogeneous" or any grammatical variation thereof refers to a population of non-identical elements, for example, having vp1, vp2, or vp3 monomers (proteins) with different modified amino acid sequences.

[0059] As used herein, the term "heterologous" or any grammatical variation thereof when used to refer to vp capsid proteins refers to a population of non-identical elements, e.g., having vp1, vp2, or vp3 monomers (proteins) with different modified amino acid sequences. The term "heterologous population" used in reference to vp1, vp2, and vp3 proteins (alternatively referred to as isoforms) refers to differences in the amino acid sequences of vp1, vp2, and vp3 proteins within a capsid. AAV capsids contain subpopulations within vp1, vp2, and vp3 proteins with predicted amino acid residue modifications. These subpopulations contain, at a minimum, specific deamidated asparagine (N or Asn) residues. For example, a particular subpopulation contains at least one, two, three, or four highly deamidated asparagine (N) positions in asparagine-glycine pairs, and optionally further contains other deamidated amino acids, where the deamidation results in an amino acid change and other optional modifications.

[0060] In certain embodiments, AAV capsids are provided that have a heterogeneous population of AAV capsid isoforms (i.e., VP1, VP2, VP3) that contain multiple highly deamidated "NG" positions. In certain embodiments, the highly deamidated positions are at the positions shown below with reference to the predicted full-length VP1 amino acid sequence. In other embodiments, the capsid gene is modified such that the referenced "NG" is removed and a mutant "NG" is engineered into another position.

[0061] As used herein, the terms "target cell" and "target tissue" may refer to any cell or tissue intended to be transduced by the subject AAV vector. The terms may refer to any one or more of muscle, liver, lung, airway epithelium, central nervous system, neurons, eye (visual cells), or heart.

[0062] As used herein, the term "vector genome" refers to a nucleic acid molecule that can be packaged into a viral capsid, e.g., an AAV capsid, and delivered to a host cell or patient's cell. In certain embodiments, the vector genome is an expression cassette, with inverted terminal repeat (ITR) sequences at the extreme 5' and 3' ends necessary for packaging the vector genome into an AAV capsid, and between them, the CDLK5 gene described herein operably linked to a sequence that directs its expression. In certain embodiments, the vector genome can include, at least from 5' to 3', an AAV 5' ITR, a coding sequence(s), and an AAV 3' ITR. In certain embodiments, the ITRs are from AAV2 (a different AAV source than the capsid), or other full-length ITRs can be selected. In certain embodiments, the ITRs are from the same AAV source as the AAV that provides the rep function or trans-complementing AAV during production. Additionally, other ITRs can be used. The vector genome is sometimes referred to herein as a "minigene."

[0063] As used herein, the term "host cell" can refer to a packaging cell line in which the rAAV is produced from a plasmid. Alternatively, the term "host cell" can refer to a target cell in which expression of the transgene is desired.

[0064] As shown above, rAAV is provided with an AAV capsid that targets desired cells, and a vector genome that includes at least the AAV ITRs that are required for packaging vector genome into capsid, hCDKL5 coding sequence, and the regulatory sequence that directs expression therefor.In certain embodiments, vector genome is a single-stranded AAV vector genome.In certain embodiments, rAAV vectors that include self-complementary (sc) AAV vector genomes can be used in the present invention.

[0065] The AAV sequence of the vector typically includes cis-acting 5' and 3' inverted terminal repeat (ITR) sequences (see, e.g., BJ Carter, in "Handbook of Parvoviruses", ed., P. Tijsser, CRC Press, pp. 155 168 (1990)). The ITR sequences are approximately 145 base pairs (bp) in length. Preferably, substantially complete sequences encoding the ITRs are used in the molecule, although some minimal modification of these sequences is tolerated. The ability to modify these ITR sequences is within the skill of the art. (See, e.g., Sambrook et al., "Molecular Cloning. A Laboratory Manual", 2d ed., Cold Spring Harbor Laboratory, New York (1989); and K. Fisher et al., J. Virol., 70:520 532 (1996)). One example of such a molecule utilized in the present invention is a "cis-acting" plasmid containing a transgene, in which the selected transgene sequence and associated regulatory elements are flanked by 5' and 3' AAV ITR sequences. In one embodiment, the ITRs are from a different AAV than the one supplying the capsid. In one embodiment, the ITR sequences are from AAV2. A shortened version of the 5'ITR, termed ΔITR, has been described, in which the D sequence and terminal separation site (trs) are deleted. In other embodiments, full-length AAV 5' and 3'ITRs are used. In certain embodiments, the vector genome comprises a shortened AAV2 ITR of 130 base pairs, in which the external A element is deleted. The shortened ITR is restored to the wild-type length of 145 base pairs during vector DNA amplification, using the internal A element as a template. In certain embodiments, the 5'ITR comprises the nucleic acid sequence of SEQ ID NO: 51. In certain embodiments, the 3'ITR comprises the nucleic acid sequence of SEQ ID NO: 54. However, ITRs from other AAV sources may be selected. When the source of the ITRs is from AAV2 and the AAV capsid is from another AAV source, the resulting vector may be referred to as pseudotyped, however, other configurations of these elements may also be suitable.

[0066] In certain embodiments, a vector genome is constructed that includes 5'AAV ITR-promoter-optional enhancer-optional intron-hCDKL5 coding sequence-polyA-3'ITR. In certain embodiments, a vector genome is constructed that includes 5'AAV ITR-promoter-optional enhancer-optional intron-hCDKL5 coding sequence-optionally repeating miR (de)targeting sequence-polyA-3'ITR. In certain embodiments, a vector genome is constructed that includes 5'AAV ITR-promoter-optional intron-hCDKL5 coding sequence-optional enhancer-polyA-3'ITR. In certain embodiments, a vector genome is constructed that includes 5'AAV ITR-promoter-optional enhancer-optional intron-hCDKL5 coding sequence-optional enhancer-optionally repeating miR (de)targeting sequence-polyA-3'ITR. In certain embodiments, the ITR is not from AAV2. In certain embodiments, two or more promoters are present. In certain embodiments, an enhancer is present in the vector genome. In certain embodiments, two or more enhancers are present. In certain embodiments, an intron is present in the vector genome. In certain embodiments, an enhancer and an intron are present. In certain embodiments, the polyA is SV40 polyA (i.e., a polyadenylation (PolyA) signal from a simian virus 40 (SV40) late gene). In certain embodiments, the polyA is rabbit beta globin (RBG) polyA. In certain embodiments, the vector genome comprises at least 5'AAV ITR-hSyn promoter-hCDKL5 coding sequence-polyA-3'ITR. In certain embodiments, the vector genome comprises 5'AAV ITR-CB7 promoter-hCDKL5 coding sequence-RBG polyA-3'ITR. In certain embodiments, the drg detargeting sequence is one, two, three, four, or more miR183 sequences described herein and is included in the expression cassette. In certain embodiments, the hCDKL5 coding sequence is for CDKL5.In certain embodiments, the hCDKL5 coding sequence is for CDKL5-2GS. In certain embodiments, the hCDKL5 coding sequence is for CDKL5-3GS. In certain embodiments, the hCDKL5 coding sequence is for CDKL5-4GS. Optionally, one or more of these vector genomes includes a WPRE element.

[0067] As used herein, a vector genome or a rAAV comprising a vector genome is exemplified herein as: AAV.promoter (optional).Kozak (optional).intron (optional).CDKL5 coding sequence (e.g., hCDKL5, hCDKL5co, CDKL5, CDKL5co).miRNA (optional).polyA (optional).Stuffer (optional). In certain embodiments, a rAAV is exemplified herein as: AAV capsid.promoter (optional).Kozak (optional).intron (optional).CDKL5 coding sequence.miRNA (optional).polyA (optional).Stuffer (optional). Optionally, one or more of these vector genomes comprises a WPRE element.

[0068] In certain embodiments, the vector genome comprises at least: 5'AAV ITR-ubiquitin C promoter-hCDKL5 coding sequence-RBG polyA-3'ITR. In certain embodiments, the vector genome comprises the nucleic acid sequence of SEQ ID NO: 58. In certain embodiments, the vector genome comprises at least: 5'AAV ITR-ubiquitin C promoter-hCDKL5 coding sequence-1, 2, 3, 4, or more miR183 sequences-RBG polyA-3'ITR. In certain embodiments, the vector genome comprises the nucleic acid sequence of SEQ ID NO: 29. In certain embodiments, the vector genome comprises the nucleic acid sequence of SEQ ID NO: 49. In certain embodiments, the vector genome comprises at least: 5'AAV ITR-chicken-beta actin hybrid promoter-hCDKL5 coding sequence-1, 2, 3, 4, or more miR183 sequences-RBG polyA-3'ITR. In certain embodiments, the vector genome comprises the nucleic acid sequence of SEQ ID NO: 31. Optionally, one or more of these vector genomes includes a WPRE element.

[0069] Additionally, provided herein is a rAAV production system useful for producing the rAAV described herein. The production system comprises a cell culture comprising (a) a nucleic acid sequence encoding an AAV capsid protein, (b) a vector genome, and (c) sufficient AAV rep and helper functions to permit packaging of the vector genome into an AAV capsid. In certain embodiments, the vector genome is SEQ ID NO: 1, 3, 5, 7, 9, 29, or 31. In certain embodiments, the cell culture is a human fetal kidney 293 cell culture. In certain embodiments, the AAV rep is from a different AAV. In certain embodiments, the AAV rep is from AAV2. In certain embodiments, the AAV2 rep is encoded by the nucleic acid sequence of SEQ ID NO: 56. In certain embodiments, the AAV rep coding sequence and the cap gene are on the same nucleic acid molecule, and optionally, there is a spacer between the rep sequence and the cap gene. In certain embodiments, the spacer is atgacttaaaccaggt (SEQ ID NO: 15).

[0070] For use in producing AAV viral vectors (e.g., recombinant (r)AAV), the vector genome can be carried on any suitable vector, e.g., a plasmid, that is delivered to a packaging host cell. Plasmids useful in the present invention can be engineered to be suitable for in vitro replication and packaging in prokaryotic, insect, mammalian cells, among others. Suitable transfection techniques and packaging host cells are known and / or can be readily designed by one of skill in the art.

[0071] Methods for generating and isolating AAV suitable for use as a vector are known in the art. See generally, for example, Grieger & Samulski, 2005, Adeno-associated virus as a gene therapy vector: Vector development, production and clinical applications, Adv. Biochem. Engin / Biotechnol. 99:119-145; Buning et al., 2008, Recent developments in adeno-associated virus vector technology, J. Gene Med. 10:717-733, and the references cited below, each of which is incorporated herein by reference in its entirety. As used herein, gene therapy vector refers to the rAAV described herein, suitable for use in treating patients. The ITRs are the only AAV components required in cis in the same construct as the nucleic acid molecule containing the gene in order to package the gene into virions. The cap and rep genes can be provided in trans.

[0072] In certain embodiments, the manufacturing process for rAAV includes the methods described in U.S. Provisional Patent Application No. 63 / 371,597, filed August 16, 2022, and U.S. Provisional Patent Application No. 63 / 371,592, filed August 16, 2022, which are incorporated by reference in their entireties.

[0073] In one embodiment, the selected genetic element can be delivered to the AAV packaging cell by any suitable method, including transfection, electroporation, liposome delivery, membrane fusion techniques, high-speed DNA-coated pellets, viral infection, and protoplast fusion. Suitable AAV packaging cells can also be made. Methods used to make such constructs are known to those skilled in the art of nucleic acid manipulation and include genetic engineering, recombinant engineering, and synthetic techniques. See, for example, Molecular Cloning: A Laboratory Manual, ed. Green and Sambrook, Cold Spring Harbor Press, Cold Spring Harbor, NY (2012).

[0074] The term "AAV intermediate" or "AAV vector intermediate" refers to assembled rAAV capsids that lack the desired genomic sequence packaged therein. These may also be referred to as "empty" capsids. Such capsids may contain no detectable genomic sequence of the expression cassette or may contain only partially packaged genomic sequence that is insufficient to achieve expression of the gene product. These empty capsids are non-functional for introducing a gene of interest into a host cell.

[0075] The recombinant adeno-associated virus (AAV) described herein can be produced using known techniques. See, for example, WO2003 / 042397, WO2005 / 033321, WO2006 / 110689, US7588772B2. Such methods include culturing a host cell that contains an expression cassette consisting of a nucleic acid sequence encoding an AAV capsid protein, a functional rep gene, at least an AAV inverted terminal repeat (ITR) and a transgene, and sufficient helper functions to allow packaging of the expression cassette into an AAV capsid protein. Methods for producing capsids, coding sequences therefor, and methods for producing rAAV viral vectors have been described. See, for example, Gao, et al, Proc. Natl. Acad. Sci. USA100(10), 6081-6086(2003) and US2013 / 0045186A1.

[0076] In one embodiment, a producer cell culture is provided that is useful for producing recombinant AAVhu68 or AAVrh91. Such cell culture comprises a nucleic acid that expresses AAVhu68 capsid protein in a host cell, a nucleic acid molecule suitable for packaging into an AAVhu68 capsid (e.g., a vector genome including AAV ITRs), and a non-AAV nucleic acid sequence that encodes a gene operably linked to a regulatory sequence that directs expression of the gene in the host cell, as well as sufficient AAV rep and adenovirus helper functions to allow packaging of the vector genome into a recombinant AAVhu68 or AAVrh91 capsid. In one embodiment, the cell culture is comprised of mammalian cells (e.g., human embryonic kidney 293 cells, among others) or insect cells (e.g., Spodoptera frugiperda (Sf9) cells). In certain embodiments, a baculovirus provides the helper functions necessary for packaging of the vector genome into a recombinant AAVhu68 or AAVrh91 capsid.

[0077] Optionally, the rep function is provided by an AAV other than hu68. In certain embodiments, at least a portion of the rep function is from AAVhu68 or AAVrh91. In another embodiment, the rep protein is a heterologous rep protein other than AAVhu68rep, such as, but not limited to, AAV1 rep protein, AAV2 rep protein, AAV3 rep protein, AAV4 rep protein, AAV5 rep protein, AAV6 rep protein, AAV7 rep protein, AAV8 rep protein, or rep78, rep68, rep52, rep40, rep68 / 78, and rep40 / 52, or fragments thereof, or another source. Any of these AAVhu68 or mutant AAV capsid sequences may be under the control of exogenous regulatory control sequences that direct their expression in the host cell.

[0078] In one embodiment, the cells are produced in a suitable cell culture (e.g., HEK293 or Sf9) or suspension. The methods for producing gene therapy vectors described herein include methods well known in the art, such as production of plasmid DNA used for the production of gene therapy vectors, production of vectors, and purification of vectors. In some embodiments, the gene therapy vector is an AAV vector, and the produced plasmids are AAV cis-plasmids encoding the AAV vector genome and gene of interest, AAV trans-plasmids containing the AAV rep and cap genes, and adenovirus helper plasmids. The vector production process may include method steps such as initiation of cell culture, passaging of cells, seeding of cells, transfection of cells with plasmid DNA, medium exchange with serum-free medium after transfection, and harvesting of vector-containing cells and culture medium. The harvested vector-containing cells and culture medium are referred to herein as crude cell harvest. In yet another system, the gene therapy vector is introduced into insect cells by infection with a baculovirus-based vector. For a review of these production systems generally, see, e.g., Zhang et al., 2009, Adenovirus-adeno-associated virus hybrid for large-scale recombinant adeno-associated virus production, Human Gene Therapy 20:922-929, the contents of each of which are incorporated herein by reference in their entireties. Methods of making and using these and other AAV production systems are also described in the following United States patents, the contents of each of which are incorporated herein by reference in their entirety: U.S. Pat. Nos. 5,139,941, 5,741,683, 6,057,152, 6,204,059, 6,268,213, 6,491,907, 6,660,514, 6,951,753, 7,094,604, 7,172,893, 7,201,898, 7,229,823, and 7,439,065.

[0079] The crude cell harvest may then be subjected to process steps such as concentration of the vector harvest, diafiltration of the vector harvest, microfluidization of the vector harvest, nuclease digestion of the vector harvest, filtration of the microfluidized intermediate, crude purification by chromatography, crude purification by ultracentrifugation, buffer exchange by tangential flow filtration, and / or formulation and filtration to prepare bulk vector.

[0080] The vector drug product is purified and empty capsids are removed using a two-step affinity chromatography purification at high salt concentration, followed by anion exchange resin chromatography. These methods are described in more detail in WO2017 / 160360, filed December 9, 2016, and its priority documents, U.S. Patent Application No. 62 / 322,071, filed April 13, 2016, and U.S. Patent Application No. 62 / 226,357, filed December 11, 2015, entitled "Scalable Purification Method for AAV9," which are incorporated herein by reference. In certain embodiments, purification of vector drug products (e.g., AAVrh91) includes that described in more detail in WO2017 / 100674, filed December 9, 2016, and its priority documents, U.S. Provisional Patent Application No. 62 / 266,351, filed December 9, 2015, and U.S. Provisional Patent Application No. 62 / 322,083, filed April 13, 2016, entitled "Scalable Purification Method for AAV1," which are incorporated herein by reference.

[0081] To calculate the content of empty and full particles, the VP3 band volume for a selected sample (e.g., in the examples herein, a preparation purified by iodixanol gradient, number of genome copies (GC) = number of particles) is plotted against the GC particles loaded. The resulting linear equation (y = mx + c) is used to calculate the number of particles in the band volume of the test article peak. The number of particles (pt) per 20 μL loaded is then multiplied by 50 to obtain particles (pt) / mL. Pt / mL is divided by GC / mL to obtain the ratio of particles to genome copies (pt / GC). Pt / mL - GC / mL gives empty pt / mL. The percentage of empty particles is obtained by dividing empty pt / mL by pt / mL and multiplying by 100.

[0082] Generally, methods for assaying AAV vector particles containing empty capsids and packaged genomes are known in the art. See, for example, Grimm et al., Gene Therapy (1999) 6:1322-1330; Sommer et al., Molec. Ther. (2003) 7:122-128. To test for denatured capsids, the method involves subjecting the cured AAV stock to SDS-polyacrylamide gel electrophoresis (e.g., gradient gels containing 3-8% Tris-acetate in buffer) of any gel capable of separating the three capsid proteins, then running the gel until the sample material is separated, and blotting the gel onto a nylon or nitrocellulose membrane, preferably nylon. An anti-AAV capsid antibody is then used as a primary antibody that binds to the denatured capsid protein, preferably an anti-AAV capsid monoclonal antibody, most preferably a B1 anti-AAV-2 monoclonal antibody (Wobus et al., J. Virol. (2000) 74:9281-9293). A secondary antibody is then used that binds to the primary antibody and includes a means for detecting binding with the primary antibody, more preferably an anti-IgG antibody containing a detection molecule covalently bound to the antibody, most preferably a sheep anti-mouse IgG antibody covalently bound to horseradish peroxidase. To semi-quantitatively determine the binding between the primary and secondary antibodies, a method for detecting binding is used, preferably a detection method capable of detecting radioisotope radiation, electromagnetic radiation, or colorimetric changes, most preferably a chemiluminescence detection kit. For example, for SDS-PAGE, samples from column fractions can be taken and heated in SDS-PAGE loading buffer containing a reducing agent (e.g., DTT), and capsid proteins resolved in precast gradient polyacrylamide gels (e.g., Novex). Silver staining may be performed using SilverXpress (Invitrogen, CA) according to the manufacturer's instructions, or other suitable staining methods, i.e., SYPRO Ruby or Coomassie staining.In one embodiment, the concentration of AAV vector genome (vg) in the column fractions can be measured by quantitative real-time PCR (Q-PCR). The samples are diluted and digested with DNase I (or another suitable nuclease) to remove exogenous DNA. After inactivation of the nuclease, the samples are further diluted and amplified using a TaqMan™ fluorogenic probe specific for the primers and the DNA sequence between the primers. The number of cycles required to reach a defined level of fluorescence (threshold cycle, Ct) is measured for each sample on an Applied Biosystems Prism 7700 sequence detection system. Plasmid DNA containing the same sequence as contained in the AAV vector is used to generate a standard curve in the Q-PCR reaction. The cycle threshold (Ct) value obtained from the sample is used to determine the vector genome titer by normalizing it to the Ct value of the plasmid standard curve. An endpoint assay based on digital PCR can also be used.

[0083] In one embodiment, an optimized q-PCR method utilizing a broad-spectrum serine protease, such as proteinase K (e.g., commercially available from Qiagen), is used. More specifically, the optimized qPCR genomic titer assay is similar to the standard assay, except that after DNase I digestion, the sample is diluted with proteinase K buffer and treated with proteinase K, followed by heat inactivation. Suitably, the sample is diluted with an amount of proteinase K buffer equal to the sample size. The proteinase K buffer may be concentrated 2-fold or more. Typically, the proteinase K treatment is about 0.2 mg / mL, but can vary from 0.1 mg / mL to about 1 mg / mL. The treatment step is generally carried out at about 55° C. for about 15 minutes, but may be carried out at lower temperatures (e.g., about 37° C. to about 50° C.) for longer times (e.g., about 20 minutes to about 30 minutes), or at higher temperatures (e.g., up to about 60° C.) for shorter times (e.g., about 5 to 10 minutes). Similarly, heat inactivation is generally at about 95° C. for about 15 minutes, but may be performed at lower temperatures (e.g., about 70 to about 90° C.) and for longer times (e.g., about 20 minutes to about 30 minutes). The sample is then diluted (e.g., 1000-fold) and subjected to TaqMan analysis as described for standard assays.

[0084] Additionally or alternatively, droplet digital PCR (ddPCR) may be used. For example, a method for determining single-stranded and self-complementary AAV vector genome titer by ddPCR has been described. See, for example, M. Lock et al, Hu Gene Therapy Methods, Hum Gene Ther Methods. 2014 Apr; 25(2): 115-25. doi: 10.1089 / hgtb.2013.131. Epub 2014 Feb 14.

[0085] Briefly, a method for separating rAAVhu68 (or AAVrh91) particles with packaged genome sequences from genome-defective AAVhu68 (or AAVrh91) intermediates involves subjecting a suspension containing recombinant AAVhu68 (or rh91) viral particles and AAVhu68 (or AAVrh91) capsid intermediates to high performance liquid chromatography, where the AAVhu68 (or AAVrh91) viral particles and AAVhu68 intermediates are bound to a strong anion exchange resin equilibrated at a pH of about 10.2 (or about 9.8 for AAVrh91) and subjected to a salt gradient while monitoring the eluate for ultraviolet absorbance at about 260 nanometers (nm) and about 280 nm. The pH can be in the range of about 10 to 10.4, although it is less optimal for rAAVhu68 and AAVrh91. In this method, AAV full capsids are collected from fractions that elute when the A260 / A280 ratio reaches the infection point. In one example, for an affinity chromatography step, the diafiltered product may be applied to an affinity resin (Life Technologies) that efficiently captures AAVhu68 or AAVrh91 serotypes. Under these ionic conditions, a significant percentage of residual cellular DNA and proteins flow through the column, and AAV particles are efficiently captured.

[0086] The rAAV.hCDKL5 is suspended in a suitable physiologically compatible composition (e.g., buffered saline). This composition can be frozen for storage, later thawed, and optionally diluted with a suitable diluent. Alternatively, the vector can be prepared as a composition suitable for delivery to a patient without undergoing a freezing and thawing step.

[0087] As used herein, the term "NAb titer" is a measure of how much neutralizing antibodies (e.g., anti-AAV Nabs) that neutralize the physiological effect of the targeted epitope (e.g., AAV) are produced. Anti-AAV NAb titers can be measured, for example, as described in Calcedo, R., et al., Worldwide Epidemiology of Neutralizing Antibodies to Adeno-Associated Viruses. Journal of Infectious Diseases, 2009.199(3):p.381-390, which is incorporated herein by reference.

[0088] The abbreviation "sc" refers to self-complementary. "Self-complementary AAV" refers to constructs in which the coding region carried by the recombinant AAV nucleic acid sequence is designed to form an intramolecular double-stranded DNA template. During infection, rather than waiting for cell-mediated synthesis of the second strand, the two complementary halves of scAAV associate to form one double-stranded DNA (dsDNA) unit capable of immediate replication and transcription. See, for example, DM McCarty et al, "Self-complementary recombinant adeno-associated virus (scAAV) vectors promote efficient transduction independently of DNA synthesis", Gene Therapy, (August 2001), Vol 8, Number 16, Pages 1248-1254. Self-complementary AAVs are described, for example, in U.S. Patent Nos. 6,596,535, 7,125,717, and 7,456,683, each of which is incorporated by reference in its entirety.

[0089] "Replication-defective virus" or "viral vector" refers to a synthetic or artificial viral particle in which an expression cassette containing a gene of interest is packaged in a viral capsid or envelope, and any viral genomic sequences packaged within the viral capsid or envelope are replication-defective, i.e., they cannot produce progeny virions, but can retain the ability to infect target cells. In one embodiment, the genome of the viral vector does not contain genes encoding enzymes required for replication (the genome can be engineered to be "gutless," containing only the gene of interest flanked by signals required for amplification and packaging of the artificial genome), but these genes can be supplied during production. Thus, it is considered safe for use in gene therapy, since replication and infection by progeny virions cannot occur except in the presence of viral enzymes required for replication.

[0090] Often, rAAV particles are referred to as DNase-resistant. However, in addition to this endonuclease (DNase), other endonucleases and exonucleases can be used in the purification steps described herein to remove contaminating nucleic acids. Such nucleases can be selected to degrade single-stranded and / or double-stranded DNA, and RNA. Such steps can include a single nuclease, or a mixture of nucleases directed to different targets, which can be endonucleases or exonucleases.

[0091] The term "nuclease resistant" indicates that the AAV capsid is constructed entirely around an expression cassette designed to deliver genes into a host cell, and protects these packaged genomic sequences from degradation (digestion) during a nuclease incubation step designed to remove contaminating nucleic acids that may be present from the production process.

[0092] IV. Other Vectors

[0093] In certain embodiments, the vector is a viral vector selected from recombinant parvovirus, recombinant lentivirus, recombinant retrovirus, or recombinant adenovirus; or a non-viral vector selected from naked DNA, naked RNA, inorganic particles, lipid particles, polymer-based vectors, or chitosan-based formulations. The selected vector can be delivered by any suitable method, including transfection, electroporation, liposome delivery, membrane fusion techniques, high-speed DNA-coated pellets, viral infection, and protoplast fusion. The methods used to create such constructs are known to those skilled in the art of nucleic acid manipulation and include genetic engineering, recombinant engineering, and synthetic techniques. See, for example, Sambrook et al, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Press, Cold Spring Harbor, NY.

[0094] "Replication-defective virus" or "viral vector" refers to a synthetic or artificial viral particle in which an expression cassette containing a gene of interest is packaged in a viral capsid or envelope, and any viral genomic sequences packaged within the viral capsid or envelope are replication-defective, i.e., they are unable to produce progeny virions, but can retain the ability to infect target cells. In one embodiment, the genome of the viral vector does not contain genes encoding enzymes required for replication (the genome can be engineered to be "gutless", containing only the transgene of interest flanked by signals required for amplification and packaging of the artificial genome), but these genes can be supplied during production. Thus, it is considered safe for use in gene therapy, since replication and infection by progeny virions cannot occur except in the presence of viral enzymes required for replication. Such replication-defective viruses can be adeno-associated viruses (AAV), adenoviruses, lentiviruses (integrating or non-integrating), or another suitable source of virus.

[0095] V. Composition Provided herein is a composition comprising an rAAV or vector as described herein and an aqueous suspension medium. In certain embodiments, the suspension is formulated for intravenous delivery, intrathecal administration, or intraventricular administration.

[0096] Provided herein is a composition comprising at least one rAAV stock, and optional carriers, excipients and / or preservatives. As used herein, a "stock" of rAAV refers to a population of rAAV. Despite the heterogeneity of capsid proteins due to deamidation, the rAAVs in the stock are expected to share the same vector genome. The stock may, for example, contain rAAVs with capsids having selected AAV capsid proteins and heterogeneous deamidation patterns characteristic of the selected production system. The stock may be produced from a single production system or may be pooled from multiple runs of a production system. A variety of production systems may be selected, including but not limited to those described herein.

[0097] As used herein, "carrier" includes any and all solvents, dispersion media, vehicles, coatings, diluents, antibacterial and antifungal agents, isotonic and absorption delaying agents, buffers, carrier solutions, suspensions, colloids, and the like. The use of such media and agents for pharma- ceutical active substances is well known in the art. Supplementary active ingredients can also be incorporated into the composition. The phrase "pharmacologically acceptable" refers to molecular entities and compositions that do not produce allergic or similar adverse reactions when administered to a host. Delivery vehicles such as liposomes, nanocapsules, microparticles, microspheres, lipid particles, vesicles, and the like can be used to introduce the compositions of the present invention into suitable host cells. In particular, the rAAV vector delivery vector genome can be formulated for delivery either encapsulated in lipid particles, liposomes, vesicles, nanospheres, nanoparticles, or the like.

[0098] In one embodiment, the composition comprises a final formulation suitable for delivery to a subject, for example, an aqueous liquid suspension buffered to a physiologically compatible pH and salt concentration.Optionally, one or more surfactants are present in the formulation.In another embodiment, the composition can be delivered as a concentrate that is diluted for administration to a subject.In other embodiments, the composition can be lyophilized and reconstituted at the time of administration.

[0099] A suitable surfactant or combination of surfactants may be selected from among non-toxic non-ionic surfactants. In one embodiment, a difunctional block copolymer surfactant terminated in a primary hydroxyl group is selected, such as Pluronic® F68 [BASF] (also known as Poloxamer 188), which has a neutral pH and an average molecular weight of 8400. Other surfactants and other poloxamers may be selected, i.e., non-ionic triblock copolymers composed of a central hydrophobic chain of polyoxypropylene (poly(propylene oxide)) flanked by two hydrophilic chains of polyoxyethylene (poly(ethylene oxide)), SOLUTOL HS 15 (macrogol-15 hydroxystearate), LABRASOL (polyoxycaprylic acid glyceride), polyoxy 10 oleyl ether, TWEEN (polyoxyethylene sorbitan fatty acid ester), ethanol, and polyethylene glycol. In one embodiment, the formulation contains a poloxamer. These copolymers are generally named with the letter "P" (for poloxamer) followed by three digits, the first two digits x 100 giving the approximate molecular mass of the polyoxypropylene core, and the last digit x 10 giving the percentage of polyoxyethylene content. In one embodiment, poloxamer 188 is selected. In one embodiment, the surfactant may be present in an amount of up to about 0.0005% to about 0.001% (w / w%, weight to weight basis) of the suspension. In another embodiment, the surfactant may be present in an amount of up to about 0.0005% to about 0.001% (v / v%, volume to volume basis) of the suspension. In yet another embodiment, the surfactant is present in an amount of up to about 0.0005% to about 0.001% of the suspension, where n% indicates n grams per 100 mL of suspension.

[0100] In another embodiment, the composition comprises a carrier, diluent, excipient and / or adjuvant. A suitable carrier can be easily selected by a person skilled in the art in view of the indication for which the introduced virus is intended. For example, one suitable carrier includes saline, which can be formulated with various buffer solutions (e.g., phosphate buffered saline). Other exemplary carriers include sterile saline, lactose, sucrose, calcium phosphate, gelatin, dextran, agar, pectin, peanut oil, sesame oil, and water. The buffer / carrier should include a component that prevents rAAV from sticking to the injection tube but does not interfere with rAAV binding activity in vivo. A suitable surfactant, or combination of surfactants, may be selected from among non-toxic non-ionic surfactants. In one embodiment, a bifunctional block copolymer surfactant terminated in a primary hydroxyl group is selected, such as Poloxamer 188 (also known under the trade names Pluronic® F68 [BASF], Lutrol® F68, Synperonic® F68, Kolliphor® P188), having a neutral pH and an average molecular weight of 8400. Other surfactants and other poloxamers may be selected, i.e., non-ionic triblock copolymers composed of a central hydrophobic chain of polyoxypropylene (poly(propylene oxide)) flanked by two hydrophilic chains of polyoxyethylene (poly(ethylene oxide)), SOLUTOL HS 15 (macrogol-15 hydroxystearate), LABRASOL (polyoxycapric acid glyceride), polyoxy-oleyl ether, TWEEN (polyoxyethylene sorbitan fatty acid ester), ethanol, and polyethylene glycol. In one embodiment, the formulation contains a poloxamer. These copolymers are generally named with the letter "P" (for poloxamer) followed by three digits, the first two digits x 100 giving the approximate molecular mass of the polyoxypropylene core and the last digit x 10 giving the percentage of polyoxyethylene content. In one embodiment, poloxamer 188 is selected. The surfactant may be present in an amount up to about 0.0005% to about 0.001% of the suspension.

[0101] In certain embodiments, compositions containing rAAV.hCDKL5 are delivered at a pH ranging from 6.8 to 8, or 7.2 to 7.8, or 7.5 to 8. For intrathecal delivery, a pH greater than 7.5, e.g., 7.5 to 8, or 7.8, may be desirable.

[0102] In certain embodiments, the formulation may contain a buffered saline solution that does not contain sodium bicarbonate. Such formulations may contain a buffered saline solution, such as Harvard buffer, which contains one or more of sodium phosphate, sodium chloride, potassium chloride, calcium chloride, magnesium chloride, and mixtures thereof, in water. The aqueous solution may further contain Kolliphor P188, a poloxamer available from BASF, formerly sold under the trade name Lutrol F68. The aqueous solution may have a pH of 7.2.

[0103] In another embodiment, the formulation contains 1 mM sodium phosphate (Na 3 PO 4 ), 150 mM sodium chloride (NaCl), 3 mM potassium chloride (KCl), 1.4 mM calcium chloride (CaCl 2 ), 0.8 mM magnesium chloride (MgCl 2 ), and 0.001% poloxamer (e.g., Kolliphor®) 188, pH 7.2. See, e.g., harvardapparatus.com / harvard-apparatus-perfusion-fluid.html. In certain embodiments, Harvard buffer is preferred, as better pH stability is observed with Harvard buffer.

[0104] In certain embodiments, the formulation buffer is an artificial CSF containing Pluronic F68. In other embodiments, the formulation may contain one or more penetration enhancers. Examples of suitable penetration enhancers may include, for example, mannitol, sodium glycocholate, sodium taurocholate, sodium deoxycholate, sodium salicylate, sodium caprylate, sodium caprate, sodium lauryl sulfate, polyoxyethylene-9-lauryl ether, or EDTA.

[0105] Optionally, the compositions of the invention may contain other conventional pharmaceutical ingredients, such as preservatives or chemical stabilizers, in addition to the rAAV and carrier(s). Suitable exemplary preservatives include chlorobutanol, potassium sorbate, sorbic acid, sulfur dioxide, propyl gallate, parabens, ethyl vanillin, glycerin, phenol, and parachlorophenol. Suitable chemical stabilizers include gelatin and albumin.

[0106] The composition according to the present invention may include a pharma- ceutically acceptable carrier as defined above.Preferably, the composition described herein includes one or more AAV suspended in an effective amount of a pharma- ceutically suitable carrier and / or mixed with a suitable excipient designed for delivery to a subject via injection, osmotic pump, intrathecal catheter, or for delivery by another device or route.In certain embodiments, an ommaya reservoir is used for delivery.In one example, the composition is formulated for intrathecal delivery.In one example, the composition is formulated for intravenous (iv) delivery.

[0107] VI.Use Provided herein is a method of treating CDD, comprising administering to a subject in need thereof an effective amount of a rAAV or vector described herein.

[0108] In certain embodiments, an "effective amount" herein is an amount that achieves alleviation of symptoms of CDD and / or slowing of progression of CDD.

[0109] The vector is administered in an amount sufficient to transfect cells and provide sufficient levels of gene transfer and expression to provide a therapeutic effect without undue adverse effects or with a medically acceptable physiological effect, which can be determined by one of skill in the art. Conventional and pharmacologic acceptable routes of administration include, but are not limited to, direct delivery to the desired organ (e.g., brain, CSF, liver (optionally via the hepatic artery), lung, heart, eye, kidney), oral, inhalation, intranasal, intrathecal, intratracheal, intraarterial, intraocular, intravenous, intramuscular, subcutaneous, intradermal, intraparenchymal, intrathecal, ICM, lumbar puncture, and other parenteral routes of administration. Routes of administration may be combined if desired.

[0110] The dosage of a viral vector (e.g., rAAV) depends primarily on factors such as the condition being treated, the age, weight, and health of the patient, and may therefore vary between patients. For example, a therapeutically effective human dosage of a viral vector is generally about 1×10 9 ~1×10 16 The concentration of vector genome copies ranges from about 25 to about 1000 microliters to about 100 mL of solution. In certain embodiments, a volume of about 1 mL to about 15 mL, or about 2.5 mL to about 10 mL, or about 5 mL of suspension is delivered. In certain embodiments, a volume of about 1, about 2, about 3, about 4, about 5, about 6, about 7, about 8, about 9, about 10, about 11, about 12, about 13, about 14, or about 15 mL of suspension is delivered. In certain embodiments, a volume of about 8.9 x 10 12 ~2.7×10 14 The total dose of GC is administered in this volume. In certain embodiments, the total dose is about 1.1×10 10 GC / g brain mass ~ approx. 3.3×10 11 In one particular embodiment, a dose of about 3.0×10 GC / g brain mass is administered in this volume. 9 , about 4.0×10 9 , about 5.0×10 9 , about 6.0×10 9 , about 7.0×10 9 , about 8.0×10 9 , about 9.0×10 9, about 1.0×10 10 , about 1.1×10 10 , about 1.5×10 10 , about 2.0×10 10 , about 2.5×10 10 , about 3.0×10 10 , about 3.3×10 10 , about 3.5×10 10 , about 4.0×10 10 , about 4.5×10 10 , about 5.0×10 10 , about 5.5×10 10 , about 6.0×10 10 , about 6.5×10 10 , about 7.0×10 10 , about 7.5×10 10 , about 8.0×10 10 , about 8.5×10 10 , about 9.0×10 10 , about 9.5×10 10 , about 1.0×10 11 , about 1.1×10 11 , about 1.5×10 11 , about 2.0×10 11 , about 2.5×10 11 , about 3.0×10 11 , about 3.3×10 11 , about 3.5×10 11 , about 4.0×10 11 , about 4.5×10 11 , about 5.0×10 11 , about 5.5×10 11 , about 6.0×10 11 , about 6.5×10 11 , about 7.0×10 11 , about 7.5×10 11 , about 8.0×10 11 , about 8.5×10 11 , about 9.0×10 11 The dose of GC is administered in this volume.

[0111] The dosage may be adjusted to balance the therapeutic effect against any side effects, and the dosage may vary depending on the therapeutic application that the recombinant vector is used for.The expression level of the transgene product can be monitored to determine the frequency of administration that results in the viral vector, preferably the AAV vector that contains a minigene.Optionally, the same administration regimen as that described for therapeutic purposes can be used for immunization using the composition of the present invention.

[0112] The replication-defective virus composition is formulated in a dosage unit, and for a human patient, the dosage is about 1.0×10 9 GC~approx. 1.0×10 16 GC, including amounts of replication defective virus in the range of all integers or subnumbers therebetween, preferably 1.0×10 12 GC~1.0×10 14 In one embodiment, the composition comprises at least 1 x 10 per dose, including all integers or decimals within the range. 9 , 2×10 9 , 3×10 9 , 4×10 9 , 5×10 9 , 6×10 9 , 7×10 9 , 8×10 9 , or 9 x 10 9 In another embodiment, the composition is formulated to contain at least 1 x 10 GC per dose, including all integers or decimals within the range. 10 , 2×10 10 , 3×10 10 , 4×10 10 , 5×10 10 , 6×10 10 , 7×10 10 , 8×10 10 , or 9 x 10 10 In another embodiment, the composition is formulated to contain at least 1 x 10 GC per dose, including all integers or decimals within the range. 11 , 2×10 11 , 3×10 11 , 4×10 11 , 5×10 11, 6×10 11 , 7×10 11 , 8×10 11 , or 9 x 10 11 In another embodiment, the composition is formulated to contain at least 1 x 10 GC per dose, including all integers or decimals within the range. 12 , 2×10 12 , 3×10 12 , 4×10 12 , 5×10 12 , 6×10 12 , 7×10 12 , 8×10 12 , or 9 x 10 12 In another embodiment, the composition is formulated to contain at least 1 x 10 GC per dose, including all integers or decimals within the range. 13 , 2×10 13 , 3×10 13 , 4×10 13 , 5×10 13 , 6×10 13 , 7×10 13 , 8×10 13 , or 9 x 10 13 In another embodiment, the composition is formulated to contain at least 1 x 10 GC per dose, including all integers or decimals within the range. 14 , 2×10 14 , 3×10 14 , 4×10 14 , 5×10 14 , 6×10 14 , 7×10 14 , 8×10 14 , or 9 x 10 14 In another embodiment, the composition is formulated to contain at least 1 x 10 GC per dose, including all integers or decimals within the range. 15 , 2×10 15 , 3×10 15 , 4×10 15 , 5×10 15 , 6×10 15 , 7×10 15 , 8×10 15 , or 9 x 10 15 It is formulated to contain GC.

[0113] In one embodiment, for human applications, the dose is 1×10 per kg of body weight, including all integers or decimals within the range. 10 ~Approx. 1×10 15 In one embodiment, the effective amount of the vector is about 1×10 per kg of body weight, including all integers or fractions within the range. 9 , 2×10 9 , 3×10 9 , 4×10 9 , 5×10 9 , 6×10 9 , 7×10 9 , 8×10 9 , or 9 x 10 9 In another embodiment, the effective amount of the vector is about 1 x 10 per kg of body weight, including all integers and decimals within the range. 10 , 2×10 10 , 3×10 10 , 4×10 10 , 5×10 10 , 6×10 10 , 7×10 10 , 8×10 10 , or 9 x 10 10 In another embodiment, the effective amount of the vector is about 1 x 10 per kg of body weight, including all integers and decimals within the range. 11 , 2×10 11 , 3×10 11 , 4×10 11 , 5×10 11 , 6×10 11 , 7×10 11 , 8×10 11 , or 9 x 10 11 In another embodiment, the effective amount of the vector is about 1 x 10 per kg of body weight, including all integers and decimals within the range. 12 , 2×10 12 , 3×10 12 , 4×10 12 , 5×10 12 , 6×10 12 , 7×10 12 , 8×10 12 , or 9 x 10 12In another embodiment, the effective amount of the vector is about 1 x 10 per kg of body weight, including all integers and decimals within the range. 13 , 2×10 13 , 3×10 13 , 4×10 13 , 5×10 13 , 6×10 13 , 7×10 13 , 8×10 13 , or 9 x 10 13 In another embodiment, the effective amount of the vector is about 1 x 10 per kg of body weight, including all integers and decimals within the range. 14 , 2×10 14 , 3×10 14 , 4×10 14 , 5×10 14 , 6×10 14 , 7×10 14 , 8×10 14 , or 9 x 10 14 In another embodiment, the effective amount of the vector is about 1 x 10 per kg of body weight, including all integers and decimals within the range. 15 , 2×10 15 , 3×10 15 , 4×10 15 , 5×10 15 , 6×10 15 , 7×10 15 , 8×10 15 , or 9 x 10 15 It's GC.

[0114] In certain embodiments, the dose is scaled by brain mass to provide an approximation of the size of the CSF compartment. Without wishing to be bound by theory, the dose conversion is based on brain masses of 0.15 g for neonatal mice (Gu et al., 2012), 90 g for juvenile NHPs (Herndon et al., 1998), 610 g for infants aged 6-8 months, 780 g for infants aged 8-12 months, and 960 g for infants over 12 months (Dekaban, 1978). Estimated brain weights for each age range of human infants were derived from the male and female brain weights presented in (Dekaban, 1978) by assuming an approximately linear increase in brain weight between neonates (370 g) and infants aged 4-8 months, resulting in an average estimated brain weight of 488 g for infants aged 1 month or older and under 4 months. The value of 610 g corresponds to the average brain weight of boys and girls aged 4-8 months (Dekaban, 1978). Examples of dose scaling from neonatal mice, juvenile NHPs, and equivalent human doses are provided in the table immediately below. Dose volumes can also be scaled from NHPs to humans based on estimated volumes of brain CSF (Matsumae et al., 1996) and spinal CSF (Rochette et al., 2016). [Table 1]

[0115] In one embodiment, for human applications, the dose is 1×10 per gram (g) of brain mass, including all integers or decimals within the range. 10 ~Approx. 1×10 15 In one embodiment, the effective amount of the vector is about 1×10 per gram (g) of brain mass, including all integers or decimals within the range. 9 , 2×10 9 , 3×10 9 , 4×10 9 , 5×10 9 , 6×10 9 , 7×10 9 , 8×10 9 , or 9 x 10 9In another embodiment, the effective amount of the vector is about 1×10 per gram (g) of brain mass, including all integers and decimals within the range. 10 , 2×10 10 , 3×10 10 , 4×10 10 , 5×10 10 , 6×10 10 , 7×10 10 , 8×10 10 , or 9 x 10 10 In another embodiment, the effective amount of the vector is about 1×10 per gram (g) of brain mass, including all integers and decimals within the range. 11 , 2×10 11 , 3×10 11 , 4×10 11 , 5×10 11 , 6×10 11 , 7×10 11 , 8×10 11 , or 9 x 10 11 In another embodiment, the effective amount of the vector is about 1×10 per gram (g) of brain mass, including all integers and decimals within the range. 12 , 2×10 12 , 3×10 12 , 4×10 12 , 5×10 12 , 6×10 12 , 7×10 12 , 8×10 12 , or 9 x 10 12 In another embodiment, the effective amount of the vector is about 1×10 per gram (g) of brain mass, including all integers and decimals within the range. 13 , 2×10 13 , 3×10 13 , 4×10 13 , 5×10 13 , 6×10 13 , 7×10 13 , 8×10 13 , or 9 x 10 13 In another embodiment, the effective amount of the vector is about 1×10 per gram (g) of brain mass, including all integers and decimals within the range. 14 , 2×10 14 , 3×10 14 , 4×10 14, 5×10 14 , 6×10 14 , 7×10 14 , 8×10 14 , or 9 x 10 14 In another embodiment, the effective amount of the vector is about 1×10 per gram (g) of brain mass, including all integers and decimals within the range. 15 , 2×10 15 , 3×10 15 , 4×10 15 , 5×10 15 , 6×10 15 , 7×10 15 , 8×10 15 , or 9 x 10 15 It's GC.

[0116] These above doses may be administered in various volumes of carrier, excipient, or buffer formulations ranging from about 25 to about 1000 microliters, or higher volumes including all numbers within that range, depending on the size of the area to be treated, the viral titer used, the route of administration, and the desired effect of the method. In one embodiment, the volume of the carrier, excipient, or buffer is at least about 25 μL. In one embodiment, the volume is about 50 μL. In another embodiment, the volume is about 75 μL. In another embodiment, the volume is about 100 μL. In another embodiment, the volume is about 125 μL. In another embodiment, the volume is about 150 μL. In another embodiment, the volume is about 175 μL. In yet another embodiment, the volume is about 200 μL. In another embodiment, the volume is about 225 μL. In yet another embodiment, the volume is about 250 μL. In yet another embodiment, the volume is about 275 μL. In yet another embodiment, the volume is about 300 μL. In yet another embodiment, the volume is about 325 μL. In another embodiment, the volume is about 350 μL. In another embodiment, the volume is about 375 μL. In another embodiment, the volume is about 400 μL. In another embodiment, the volume is about 450 μL. In another embodiment, the volume is about 500 μL. In another embodiment, the volume is about 550 μL. In another embodiment, the volume is about 600 μL. In another embodiment, the volume is about 650 μL. In another embodiment, the volume is about 700 μL. In another embodiment, the volume is about 700-1000 μL.

[0117] In certain embodiments, the dose is about 1×10 9 Approximately 1 × 10 GC / g brain mass 12 GC / g brain mass. In certain embodiments, the dose is about 1×10 10 GC / g brain mass ~ approx. 3×10 11 GC / g brain mass. In certain embodiments, the dose is about 1×10 10 GC / g brain mass ~ approx. 2.5×10 11 GC / g brain mass. In certain embodiments, the dose is about 5×10 10The range may be in the range of GC / g brain mass.

[0118] In one embodiment, the viral construct comprises at least about 1×10 9 GC~approx. 1×10 15 , or about 1 × 10 11 ~5×10 13 The GC may be delivered in a dose of 100 mL. Suitable volumes for delivery of these doses and concentrations may be determined by one of skill in the art. For example, a volume of about 1 μL to 150 mL may be selected, with larger volumes being selected for adults. Typically, for newborns, a suitable volume may be selected from about 0.5 mL to about 10 mL, and for older infants, from about 0.5 mL to about 15 mL. For infants, a volume of about 0.5 mL to about 20 mL may be selected. For children, a volume of up to about 30 mL may be selected. For pre-teens and teenagers, a volume of up to about 50 mL may be selected. In yet other embodiments, patients may receive intrathecal administration in a volume of about 5 mL to about 15 mL, which may be selected, or may receive about 7.5 mL to about 10 mL. Other suitable volumes and dosages may be determined. Dosages may be adjusted to balance the therapeutic benefit against any side effects, and such dosages may vary depending on the therapeutic application for which the recombinant vector is utilized.

[0119] The recombinant vectors described above can be delivered to host cells according to published methods. The rAAV, preferably suspended in a physiologically compatible carrier, can be administered to a human or non-human mammalian patient. In certain embodiments, for administration to a human patient, the rAAV is suitably suspended in an aqueous solution containing saline, a surfactant, and a physiologically compatible salt or mixture of salts. Suitably, the formulation is adjusted to a physiologically acceptable pH, for example, in the range of pH 6-9, or pH 6.5-7.5, pH 7.0-7.7, or pH 7.2-7.8. The pH of cerebrospinal fluid is about 7.28 to about 7.32, so for intrathecal delivery, a pH within this range may be desirable, and for intravenous delivery, a pH of about 6.8 to about 7.2 may be desirable. However, other pHs within the broadest range, and subranges of these, may be selected for other delivery routes.

[0120] As used herein, the term "intrathecal delivery" or "intrathecal administration" refers to the route of administration of a drug by injection into the spinal canal, more specifically, by injection into the subarachnoid space to reach the cerebrospinal fluid (CSF). Intrathecal delivery may include lumbar puncture, intraventricular (including intraventricular cerebroventricular (ICV)), suboccipital / intracisternal, and / or C1-2 puncture. For example, the material may be introduced by lumbar puncture for diffusion throughout the subarachnoid space. In another example, it may be an injection into the cisterna magna. In certain embodiments, the rAAV, vector, or composition described herein is administered to a subject in need thereof via intrathecal administration. In certain embodiments, intrathecal administration is performed as described in U.S. Patent Publication No. 2018-0339065A1, published November 29, 2019, which is incorporated herein by reference in its entirety.

[0121] As used herein, the term "intracisternal delivery" or "intracisternal administration" refers to the route of administration of a drug directly into the cerebrospinal fluid of the cerebellum-medullary cisterna magna, more specifically, by suboccipital puncture or by direct injection into the cisterna magna, or by a permanently placed tube.

[0122] In certain embodiments, treatment with the compositions described herein is well tolerated with respect to sensory neurotoxicity and subclinical sensory neuronal pathology, with minimal to mild subclinical degeneration of DRG sensory neurons in animals and / or human patients.

[0123] VII. Devices and methods for delivery of pharmaceutical compositions into the cerebrospinal fluid In one aspect, the vectors provided herein may be administered intrathecally via the methods and / or devices provided in this section and described in WO2018 / 160582 (incorporated herein by reference). Alternatively, other devices and methods may be selected. In certain embodiments, the method includes a step of CT-guided suboccipital injection into the cisterna magna of a patient via a spinal needle. As used herein, the term computed tomography (CT) refers to radiography in which a three-dimensional image of a body structure is constructed by a computer from a series of planar cross-sectional images made along an axis. In certain embodiments, the vectors and / or compositions thereof described herein are administered via computed tomography (CT)-guided suboccipital injection into the cisterna magna (intracisternomagna [ICM]). In certain embodiments, an Ommaya reservoir is used for delivery of the pharmaceutical composition. In certain embodiments, the device is described in U.S. Patent Publication No. 2018-0339065A1, published November 29, 2019, which is incorporated herein by reference in its entirety.

[0124] In certain embodiments, AAVhu68.UbC.hCDKL5-1co.miR183.rBG is administered as a single dose to hospitalized subjects on day 1 via CT-guided suboccipital injection into the cisterna magna. On day 1, a syringe containing AAVhu68.UbC.hCDKL5-1co.miR183.rBG (final volume ≦5 ml) at the appropriate potency is prepared by the Investigational Pharmacy associated with the study and sent to the procedure room. Prior to administration of the investigational drug, subjects are anesthetized, intubated, and the injection site is prepared and draped using aseptic technique. A lumbar puncture is performed to remove a predetermined volume of CSF, followed by an IT injection of iodinated contrast to aid in visualization of the relevant anatomical structures of the cisterna magna. Intravenous (IV) contrast may be administered before or during the needle puncture as an alternative to IT contrast. The decision to use IV or IT contrast is left to the discretion of the interventionalist. Under fluoroscopic guidance, a spinal needle (22-25G) is advanced into the cisterna magna. A larger introducer needle may be used to aid in needle placement. After confirmation of needle placement, an extension set is connected to the spinal needle and filled with CSF. At the interventionalist's discretion, a syringe containing contrast material may be connected to the extension set and a small volume injected to confirm needle placement in the cisterna magna. After needle placement is confirmed, a syringe containing rAAV is connected to the extension set. The contents of the syringe are injected slowly over 1-2 minutes, delivering a volume of 5.0 ml or less.

[0125] VIII. CDD As used herein, "patient" or "subject" refers interchangeably to a mammal, male or female, including humans, veterinary or agricultural animals, domestic or pet animals, and animals typically used in clinical research. In one embodiment, the subject of these methods and compositions is a human patient. In one embodiment, the subject of these methods and compositions is a human, male or female. In certain embodiments, the subject of these methods and compositions is diagnosed with CDD and / or symptoms of CDD.

[0126] The methods and compositions may be used for the treatment of any of the stages of CDD. In certain embodiments, the patient is about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 months old, or about 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10, 10.5, 11, 12, 13, 14, 15, 16, 17, 18 years old. In certain embodiments, the patient is a young child, e.g., 18 months to 3 years old. In certain embodiments, the patient is 3 to 6 years old, 3 to 12 years old, 3 to 18 years old, 3 to 30 years old. In certain embodiments, the patient is 18 years old or older.

[0127] Symptoms of CDD include seizures that usually begin within the first 3 months of life and can appear as early as the first week after birth. Seizures vary with age and may follow predictable patterns. The most common types are generalized tonic-clonic seizures with loss of consciousness, muscle rigidity, and convulsions; tonic seizures, characterized by abnormal muscle contractions; and epileptic spasms, with brief episodes of muscle jerks. Seizures occur daily in most people with CDKL5 deficiency, but they may have seizure-free periods. Seizures in CDKL5 deficiency are typically resistant to treatment.

[0128] Children with CDKL5 deficiency have developmental disabilities. Most have severe intellectual disability and little or no speech. Gross motor skills such as sitting, standing, and walking are delayed or never achieved. About one-third of affected people are able to walk independently. Fine motor skills such as picking up small objects with the fingers are also impaired, and about half of affected people use their hands purposefully. Most people with the condition have impaired vision (cortical visual impairment).

[0129] Other common features of CDKL5 deficiency include repetitive hand movements such as clapping, licking, and sucking (stereotypic), teeth grinding (bruxism), sleep disorders, difficulty eating, and gastrointestinal problems including constipation and reflux of acidic stomach contents into the esophagus (gastroesophageal reflux). Some affected individuals have episodes of irregular breathing. Characteristic facial features in some people with CDKL5 deficiency include a high, wide forehead, large, sunken eyes, a distinct space between the nose and upper lip (filtrum), full lips, widely spaced teeth, and a high roof of the mouth (palate). Other physical differences may also occur, such as an abnormally small head size (microcephaly), a side-to-side curvature of the spine (scoliosis), and tapered fingers.

[0130] As described above, "increase," "decrease," "reduction," "mitigation," "improvement," "delay," or any grammatical variations thereof, or any similar term, unless otherwise specified, means a change of about 5-fold, about 2-fold, about 1-fold, about 90%, about 80%, about 70%, about 60%, about 50%, about 40%, about 30%, about 20%, about 10%, about 5% compared to a corresponding reference (e.g., an untreated control or a healthy subject not having CDD).

[0131] In certain embodiments, patients receive medications to control some of the signs and symptoms associated with CDD, such as seizures, muscle stiffness, or breathing, sleep, gastrointestinal, or cardiac problems.

[0132] In certain embodiments, a diuretic may be used in the combination therapy in a subject in need thereof. The diuretic used may be acetazolamine (Diamox) or other suitable diuretic. In some embodiments, the diuretic is administered at the time of gene therapy administration. In some embodiments, the diuretic is administered prior to gene therapy administration. In some embodiments, the diuretic is administered when the injection volume is 3 mL.

[0133] In certain embodiments, combination therapy may be utilized, including co-administration of Cdkl5-isoform 1, isoform 2, isoform 3, and / or isoform 4 expression vectors, or various two-way or three-way combinations thereof. Optionally, the combination therapy may further include administration of another active agent. In certain embodiments, the combination therapy may include enzyme replacement therapy.

[0134] Optionally, immunosuppressive combination therapy may be used in subjects in need thereof. Immunosuppressants for such combination therapy include, but are not limited to, glucocorticoids, steroids, antimetabolites, T-cell inhibitors, macrolides (e.g., rapamycin or rapalogs), and cytostatic agents (including alkylating agents, antimetabolites, cytotoxic antibiotics, antibodies, or agents active against immunophilins). Immunosuppressants may include nitrogen mustards, nitrosoureas, platinum compounds, methotrexate, azathioprine, mercaptopurine, fluorouracil, dactinomycin, anthracyclines, mitomycin C, bleomycin, mithramycin, IL-2 receptor (CD25)-specific antibodies or CD3-specific antibodies, anti-IL-2 antibodies, cyclosporine, tacrolimus, sirolimus, IFN-β, IFN-γ, opioids, or TNF-α (tumor necrosis factor-α) binding agents. In certain embodiments, immunosuppressive therapy may be initiated 0, 1, 2, 3, 4, 5, 6, 7 days prior to or after gene therapy administration, or earlier or later. Such immunosuppressive therapy may involve administration of one, two, or more drugs (e.g., glucocorticoids, prenelisone, mycophenolate mofetil (MMF) and / or sirolimus (i.e., rapamycin)). Such immunosuppressive drugs may be administered to a subject in need once, twice, or more times at the same dose or adjusted doses. Such therapy may involve simultaneous administration of two or more drugs (e.g., prednisone, mycophenolate mofetil (MMF) and / or sirolimus (i.e., rapamycin)) on the same day. One or more of these drugs may be continued at the same dose or adjusted doses after gene therapy administration. Such therapy may be for about one week (7 days), about 60 days, or more, as needed. In certain embodiments, a tacrolimus-free regimen is selected.

[0135] The words "comprise", "comprises", and "comprising" are to be interpreted inclusively rather than exclusively. The words "consist", "consisting", and variations thereof are to be interpreted exclusively rather than inclusively. Although various embodiments herein are set forth using the word "comprising", it is also intended that in other circumstances the relevant embodiment should be construed and described using the word "consisting of" or "consisting essentially of".

[0136] The term "expression" is used herein in the broadest sense and includes the production of RNA or RNA and protein. With respect to RNA, the terms "expression" or "translation" specifically relate to the production of peptides or proteins. Expression can be transient or stable.

[0137] It should be noted that the terms "a" or "an" refer to one or more, e.g., "an enhancer" is understood to refer to one or more enhancer(s). Thus, the terms "a" (or "an"), "one or more," and "at least one" are used interchangeably herein.

[0138] Throughout this specification, exponents are referred to using the term "e" followed by a number (n). This means "×10 n )". For example, "3e9" is 3×10 9 "1e13" is the same as 1×10 13 is the same as:

[0139] As used above, the term "about," when used to modify a numerical value, means a variance of ±10%, unless otherwise specified.

[0140] Unless otherwise defined herein, technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art and by reference to published documents which provide general guidance to those of ordinary skill in the art for many of the terms used herein. EXAMPLES

[0141] IX. Working Examples The following examples are merely illustrative and are not intended to limit the invention. The following table provides a list of certain abbreviations used in the following examples. Other abbreviations or meanings may be readily apparent to one of ordinary skill in the art. [Table 2-1] [Table 2-2] [Table 2-3] [Table 2-4] [Table 2-5]

[0142] Example 1. Mouse model and preliminary studies CDD is caused by a splice site mutation in the CDKL5 gene, which encodes a phosphorylated serine / threonine protein kinase highly expressed in the brain. This mutation causes disruption of CDKL5 kinase activity, leading to reduced signaling of AKT-mTOR and other related pathways, along with defects in neural circuit communication.

[0143] Several models for CDD have been developed and can be selected for use in assessing therapeutic efficacy. These following models are null for CDKL5 expression: e.g., Cdkl5-ko mice with a deletion in exon 6 (Δexon6) (Wang et al, Proceedings of the National Academy of Sciences Dec 2012, 109(52)21516-21521; DOI:10.1073 / pnas.1216988110); Cdkl5-ko mice with a deletion in exon 4 (Δexon4) (see Amendola et al. (2014) Mapping Pathological Phenotypes in a Mouse Model of CDKL5 Disorder. PLoS ONE 9(5):e91613.doi.org / 10.1371 / journal.pone.0091613(2014)); Cdkl5(R59X knock-in) model (available from the Jackson Laboratory, Tang et al. al(2019), Tang S,et al.Altered NMDAR signaling underlies autistic-like features in mouse models of CDKL5 deficiency disorder.Nat Commun.2019;10(1):2655.Published 2019 Jun 14.doi:10.1038 / s41467-019-10689-w), and Cdkl5(D471fs) model (Rodney Samaco, Baylor College of Medicine, Houston, TX).

[0144] CDD mice carrying a truncated CDKL5 gene lacking exon 6, which corresponds to exon 7 of human CDKL5, display a phenotype at 10-11 weeks with clinically relevant symptoms observed in humans with CDD, including impaired motor coordination, cognitive dysfunction and social behavior, as well as AKT and mTOR signaling. Notably, hindlimb clasping and autistic-like and antisocial behaviors were observed in CDD mice, indicating the development of neurodegenerative disorders within 2 or 2.5 months.

[0145] However, current CDD mouse models do not display any form of spontaneous or refractory epilepsy, in contrast to the disease phenotype caused by pathogenic variants in CDKL5 in humans. This lack of phenotype in CDD mice may be due to the age of the animals, the duration of the study, and the increased seizure resistance conferred by the genetic background of the C57BL / 6 mice (Wang et al., 2012 and Amendola et al., 2014). Based on this observation, it is not possible to observe a seizure phenotype in neonatal CDD mice. In addition, CDD primarily affects heterozygous female individuals, and therefore, although the proposed clinical trial patient population includes heterozygous females, signs of CDD development are not observed in heterozygous male Cdkl5 mice. KO / Y In mice, heterozygous female Cdkl5 KO / + This is not clear in mice. Therefore, heterozygous male Cdkl5 KO / YMice were used in the current preclinical program because they best represent the disease phenotype of CDD in humans, which tends to be more severe in male patients. Several other mouse models of CDKL5 deficiency have been reported as described herein. All of these mice lack CDKL5 protein expression, exhibit normal life spans, and display a wide range of mild behavioral abnormalities. CDKL5 expression is developmentally regulated in the mouse brain. Thus, CDKL5 was found to be highly expressed throughout the mouse brain, including the cortex and hippocampus, suggesting that Cdkl5 deficiency in these regions may be related to the phenotypes observed in Cdkl5-deficient mice. The most prominent phenotypes have been observed in male Cdkl5-ko mice, which appear at approximately 11 weeks of age, and most studies involving neurobehavioral phenotypes have been performed in male Cdkl5-ko mice.

[0146] In this study, we show that restoring CDKL5 expression in the CNS of three CDD mouse models (Cdkl5-ko(exon 6), Cdkl5(R59X), and Cdkl5(D471fs)) significantly ameliorates disease symptoms. We developed an AAV gene therapy vector composed of an AAVhu68 capsid and an expression cassette with the human synapsin promoter and an engineered human CDKL5 transgene. When the AAV-CDKL5 vector was administered to Cdkl5 knockout mice via injection into the neonatal lateral ventricle, we detected robust expression in up to 50% of neurons throughout the brain. AAV administration and human CDKL5 expression were well tolerated. Human CDKL5 was mostly localized in the cytoplasm, and protein expression and kinase activity persisted for more than 4 months. We subjected a cohort of treated Cdkl5-ko mice to a battery of neurobehavioral tests and found a striking improvement from the untreated Cdkl5-ko phenotype in behavioral outcomes observed in wild-type mice. We then repeated the same study in two different CDD mouse models harboring a patient-derived frameshift mutation instead of a gene knockout (Cdkl5(R59X) and CDKL5(D471fs) models). We obtained very similar results, thus reaffirming the curative benefit of our CDKL5 gene therapy vector.

[0147] CDKL5 is expressed as at least four different isoforms in the human brain. We generated similar AAV gene therapy vectors for the three alternatively spliced ​​isoforms, all of which were found to express and exhibit kinase activity in transduced mouse brain.

[0148] To test the expression of our AAV-CDKL5 gene therapy vector in larger animals, we performed studies with rhesus macaques. Through injection into the cerebrospinal fluid via the cisterna magna, we were able to achieve vector distribution throughout the brain, at 0.1–1 vector copy per diploid genome. Much higher vector transduction was observed in the dorsal root ganglion (DRG). Accordingly, in situ hybridization with a probe specific for the human CDKL5 sequence demonstrated abundant expression in DRG neurons, but much sparser expression in cortical gray matter neurons. CDKL5 administration and expression was generally well tolerated in all six rhesus macaques over a 60-day period, although we did notice mild axonopathy in the white matter spinal tracts.

[0149] In summary, we provide promising evidence that CDKL5 gene therapy confers durable curative benefit in mouse models and is well tolerated in non-human primates. Further optimization of this approach may ultimately provide an option for clinical intervention in children affected by CDD.

[0150] Materials and Methods. Plasmid Human Brain 1 The amino acid sequences of four CDKL5 (cyclin-dependent kinase-like 5, Uniprot ID O76039) expressed in C. cerevisiae were reverse translated into DNA sequences. The coding sequences were further engineered by taking into account, for example, the codon frequencies found in the human genome, codifying RNA splice sites, and alternative reading frames. The engineered sequences were cloned into the human synapsin promoter 2The CDKL5 vector was cloned into an AAV expression plasmid under the control of the AAV183 gene. The coding sequence is framed by a Kozak sequence, followed by a WPRE enhancer cassette (woodchuck hepatitis virus posttranscriptional regulatory element), followed by an SV40 polyA sequence, and then by AAV2 inverted terminal repeats (ITRs). To suppress expression in the dorsal root ganglion (DRG), in some experiments the above plasmid was modified to contain four repeats of the miR183 binding site (agtgaattctaccagtgccata) (SEQ ID NO: 11) immediately after the CDKL5 coding sequence and immediately before the WPRE sequence. The AAV CDKL5 vector was modified to contain the capsid PHP.B for mouse studies or the AAVhu68 for mouse and non-human primate studies. 4 or AAVrh91.

[0151] Mouse studies. All studies involving mice were approved by the University of Pennsylvania IACUC. We obtained Cdkl5-ko mice (B6.129(FVB)-Cdkl5tm1.1Joez / J, strain no. 021967) from Jackson Laboratories and mated heterozygous female ko mice with wt C57Bl6 males to obtain littermates of the following genotypes: male hemizygous Cdkl5-ko (also referred to as KO mice or mice), male wt. Female heterozygous Cdkl5-ko, and female wt mice were injected with AAV Cdkl5 vectors (dose range 1 × 10) in a total volume of 100 μl by retro-orbital injection at 18–21 days of age. 11 GC~5×10 11 Alternatively, mice were administered 1 × 10 10 GC~5×10 10 Doses of GC were injected intracerebroventricularly in a total volume of 2 μl. Mice mixed in genotype and injection product were housed, weighed, observed at least twice a week, and subjected to behavioral testing until aged 11 weeks for males and 14 weeks for females. No treatment-related morbidity was observed.

[0152] Western blotting and tissue staining. After euthanasia, one cortical hemisphere was flash frozen and subsequently protein lysates were generated using RIPA buffer. Western blotting was performed using antibodies against human Cdkl5 (S957D, University of Dundee, UK), EB2 (ab45767, Abcam) or EB2 phospo222 (pab01032-P, Covalab, UK). The other brain half was fixed overnight in formalin, paraffin embedded, and thin sections were processed for immunofluorescence staining using the same CDKL5 antibody.

[0153] Behavioral testing. Mice were tested once a day. Testing time, operator, and environment were kept the same (60 dB white noise background, and 1000 lumens incandescent indirect lighting). Before each test, mice were habituated for 30 min in their home cage. For the open field assay, a new cage with a minimal amount of bedding was placed in an infrared beam array (MedAssociates, Inc.). A single mouse was added to the center of the cage, and over the next 30 min, the number of beam breaks was automatically recorded and separated into beam breaks close to the ground (general activity) and beam breaks 3 inches above the ground (hindlimb rearing activity). For the elevated zero maze (EZM, Stoelting Co.), an elevated circular platform with two opposing closed arms and two open arms was used to allow uninterrupted exploration. A single mouse was placed in the center of the open arms, and its movements were videotaped for 15 min. For the Y-maze (Stoelting Co.), an enclosed platform containing three identical arms in a Y shape was used. A single mouse was placed in the arm closest to the operator and its movements were video recorded for 5 min. For the marble burying assay, a new cage was filled with 3 inches of AlphaDri bedding (Shepherd Specialty Papers) and compressed by gentle pressing. Twelve solid blue marbles were placed evenly spaced on the bedding and a single mouse was placed in the center of the cage. The number of marbles that were at least half buried by the bedding was recorded after 30 min.

[0154] Data analysis. Data were graphed and analyzed using GraphPad Prims software. Video files were recorded in mp4 format at 20 fps and analyzed using EthoVision XT software (version 14, Noldus Information Technology). For the nest building assay, mice were housed singly overnight in novel cages and provided with a standardized 2 x 2 inch square nestlet (cotton based). After 24 h, nest quality was scored on a scale of 1 to 5 and remaining untouched nestlet material was weighed.

[0155] Non-human primate experiments. All studies involving non-human primates were approved by the University of Pennsylvania IACUC and performed in accordance with USDA regulations. Non-human primates (NHPs) of the species Macaca mulatta (rhesus monkey) were obtained from Covance Research Products, Inc. Isolation and animal housing were performed in accordance with gene therapy program SOPs. Weight, temperature, respiratory rate, and heart rate were monitored periodically, and blood and CSF samples were obtained, during the month prior to AAV vector administration and throughout the study. Whole blood was used for cell count and differential, and a clinical blood chemistry panel was used. CSF samples were used for blood cell count and differential, and total protein quantification. For AAV vector delivery to the CSF via cisternal puncture, anesthetized monkeys were positioned on a procedure table in a lateral decubitus position with the head flexed forward. Using aseptic technique, a 21-27 gauge, 1-1.5 inch Quincke spinal needle (Becton Dickinson) was advanced into the suboccipital space until CSF flow was observed. The needle is directed toward the larger upper space of the cisterna magna to avoid blood contamination and potential brainstem injury. Correct placement of the needle puncture is confirmed by myelography using a fluoroscopic device. 1 mL of CSF was collected for baseline analysis prior to administration. After CSF collection, a Luer access extension catheter was connected to the spinal needle to facilitate 1 mL of iohexal (Omnipaque 180 mg / mL, General Electric Healthcare) contrast medium. After needle placement was confirmed, a syringe containing the test article (volume equivalent to 1 mL plus the dead space volume of the syringe and linker) was connected to a flexible linker and injected over 30 ± 5 seconds. The needle was removed and direct pressure was applied to the puncture site. AAVhu68.hSyn.Cdkl5-1co and AAVhu68.hSyn.Cdkl5-1co vectors were administered in a dose of up to 3 × 10 13The rhesus monkeys were injected with a dose of GC / NHP. Neurological evaluations were performed on all rhesus monkeys on study days 0, 14, 18, 41, and the final study day for detailed assessment of neurological function. Briefly, evaluations included posture and gait evaluation, cranial nerve evaluation, proprioceptive evaluation, and spinal / nerve reflexes. On study day 56, the rhesus monkeys were euthanized and extensive necropsy and autopsy were performed. Twenty-five major tissues were collected in duplicate from each rhesus monkey for either snap freezing or formalin fixation. DNA or RNA was purified from the snap frozen tissues and used for vector biodistribution or transgene expression analysis, respectively. For vector biodistribution, genome copies (GC) per total DNA weight were determined using TaqMan qPCR assays with a probe redirected to the polyA region of the transgene cassette and an internal standard. To quantify transgene expression, total RNA was used to generate cDNA via first strand synthesis with poly-T oligonucleotides, followed by TaqMan qPCR using a probe specific for the transgene that does not cross-react with endogenous rhesus CDKL5 sequences. For histopathological analysis, macaque tissues were paraffin embedded and sections were stained with H&E solution or CDKL5 antibody, respectively. The same spinal cord sections were incubated with Luxol Fast Blue to stain myelin. All stained tissue sections were reviewed by a board-certified veterinary pathologist and abnormal findings were verified by peer review.

[0156] References: 1.Hector,RDet al.Characterization of CDKL5 Transcript Isoforms in Human and Mouse.PloS one 11,e0157758,(2016). 2. Thiel, G., Greengard, P. & Sudhof, T. C. Characterization of tissue-specific transcription by the human synapsin I gene promoter. Proc Natl Acad Sci U S A 88, 3431-3435, (1991). 3. Deverman, B. E. et al. Cre-dependent selection yields AAV variants for widespread gene transfer to the adult brain. Nat Biotechnol 34, 204-209, (2016). 4. Hinderer, C. et al. Severe Toxicity in Nonhuman Primates and Piglets Following High-Dose Intravenous Administration of an Adeno-Associated Virus Vector Expressing Human SMN. Human Gene Therapy 29, 285-298, (2018).

[0157] Example 2: Gene Therapy AAV Vector The expression construct between the ITRs is composed of the human synapsin promoter (SEQ ID NO:23), the engineered coding sequence of human CDKL5, isoform 1 (SEQ ID NO:22), the WPRE expression enhancer (SEQ ID NO:27), and the SV40 polyA sequence (SEQ ID NO:28) (Figure 1A). Similarly, the alternative expression constructs contain the engineered coding sequence of human CDKL5, isoform 2 (CDKL5-2GS or hCDKL5-2GS, SEQ ID NO:24), 3 (CDKL5-3GS or hCDKL5-3GS, SEQ ID NO:25), or 4 (CDKL5-4GS or hCDKL5-4GS, SEQ ID NO:26), respectively, instead of isoform 1. All tested plasmids express well in mouse brain and show minor differences in their ability to phosphorylate EB2 (a measure of CDKL5 kinase activity). We found that the WPRE enhancer is required to obtain human CDKL5 expression levels in mouse brain similar to wild-type mouse Cdkl5 expression (Figure 2). CDKL5 was fully active, as determined by its ability to phosphorylate its endogenous target EB2 protein (Figure 2). CDKL5 expression and localization was confirmed via IHC.

[0158] Alternative expression constructs were also generated by exchanging human synapsin with the human ubiquitin C promoter (Ubc) (Figure 1B) or the chicken β-actin hybrid promoter (Figure 1C). The AAV vectors AAVrh91.UbC.CDKL5-1co.miR183 and AAVrh91.CBh.CDKL5-1co.miR183 in AAVrh91 capsids were delivered to 3 × 10 10 MiR183 was administered at a dose of 100 mg / kg / day and necropsy was performed at P14. Western blot analysis of mouse brain tissue confirmed the expression of CDKL5 after transduction with AAV vector genomes containing the UbC or CBh promoter (Figure 20). Figure 20 shows CDKL5 expression, as determined qualitatively by Western blot, 14 days after administration of either AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183. Overall, 3x10 10We observed robust restoration of CDKL5 expression and kinase function in CDKL5-ko mouse brains (cortical tissue analyzed) when treated with AAVrh91.UbC.CDKL5-1co.miR183 at a GC dose of 3×10 compared to PBS-treated WT and knockout mice (control group) (Figures 29A, 29B, and 30). 10 Figure 29B shows CDKL5 expression quantified from Western blot analysis charted as CDKL5 / tubulin levels in knockout mice administered AAVrh91.UbC.CDKL5-1co.miR183 at a GC dose of 3×10 compared to PBS-treated WT and knockout mice (control group). 10 Figure 30 shows kinase activity quantified from Western blot analysis charted as pEB2pS222 / total EB2 levels in knockout mice treated with AAVrh91.UbC.CDKL5-1co.miR183 at a GC dose of 3×10 compared to PBS-treated WT and knockout mice (control group). 10 Figure 1 shows kinase activity, as measured qualitatively by Western blotting (using pEB-S222 antibody, Baltussen et al., (2018)), in knockout mice administered AAVrh91.UbC.CDKL5-1co.miR183 at a dose of GC.

[0159] Next, CDKL5 expression levels were examined after administration of various doses of AAVrh91.UbC.CDKL5-1co.miR183 and AAVrh91.CBh.CDKL5-1co.miR183. In this study, mice were administered 1 × 10 10 , 3×10 10 , 6×10 10Mice were administered AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183 via neonatal ICV at a dose of 3×10 GC. Cortical tissue samples were taken from 4-month-old mice and CDKL5 expression was examined via Western blotting (Figures 21A, 21B, and 21C). Figure 21A shows mice were administered 3×10 GC via neonatal ICV. 10 , 6×10 10 Figure 21C shows CDKL5 expression as qualitatively measured by Western blotting at 4 months of age, administered AAVrh91.UbC.CDKL5-1co.miR183 at a dose of GC, 3x10 via neonatal ICV. 10 , 6×10 10 5 × 10 in wild-type and knockout mice treated with AAVrh91.UbC.CDKL5-1co.miR183 at a dose of GC 10 Figure 1 shows CDKL5 expression quantified from Western blot analysis, charted as CDKL5 / tubulin levels, compared to AAVhu68.hSyn-CDKL5 at doses of GC. Western blot analysis of the cortex demonstrated robust hCDKL5 transgene expression. These results show dose-dependent expression, 6x10 10 We observed saturation at doses of GC of 100-200 mg / kg / day. Comparing the results with those observed from CDKL5 expression driven by the hSyn promoter, the UbC promoter allowed CDKL5 expression to reach levels close to wild-type (WT) CDKL5 expression levels in healthy subjects. Western blot analysis of CDKL5 expression after administration of AAVrh91.CBh.CDKL5-1co.miR183 at various doses showed less robust hCDKL5 transgene expression (Figure 21B). Figure 21B shows mice administered 3 × 10 6-HT ... 10 , 1×10 10 Shows CDKL5 expression, as measured qualitatively by Western blotting, at 4 months of age administered AAVrh91.CBh.CDKL5-1co.miR183 at a dose of GC.

[0160] The altered expression of CDKL5 following administration of AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183, as observed by Western blot analysis, was further confirmed by fluorescence microscopy. Representative immunofluorescence microscopy images show significant expression of CDKL5 throughout the brain following administration of AAVrh91.UbC.CDKL5-1co.miR183, with dim expression observed following administration of AAVrh91.CBh.CDKL5-1co.miR183, mostly visible in the cortex (Figures 22A and 22B). Figure 22A shows that 3x10 CDKL5 was administered via neonatal ICV. 10 Figure 22B shows representative images from immunofluorescence microscopy analysis of CDKL5 expression after administration of AAVrh91.UbC.CDKL5-1co.miR183 at a dose of 3 × 10 GC via neonatal ICV. 10 Representative images are shown from immunofluorescence microscopy analysis of CDKL5 expression after administration of AAVrh91.CBh.CDKL5-1co.miR183 at a dose of 3×10 GC. Upon further analysis, we observed that the UbC promoter drives CDKL5 expression in more cells than the CBh promoter, which is mostly localized to neurons (Figures 23A and 23B). Figure 23A shows the expression of 3×10 GC via neonatal ICV. 10 Figure 23B shows representative images (zoomed in) from immunofluorescence microscopy analysis of CDKL5 expression (samples were also probed for NeuN, a neuronal marker) following administration of AAVrh91.UbC.CDKL5-1co.miR183 at a dose of 3 × 10 GCs via neonatal ICV. 10 Representative images (enlarged) from immunofluorescence microscopy analysis of CDKL5 expression (samples were also probed for NeuN, a neuronal marker) following administration of AAVrh91.CBh.CDKL5-1co.miR183 at a dose of GCs.

[0161] In summary, we observed expression of the CDKL5 transgene in mouse brain after transduction with AAVrh91.UbC.CDKL5-1co.miR183 and AAVrh91.CBh.CDKL5-1co.miR183. Upon further analysis, higher CDKL5 expression per cell was observed in the hippocampus of mouse brain after transduction with AAVrh91.UbC.CDKL5-1co.miR183. Meanwhile, lower CDKL5 expression per cell was observed in the hippocampus of mouse brain after transduction with AAVrh91.CBh.CDKL5-1co.miR183. Higher CDKL5 expression per cell was observed in the cortex of mouse brain after transduction with AAVrh91.UbC.CDKL5-1co.miR183. After transduction with AAVrh91.CBh.CDKL5-1co.miR183, lower CDKL5 expression per cell was observed in the cortex of mouse brain. Furthermore, AAVrh91.UbC.CDKL5-1co.miR183 (3 × 10 10 GC, neonatal ICV) and AAVhu68.hSyn.CDKL5-1co.miR183 (2.5 × 10 10 We compared CDKL5 expression in Cdkl5-ko mouse brains following administration of AAVrh91.UbC.CDKL5-1co.miR183 (AAVrh91 capsid, ubiquitin C promoter) or AAVhu68.hSyn.CDKL5-1co.miR183 (AAVhu68 capsid, synapsin promoter). A very similar expression pattern was observed in mouse brains following administration of AAV, as indicated. CDKL5 expression was observed in mouse brains following transduction with either AAVrh91.UbC.CDKL5-1co.miR183 (AAVrh91 capsid, ubiquitin C promoter) or AAVhu68.hSyn.CDKL5-1co.miR183 (AAVhu68 capsid, synapsin promoter).

[0162] Example 3: Preclinical therapeutic efficacy of CDKL5 gene therapy in CDD mouse models To test the therapeutic efficacy of CDKL5 gene therapy on Cdkl5-deficient mice, we administered 5 × 10 to 1000 young Cdkl5-ko (also referred to as KO mice or mice) and wild-type (wt) littermates by retro-orbital IV injection. 11GC(5e11gc) were treated with the AAV9-PHP.B-hSyn-hCDKL5-1co.WPRE vector. All treatment groups continued to grow at the same rate and no treatment-related deaths were observed. At 10 weeks of age, mice were subjected to a battery of behavioral tests. Robust and statistically significant normalization was observed for the treatment groups in the elevated zero maze and open field activity tests. The same groups showed slight improvements in the rotarod and Y-maze tests, but no improvement was observed in thermal sensitivity.

[0163] Given that Cdkl5 is important for neurodevelopment, we wondered whether early administration of gene therapy might improve treatment outcomes in mice. We then administered 6 × 10 Cdkl5 per mouse by intracerebroventricular (ICV) injection. 9 GC~5×10 10 Cohorts of newborn Cdkl5-ko or wt littermates were treated with a range of doses of the AAVhu68-hSyn-hCDKL5-1co.WPRE vector for GC. Behavioral testing as described at 10 weeks or 11-14 weeks (where indicated) of age. The overall observation across all groups was that treatment was well tolerated, there were no treatment-related deaths, and normal weight gain (Figures 9A and 9B) and overall development was observed. 5 x 10 10 GC dose, 2.5×10GC dose, and 1×10 10At doses of GC, a dose-dependent expression of the CDKL5 transgene in the Cdkl5-ko mouse brain was observed. At 10 weeks of age, KO mice showed a characteristic hindlimb clasping phenotype, which was substantially ameliorated in treated ko mice. The therapeutic efficacy of CDKL5 gene therapy was measured by the hindlimb clasping test. A dose-dependent improvement was observed in the severity scores of treated CDKL5-ko mice (Figures 9C-9F). Similarly, the persistent hyperactivity and hindlimb rearing phenotype seen in ko mice was normalized toward wild-type activity. Hippocampal learning and memory were highly improved in treated ko mice in the Y-maze test (spontaneous change index). The therapeutic efficacy of CDKL5 gene therapy was further measured by the open field activity test, which measures the correction of hyperactivity in KO and AAV-treated mice. A dose-dependent resolution of hyperactivity was observed in treated Cdkl5-ko mice (Figures 11A-11F). KO mice are poor nest builders and do not tear all nest material when housed alone overnight in fresh cages. Treated KO mice significantly improved nest building ability in a dose-dependent manner (Figure 10A, 10C, 10E). Two additional other neurobehavioral assays (marble burying and Y-maze) showed a strong trend for phenotypic improvement in treated Cdkl5-ko mice (Figure 10B and 10D). The EEG phenotype was rescued in a preliminary collaborative study and may be useful as a translational biomarker. Validation of outcomes of the primary assays (hindlimb clasping, locomotor activity) will be additionally performed using independent experimental cohorts. Larger cohort sizes (N=18 / group) would have yielded more robust statistical significance. Validation of outcomes of the primary assays (hindlimb clasping, locomotor activity) using alternative CDD mouse models (R59X, D471fs) will also be performed. In summary, CDKL5 gene therapy delivers functional CDKL5 protein to the mouse brain and has dose-dependent therapeutic effects on several neurobehavioral outcomes in a male mouse model of CDD.

[0164] After completing behavioral testing, brains were harvested when the mice were approximately 3 months old. Western blotting revealed robust expression of human CDKL5 in Cdkl5-ko brains, even 3 months after AAV administration. Finally, human CDKL5 showed robust kinase activity when blotted for phosphorylated EB2 protein.

[0165] We also investigated the outcome of CDKL5 gene therapy using an alternative CDD mouse model. Cdkl5(D471fs) carries a patient point mutation that leads to a premature stop codon. Similar to patients, no CDKL5 protein and very reduced EB phosphorylation levels were found in the brains of these mice. Thus, Cdkl5(D471fs) shares the absence of Cdkl5 protein with previously used Cdkl5-ko mice, but with a slightly different genetic background due to the method of generating the mouse model. Newborns were treated with the same concentration of Cdkl5 as before (5 × 10 10 GC, neoICV) and showed robust hCDKL5 protein expression and EB2 phosphorylation after 3 months. A small pilot cohort was subjected to behavioral testing. AAV treatment was well tolerated and no morbidity was observed. Hindlimb clasping was significantly corrected in treated mutant mice, and similarly, the mutant phenotype in the elevated zero maze test was corrected to wild-type behavior in treated mutant mice (open zone in, open zone entry, Figures 14, 15, 16). Mutant mice show poor interaction and burying behavior with marbles placed on a grid in a new cage, whereas wild-type mice typically bury almost everything. Treated mutant mice show a strong correction of their behavior relative to wild-type mice. See also Examples 8 and 9 described below. In the treated group of Cdkl5-ko mice, 5 × 10 10 Significant improvements were observed at higher doses of GC / mouse.

[0166] We also investigated the outcome of CDKL5 gene therapy using an alternative CDD mouse model. Cdkl5(R59X) carries a patient point mutation that introduces a premature stop codon. Similar to patients, no CDKL5 protein and very reduced EB phosphorylation levels were found in the brains of these mice. Thus, Cdkl5(R59X) shares the absence of Cdkl5 protein with previously used Cdkl5-ko mice, but with a slightly different genetic background due to the method of generation of the mouse model. Newborns were treated with the same concentration of Cdkl5 as before (5 × 10 10 Injection of AAV-treated mice with 100-fold increased the expression of hCDKL5 protein and EB2 phosphorylation in the 6-HT2C1 / 2 mice (GC, neoICV) and demonstrated robust hCDKL5 protein expression and EB2 phosphorylation after 3 months. A small pilot cohort was subjected to behavioral testing. AAV treatment was well tolerated and no morbidity was observed. Hindlimb clasping was significantly corrected in treated mutant mice.

[0167] We also investigated the outcome of CDKL5 gene therapy using an alternative CDD mouse model in heterozygous female Cdkl5-ko mice. Such a model reflects most CDD patients (CDD females). The overall evaluation showed that the heterozygous female Cdkl5-ko mouse neurobehavioral phenotype is much milder and develops later, making a robust evaluation of treatment outcome more difficult. However, at high doses (5 × 10 10 Representative data for hindlimb clasping and locomotor activity in GC (neonatal ICV) show significant improvements (Figures 14B and 14C). See also Examples 8 and 9.

[0168] Additionally, we investigated dose escalation of CDKL5 gene therapy in WT mice. WT (C57Bl6 / J) mice were administered 7.5 × 10 10 GC and 1×10 11 GC (i.e., 1.5-fold or 2-fold the highest dose previously used). No obvious effects on body weight, development, and survival were observed. Mice appeared normal and did not exhibit hindlimb clasping or activity changes. No effects were observed upon pathological review of CNS tissues (Figures 19A and 19B).

[0169] Example 4: Preclinical therapeutic efficacy of CDKL5 gene therapy (AAVrh91.UbC.CDKL5-1co.miR183) in CDD mouse models Additionally, 1×10 10 GC, 3×10 10 GC, and 6 × 10 10 Various doses of AAVrh91.UbC.CDKL5-1co.miR183 and AAVrh91.CBh.CDKL5-1co.miR183 in GC were examined in Cdkl5-ko and wild-type mice for behavioral (i.e., therapeutic) effects. In this study, Cdkl5-ko (also referred to as KO mice or mice) and wild-type (WT) mice were administered 3 × 10 10 GC doses of 3×10. Figure 24 shows quantification of CDKL5-expressing neurons (above background levels) compared to previous results after administration of AAVhu68.hSyn.CDKL5. By comparison, administration of AAVrh91.UbC.CDKL5-1co.miR183 was observed to achieve expression within the range of WT expression in normal subjects. On the other hand, administration of AAVrh91.CBh.CDKL5-1co.miR183 increased CDKL5 expression by 3×10. 10 We observed that a dose of GC of 100 mg / kg / day achieved suboptimal expression levels in mouse brains. As expected, the number and intensity of neurons was increased in WT brains.

[0170] Via neonatal ICV, 1 x 10 10 , 3×10 10 , 6×10 10 We further confirmed CDKL5 expression in mice treated with AAVrh91.UbC.CDKL5-1co.miR183 at a GC dose of 1×10 compared to PBS-treated WT. 10 , 3×10 10 , 6×10 10Figure 31B shows the results of the percentage of neurons with CDKL5 protein expression in mouse cortical and hippocampal tissues following neonatal ICV administration of AAVrh91.UbC.CDKL5-1co.miR183 at a GC dose of 3×10 10 Representative microscopy images from immunofluorescence analysis staining with DAPI (nuclei), CDKL5, and NeuN (neuronal marker) of cortical section tissue following neonatal ICV administration of AAVrh91.UbC.CDKL5-1co.miR183 at a dose of GC of 100 mg / kg. Abundant, dose-dependent CDKL5 expression in the cortex and hippocampus was observed.

[0171] The results of the survival and viability studies showed a trend towards dose-limiting survival after neonatal ICV injection (Figure 25). 10 , 3×10 10 , 6×10 10 GC dose of AAVrh91.CBh.CDKL5-1co.miR183 at 100 mg / kg / day and 100 mg / kg / day, respectively.

[0172] Mice receiving AAVrh91.UbC.CDKL5-1co.miR183 showed increased expression of IFN-γ at all doses (i.e., 1 × 10 10 , 3×10 10 , 6×10 10 Normal development was observed in all cohorts (1×10 GC). Normal weight gain was observed in all cohorts. Additionally, no treatment-related morbidity was observed. The control group of WT mice tolerated CDKL5 expression well. Figure 32 shows the effect of PBS or 1×10 10 , 3×10 10 , 6×10 10 Figure 33A shows an analysis of the measured body weights of wild type and CDKL5-ko mice treated with AAV.UbC.CDKL5-1co.miR183 at a dose of 3×10 GC compared to untreated mice in Cdkl5-ko and WT mice. Furthermore, an improvement in hindlimb clasping was observed (Figure 33A). Figure 33A shows an analysis of the measured body weights of wild type and CDKL5-ko mice treated with AAV.UbC.CDKL5-1co.miR183 at a dose of 3×10 GC compared to untreated mice in Cdkl5-ko and WT mice. 10 1 shows the results of the hind paw clasping test for the AAV.UbC.CDKL5-1co.miR183 treatment group at a dose of GC of 100 mg / kg.

[0173] Figure 33B shows the 1×10 10 , 3×10 10 , 6×10 10 Figure 34A shows the dose-dependent effect on hyperactivity, measured in the open field activity test and plotted as locomotor activity (beam breaks), in Cdkl5-ko and WT mice following administration of AAV.UbC.CDKL5-1co.miR183 at a GC dose of 1 x 10 10 Figure 34B shows binned locomotor activity results for groups of WT and Cdkl5-ko mice administered AAV.UbC.CDKL5-1co.miR183 with a low dose of GC of 3×10 10 Figure 34C shows binned locomotor activity results for groups of WT and Cdkl5-ko mice administered AAV.UbC.CDKL5-1co.miR183 at a medium dose of 6×10 GC. 10 Figure 1 shows binned locomotor activity results for groups of WT and Cdkl5-ko mice administered AAV.UbC.CDKL5-1co.miR183 at a high dose of GC. These results show a dose-dependent improvement in hyperactivity in Cdkl5-ko mice administered AAV.UbC.CDKL5-1co.miR183.

[0174] Furthermore, we observed an improvement in nest construction scores when mice were administered the lowest dose of AAV.UbC.CDKL5-1co.miR183 (Figure 35). 10 , 3×10 10 , 6×10 10 Figure 1 shows nest construction (nesting quality / score) results for AAV.UbC.CDKL5-1co.miR183-treated WT and Cdkl5-ko mice with a GC dose of 100 mg / kg.

[0175] In summary, we observed significant improvement in Cdkl5-ko mice when treated with AAV.CDKL5 (hSyn, under UbC) compared to Cdkl5-ko mice treated with PBS. [Table 3]

[0176] The AAV.UbC.CDKL5-1co.miR183 vector demonstrated therapeutic utility in mice, similar to when CDKL5 expression was driven by hSyn. The rAAV.CDLK5 vector, carrying an AAVrh91 capsid, used with an AAV vector genome containing an engineered nucleic acid sequence achieved CDKL5 expression levels similar to rAAVhu68.CDKL5. CDKL5 protein levels in mouse brain were higher with the Ubc promoter compared to the hSyn promoter.

[0177] Example 5: Usefulness of human CDKL5 isoforms 2 to 4 It has been shown that there are at least four detectable Cdkl5 mRNA splice variants in human and mouse brain, although the existence of isoforms 2-4 as stable proteins has not been established. Isoform 1 accounts for >85% of brain CDKL5.

[0178] We codon engineered the coding sequences for human CDKL4 isoforms 2-4 and cloned the coding sequences into the same AAV vector as before. AAV9-PHP.B vector coding for isoforms 1-4 was tail IV injected into adult Cdkl5-ko mice and brains harvested 2 weeks later for Western blot analysis. We found all four isoforms to be robustly expressed and display the expected gel migration pattern. For follow-up, we selected the engineered sequences that resulted in expression of isoforms 2-4 that closely resembled the expression levels of isoform 1. EB2 phosphorylation levels produced by isoform 2 were slightly higher compared to isoform 1 and slightly lower by isoforms 3 or 4.

[0179] Three alternative CDKL5 isoforms were injected into neonatal Cdkl5-ko mice to test how their potential therapeutic effects compare to isoform 1. FIG. 12 shows significant correction of the clasping phenotype with treatment with alternative CDKL5 isoforms (2, 3, and 4). FIGS. 8A-8D provide CDKL5 expression levels or activity of AAV.CDLK5 vector constructs for expressing isoform 1, isoform 2, isoform 3, or isoform 4. FIG. 8A shows the CDKL5 expression levels or activity of AAV vectors (5×10) expressing each of these isoforms compared to vehicle-injected wild-type mice and vehicle-injected knockout mice. 10 Figure 8B shows expression levels in knockout mice injected with vehicle (PBS), vehicle injected, or AAV.CDKL5-1co using pS222EB2. Figure 8C shows CDKL5 activity determined using pS222EB2 for the groups in Figure 8A. Figure 8D shows CDKL5 expression levels for the groups in Figure 8B.

[0180] In summary, all isoforms are expressed as proteins and have equivalent catalytic activity. Therapeutic outcomes with isoforms 2, 3, or 4 in CDD mouse models are comparable to isoform 1 when tested head-to-head (5×10 10 GC, neonatal ICV). Overall, CDKL5 gene therapy with CDKL5 isoform 1 is a promising and safe approach in male CDD model mice with significant therapeutic efficacy.

[0181] Example 6: NHP Pilot Study for Toxicity and Safety Testing of hCDKL5 Gene Therapy We wished to investigate the expression pattern and safety profile of the AAV-hSyn-CDKL5-1co.WPRE construct packaged in AAVhu68 capsids in NHPs. The vector was generated using the vector genome described herein and the production methods previously described. See, for example, WO2018 / 160582 (hereby incorporated by reference). Groups of six rhesus macaques (4-6 years old) were injected via the cisterna magna (ICM). Different conditions were tested, A. Dose 1×10 14 GC, injected in 1 ml of buffer B. Dose 1 x 10 14 GC, injected in 3 or 5 ml of buffer C. Dose 3 x 10 14 GC, injected in 3 ml of buffer D. Dose 1 x 10 14 GC was injected in 1 ml of buffer, following pretreatment for 2 days with the diuretic acetazolamine (Diamox®), which reduces CSF production.

[0182] Figure 36A shows a schematic of the intracisternal cyst (ICM) administration procedure. Figure 36B shows a more detailed overview of ICM administration as a fluorescence-guided procedure. Injection into the CSF via the intracisternal cyst (ICM) provides the best access to the brain.

[0183] Briefly, toxicity and safety testing of hCDKL5 gene therapy was performed using the AAVhu68-hSyn-Cdkl5-1co-WPRE vector for non-human primate (NHP) studies. In preliminary studies, three different doses, 3 × 10 12 GC / Animal, 1×10 13 GC / animal, and 3 x 10 13 GC / animal was evaluated. For the pilot study, 1 × 10 14 The dose of GC / animal was selected and two different volumes (3 mL and 5 mL) were evaluated for delivery of AAV vectors into the cerebrospinal fluid (CSF) via the cisterna magna. The other study group received 1 × 10 14 GC / animal or 3×10 14GC / animal (subject) was used.

[0184] Additionally, 2×10 12 , 1×10 13 , and 3 × 10 13 Doses of GC / subject were tested. Treatment with CDKL5 vectors was well tolerated and no signs of altered clinical blood chemistry were observed. No changes from baseline were noted in observations from cage-side neurological examination.

[0185] Necropsy was performed 28 days after injection, followed by molecular analysis, histology, and pathology. Overall, no major transduction differences were found in major organs other than the CNS (e.g. liver transduction was likely already at a maximum). No major transduction differences (still very high transduction rates) were observed in the spinal cord and DRG, but significant changes in brain tissue were evident depending on the injection parameters. The highest transduction increase was seen in the cortex. Diamox leads to a slight increase in transduction efficiency through the brain.

[0186] Figure 17 shows the vector biodistribution for each NHP across non-neural, spinal tract, and brain tissues. In Figure 18, only the brain vector biodistribution data is shown. The results show that strong transduction of the dorsal root ganglion (DRG), moderate to low transduction of brain tissue, and transduction leakage into neural tissue is observed. Pathological results showed mild axonal damage in the dorsal white matter tract. Small differences in hCDKL5 mRNA levels were observed. There is a trend towards higher mRNA expression when a 3ml injection volume was used. We also visualized the distribution of hCDKL5 mRNA by in situ hybridization (ISH). The dorsal root ganglion (DRG) showed very strong mRNA expression across all six NHPs. In the motor cortex, a small number of transduced neurons were observed (<10%). Occasionally, small clusters of transduced neurons were found. There were no significant differences in hCDKL5 mRNA positive neurons in the motor cortex between NHPs in this study.

[0187] Pathological review did not reveal any gross lesions across tissues or NHPs. NHPs receiving the highest dose showed mild signs of inflammatory cell infiltration in the liver. Additionally, all NHPs showed mild to moderate spinal cord axonopathy and DRG satellitosis.

[0188] Additionally, we investigated vector expression and safety studies in NHPs of the AAVrh91.UbC.CDKL5-1co.miR183 and AAVrh91.CBh.CDKL5-1co.miR183 vectors. The vectors were delivered at 3 × 10 10 GC was administered to C. macaques. To evaluate the effect of CDKL5 expression after AAV vector administration, pathological analysis and neurological examination were performed.

[0189] 3 × 10 via the ICM pathway 10 Toxicity and safety were evaluated after administration of AAVrh91.UbC.CDKL5-1co.miR183 and AAVrh91.CBh.CDKL5-1co.miR183 vectors to NHPs at a dose of 1 × 10 GC. Clinical blood chemistry tests showed normal results throughout the study, with no elevations in ALT or AST observed. No signs of safety or toxicity concerns were observed. Analysis of cerebrospinal fluid (CSF) showed two NHPs with persistent mild pleocytosis, but no evidence of inflammation at the ICM injection site during necropsy. Cageside neurological examination showed no obvious changes or defects. Histopathology slide review examined DRG / SpC, peripheral nerves, and other organs. Figure 26A shows that 3 × 10 GC administered via the ICM route was associated with increased cellularity in the NHPs. 10 Figure 26B shows the severity scores observed in DRG neurons from tissue collected in the cervical, thoracic, and lumbar regions from NHPs treated with AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183 vectors at a GC dose of 3×10 via the ICM route. 10Figure 26C shows the severity scores observed in spinal cord neurons from tissue collected in the cervical, thoracic, and lumbar regions from NHPs treated with AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183 vectors at a GC dose of 3×10 via the ICM route. 10 Figure 1 shows the severity scores observed in the sural nerve from proximal and distal collected tissues from NHPs treated with AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183 vectors at a GC dose of 100 mg / kg.

[0190] Next, we investigated the vector copy number in different tissues (Figures 27 and 37). Figure 27 shows the results of vector copy number plotted as GC / diploid genome in different tissues of NHPs after administration of AAVrh91.UbC.CDKL5-1co.miR183 or AAVrh91.CBh.CDKL5-1co.miR183 vectors. Figure 37A shows the analysis of brain transduction as measured by vector genome copies via qPCR of DNA / RNA extracted from different brain regions of NHPs after administration of AAVrh91.UbC.CDKL5-1co.miR183. We observed good transduction in the cortex (about 10 GC / cell) and high transduction in DRG neurons and hepatocytes (about 100 gc / cell). Furthermore, we analyzed the expression of the transgene in different tissues (Figures 28 and 37B). Figure 28 shows the relative expression of CDKL5 plotted per 100 ng of cDNA in various CNS tissues of NHPs (motor cortex, somatosensory cortex, parietal cortex, hippocampus, thalamus) compared to the results observed in mouse brain. Figure 37B shows the relative CDKL5 transgene expression (mRNA) measured via qPCR of RNA extracted from different NHP brain regions after administration of AAVrh91.UbC.CDKL5-1co.miR183 (3×10 10 (compared to expression in mouse brain when administered at a dose of GC).

[0191] We next performed in situ hybridization (ISH) microscopy to examine transgene expression in DRG tissues. We observed that CDKL5 transgene mRNA was sparsely detected throughout the NHP brain. Although we observed sporadic transgene expression, it should be noted that some DRG neurons were overexposed during analysis (data not shown). Also, the sensitivity of the detection technique may underestimate the number of neurons expressing the CDKL5 transgene.

[0192] Additionally, single neuron analysis of brain tissue after administration of AAVrh91.UbC.CDKL5-1co.miR183 was performed via two-step PCR detection of the presence of vector genome and CDKl5 expression, and confirmed by PCR detection of vector genome copies or CDKL5 mRNA from bulk tissue (Figures 38A and 38B). Figure 38A shows the results of molecular analysis of CDKL5 gene therapy outcome on a single neuron basis plotted as percentage of transduced neurons measured by vector genome copies. Figure 38B confirms the results obtained from the single neuron analysis. Figure 38B shows CDKL5 transgene expression levels measured from bulk mRNA plotted as percentage of transgene-expressing neurons. From these results, we observed many neurons expressing the CDKL5 transgene, but at modest levels.

[0193] In conclusion, stable CDKL5 protein expression in neurons can be achieved using the AAV-CDKL5 vector (SEQ ID NO: 1) examined in our study. AAV-CDKL5 gene therapy significantly improved the phenotype of CDD mouse models. Additionally, AAV-CDKL5 vectors can be efficiently delivered to non-human primates via the cisterna magna and expressed throughout the CNS.

[0194] Example 7: hCDKL5 Gene Therapy AAVhu68.UbC.hCDKL5-1co.miR183.rBG AAVhu68.UbC.hCDKL5-1co.miR183.rBG is an AAV serotype hu68 (AAVhu68) vector expressing a mutant coding sequence of the human CDKL5 isoform 1 gene. AAVhu68.UbC.hCDKL5-1co.miR183.rBG addresses a significant unmet need by providing functional CDKL5 protein in the CNS, thereby correcting the underlying cause of the disease, as described below. The first-in-human (FIH) study described is an open-label, multicenter, dose-escalation study of AAVhu68.UbC.hCDKL5-1co.miR183.rBG administered via intracisternal (ICM) injection to evaluate safety, tolerability, and exploratory efficacy endpoints in pediatric (age ≥ 30 days) and adult subjects with CDKL5 deficiency (CDD).

[0195] Route of administration Many animal models of monogenic central nervous system (CNS) diseases have been successfully treated using AAV-mediated gene transfer, and several early human studies using first-generation AAV vectors demonstrated the safety of vector delivery to the brain (Janson et al., 2002, Mandel and Burger, 2004, Kaplitt et al., 2007, Mittermeyer et al., 2012, Bartus et al., 2014). However, the low efficiency of these vectors hindered the translation of efficacy in animal models into clinical benefit. The advent of second-generation AAV vectors has significantly increased the feasibility of gene transfer to the brain. Specifically, several clade F isolates, such as AAV9, have shown highly efficient brain transduction (Gray et al., 2013, Haurigot et al., 2013, Hinderer et al., 2014b, Hinderer et al., 2015). Using these more efficient vectors, gene therapy has shown greatly enhanced potential for treating various neurological disorders, and several programs utilizing second-generation vectors have progressed to the clinic ( Haurigot et al., 2013 , Hinderer et al., 2014b , Hinderer et al., 2015 , Gurda et al., 2016 ).

[0196] Early studies of CNS gene transfer were challenged not only by the low gene transfer efficiency of first generation AAV vectors, but also by the limitations of available delivery methods. Most early preclinical and clinical studies utilized vector injections directly into the brain or spinal cord parenchyma (Vite et al., 2005; Worgall et al., 2008; Colle et al., 2010; Ellinwood et al., 2011; Tardieu et al., 2014). While this method results in robust transduction near the injection site, translating this approach to diseases affecting cells throughout the CNS has been challenging, as multiple vector injections were required to achieve widespread transgene delivery. An additional obstacle to CNS gene transfer is the finding that intraparenchymal vector injections can cause inflammation at the injection site, which can promote an adaptive immune response to the transgene product (Worgall et al., 2008; Colle et al., 2010; Ellinwood et al., 2011; Ciesielska et al., 2013). Two alternative vector delivery methods have been developed to more safely and effectively target large areas of the CNS.

[0197] The first was based on the discovery that some AAV vectors, including AAV9, can transduce cells within the CNS after IV delivery (Foust et al., 2009). However, IV vector delivery has two significant limitations. First, the low efficiency of vector penetration into the CNS requires very large vector doses to achieve therapeutic levels of transgene expression, potentially increasing the risk of systemic toxicity and requiring quantities of vector that may be impossible to manufacture for many patient populations (Gray et al., 2011; Hinderer et al., 2014b; Gurda et al., 2016). Second, gene transfer into the CNS after IV vector delivery is notably limited by pre-existing NAbs against the vector capsid (Gray et al., 2011). Given the high prevalence of AAV NAbs in humans, this leaves a significant population of patients who are not candidates for IV AAV therapy. To circumvent the limitations of IV AAV for targeting the CNS, an alternative approach, intrathecal (IT) vector delivery, has been developed. Using cerebrospinal fluid (CSF) as a vehicle for vector dispersion, IT ROA has the potential to achieve transgene delivery throughout the CNS and peripheral nervous system (PNS) with a single, minimally invasive injection. Animal studies have shown that by eliminating the need to cross the blood-brain barrier, IT delivery results in substantially more efficient CNS gene transfer at much lower vector doses than those required for IV approaches (Gray et al., 2011; Hinderer et al., 2014b). Antibodies are present at very low levels in the CSF, and IT vector delivery is not affected by pre-existing NAbs to the AAV capsid, making this approach applicable to a broader patient population (Haurigot et al., 2013). IT AAV delivery can be performed using a variety of routes for CSF access. Lumbar puncture (LP) is the most common method for accessing the CSF and was therefore evaluated as a route for AAV administration in NHPs.Delivery of AAV9 vectors to the CSF via the LP was found to be at least 10-fold less efficient in transducing brain and spinal cord cells compared to injection of higher vectors at the level of the cisterna magna (Hinderer et al., 2014b).

[0198] Due to the excellent brain transduction achieved by a single ICM injection in NHPs, this ROA was selected for clinical trials of AAVhu68.CB7.CI.hARmiR3610.WPRE.rBG. In common procedures, ICM injection (also known as suboccipital puncture) was eventually replaced by LP in the pre-imaging era due to rare cases of damage to the brainstem or nearby blood vessels (Saunders and Riordan, 1929). Today, the procedure can be performed under real-time computed tomography (CT) guidance, allowing visualization of important structures such as the medulla, vertebral artery, and posterior inferior cerebellar artery during needle insertion (Pomerantz et al., 2005; Hinderer et al., 2014b).

[0199] The AAVhu68.UbC.hCDKL5-1co.miR183.rBG filled drug product (FDP) consists of a non-replicating recombinant adeno-associated virus (rAAV) vector active ingredient and formulation buffer. The rAAV vector is produced at a contract manufacturing organization (CMO). The AAVhu68.UbC.hCDKL5-1co.miR183.rBG is produced using procedures that ensure the safety, identity, quality, purity, and strength of the product, with practices in compliance with both the "US Food and Drug Administration's (FDA's) Guidance for Industry cGMP for Phase 1 Investigational Drugs" (July 2008) and the "FDA Guidance for Industry: Chemistry, Manufacturing, and Control (CMC) Information for Human Gene Therapy Investigational New Drug Applications (INDs)" (January 2020).

[0200] The manufacturing process for AAVhu68.UbC.hCDKL5-1co.miR183.rBG involves transient transfection of human embryonic kidney 293 (HEK293) cells with plasmid DNA. To support clinical development, single or multiple batches of bulk drug substance (BDS) are produced by polyethyleneimine (PEI-)-mediated triple transfection of HEK293 cells in a bioreactor. Harvested AAV material is sequentially purified by clarification, tangential flow filtration (TFF), affinity chromatography, and anion exchange chromatography in a single-use closed bioprocessing system, when possible. Drug substance (DS) and drug product (DP) are formulated in intrathecal final formulation buffer (ITFFB; artificial CSF with 0.001% poloxamer 188). One or more batches of BDS are frozen, then thawed, pooled as necessary, adjusted to target concentration, sterile filtered through a 0.2 μm filter, and filled into vials. Filling data will be provided as part of the lot documentation package.

[0201] The scale-independent manufacturing process will be monitored using a droplet digital polymerase chain reaction (ddPCR) potency assay that will also be used to define the clinical dose. The assay was developed and qualified for use by assessing linearity, accuracy and precision. The same assay will be used throughout program development.

[0202] Description of biological products The biological product includes hCDKL5-1co, human cyclin-dependent kinase-like 5 isoform 1 (engineered mutant); ITR, inverted terminal repeat; miR183, microRNA-183; polyA, polyadenylation; rBG, rabbit β-globin; UbC, ubiquitin C), the sequence elements of which are detailed below (see also SEQ ID NO: 49 (vector genome), and SEQ ID NO: 50 (expression cassette)).

[0203] Manufacturing: Ingredients and Materials AAVhu68.UbC.hCDKL5-1co.miR183.rBG is produced by triple plasmid transfection of HEK293 cells with an AAV cis plasmid (pAAV.UbC.hCDKL5-1co.miR183.rBG.KanR (vector genome of SEQ ID NO:49)), an AAV trans plasmid encoding the AAV2 rep and AAVhu68 cap genes (pAAV2 / hu68n.KanR (comprising SEQ ID NO:55)), and a helper adenovirus plasmid (pAdΔF6.KanR). The size of the vector genome packaged in AAVhu8.UbC.hCDKL5-1co.mirR183.RBG is 4857 bases (including wild-type length ITRs). As set forth in SEQ ID NO:49, the vector genome in the plasmid is a 130 base pair truncated AAV2 The shortened ITRs are restored to their wild-type length of 145 base pairs during vector DNA amplification using the internal A element as a template.

[0204] The cis-plasmid contains the following vector genome sequence elements: Inverted terminal repeats (ITRs): ITRs are identical reverse complement sequences derived from AAV2 (130 base pairs [bp] (SEQ ID NO: 51), GenBank: NC_001401) and flank all components of the vector genome. The ITRs function as both an origin of vector DNA replication and a packaging signal for the vector genome when AAV and adenovirus helper functions are provided in trans. Thus, the ITR sequences represent the only cis sequences required for vector genome replication and packaging.

[0205] - Human Ubiquitin C (UbC) promoter: This ubiquitous promoter (1229 bp, GenBank: D63791.1) was selected to drive expression of the transgene product (SEQ ID NO: 52) in CNS cell types.

[0206] ● Coding sequence: The coding sequence is an engineered embodiment of the human CDKL5 isoform 1 gene (2883 bp, GenBank: NP_001310218.1 (SEQ ID NO: 20)). This isoform comprises over 85% of total brain CDKL5 expression and is considered the primary brain isoform of CDKL5. (Note: The original name of the predominant brain isoform of CDKL5 was isoform 2. However, CDKL5 isoform 2 was recently redesignated CDKL5 isoform 1 in Hector et al 2016. The GenBank sequence listed above has not been updated to reflect this change in nomenclature and is still designated CDKL5 isoform 2, but refers to the coding sequence of AAVhu68.UbC.hCDKL5-1CO.miR183.RBG as CDKL5 isoform 1 to reflect the current nomenclature used in the field for the CDKL5 predominant brain isoform) (SEQ ID NO: 22).

[0207] ● MicroRNA-183 (miR183): The four 22 bp target sequences of miR183 (GenBank: NR_029615.1) are contained in the 3' untranslated region of the human CDKL5 sequence. MicroRNAs post-transcriptionally downregulate the expression of target messenger ribonucleic acid (mRNA) in multicellular organisms by influencing both mRNA stability and translation. Since miR183 expression is mainly restricted to the DRG, the miR183 target sequences allow DRG-specific downregulation of the human CDKL5 transgene product. (SEQ ID NO: 11).

[0208] Rabbit β-globin polyadenylation signal (rBG polyA): The rBG polyA signal (127 bp, GenBank: V00882.1) promotes efficient polyadenylation of the transgene mRNA in cis. This element functions as a signal for transcription termination, a specific cleavage event at the 3' end of the nascent transcript, and the addition of a long polyadenylation tail (SEQ ID NO: 53).

[0209] The AAV2 / hu68 transplasmid pAAV2 / hu68 (comprising SEQ ID NO: 55) is used. The AAVhu68 transplasmid encodes four WT AAV serotype 2 (AAV2) Rep proteins and three WT AAV VP capsid proteins from AAVhu68. An adenovirus helper plasmid was used that contains the regions of the adenovirus genome important for AAV replication, namely E2A, E4, and VA RNA (the function of adenovirus E1 is provided by HEK293 cells). However, this plasmid does not contain other adenovirus replication or structural genes. The plasmid does not contain cis elements important for replication, such as adenovirus ITRs. The adenovirus genes E2, E4, and VA remaining in this plasmid, together with E1 present in HEK293 cells, are necessary for the production of AAV vectors.

[0210] Manufacturing process overview AAVhu68.UbC.hCDKL5-1co.miR183.rBG for FIH clinical trials will be produced by transient transfection of HEK293 cells with plasmid DNA followed by downstream purification. Figures 47A, 47B show a flow diagram of the manufacturing process. Figure 46A shows a flow diagram of the upstream manufacturing process for the drug substance. Figure 46B shows a flow diagram of the downstream manufacturing process for the drug substance. Proposed in-process testing is shown on the right side of the diagram. A description of each production and purification step is also provided.

[0211] ITFFB manufacturing Intrathecal Final Formulation Buffer (ITFFB) solution is used at the clinical site to dilute the drug product prior to administration according to the study protocol. ITFFB diluent is a sterile aqueous solution containing the same excipients as the drug product without the active agent. ITFFB solution is stored frozen at -60°C or below.

[0212] Example 8 - Further evaluation of the pharmacology, safety, and toxicity of AAVhu68.UbC.hCDKL5-1co.miR183.rBG: The pharmacology, safety, and toxicity of AAVhu68.UbC.hCDKL5-1co.miR183.rBG, an AAV serotype hu68 (AAVhu68) vector encoding an engineered mutant version of the human cyclin-dependent kinase-like 5 (CDKL5) isoform 1 gene, have been evaluated in various studies with other candidates as described herein.

[0213] The improvement in behavioral phenotype correlated with an increase in transgene product expression (CDKL5 protein) and activity (phosphorylation of EB2 substrate) to wild-type levels in disease-relevant target tissues (brain) 14 weeks after treatment. Because the open field test, nest building test, and hind paw clasping test were found to be the most sensitive assays for evaluating the efficacy of AAV administration in mouse models of CDD, these assays were selected as readouts for future subsequent mouse pharmacology studies.

[0214] Pharmacology studies will evaluate the efficacy of AAVhu68.UbC.hCDKL5-1co.miR183.rBG in neonatal (PND0-1) male Cdkl5KO / Y mice and assess the MED. The MED will be determined based on expression of the transgene product (human CDKL5) and effects on neurological and behavioral phenotypes reminiscent of clinical features observed in CDD patients. Seizures, a clinical feature of CDD in humans, will not be evaluated in the MED study. Finally, toxicology studies will evaluate the safety, tolerability, pharmacology (expression of the transgene product), biodistribution, and excretion of AAVhu68.UbC.hCDKL5-1co.miR183.rBG following ICM administration to juvenile male and female rhesus monkeys.

[0215] Methods: Pharmacology study In vivo pharmacology studies will be performed in C57BL / 6J (wild type) mice, mouse models of CDD (male Cdkl5 KO mice and female Cdkl5 HET mice), and two species of NHP (rhesus and African green monkeys).

[0216] A. Mouse Models of CDD The nonclinical pharmacology studies described herein utilize a knockout mouse model of CDD in which exon 6 of X-linked mouse Cdkl5 is deleted, resulting in a significant reduction in Cdkl5 mRNA and the absence of detectable CDKL5 protein (Wang et al. 2012). This knockout mutation in the CDD mouse model recapitulates a CDKL5 patient-associated splice site mutation that causes skipping of human exon 7 (homologous to mouse exon 6), resulting in a premature stop codon in human exon 8 and leading to a lack of residual CDKL5 protein expression in humans carrying this mutation. Male mice are commonly utilized by groups studying the CDD mouse phenotype. This is because male Cdkl5KO / Y mice (which are hemizygous for the X-linked Cdkl5 knockout allele) typically display more severe and consistent phenotypes than female Cdkl5KO / X mice (which are heterozygous for the X-linked Cdkl5 knockout allele and also display variable tissue mosaicism for expression of the wild-type Cdkl5 gene due to random X-chromosome inactivation). Many of the phenotypes observed in male Cdkl5KO / Y mice are also reminiscent of those seen in CDD patients, including social behavioral phenotypes, motor coordination abnormalities / hyperactivity, cognitive dysfunction, deficits in neural circuit communication, and biochemical defects resulting from impaired CDKL5 kinase activity. Phenotypes observed in male Cdkl5KO / Y mice are discussed below.

[0217] Social behavior phenotype: Antisocial and autistic-like behaviors reminiscent of those observed in CDD patients have been observed in male Cdkl5KO / Y mice at approximately 8 weeks of age. In a three-chamber social approach test, male Cdkl5KO / Y mice spent more time occupying and sniffing novel objects in the nonsocial chamber compared to the time spent in the social chamber with a novel stimulus mouse, and when exposed to the barrier-free chamber, the interaction time of male Cdkl5KO / Y mice with other mice in the social chamber was significantly reduced compared to wild-type mice, suggesting a reduced social preference. Nest-building behavior of Cdkl5KO / Y mice was also impaired in the home cage environment by this age, suggesting a deficit in social behavior and not due to a defect in the olfactory system (Wang et al., 2012).

[0218] Motor Coordination / Anxiety-Like Phenotype: Male Cdkl5KO / Y mice exhibit various motor phenotypes reminiscent of symptoms in CDD patients, which appear by approximately 10-11 weeks of age. For example, by this age, decreased latency to fall in the rotarod assay and abnormal hindlimb clasping were observed in male Cdkl5KO / Y mice, indicating loss of motor coordination. Anxiety-like motor behaviors (e.g., obsessive-compulsive, hyperactive, and / or risk-prone behaviors) have also been observed in male Cdkl5KO / Y mice in the elevated zero maze test and open field test (Wang et al., 2012).

[0219] Cognitive phenotype and neural circuit deficits: Male Cdkl5KO / Y mice exhibit impaired cognitive function, reminiscent of the cognitive symptoms observed in patients with CDD. The cognitive phenotype includes deficits in motor activity along with impaired learning and memory. For example, significant deficits in context- and cue-dependent behavioral responses in male Cdkl5KO / Y mice were observed by 9–12 weeks of age based on contextual fear conditioning assays. Furthermore, male Cdkl5KO / Y mice also exhibit deficits in auditory evoked event-related potentials (ERPs), indicative of impaired neural circuit activity, reminiscent of those seen in patients with CDD. ERPs are stereotyped electrophysiological responses to specific sensory, cognitive, or motor stimuli. As a measure of sensory information processing, ERPs have been exploited as a readout of neural circuit communication and have been shown to be altered in cognitive disorders such as schizophrenia and autism. Disruptions within neuronal networks may contribute to the delayed behavioral responses observed in Cdkl5 KO mice. Circuit communication depends on oscillations spanning low or high frequencies, with low frequencies being associated with long-distance neuronal circuit communication. Similar to the neuronal deficits reported in patients with autism spectrum disorder, oscillation strength at low delta, theta, and alpha frequencies is attenuated in Cdkl5 KO mice (Wang et al., 2012).

[0220] Biochemical defects: In contrast to patients with CDD, Cdkl5 KO mice do not exhibit spontaneous or refractory epilepsy. This lack of phenotype in CDD mice may be due to the increased seizure resistance conferred by the age of the animals, the duration of the study, and the genetic background of the Cdkl5 KO mouse model (C57BL / 6) (Wang et al., 2012 and Amendola et al., 2014). No abnormal EEG patterns were observed in Cdkl5 KO mice before the age of 12 weeks. Several mouse models of CDD have been generated through heterozygous mutations in mouse strains. Recently, a high frequency of epileptic events was observed in aged female Cdkl5 KO mice at 42 weeks of age (Mulcahey et al., 2020). Age-dependent effects on seizure-like events were also observed in a specific strain of Cdkl5 KO mice (Cdkl5RS9X / + females), which showed earlier onset of seizures at 16 weeks of age, myoclonic seizure-like events (Racine Stage 3-5) at 32 weeks of age, and severe seizure-like events (Racine Stage 3-5) at 57 weeks of age (Terzic et al. 2021). The seizure phenotype is age-dependent, requiring evaluation to be performed in older mice. Based on this observation, no seizure phenotypes are observed in completed and planned mouse pharmacology studies utilizing neonatal Cdkl5 KO mice.

[0221] Given the similarities with human CDD, neonatal male Cdkl5KO / Y mice treated at the presyndromic stage of the disease represent the most relevant animal model of the intended patient population for evaluation of the potential efficacy of AAVhu68.UbC.hCDKL5-1co.miR183.rBG in MED trials.

[0222] B. Non-human primates NHPs have been selected for POC large animal pharmacology studies. These studies include both rhesus and African green monkeys. NHPs were selected for pilot pharmacology studies because the toxicity and immune responses of NHPs closely represent those of humans. Furthermore, the dimensions of the NHP central nervous system (CNS) of both rhesus and African green monkeys serve as a representative model of the target clinical population, allowing administration of AAVhu68.UbC.hCDKL5-1co.miR183.rBG via the intended clinical route (ICM administration).

[0223] Genotype Because Cdkl5 is an X-linked gene, initial POC pharmacology studies utilize female Cdkl5KO / X mice (heterozygous for the Cdkl5 KO allele) and male Cdkl5KO / Y mice (hemizygous for the Cdkl5 KO allele) to model CDD in humans. MED studies are performed in male mice, therefore male Cdkl5KO / Y mice are evaluated. However, genotype and sex are difficult to determine in newborn mice (PND0-1), so in MED studies, whole fetuses of newborn mice, including female Cdkl5KO / X, are dosed. However, only male Cdkl5KO / Y mice are enrolled and analyzed to determine MED. Wild-type C57BL / 6 mice have also been selected for initial POC and MED studies, as they are of a similar genetic background to the Cdkl5 KO mouse model and therefore useful as a healthy control group.

[0224] sex Male Cdkl5KO / Y mice, female Cdkl5KO / X mice, and gender-matched C57BL / 6J wild-type controls are included in the initial POC study to characterize the severity and progression of the CDD phenotype. Male Cdkl5KO / Y mice have been selected for MED study because the data obtained and described herein show that male Cdkl5KO / Y mice have a more severe phenotype than female Cdkl5KO / X mice in certain assessments (e.g., open field test) that are important for determining MED. Male Cdkl5KO / Y mice are also preferred for the planned MED study because random X-chromosome inactivation leads to mosaic Cdkl5 expression in females due to Cdkl5 being an X-linked gene. Mosaicism for Cdkl5 expression can lead to phenotypic variation due to animal-to-animal variability in the overall percentage of cells expressing wild-type Cdkl5 alleles versus cells expressing Cdkl5 knockdown alleles, and female mice are not optimal for use in MED studies.

[0225] Male and female rhesus and African green monkeys have been used in POC pharmacology studies. Both sexes were selected to model the intended patient population (male and female CDD patients) in the planned clinical trials.

[0226] age All mouse pharmacology studies evaluated neonatal mice administered vector on PND0-1. This age was chosen because it is the earliest feasible treatment time point and represents a presymptomatic stage of the disease prior to the onset of prominent clinical symptoms, which typically begin to appear at approximately 8-10 weeks of age in male Cdkl5KO / Y mice, depending on the assay used. Thus, the neonatal (PND0-1) mouse model reflects, to the greatest extent practicable, the disease stage of the youngest intended patient population.

[0227] POC studies conducted in NHPs evaluated adult (3-10 years old) animals. This age range was deemed sufficient for initial comparisons of expression profiles and safety / toxicity of lead candidate vectors, as well as modeling, to the greatest extent practicable, the cisterna magna size and anatomy of the youngest intended patient population.

[0228] Dose Selection Completed POC mouse pharmacology studies have determined that 5.0 x 10 is close to the highest feasible dose for ICV administration in mice based on expected vector titers and volume constraints. 10 A separate POC mouse pharmacology study evaluated the test article at a 2-fold lower dose (2.5 × 10 10 AAVhu68.UbC.hCDKL5-1co.miR183.rBG will be evaluated in a 10-mL GC study because studies performed with other candidate AAV vectors expressing human CDKL5 have shown efficacy at this dose comparable to that of the highest viable dose (data not shown). The MED study will then include a high dose, two medium doses, and a low dose, selected based on the results of the aforementioned POC study. The doses selected will allow for the evaluation of dose-dependent efficacy while ensuring that the dose levels evaluated in the MED study are distinct.

[0229] NHP POC pharmacology studies conducted in rhesus monkeys showed a 3.0 × 10 13 A high dose of GC was utilized, which is close to the maximum feasible dose for ICM administration in NHPs based on expected vector titers and volume constraints. The medium and low doses were approximately 3- and 10-fold lower than the maximum feasible dose, respectively. This range was chosen to ensure that the doses were distinct and encompassed a dose range similar to that which may be assessed in mouse MED pharmacology studies and GLP-compliant NHP toxicity studies. NHP POC pharmacology studies conducted in African green monkeys used a dose of 5.0 × 10 13 A high dose of GC was utilized because this dose approximates the maximum feasible dose for ICM administration in NHPs based on expected vector titers and volume constraints.

[0230] Test period All pharmacological studies performed in neonatal mice were 13–14 weeks in length, which is a sufficient duration to assess the behavioral defects that begin to appear at 8–11 weeks of age along with any biochemical phenotypes associated with loss of Cdkl5 kinase activity.

[0231] Both POC pharmacology studies conducted in NHPs are 56 days long, a duration sufficient to evaluate animals during the expected onset, peak, and plateau of transgene product expression and to evaluate potential acute safety signals.

[0232] Route of administration The ICV route (i.e., administering the vector directly into the cerebral ventricles) has been chosen for pharmacology studies in mice because it allows for efficient delivery of AAV vectors to disease-relevant target tissues (brain). Use of the intended clinical route (ICM administration into the cisterna magna), which uses CSF as a vehicle for vector dispersion with the potential to achieve transgene product expression throughout the CNS via a single minimally invasive injection, is not feasible in mice due to the small size of the animals. However, the ICM route has been chosen for NHP POC pharmacology studies because it reflects the intended clinical route and allows for the use of a clinical administration system comparable to that utilized in planned clinical trials.

[0233] C. Pharmacological Endpoints Open field test (hyperactivity test) The open field test can be used to measure locomotor activity and anxiety-like behavior in rodents. It consists of a circular or square enclosure with an open, unobstructed field in the center of the apparatus. The open field arena transmits infrared light from one side of the enclosure to the other. When the beam is interrupted by an animal moving through it, it is counted as a "beam break." During testing, mice are placed in the enclosure and the tester leaves the room. The mouse's behavior is recorded for 30 minutes using video tracking software and beam breaks are quantified. Activity, and anxiety, is assessed based on overall locomotor activity in the center of the field (horizontal activity based on the number of x / y-axis beam breaks, along with percent center beam breaks), as well as movements that occur away from the walls in the enclosure, including rearing behavior (z-axis beam breaks). Cdkl5KO / Y mice have been shown to exhibit hyperactivity in the open field test, consisting of increased locomotor activity in the center of the field (increased x / y-axis beam breaks) and increased rearing (increased z-axis beam breaks). Movement within the center of the arena (determined as the percent of beam breaks occurring in the center of the maze [i.e., as opposed to the percent of beam breaks occurring on the periphery of the maze closer to the walls]) indicates reduced anxiety-like behavior. Reductions in activity in the center of the field (x / y-axis beam breaks and percent center beam breaks) and / or reduced rearing behavior (z-axis beam breaks) are expected to indicate amelioration of the hyperactivity phenotype of Cdkl5 KO mice following AAV administration.

[0234] Marble burying (hyperactivity test) The marble burying assay evaluates the obsessive-compulsive and hyperactive phenotype in rodents. Marble burying was evaluated in Cdkl5 KO mice because these mice showed a hyperactive phenotype in other assays (e.g., open field test). To perform the marble burying assay, mice are first adapted for 30 minutes in their home cage. Then, mice are placed in a test cage in which 12 marbles are pre-placed (3 marbles x 4 marbles) on a horizontal pile on the dry cage bedding. The tester leaves the room and leaves the mouse in the cage for 30 minutes. After 30 minutes, the mouse is returned to its home cage and the number of marbles buried 50% or more in the cage bedding is counted. A reduction in the number of buried marbles is expected to indicate an improvement in the hyperactive phenotype of Cdkl5 KO mice after AAV administration.

[0235] Elevated zero maze (risky behavioral test) The elevated zero maze measures rodent behavior based on the animal's ability to balance exploration / gathering behavior (curiosity) with avoidance of potential danger (risk-taking) (a test of anxiety-like behavior). The elevated zero maze is a circular maze with alternating "open" and "closed" quadrants. Elevated zero maze testing is performed by first allowing the mouse to acclimate for 30 minutes in its home cage. The animal is then placed in the open area of ​​the maze, which is illuminated by a lamp at one end of the apparatus, creating a brightly lit open area and a poorly lit closed area. The tester leaves the room and the animal is videotaped for approximately 15 minutes. The number of entries into the illuminated open zone, the time spent in the open zone, and the total distance traveled are then determined with EthoVisionXT video tracking software. Cdkl5 KO mice have been shown to spend more time in the brighter open zone and are more risk-prone in the elevated zero maze test compared to normal control mice that spent more time in the darker closed area of ​​the apparatus (Wang et al., 2012). A reduction in the number of open zone entries, time spent in the open zone, and / or total distance traveled while in the maze by Cdkl5 KO mice would be expected to indicate an amelioration of the disease phenotype following AAV administration.

[0236] Nest building (social behavior test) Nest building is a social behavior in rodents' home cages that is important for shelter, warmth, and reproduction. Nest building involves mice chopping up materials, such as tightly packed cotton or thread, placed in the cage and then arranging it into a nest (Deacon 2006). Cdkl5 KO mice exhibit impaired nest building behavior, characterized by either failure to build a nest (i.e., no chopping of nestlets to create materials for nest building) or poor quality nest building, suggesting a social behavioral deficit. Nest building is assessed by first acclimating mice in the testing room for approximately 24 hours. Mice are then housed singly with pre-weighed cotton square nestlets in the late afternoon. After approximately 20 hours, the quality of the newly created nest is scored according to the 5-point scoring system presented in the table immediately below. Additionally, remaining unchopped nestlets are weighed and recorded (i.e., approximately 0.1 g or more of nestlets remain). An increase in fragmented nestlets (i.e., a decrease in the percent of intact nestlets) and an increase in nest building quality scores suggest improved quality of nest building and would be expected to indicate an overall improvement in the social behavioral phenotype of Cdkl5 KO mice following AAV administration. [Table 4]

[0237] Hindlimb clasping (motor control test) Hindlimb clasping is a motor control phenotype observed in Cdkl5 KO mice in which animals retract their hindlimbs towards the body and clasp the hindlimbs together when held upside down (Wang, 2012). Hindlimb clasping is assessed by hanging a mouse by its tail over a cage for 20-30 seconds and observing the behavior of its hindlimbs according to the scoring system shown in the table immediately below. A reduction in cumulative hindlimb clasping scores is expected to indicate an improvement in the motor phenotype of Cdkl5 KO mice following AAV administration. [Table 5]

[0238] Y-maze spontaneous change test (spatial working memory test) The Y-maze spontaneous change test evaluates the exploratory activity of mice and evaluates spatial working memory. The apparatus for the test is an opaque Y-shaped enclosed maze with three arms arranged at an angle of 120° to each other. In this test, the animal is first placed in the center of the maze. The tester leaves the room and the mouse is allowed to freely explore the maze for about 5 minutes. The animal's movements are video recorded to evaluate arm entries, which are defined as when the animal moves all four limbs completely into an arm of the maze. Since normal mice usually prefer to explore novel arms of the maze instead of returning to parts of the maze that they have previously explored, it is expected that normal mice will show a tendency to explore arms of the maze that have not been visited recently. The tendency to visit arms of the maze that have not been explored recently is called spontaneous change, and the spontaneous change percentage is calculated by dividing the number of spontaneous changes by the total number of entries minus 2 and multiplying the result by 100. Because Cdkl5 KO mice show reduced spontaneous changes in the Y-maze test compared to WT, an increased percentage of spontaneous changes would be expected to indicate an improved exploratory / spatial working memory phenotype in Cdkl5 KO mice following AAV administration.

[0239] Contextual fear conditioning test (learning and memory test) The contextual fear conditioning test evaluates the learning and memory abilities of rodents. For this test, mice are placed in a conditioning chamber for 3 minutes, at the end of which time a training phase is performed in which the mice receive a 1.5 mA shock to the foot. The mice are then left in the chamber for 1 minute after shocking. The next day, the mice are returned to the conditioning chamber again for 5 minutes for the test phase. The animals are videotaped to determine the percentage of time the animals "freeze" (i.e., no movement is detected except for respiratory movements) during the test phase. Cdkl5 KO mice typically spend a shorter percentage of the test phase freezing, indicating a deficit in learning and memory. Thus, an increase in the percentage of test time spent freezing would indicate an improvement in the learning and memory deficits of Cdkl5 KO mice following AAV administration (Yennawar, 2019).

[0240] Transgene product expression - CDKL5 protein (western blot, immunofluorescence), CDKL5 mRNA (in situ hybridization, qPCR, single cell RNAseq) For NHP pharmacology studies, expression of the transgene product in disease-relevant target tissues (brain) has been assessed at the mRNA level by human CDKL5 in situ hybridization, human CDKL5 qPCR, and single-cell RNAseq. For mouse pharmacology studies, expression of the transgene product in disease-relevant target tissues (brain) has been assessed at the protein level by CDKL5 Western blotting, which detects both human CDKL5 and endogenous mouse CDKL5, and human CDKL5 immunofluorescence using an anti-human CDKL5 antibody. AAV administration is expected to increase CDKL5 expression in the brain, where the protein is required for normal neuronal function.

[0241] Transgene product activity - CDKL5 substrate phosphorylation (phospho-EB2 Western blot) The kinase activity of CDKL5 can be assessed by measuring the phosphorylation of its substrates, including the microtubule-associated protein EB2. CDKL5 kinase activity is assessed in disease-relevant target tissues (brains) of Cdkl5 KO mice by phospho-EB2 Western blotting using an antibody that recognizes phosphorylation of serine 222. AAV administration is expected to increase the abnormally low levels of EB2 phosphorylation typically observed in the brains of Cdkl5 KO mice.

[0242] Example 9 - Further evaluation of the pharmacology, safety, and toxicity of AAVhu68.UbC.hCDKL5-1co.miR183.rBG: A. Proof-of-concept pharmacology and assay development study to evaluate the efficacy of rAAV vectors in a mouse model of CDD This POC pharmacology study examined neonatal male Cdkl5 KO / Y Mouse and female Cdkl5 KO / X We evaluated the efficacy of AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 following ICV administration in mice. AAVhu68.hSyn.hCDKL5 1co.WPRE.SV40 utilizes the same capsid (AAVhu68) and expresses the same transgene product (human CDKL5 isoform 1). However, AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 contains a different promoter (hSyn vs. UbC) and polyA (SV40 vs. rBG), incorporates a WPRE sequence 3' of the transgene sequence, and lacks the miR183 target sequence for DRG detargeting.

[0243] Neonatal (PND0-1) male Cdkl5 KO / Y Mouse and female Cdkl5 KO / X Mouse, 5.0 x 10 10100 mice received a single ICV dose of AAVhu68.hSyn.hCDKL5 1co.WPRE.SV40 at a dose of 1000 mg / kg / day. Age-matched male and female wild-type C57BL / 6 mice were also administered vehicle (PBS) as a control. In-life assessments included daily survival checks, weekly weight measurements starting 4 weeks after treatment, and behavioral assessments (open field test, nest building test, marble burying test, hind paw clasping test, Y-maze test, elevated zero maze test, and contextual fear conditioning test) performed 11–14 weeks after treatment. At 14 weeks after treatment, mice were necropsied and Western blot analysis was performed to assess the effect on CDKL5 knockdown and substrate phosphorylation (phospho-EB2) in disease-relevant target tissues (brain).

[0244] AAV treatment was well tolerated. Male Cdkl5 mice treated with either AAV or vehicle throughout the course of the study KO / Y Mouse and female Cdkl5 KO / X The mice gained weight comparably to sex-matched wild-type mice administered either AAV or vehicle (data not shown), confirming observations in the published literature that defects in postnatal weight gain (i.e., failure to thrive) are not a feature of the Cdkl5 knockout mouse phenotype.

[0245] In the open field test, AAV-treated male Cdkl5 mice KO / Y Mouse and female Cdkl5 KO / X Mice were treated with vehicle-treated male Cdkl5 KO / Y Mouse and female Cdkl5 KO / X Compared to female Cdkl5 mice, they showed reduced hyperactivity (significantly reduced horizontal activity and rearing, as well as reduced tendency for central activity), and AAV treatment normalized activity to near wild-type levels for all three parameters assessed. KO / X Correction of the hyperactivity phenotype was not observed in male Cdkl5 mice, as males generally showed a more severe hyperactivity phenotype for all three parameters assessed compared to male Cdkl5 mice. KO / YThis was most evident in mice (FIGS. 15A-F, see also Example 3).

[0246] 15A-15F show male Cdkl5 mice after ICV administration of an AAV vector expressing human CDKL5. KO / Y Mouse and female Cdkl5 KO / X Figure 15 shows the results of the open field test in mice. Figure 15A shows the results of the horizontal activity open field test in males plotted as X / Y axis beam break. Figure 15B shows the results of the horizontal activity open field test in females plotted as X / Y axis beam break. Figure 15C shows the results of the rearing open field test in males plotted as Z axis beam break. Figure 15D shows the results of the rearing open field test in females plotted as Z axis beam break. Figure 15E shows the results of the central activity open field test in males plotted as central beam break percentage. Figure 15F shows the results of the central activity open field test in females plotted as central beam break percentage. Briefly, neonatal (PND0-1) male Cdkl5 mice were treated with 100 mg / kg / day of ... KO / Y Mouse and female Cdkl5 KO / X Mouse, 5.0 x 10 10 Additional age-matched C57BL / 6 wild-type mice received a single ICV dose of either AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 (N=16 males, 10 females) or vehicle (PBS, N=16 males, 13 females) at a dose of 5.0 × 10 GC. 10Mice were administered either GC (N = 10 males, 11 females) or vehicle (PBS, N = 15 males, 12 females) ICV as a control. At 11-14 weeks after treatment, an open field test was performed. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 based on two-way ANOVA followed by Tukey's multiple comparison test comparing AAV-treated Cdkl5 KO mice (dark blue line) to vehicle-treated Cdkl5 KO (red line). Abbreviations: AAV: adeno-associated virus, ANOVA: analysis of variance, Cdkl5: cyclin-dependent kinase-like 5 (gene, mouse), GC: genome copies, ICV: intracerebroventricular, KO: knockout, N: number of animals, PBS: phosphate-buffered saline, PND: postnatal days, WT: wild type.

[0247] Vehicle-treated male Cdkl5 KO / Y Mouse and female Cdkl5 KO / X Although mice tended to bury more marbles than healthy vehicle-treated, gender-matched wild-type controls in the marble-burying assay, the difference in the number of buried marbles was not statistically significant, thus precluding the use of this assay for the evaluation of the efficacy of AAV treatment (Figure 10B, see also Example 3).

[0248] In the elevated zero maze test, AAV-treated male Cdkl5 KO mice were significantly superior to vehicle-treated male Cdkl5 KO / Y Compared with female Cdkl5 mice, they showed a trend toward reduced risk-taking behavior, as indicated by spending less time on average in the open zone of the maze, but the reduction was not statistically significant. KO / X In mice, female Cdkl5 mice showed only a minor increase in risk-prone behavior compared with healthy sex-matched vehicle-treated wild-type controls. KO / X A trend toward decreased open zone time following AAV administration in mice was observed, but the differences were not significant (Figures 39A, 39B, 40A, 40B, 41A, 41B).

[0249] AAV-treated female Cdkl5 mice exhibited open zone entry in the elevated zero maze test. KO / X Mice were treated with vehicle-treated female Cdkl5 KO / X Male Cdkl5 mice showed a marked reduction in risk-prone behavior, as indicated by significantly fewer open zone entries, compared with male Cdkl5 mice, and AAV treatment normalized open zone entries to near wild-type levels. In contrast, the risk-prone phenotype, as assessed by this parameter, was not significantly different in male Cdkl5 mice. KO / Y Not evident in vehicle-treated male Cdkl5 mice KO / Y Mice exhibited a similar number of entries into the open zone as healthy gender-matched vehicle-treated wild-type controls, thus precluding the use of this parameter in males for the assessment of the efficacy of AAV treatment (Figures 39A, 39B, 40A, 40B, 41A, 41B).

[0250] Vehicle-treated male Cdkl5 mice were assessed for distance traveled in the elevated zero maze test. KO / Y Mouse or female Cdkl5 KO / X No significant differences were observed in mice compared to healthy gender-matched vehicle-treated wild-type controls, precluding the use of this parameter in both genders for the evaluation of the efficacy of AAV therapy (Figures 39A, 39B, 40A, 40B, 41A, 41B).

[0251] FIG. 39A shows male Cdkl5 mice following ICV administration of an AAV vector expressing human CDKL5 plotted as time in the open zone (seconds). KO / Y Figure 39B shows the results of the elevated zero maze test in mice. Figure 39B shows the results of the elevated zero maze test in mice. KO / X Figure 40A shows the results of the elevated zero maze test in mice. Figure 40B shows the results of the elevated zero maze test in mice. KO / YFigure 40B shows the results of the elevated zero maze test in mice. Figure 40B shows the results of the elevated zero maze test in mice. KO / X Figure 41A shows the results of the elevated zero maze test in mice. Figure 41B shows the results of the elevated zero maze test in mice. Figure 41A shows the results of the elevated zero maze test in mice. KO / Y Figure 41B shows the results of the elevated zero maze test in mice. Figure 41B shows the results of the elevated zero maze test in mice. KO / X Results of the elevated zero maze test in mice are shown. Briefly, neonatal (PND0-1) male Cdkl5 KO / Y Mouse and female Cdkl5 KO / X Mouse, 5.0 x 10 10 Additional age-matched C57BL / 6 wild-type mice received a single ICV dose of either AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 (N=16 males, 10 females) or vehicle (PBS, N=12 males, 10 females) at a dose of 5.0 × 10 GC. 10 Mice were administered either GC (N = 10 males, 11 females) or vehicle (PBS, N = 14 males, 11 females) ICV as a control. Elevated zero maze testing was performed 11-14 weeks after treatment. *p<0.05, **p<0.01, ***p<0.001 based on one-way ANOVA followed by Sidak's multiple comparison test comparing all groups against each other except for AAV-treated wild-type mice. Abbreviations: AAV: adeno-associated virus, ANOVA: analysis of variance, Cdkl5: cyclin-dependent kinase-like 5 (gene, mouse), GC: genome copies, ICV: intracerebroventricular, KO: knockout, N: number of animals, PBS: phosphate-buffered saline, PND: postnatal days, WT: wild-type.

[0252] In the nest building assay, AAV-treated male Cdkl5 KO / Y Mouse and female Cdkl5 KO / X Mice were treated with vehicle-treated male Cdkl5 KO / YMouse and female Cdkl5 KO / X Compared to mice, rats showed improved nest quality, as indicated by a significant increase in nest quality score and a significant decrease in the percentage of intact nestlets. Notably, AAV treatment normalized both nest quality score and intact nestlet size to wild-type levels (Figures 10A and 10F).

[0253] FIG. 10F shows male Cdkl5 mice after ICV administration of an AAV vector expressing human CDKL5. KO / Y Mouse and female Cdkl5 KO / X Results of the nest building assay in mice are shown, plotted as a percentage of intact original nestlet weight. Briefly, neonatal (PND0-1) male Cdkl5 mice were KO / Y Mouse and female Cdkl5 KO / X Mouse, 5.0 x 10 10 Additional age-matched male and female C57BL / 6 wild-type mice received a single ICV administration of either AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 (5.0 × 10 10 Mice were administered ICV with either GC (AAV-treated, N=11) or vehicle (PBS, N=11) as a control. At 11-14 weeks after treatment, nest construction tests were performed. Nest quality scores were assigned and the percent of original intact nestlets based on weight was measured. Based on one-way ANOVA followed by Sidak's multiple comparison test comparing all groups against each other except AAV-treated wild-type mice (for nest quality scores), based on one-way ANOVA followed by Tukey's multiple comparison test comparing all groups against each other (for percent of original nestlets), *p<0.05, **p<0.01, ***p<0.001. Abbreviations: AAV: adeno-associated virus, ANOVA: analysis of variance, Cdkl5: cyclin-dependent kinase-like 5 (gene, mouse), GC: genome copies, ICV: intracerebroventricular, KO: knockout, N: number of animals, PBS: phosphate-buffered saline, PND: postnatal days, WT: wild-type.

[0254] AAV-treated male Cdkl5 KO / Y Mouse and female Cdkl5 KO / X Mice were treated with vehicle-treated male Cdkl5 KO / Y Mouse and female Cdkl5 KO / X Compared with mice, AAV-treated male Cdkl5 mice showed significantly reduced hindlimb clasping scores, indicating a significant improvement in motor control. KO / Y Mouse and female Cdkl5 KO / X AAV treatment did not completely normalize this motor phenotype, as the hindlimb clasping scores of the mice remained higher than healthy gender-matched wild-type controls (Figures 14A and 14B, also see Example 3).

[0255] In the Y-maze test, AAV-treated female Cdkl5 KO / X Mice were treated with vehicle-treated female Cdkl5 KO / X The results showed an increase in the percent spontaneous change when compared to female Cdkl5 mice, and AAV treatment normalized the percent change to near wild-type levels. KO / X Mice showed an increased tendency to explore less recently visited arms of the maze, suggesting improved spatial learning / memory. In contrast, this phenotype was not observed in vehicle-treated male Cdkl5 mice. KO / Y Male Cdkl5 mice showed similar percent spontaneous changes to healthy, gender-matched, vehicle-treated wild-type controls. KO / Y This was not evident in mice, thus precluding the use of this parameter in males for the evaluation of the efficacy of AAV therapy (Figures 42A and 42B).

[0256] FIG. 42A shows male Cdkl5 expression after ICV administration of an AAV vector expressing human CDKL5 plotted as percent spontaneous change. KO / Y Figure 42B shows the results of the Y-maze test in mice. Figure 42B shows the results of the Y-maze test in mice. KO / XResults of the Y-maze test in mice are shown. Briefly, neonatal (PND0-1) male Cdkl5 KO / Y Mouse and female Cdkl5 KO / X Mouse, 5.0 x 10 10 Additional age-matched C57BL / 6 wild-type mice received a single ICV dose of either AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 (N=16 males, 10 females) or vehicle (PBS, N=16 males, 13 females) at a dose of 5.0 × 10 GC. 10 Mice were administered either GC (N = 10 males, 11 females) or vehicle (PBS, N = 15 males, 12 females) ICV as a control. Y-maze testing was performed 11-14 weeks after treatment. *p<0.05 based on one-way ANOVA followed by Sidak's multiple comparison test comparing all groups against each other except for AAV-treated wild-type mice. Abbreviations: AAV: adeno-associated virus, ANOVA: analysis of variance, Cdkl5: cyclin-dependent kinase-like 5 (gene, mouse), GC: genome copies, ICV: intracerebroventricular, KO: knockout, N: number of animals, PBS: phosphate-buffered saline, PND: postnatal days, WT: wild-type.

[0257] In the contextual fear conditioning test, AAV-treated female Cdkl5 KO / X Mice were treated with vehicle-treated female Cdkl5 KO / X The percentage of freezing behavior was significantly increased compared to female Cdkl5 mice, indicating a significant improvement in the phenotype after AAV administration. KO / X AAV treatment in mice increased the percent freezing to the level of healthy, gender-matched wild-type controls, indicating normalization of the phenotype. KO / Y AAV treatment in mice significantly reduced the expression of vehicle-treated male Cdkl5 KO / Y The freezing percentage did not increase significantly in female Cdkl5 mice compared to female Cdkl5 mice, indicating that AAV administration did not significantly increase the freezing percentage in female Cdkl5 mice compared to female Cdkl5 mice. KO / X Despite significant efficacy in mice, male Cdkl5 KO / YThe results show that treatment with IFN-γ did not ameliorate this phenotype in mice (Figures 43A and 43B).

[0258] FIG. 43A shows male Cdkl5 mice following ICV administration of an AAV vector expressing human CDKL5 plotted as percent freezing. KO / Y Figure 43B shows the results of a contextual fear conditioning test in mice. Figure 43B shows the results of a contextual fear conditioning test in mice. KO / X Results of the contextual fear conditioning test in mice are shown. Briefly, neonatal (PND0-1) male Cdkl5 mice were KO / Y Mouse and female Cdkl5 KO / X Mouse, 5.0 x 10 10 Additional age-matched C57BL / 6 wild-type mice received a single ICV dose of either AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 (N=16 males, 10 females) or vehicle (PBS, N=16 males, 13 females) at a dose of 5.0 × 10 GC. 10 Mice were administered either GC (N = 10 males, 11 females) or vehicle (PBS, N = 15 males, 12 females) ICV as a control. Contextual fear conditioning tests were performed 11-14 weeks after treatment. **p<0.01, ***p<0.001, ****p<0.0001 based on one-way ANOVA followed by Sidak's multiple comparison test comparing all groups against each other except for AAV-treated wild-type mice. Abbreviations: AAV: adeno-associated virus, ANOVA: analysis of variance, Cdkl5: cyclin-dependent kinase-like 5 (gene, mouse), GC: genome copies, ICV: intracerebroventricular, KO: knockout, N: number of animals, PBS: phosphate-buffered saline, PND: postnatal days, WT: wild-type.

[0259] Improved behavioral phenotype in male Cdkl5 mice KO / Y Mouse and female Cdkl5 KO / X This correlated with normalization of transgene product expression and activity following AAV administration in mice. Specifically, vehicle-treated male Cdkl5KO / Y Mouse and female Cdkl5 KO / X The absence of detectable brain CDKL5 protein expression observed in mice was compared with that observed in male Cdkl5 mice 14 weeks after AAV administration. KO / Y Mouse and female Cdkl5 KO / X In vehicle-treated male Cdkl5 mice, the expression of Cdkl5 was restored to wild-type levels. KO / Y Mouse and female Cdkl5 KO / X The observed decrease in substrate phosphorylation by CDKL5 in mouse brain (as determined by phospho-EB2 levels) was compared with that in male Cdkl5 mice 14 weeks after AAV administration. KO / Y Mouse and female Cdkl5 KO / X The expression of IFN-γ was restored to wild-type levels in mice (Figures 44A and 44B).

[0260] FIG. 44A shows male Cdkl5 mice following ICV administration of an AAV vector expressing human CDKL5 (CDKL5 / tubulin). KO / Y Mouse and female Cdkl5 KO / X The results of transgene product expression in mice are shown. Figure 44B shows male Cdkl5 mice after ICV administration of an AAV vector expressing human CDKL5 (pS222 / total EB2). KO / Y Mouse and female Cdkl5 KO / X Activity results in mice are shown. Briefly, neonatal (PND0-1) male Cdkl5 KO / Y Mouse and female Cdkl5 KO / X Mouse, 5.0 x 10 10 Additional age-matched C57BL / 6 wild-type mice received a single ICV administration of either AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 (5.0 × 10 10Mice were administered either ICV with GC (at 400 nmol / kg / day) or vehicle (PBS) as a control. At necropsy 14 weeks after treatment, brain tissue was collected for evaluation of transgene product expression (CDKL5 protein expression) and transgene product activity (phosphorylation of EB2) by Western blot. **p<0.01, ***p<0.001 based on one-way ANOVA followed by Sidak's multiple comparison test comparing all groups against each other except AAV-treated wild-type mice. Abbreviations: AAV: adeno-associated virus, ANOVA: analysis of variance, Cdkl5: cyclin-dependent kinase-like 5 (gene, mouse), GC: genome copies, ICV: intracerebroventricular, KO: knockout, N: number of animals, PBS: phosphate-buffered saline, PND: postnatal days, pS222: phospho-serine 222, WT: wild-type.

[0261] Cumulatively, this POC pharmacology and assay development study demonstrated that a single ICV administration of an AAV vector similar to AAVhu68.UbC.hCDKL5-1co.miR183.rBG utilizing the same capsid (AAVhu68) and transgene (human CDKL5 isoform 1) resulted in significant improvement of the behavioral phenotype in a neonatal mouse model of CDD. The improvement in behavioral phenotype in this mouse model correlated with an increase in transgene product expression (CDKL5 protein) and activity (phosphorylation of EB2 substrate) to wild-type levels in disease-relevant target tissues (brain) 14 weeks after treatment.

[0262] The most sensitive assays to evaluate the efficacy of AAV administration in CDD mouse models were the open field test, the nest building test, and the hindlimb clasping test. Specifically, in the open field test, male Cdkl5 mice KO / Y Mouse and female Cdkl5 KO / X Both mice showed normalization of hyperactivity after AAV treatment, as indicated by a significant reduction in horizontal activity and rearing to wild-type levels. The effect of treatment on the open field test was significantly greater in males than in females in this assay. KO / X The male Cdkl5 mice showed significantly more severe phenotypes than the female mice. KO / Y This is most evident in mice, where male Cdkl5KO / Y In mice, this resulted in increased sensitivity of the test. KO / Y Mouse and female Cdkl5 KO / X Both male and female Cdkl5 mice showed normalization of nest building defects after AAV administration, characterized by a significant increase in nest building score and a significant decrease in nestlet weight, both of which were normalized to wild-type levels. KO / Y Mouse and female Cdkl5 KO / X Both mice also showed improved motor coordination phenotype following AAV administration, characterized by a significant decrease in hindlimb clasping scores, although the phenotype did not completely normalize to wild-type levels.

[0263] Several tests (marble burying test and Y-maze test) demonstrated that vehicle-treated male Cdkl5 KO / Y Mouse and / or female Cdkl5 KO / X Minimal or no phenotypic abnormalities were observed in the mice, indicating that the CDD mouse model phenotypic assessments were ineffective, making future evaluation of dose-dependent therapeutic effects difficult. Furthermore, one additional study (contextual fear conditioning) showed AAV therapeutic effects in only one sex (female Cdkl5 mice). KO / X KO mice, but male Cdkl5 KO / Y (not shown in mice), which precludes the use of this assessment in future pharmacological studies.

[0264] Based on the results of this study, the open field test, nest building test, and hindlimb clasping test were performed using male Cdkl5 mice. KO / Y Mouse and female Cdkl5 KO / X This assay was found to be the most sensitive assay for assessing the efficacy of AAV administration in both naive and non-naive mice and was therefore selected for use in future pharmacology studies.

[0265] B. Proof-of-concept vector comparative pharmacology study of clinical candidate leads following ICM administration to adult rhesus monkeys This POC vector comparison study was aimed at evaluating the safety, tolerability, and transgene product expression of two lead candidates (AAVhu68.hSyn.hCDKL51co.WPRE.SV40 [evaluated in Example 3 and Example 9A] and AAVhu68.UbC.hCDKL5-1co.SV40) following ICM administration to adult rhesus macaques. AAVhu68.hSyn.hCDKL51co.WPRE.SV40 and AAVhu68.UbC.hCDKL51co.SV40 utilize the same capsid (AAVhu68) and express the same transduction product (human CDKL5). AAVhu68.UbC.hCDKL5-1co.SV40 also contains the same promoter as AAVhu68.UbC.hCDKL5-1co.miR183.rBG(UbC). However, although AAVhu68.hSyn.hCDKL5 1co.WPRE.SV40 contains a different promoter (hSyn vs. UbC) and a WPRE sequence 3' to the transduced gene, both vectors have a different polyA (SV40 vs. rBG) and lack the miR183 target sequence for DRG detargeting found in AAVhu68.UbC.hCDKL5-1co.miR183.rBG.

[0266] Briefly, adult (3-10 years) male and female rhesus monkeys were administered a low dose (3.0 × 10 12 of GC), medium dose (1.0 × 10 13 of GC), or high dose (3.0 × 10 13NHPs received a single ICM dose of either AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 or AAVhu68.UbC.hCDKL5-1co.SV40 at 100 ng / mL (GC). In-life assessments included daily observations, body weight, neurological monitoring, and clinical pathology of blood (CBC, coagulation panel, serum chemistry) and CSF. All NHPs were necropsied on day 56. At necropsy, disease-relevant target tissue (brain) was collected along with additional highly perfused CNS (spinal cord), PNS (DRG, TRG, and sciatic nerve), and peripheral tissues to assess vector biodistribution. Brain, spinal cord, and PNS tissues were evaluated for histopathology as these are tissues highly transduced by the ICM pathway. Pituitary tissue was also evaluated for histopathology. Additional brain tissue was collected to assess transgene product expression (human CDKL5 mRNA expression by ISH and qPCR) in this disease-relevant target tissue. Serum was collected and stored for possible future evaluation of NAbs to the vector capsid. PBMCs and tissue-resident lymphocytes were also collected and stored for possible future evaluation of T cell responses to the vector capsid and / or transgene product (IFN-γ ELISpot).

[0267] AAVhu68.UbC.hCDKL5-1co.miR183.rBG was well tolerated, with no test article-related findings observed on cage side observations, neurological monitoring, or hematological clinical pathology. 13 A transient mild CSF lymphocytosis (6 or more white blood cells [WBCs] / μL) that was considered test article related was observed in a single animal receiving AAVhu68.UbC.hCDKL5 1co.SV40 at the mid-dose (1.0 × 10 13 Additional animals receiving AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 (GC) showed significant lymphocytosis on day 8. However, this subclinical finding was likely due to hemodilution in the sample (1780 RBC) and was not observed at other time points evaluated.

[0268] Histopathology on day 56 did not observe DRG sensory neuron degeneration for any vector at any dose (N=0 / 3 DRG segments per animal). However, in most animals, axonopathy was observed in both the dorsal white matter tracts of the spinal cord and the peripheral nerve (sciatic nerve). The findings of axonopathy were suggestive of DRG sensory neuron pathology as axons from these DRG neurons project to this region of the spinal cord and to the peripheral nerves. In all cases, spinal cord and peripheral nerve axonopathy were asymptomatic and no clinical abnormalities were noted on daily observation or neurological examination.

[0269] Regarding axonopathy in the dorsal white matter tracts of the spinal cord, the incidence and severity appeared to be largely dose-dependent for both vectors, increasing from no axonopathy at the lowest dose (N=0 / 3 segments for both vectors) to minimal severity (grade 1) at the mid and high doses (N=1 / 3 and N=3 / 3 segments for AAVhu68.hSyn.hCDKL5 1co.WPRE.SV40, respectively, and N=2 / 3 segments for AAVhu68.UbC.hCDKL5 1co.SV40 in mice at both doses). When compared between vectors, both vectors showed a similar severity of spinal cord axonopathy, with minimal (grade 1) lesions observed in all cases (N=2 / 3 animals for each vector). Moreover, no clear differences in the incidence of spinal cord axonopathy between these vectors were observed. At the mid-dose, AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 reduced the incidence of spinal cord axonopathy (N=1 / 3 segments) compared to AAVhu68.UbC.hCDKL5 1co.SV40 at the same dose (N=2 / 3 segments), but the opposite was observed at the high dose, with AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 showing a higher incidence of spinal cord axonopathy (N=3 / 3 segments) compared to AAVhu68.UbC.hCDKL5 1co.SV40 (N=2 / 3 segments).

[0270] Regarding axonopathy in the peripheral nerve (sciatic nerve), the severity and incidence did not appear to be dose-dependent for AAVhu68.UbC.hCDKL5 1co.SV40, with minimal (grade 1) peripheral nerve axonopathy observed for each animal at all doses (N=3 / 3 sciatic nerves, N=3 / 3 animals). In contrast, AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 caused peripheral nerve axonopathy that was dose-dependent in terms of both severity and incidence, with no axonopathy observed at the low or mid dose (N=2 / 2 sciatic nerves, N=2 animals) and mild axonopathy (grade 2) observed at the high dose (N=1 / 1 sciatic nerve, N=1 / 1 animals). Comparing the vectors, AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 resulted in a reduced overall incidence of peripheral axonopathy, while AAVhu68.UbC.hCDKL5 1co.SV40 resulted in a reduced overall severity of axonopathy.

[0271] Evaluation of vector biodistribution on day 56 revealed high levels of transduction throughout the brain for both vectors. Both vectors also showed relatively high levels of transduction in the spinal cord, DRG, peripheral nerves (trigeminal nerve), and spleen. Relatively low transduction was observed for both vectors in peripheral tissues including lung, muscle, heart, kidney, liver, and eye. A dose response was not evident for either vector, likely due to the low number of animals evaluated for this POC study. Comparisons between vectors showed that transduction levels in each tissue, including each brain region evaluated, were similar for each dose, considering expected animal-to-animal variability. Thus, ICM administration of AAVhu68.hSyn.hCDKL5 1co.WPRE.SV40 or AAVhu68.UbC.hCDKL5 1co.SV40 to NHPs resulted in comparable biodistribution profiles with both vectors efficiently transducing disease-relevant target tissues (brain).

[0272] Figure 45A shows vector biodistribution results in adult rhesus macaques after ICM administration of AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 at a low dose. Figure 45B shows vector biodistribution results in adult rhesus macaques after ICM administration of AAVhu68.UbC.hCDKL5-1co.SV40 at a low dose. Figure 45C shows vector biodistribution results in adult rhesus macaques after ICM administration of AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 at a medium dose. Figure 45D shows vector biodistribution results in adult rhesus macaques after ICM administration of AAVhu68.UbC.hCDKL5-1co.SV40 at a medium dose. Figure 45E shows vector biodistribution results in adult rhesus macaques following ICM administration of AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 at a high dose. Figure 45F shows vector biodistribution results in adult rhesus macaques following ICM administration of AAVhu68.UbC.hCDKL5-1co.SV40 at a high dose. Briefly, adult (3-10 years old) male and female rhesus macaques received a low dose (3.0 x 10 12 of GC), medium dose (1.0 × 10 13 of GC), or high dose (3.0 × 10 13 NHPs received a single ICM dose of either AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 or AAVhu68.UbC.hCDKL5-1co.SV40 at 100 GCs (N=1 animal per vector per dose). All NHPs were necropsied on day 56±4 and the indicated tissues were collected for assessment of AAVhu68 vector biodistribution (TaqMan qPCR). The dashed line represents the limit of detection (50 GCs / μg DNA). Abbreviations: AAVhu68: adeno-associated virus hu68, DNA: deoxyribonucleic acid, GC: genome copies, ICM: intracisternae, N: number of animals, NHP: non-human primate, qPCR: quantitative polymerase chain reaction.

[0273] Consistent with the biodistribution profiles observed for each vector, transgene product expression (human CDKL5 isoform 1 mRNA) was detectable at all doses evaluated for both vectors in brain regions relevant for the treatment of CDD, including the cerebellum and entire cortex (Figure 46). Transgene product expression was largely dose-dependent for both vectors, with low levels of expression observed for most brain regions at low doses compared to the mid and high doses for each vector. Comparing between vectors, expression levels for each brain region were largely similar at each respective dose, given the expected inter-animal variability. Thus, ICM administration of AAVhu68.hSyn.hCDKL5 1co.WPRE.SV40 or AAVhu68.UbC.hCDKL5 1co.SV40 to NHPs resulted in similar levels of dose-dependent transgene product expression in the disease-relevant target tissue (brain).

[0274] Figure 46 shows the results of transgene product expression in the brains of adult rhesus macaques following ICM administration of an AAV vector expressing human CDKL5. Adult (3-10 years old) male and female rhesus macaques were administered a low dose (3.0 x 10 12 of GC), medium dose (1.0 × 10 13 of GC), or high dose (3.0 × 10 13 All NHPs received a single ICM dose of either AAVhu68.hSyn.hCDKL5-1co.WPRE.SV40 or AAVhu68.UbC.hCDKL5-1co.SV40 with 100% GC (N=1 animal per vector per dose). All NHPs were necropsied on day 56±4 and brains were collected for assessment of transgene product expression (human CDKL5 isoform 1 mRNA qPCR). Abbreviations: CDKL5-1: cyclin-dependent kinase-like 5 (isoform 1); GC: genomic copies; ICM: intracisternae; mRNA: messenger ribonucleic acid; N: number of animals; NHP: non-human primate; qPCR: quantitative polymerase chain reaction.

[0275] Cumulatively, ICM administration of either AAVhu68.hSyn.hCDKL5 1co.WPRE.SV40 or AAVhu68.UbC.hCDKL5 1co.SV40 to adult male and female rhesus macaques was well tolerated, with no test article-related findings observed in cage-side observations, neurological monitoring, or hematologic clinical pathology for either vector. A transient, mild, asymptomatic CSF lymphocytosis likely related to the test article was observed at day 28 after the medium dose (1.0 × 10 13 A single animal receiving AAVhu68.UbC.hCDKL5 1co.SV40 (100% GC) was observed and resolved without treatment by day 42. Histopathological evaluation of both vectors on day 56 showed asymptomatic axonopathy in the dorsal white matter tracts of the spinal cord and peripheral nerves that was thought to be secondary to DRG sensory neuron degeneration. The incidence and severity of axonopathy in the spinal cord was similar for both vectors, but the severity in the peripheral nerves was slightly lower for AAVhu68.UbC.hCDKL5 1co.SV40 compared to AAVhu68.hSyn.hCDKL5 1co.WPRE.SV40 (grade 1 versus grade 2, respectively). Both vectors also showed similarly robust vector transduction profiles, resulting in transduction product (human CDKL5 mRNA) in disease-relevant target tissues (brain) at day 56.

[0276] Ultimately, this study led to the selection of a hCDKL5 1co transgene sequence lacking the AAVhu68 capsid, UbC promoter, and 3'WPRE elements for further evaluation. These elements were selected based on favorable vector biodistribution and transgene product expression profile in disease-relevant target tissue (brain) following ICM administration of AAVhu68.UbC.hCDKL5 1co.SV40 to NHPs, along with the observation that animals treated with AAVhu68.UbC.hCDKL5 1co.SV40 exhibited more severe peripheral nerve pathology than those administered other vectors.

[0277] C. Proof-of-concept pharmacology and assay development study of AAVhu68.UbC.hCDKL5-1co.miR183.rBG in a mouse model of CDD This POC pharmacology study was conducted in neonatal male Cdkl5 mice to optimize the study design and assays to be used in the planned MED pharmacology study. KO / Y Mouse and female Cdkl5 KO / X To evaluate the therapeutic efficacy of AAVhu68.UbC.hCDKL5-1co.miR183.rBG following ICV administration in mice.

[0278] Briefly, newborn (PND0–1) male Cdkl5 KO / Y Mouse and female Cdkl5 KO / X Mouse, 2.5 x 10 10 Mice received a single ICV administration of either AAVhu68.UbC.hCDKL5-1co.miR183.rBG or vehicle (PBS) at a dose of 100 mg / kg / day (N=12 / group). Additional age-matched male and female C57BL / 6J wild-type mice received vehicle (PBS) as a control (N=12). Ongoing in-life evaluations included daily survival checks, weekly body weight measurements, and behavioral evaluations (open field, nest building, hind paw clasping test) performed 10-11 weeks after treatment. Necropsy was performed 13-14 weeks after treatment. At necropsy, blood was collected for CBC / differential and serum clinical chemistry analysis. A list of tissues was collected for histopathological evaluation. Transgene product expression (CDKL5 Western blot, CDKL5 immunofluorescence) and transgene product activity (phosphorylation of EB2 [phospho-EB2 Western blot]) will be assessed in disease-relevant target tissues (brain) and highly transduced peripheral tissues.

[0279] Efficacy of AAVhu68.UbC.hCDKL5-1co.miR183.rBG following ICV administration to neonatal male Cdkl5KO / Y mice to determine D.MED This pharmacological study examined neonatal male Cdkl5 KO / YEvaluate the efficacy of ICV administered AAVhu68.UbC.hCDKL5-1co.miR183.rBG in mice and determine the MED. The vector used in this study is a toxic vector lot manufactured for a planned GLP-compliant NHP toxicity study.

[0280] Briefly, this study included N=60 neonatal (PND 0–1) AAVhu68.UbC.hCDKL5-1co.miR183.rBG-treated male Cdkl5 mice. KO / Y Evaluation of mice and N=12 age-matched vehicle-treated male C57BL / 6J wild-type controls will be performed. The study includes one necropsy time point (13-14 weeks post-treatment). Four dose levels of AAVhu68.UbC.hCDKL5-1co.miR183.rBG will be evaluated using ICV administration. Dose levels will be selected based on results from an ongoing POC pharmacology study evaluating the efficacy of AAVhu68.UbC.hCDKL5-1co.miR183.rBG administration in this mouse model of CDD (as described above) in addition to POC safety and pharmacology studies of AAVhu68.UbC.hCDKL5-1co.miR183.rBG conducted in adult African green monkeys (also as described above).

[0281] In-life assessments include daily viability checks, weight measurements, and behavioral assessments (open field, nest building, hind paw clasping test). Necropsy is performed after 13-14 weeks of treatment. At necropsy, blood is collected for CBC / differential and serum clinical chemistry analysis. A list of tissues is collected for histopathological evaluation. Transgene product expression (CDKL5 Western blot, CDKL5 immunofluorescence) and activity (phosphorylation of EB2) are assessed in disease-relevant target tissues (brain) and highly transduced peripheral tissues.

[0282] E. Toxicity study of ICM administration of AAVhu68.UbC.hCDKL5-1co.miR183.rBG to juvenile rhesus macaques The 180-day GLP-compliant toxicity study will evaluate the safety, tolerability, pharmacology (CDKL5 mRNA expression), biodistribution, and excretion profile of GTP-213 following single ICM administration at low, medium, or high doses (N=4 / dose) to juvenile (1.5-2 years) male and female rhesus monkeys. Additional age-matched male and female NHPs will receive vehicle (intral final formulation buffer [ITFFB]) as controls (N=2).

[0283] NHPs (rhesus monkeys) were selected for the planned toxicity studies (genotoxicity, carcinogenicity, reproductive toxicity, and developmental toxicity assessments). The highest dose evaluated is the maximum feasible dose based on expected vector titers and maximum administration volume. The medium and low doses are approximately 3- and 10-fold lower than the maximum feasible dose, respectively. This range was chosen to ensure that doses were distinct and encompassed the dose range evaluated in the mouse MED pharmacology study. 180-day study duration with a tentative 90-day necropsy time point for this toxicity study.

[0284] Using CSF as a vehicle for vector dispersion, intrathecal (IT) ROA has the potential to achieve transgene delivery throughout the CNS. Studies in large animal models of lysosomal storage diseases (such as mucopolysaccharidosis [MPS] type I and MPS type VII) have shown that CSF delivery of AAV results in widespread transduction of neurons throughout the brain, a major target tissue for the treatment of CDD (Hinderer et al., 2014a, Hinderer et al., 2015, Gurda et al., 2016). Recent studies examining different routes for CSF access have shown that delivery of AAV vectors via ICM administration is at least 10-fold more efficient in transducing cells of the brain, spinal cord, and spinal motor neurons compared to injection of vectors via lumbar puncture (Hinderer et al., 2014b). Thus, ICM administration has been selected for the planned clinical trial, and the ICM route will be utilized in the planned NHP toxicity study to replicate the intended clinical ROA.

[0285] Methods for scaling from non-clinical to clinical doses Administration of ICM vectors results in immediate vector distribution within the CSF compartment, and both efficacy and toxicity are expected to be related to CNS vector exposure. Thus, doses are scaled by brain mass to provide an approximation of the size of the CSF compartment. Dose conversions are based on brain masses of 0.15 g for neonatal mice (Gu et al., 2012), 90 g for juvenile NHPs (Herndon et al., 1998), 610 g for infants aged 6-8 months, 780 g for infants aged 8-12 months, and 960 g for infants older than 12 months (Dekaban, 1978). Estimated brain weights for each age range of human infants were derived from the male and female brain weights presented in (Dekaban, 1978) by assuming an approximately linear increase in brain weight between neonates (370 g) and infants aged 4-8 months, resulting in a mean estimated brain weight of 488 g for infants aged ≥1 month and <4 months. The value of 610 g corresponds to the mean brain weight of boys and girls aged 4 to 8 months (Dekaban, 1978).

[0286] Examples of dose scaling from neonatal mice, juvenile NHPs, and equivalent human doses are provided in the table immediately below. Dose volumes are also scaled from NHPs to humans based on estimated volumes of brain CSF (Matsumae et al., 1996) and spinal CSF (Rochette et al., 2016). [Table 6]

[0287] Example 10: First-in-human clinical trial protocol overview Overview of First-in-Human Clinical Trials The FIH study is an open-label, multicenter, dose-escalation study of AAVhu68.UbC.hCDKL5-1co.miR183.rBG administered via intracisternal (ICM) injection to evaluate safety, tolerability, and exploratory efficacy endpoints in pediatric (age 30 days or older) and adult subjects with CDKL5 deficiency (CDD). Up to 36 subjects with CDD may be enrolled in the study. The study will initially enroll subjects aged 12 years or older in the first dose-escalation cohort. Staggered enrollment and treatment of younger age groups (ages 2 years or older and <12 years, ages 30 days or older and <2 years) will begin only after available safety, laboratory, and clinical data from the next higher age group has been reviewed by an independent Data Safety Monitoring Board (DSMB). Each age group has a dose escalation where progression to the next dose level requires consent from the DSMB.

[0288] Dose escalation (cohorts 1 and 2) evaluates a single ICM administration of two dose levels of AAVhu68.UbC.hCDKL5-1co.miR183.rBG. The AAVhu68.UbC.hCDKL5-1co.miR183.rBG dose levels to be tested were determined based on data from mouse MED studies and GLP NHP toxicity studies and consist of a low dose (administered to cohort 1) and a high dose (administered to cohort 2). Both dose levels are expected to confer therapeutic benefit, with the understanding that if tolerated, the higher dose is expected to be advantageous and a step forward. Our standard approach is that a safety margin is applied such that the high dose selected for human subjects is 30-50% of the equivalent MTD in NHPs. The low dose is typically 2-3 times lower than the high dose selected, as long as it is a dose that exceeds the equivalent scaled MED in animal studies.

[0289] The dose escalation portion of the study follows a 3+3 design. For each age group, three subjects are enrolled in a dose cohort. If safety data are deemed acceptable by the DSMB, that age cohort may progress to the next dose level. Additionally, the next younger cohort may begin enrolling at the same dose level tested in the older cohort. If one of the first three subjects develops a safety review trigger (SRT) or based on DSMB guidance, up to three additional subjects will be enrolled in the same age and dose cohort.

[0290] During the dose escalation phase, two dose level cohorts (up to 12 subjects) are planned, with performance of the second dose level cohort dependent on available evolving safety, tolerability, and efficacy data. The dose levels, cohort sizes, and safety monitoring of subsequent cohorts will be confirmed by the DSMB prior to enrollment.

[0291] Because the proposed clinical trial is the first to evaluate AAVhu68.UbC.hCDKL5-1co.miR183.rBG in humans, IP dosing of subjects with AAVhu68.UbC.hCDKL5-1co.miR183.rBG will be staggered by at least 6 weeks between subjects to monitor for elevated liver enzymes and assessment of adverse events (AEs) indicative of complications of ICM administration, immune reactions, or other dose-limiting toxicities. Furthermore, this 6-week window captures the time of expected maximal gene expression based on nonclinical data. This duration between IP dosing of subjects may be further refined based on emerging nonclinical data to shorten or lengthen the interval between IP dosing of subjects in the final protocol.

[0292] All treated subjects will be followed for two years to assess the safety profile and characterize the pharmacokinetic and efficacy properties of AAVhu68.UbC.hCDKL5-1co.miR183.rBG in a Phase 1 FIH study. Subjects will then be followed for an additional three years (for a total of five years post-dose) in a subsequent long-term follow-up study to evaluate long-term clinical outcomes. This is in accordance with the "FDA Guidance for Industry: Long Term Follow-Up after Administration of Human Gene Therapy Products" (January 2020). [Table 7-1] [Table 7-2] [Table 7-3] [Table 7-4]

[0293] References 1. Bahi-Buisson, N. et al. Key clinical features to identify girls with CDKL5 mutations. Brain 131, 2647-2661, (2008). 2.Fehr,S.et al.The CDKL5 disorder is an independent clinical entity associated with early-onset encephalopathy.Eur J Hum Genet 21,266-273,(2013). 3.Kalscheuer,V.M.et al.Disruption of the serine / threonine kinase 9 gene causes severe X-linked infantile spasms and mental retardation.American journal of human genetics 72,1401-1411,(2003). 4.Tao,J.et al.Mutations in the X-linked cyclin-dependent kinase-like 5 (CDKL5 / STK9) gene are associated with severe neurodevelopmental retardation.American journal of human genetics 75,1149-1154,(2004). 5.Weaving,L.S.et al.Mutations of CDKL5 cause a severe neurodevelopmental disorder with infantile spasms and mental retardation.American journal of human genetics 75,1079-1093,(2004). 6.Hector,R.D.et al.Characterization of CDKL5 Transcript Isoforms in Human and Mouse.PloS one 11,e0157758,(2016). 7.Baltussen,L.L.,et al.(2018).“Chemical genetic identification of CDKL5 substrates reveals its role in neuronal microtubule dynamics.”EMBO Journal 37(24). 8. Munoz, IM, et al. (2018). “Phosphoproteomic screening identifies physiological substrates of the CDKL5 kinase.” EMBO Journal.

[0294] All documents cited herein are incorporated by reference, as well as U.S. Provisional Patent Application No. 63 / 016,036, filed April 27, 2020, U.S. Provisional Patent Application No. 63 / 091,032, filed October 13, 2020, U.S. Provisional Patent Application No. 63 / 109,608, filed November 4, 2020, International Patent Application No. PCT / US21 / 29185, filed April 26, 2021, and U.S. Provisional Patent Application No. 63 / 256,827, filed October 18, 2021. The electronic sequence listing entitled "UPN-22-9863PCT_SequenceListing_20221018.xml", filed herewith and created on October 18, 2022, of size 281,561 bytes, and the contents of the electronic sequence listing (e.g., sequences and text therein) are hereby incorporated by reference in their entirety. The invention has been described with reference to specific embodiments, but it will be understood that modifications can be made without departing from the spirit of the invention. Such modifications are intended to be within the scope of the appended claims.

Claims

1. 1. A recombinant adeno-associated virus (rAAV) useful for treating CDKL5 deficiency disorder (CDD), comprising: (a) an AAVhu68 or AAVrh91 capsid; (b) a vector genome in the AAV capsid of (a), said vector genome comprising a 5' AAV inverted terminal repeat (ITR), an expression cassette comprising the human CDKL5 sequence from nucleotides 1 to 2883 of SEQ ID NO: 22 operably linked to regulatory sequences that direct its expression and further comprise four tandem miR183 targeting sequences, and a 3' AAV ITR.

2. The rAAV of claim 1, wherein the regulatory sequence further comprises a UbC promoter or an hSyn promoter.

3. 3. The rAAV of claim 1 or 2, wherein the expression cassette comprises the nucleic acid sequence of nucleotides 220 to 4609 of SEQ ID NO:49 (or SEQ ID NO:50), the nucleic acid sequence of nucleotides 226 to 4608 of SEQ ID NO:29 (or SEQ ID NO:59), or the nucleic acid sequence of nt 224 to 4191 of SEQ ID NO:31 (or SEQ ID NO:60).

4. 3. The rAAV of claim 1 or 2, wherein the AAV capsid is an AAVhu68 capsid, and optionally, the AAVhu68 capsid comprises a nucleic acid molecule comprising a vector genome of SEQ ID NO:

49.

5. 3. The rAAV of claim 1 or 2, wherein the vector genome comprises an AAV 5' ITR, a UbC promoter, a hCDKL5 coding sequence, four miR183 targeting sequences, a rabbit globin polyA signal, and an AAV 3' ITR, and optionally, the vector genome further comprises a Kozak sequence.

6. The rAAV of claim 2, wherein the UbC promoter has the sequence of SEQ ID NO:

52. (a) at least one of the miR183 targeting sequences has the sequence AGTGAATTCTACCAGTGCCATA (miR183, SEQ ID NO: 11); and / or (b) the four miR183 targeting sequences are located in tandem and separated by spacer sequences; 3. The rAAV of claim 1 or 2.

8. A pharmaceutical composition comprising the rAAV of claim 1 or 2 and one or more of a carrier, preservative, excipient, or aqueous diluent.

9. 9. The pharmaceutical composition of claim 8, comprising an aqueous liquid suitable for intraventricular or intracisternal injection.

10. 3. The rAAV of claim 1 or 2, which is suitable for treating CDKL5 deficiency.

11. A pharmaceutical composition according to claim 8, suitable for treating CDKL5 deficiency.

12. 3. The rAAV of claim 1 or 2 for use in the preparation of a medicament.

13. The rAAV of claim 1 or 2 for use in treating CDKL5 deficiency.

14. A nucleic acid molecule useful for producing an rAAV vector, comprising a vector genome comprising a 5' AAV inverted terminal repeat (ITR), an expression cassette comprising the human CDKL5 sequence of nucleotides 1 to 2883 of SEQ ID NO:22 operably linked to regulatory sequences that direct its expression and further comprise four tandem miR183 targeting sequences, and a 3' AAV ITR.

15. (a) the vector genome comprises SEQ ID NO: 49; and / or (b) the nucleic acid molecule is a plasmid The nucleic acid molecule of claim 14.

16. 1. An rAAV-producing host cell comprising: (a) a nucleic acid molecule according to claim 14 or 15; (b) a nucleic acid molecule comprising an AAV capsid coding sequence, and optionally further comprising an AAV rep coding sequence; (c) adenovirus helper genes.

Images