WO2024218192A1

WO2024218192A1 - Novel neurotropic adeno-associated virus capsids with detargeting of peripheral organs

Info

Publication number: WO2024218192A1
Application number: PCT/EP2024/060496
Authority: WO
Inventors: - Anggakusuma; Elena Maria SANTIDRIAN YEBRA-PIMENTEL; Sebastiaan Menno Bosma
Original assignee: Uniqure Biopharma BV
Current assignee: Uniqure Biopharma BV
Priority date: 2023-04-18
Filing date: 2024-04-18
Publication date: 2024-10-24
Anticipated expiration: 2025-10-18
Also published as: CN121358754A; AU2024258047A1

Abstract

The present invention relates to novel neurotropic adeno-associated virus (AAV) capsid variants. In particular, the invention relates to novel AAV capsid variants that efficiently transduce cells in the CNS upon systemic administration but that show reduced transduction of peripheral organs. The invention further relates to method for identifying AAV capsid variants with one or more desired characteristics, such as e.g. the combination of CNS tropism and detargeting of peripheral organs.

Description

Novel neurotropic adeno-associated virus capsids with detargeting of peripheral organs

Field of the invention

The present invention relates to the fields of medicine, molecular biology, gene therapy and insect cell culture. In particular, the invention relates to novel neurotropic adeno-associated virus capsids. The invention further relates to method for identifying adeno-associated virus capsid variants with desired characteristics.

Background of the invention

Adeno-associated virus (AAV) has emerged as a preferred platform for gene therapy. Yet, the success of AAV-based therapeutic strategies has been limited amongst others by nonspecific tropism. Specifically, developing novel adeno-associated virus (AAV) capsids that can transduce cells in the central nervous system (CNS) via systemic administration is one of the most intriguing challenges in the field of AAV-based gene therapy of CNS diseases. The inefficiency of blood-brain barrier (BBB) crossing and high-level sequestration and transduction of peripheral organs limit the CNS application of first generation AAV-based vectors due to the narrow therapeutic window. Therefore, several next generation AAV capsids have been engineered in recent years resulting in CNS capsid variants with a different mode of action.

For example, Hanlon et al. (2019, Mol Ther Methods Clin Dev. 15:320-332) and Nonnenmacher et al., (2020, Mol Ther Methods Clin Dev. 20:366-378) report variant AAV9 capsids, with increased CNS transduction efficiency compared to wild type AAV9. However, upon systemic administration, these variant AAV9 capsids still transduce peripheral organs such as liver, skeletal muscle and heart, with similar efficiencies as wild type AAV9. Goersten et al. (2022, Nat Neurosci. 25(1 ): 106-115) disclose AAV capsid variants with brain-wide transgene expression and decreased liver targeting after intravenous delivery. Pulicherla et al. (2011 , Mol Ther. 19(6):1070-8) report the generation of variant AAV9 capsids using random mutagenesis of residues within a surface- exposed region of the major AAV9 capsid proteins. Their capsid variants were classified into three functional subgroups, with respect to parental AAV9 displaying: (i) a functionally defective phenotype with decreased transduction efficiency across multiple tissues; (ii) a selective decrease in liver transduction, or (iii) a similar transduction profile.

There is however, still a need in the art for further improved AAV capsids that efficiently transduce cells in the CNS upon systemic administration but that show reduced transduction of peripheral organs. It is an object of the invention to provide for such AAV capsids with CNS tropism and detargeting of peripheral organs and to provide for means and methods for screening such AAV capsids.

It is further an object of the invention to provide for means and methods for screening and identifying an AAV capsid variant with a desired characteristic, such as e.g. the combination of CNS tropism and detargeting of peripheral organs. Summary of the invention

In a first aspect, there is provided an adeno-associated virus (AAV) serotype 9 (AAV9) or Clade F AAV capsid protein variant comprising a mutation in one or more amino acids, which mutation(s) confer to the capsid protein variant at least one of: i) a phenotype of decreased liver transduction as compared to a control capsid protein not having the mutation(s); and, ii) a phenotype of decreased overall transduction as compared to a control capsid protein not having the mutation(s), and wherein the capsid protein variant further comprises an amino acid sequence that confers increased CNS transduction as compared to a control capsid protein not having the amino acid sequence.

In one embodiment, the AAV capsid protein variant is a variant, wherein at least one of: i) the variant comprises a mutation in one or more amino acids in amino acid regions 452-458, 498-504, 590-595 and/or 582-589, and wherein the mutation(s) result in a phenotype of decreased liver transduction as compared to a control not having the mutation(s); ii) the variant comprises a mutation at T138, S414, G453, K557, T568, T582 and Q590 or any combination thereof, and wherein the mutation(s) result in a phenotype of decreased overall transduction as compared to a control capsid protein not having the mutation(s); and, iii) the amino acid sequence that confers increased CNS transduction to the variant comprises an amino acid sequence selected from the group consisting of STTLYSP, FVVGQSY, DGTLAVPFK and WPTSYDA, and sequences differing by no more than four, three, two or one amino acid(s) therefrom.

In one embodiment, the AAV capsid protein variant is a variant, wherein at least one of: i) the variant comprises at least one mutation selected from the group consisting of mutations at positions W595, Q592, W503, N498, E500, and the amino acid sequence DGAATKN at positions 452 - 458, and wherein the mutation(s) result in a phenotype of decreased liver transduction as compared to a control not having the mutation(s); ii) the variant comprises mutations at both T568 and Q590, or at both S414 and G453, or at all of T138, S414, G453, K557 and T582, or at all of S414, G453, K557 and T582, and wherein the mutation(s) result in a phenotype decreased overall transduction as compared to a control capsid protein not having the mutation(s); and, iii) the amino acid sequence that confers increased CNS transduction to the variant comprises an amino acid sequence selected from the group consisting of FVVGQSY, FWAQSY, FWVQSY, FVVLQSY, FVVIQSY, FVVNQSY, FVVSQSY, FVVEQSY, FVVQQSY, FVVCQSY, FVVTQSY, FVVPQSY, FWQSY, FVGVQSY, FVVQGSY, FVQGVSY, FSVGQVY and FSQGVVY, preferably selected from the group consisting of FVVGQSY, FWAQSY, FVVPQSY, FVGVQSY and FVVQGSY.

In one embodiment, the AAV capsid protein variant is a variant, wherein at least one of: i) the variant comprises a W595C mutation, a Q592L mutation, a W503R mutation, a N498Y mutation, a E500D mutation or any combination thereof, wherein the mutation(s) result in a phenotype of decreased liver transduction as compared to a control not having the mutation(s); ii) the variant comprises the T568P and Q590L mutations, or the S414 and G453D mutations, or the T138A, S414N, G453D, K557E and T582I mutations, or the S414N, G453D, K557E and T582I mutations; and, iii) the inserted amino acid sequence is inserted in the variant, in a hypervariable region of the AAV capsid protein sequence, preferably in a hypervariable region selected from Loop IV or Loop VIII.

In one embodiment, the AAV capsid protein variant is a variant comprising: a) the T138A, S414N, G453D, K557E and T582I mutations and the amino acid sequence FVVGQSY inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589; b) the S414N, G453D, K557E and T582I mutations and the amino acid sequence FVVGQSY or FVVAQSY inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589; c) the T138A, S414N, G453D, K557E and T582I mutations and the amino acid sequence WPTSYDA inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589; d) the S414N, G453D, K557E and T582I mutations and the amino acid sequence WPTSYDA inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589; e) the amino acid sequence DGAATKN at positions 452 - 458 and the amino acid sequence FVVGQSY inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589; or, f) the amino acid sequence DGAATKN at positions 452 - 458 and the amino acid sequence WPTSYDA inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589.

In one embodiment, the AAV capsid protein variant is a variant comprised in an amino acid sequence selected from the group consisting of SEQ ID NO.’s: 6, 7, 9, 10 and 96.

In one embodiment, the AAV capsid protein variant is a variant of at least one of a VP1 , VP2 and VP3 capsid protein.

In a second aspect, there is provided a nucleic acid encoding an AAV capsid protein variant as described herein.

In a third aspect, there is provided a host cell comprising a nucleic acid molecule or expression construct for expression of the capsid protein variants as described herein.

In a fourth aspect, there is provided a method for producing an AAV vector virion as described herein, comprising capsid protein variant as described herein.

In a fifth aspect, there is provided a recombinant AAV (rAAV) vector virion comprising at least one AAV capsid protein variant as described herein, wherein the rAAV vector virion preferably does not comprise at least one of a wild type VP1 , VP2, and VP3 capsid protein.

In one embodiment, the rAAV vector virion comprises a nucleic acid molecule encapsidated by the at least one capsid protein variant, and wherein the nucleic acid molecule comprises a transgene flanked by at least one AAV inverted terminal repeat (ITR).

In a sixth aspect, there is provided a composition comprising an AAV vector virion as described herein, wherein preferably the composition is a pharmaceutical composition comprising the AAV vector virion and at least one pharmaceutically acceptable carrier.

In a seventh aspect, there is provided an AAV vector virion according as described herein, or a composition as described herein, for use as a medicament, wherein preferably the medicament is used in the treatment of a condition of the central nervous system.

In an eighth aspect, there is provided an AAV vector virion as described herein, or a composition comprising the AAV vector virion, for use (as a medicament) in gene therapy. In one embodiment, there is provided a method of gene therapy comprising the step of administering an effective amount of an AAV vector virion as described herein, or a composition comprising the AAV vector virion, to a subject in need of gene therapy. In one embodiment, the gene therapy is for the treatment of a disease defined herein above (e.g. a disease or condition that can be treated by gene therapy of the central nervous system).

In an nineth aspect, there is provided a method for identifying an AAV capsid variant with a desired characteristic, the method comprising: a) providing a library comprising a plurality of individually produced AAV capsid variants, wherein each member in the library differs by at least one amino acid from amino acid sequences of AAV capsid variants of other members in the library, and wherein each member in the library comprises a DNA construct, wherein the DNA construct comprises: i) a unique molecular identifier (UMI) that is unique for the member in the library; ii) a reporter gene operably linked to a promoter for driving expression in a mammalian cell; and, iii) at least one AAV ITR; b) contacting the library with a culture of cells, an organoid or a tissue, or administering the library to a non-human animal; c) allowing AAV capsid variants in the library to transduce the cells, the organoid, the tissue or cells in the animal; and, d) identifying at least one AAV capsid variant that has transduced at least one cell of a desired cell-type in the cells, the organoid, the tissue or in the animal in b), as an AAV capsid variant with the desired characteristic by sequencing its UMI, and optionally recovery of the AAV capsid variant with the desired characteristic from a cell of the desired cell-type.

In one embodiment of the method, step d) comprises detection of transduction of a cell of the desired cell-type by detecting expression of the reporter gene in at least one cell of the desired celltype.

In one embodiment of the method, in step d) in at least one cell of the desired cell-type, for at least two members in the library, at least one of the mRNA expressed from and the genome copy-number of the least two members is quantified and identified by sequencing its UMI, and wherein the member with at least one of the highest mRNA expression level and genome copynumber is identified as the AAV capsid variant with the desired characteristic.

In one embodiment the method is a method, wherein in cells of more than one cell-types including the at least desired cell-type, for at least two members in the library, at least one of the mRNA expressed from and the genome copy-number of the least two members is quantified and identified by sequencing its UMI, and wherein the member with the most desired distribution over the more than one cell-types is identified as the AAV capsid variant with the desired characteristic.

Description of the invention

Definitions

Unless defined otherwise, technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. One skilled in the art will recognize many methods and materials similar or equivalent to those described herein, which could be used in the practice of the present invention. Indeed, the present invention is in no way limited to the method. In this document and in its claims, the verb "to comprise" and its conjugations is used in its non-limiting sense to mean that items following the word are included, but items not specifically mentioned are not excluded. In addition, reference to an element by the indefinite article "a" or "an" does not exclude the possibility that more than one of the elements is present, unless the context clearly requires that there be one and only one of the elements. The indefinite article "a" or "an" thus usually means "at least one".

As used herein, the term "and/or" indicates that one or more of the stated cases may occur, alone or in combination with at least one of the stated cases, up to with all of the stated cases.

As used herein, with "At least" a particular value means that particular value or more. For example, "at least 2" is understood to be the same as "2 or more" i.e., 2, 3, 4, 5, 6, 7, 8, 9, 10, 1 1 , 12, 13, 14, 15, ... ,etc.

The word “about” or “approximately” when used in association with a numerical value (e.g. about 10) preferably means that the value may be the given value (of 10) more or less 0.1 % of the value.

As used herein, "an effective amount" is meant the amount of an agent required to ameliorate the symptoms of a disease relative to an untreated patient. The effective amount of active agent(s) used to practice the present invention for therapeutic treatment of, for example a cancer or a neurologic disease, varies depending upon the manner of administration, the age, body weight, and general health of the subject. Ultimately, the attending physician or veterinarian will decide the appropriate amount and dosage regimen. Such amount is referred to as an "effective" amount, which may be determined for instance as genome copies per kilogram (GC/kg) or as GC per dose. Thus, in connection with the administration of a drug which, in the context of the current disclosure, is "effective against" a disease or condition indicates that administration in a clinically appropriate manner results in a beneficial effect for at least a statistically significant fraction of patients, such as an improvement of symptoms, a cure, a reduction in at least one disease sign or symptom, extension of life, improvement in quality of life, or other effect generally recognized as positive by medical doctors familiar with treating the particular type of disease or condition.

The use of a substance as a medicament as described in this document can also be interpreted as the use of said substance in the manufacture of a medicament. Similarly, whenever a substance is used for treatment or as a medicament, it can also be used for the manufacture of a medicament for treatment. Products for use as a medicament described herein can be used in methods of treatments, wherein such methods of treatment comprise the administration of the product for use.

The terms “homology”, “sequence identity” and the like are used interchangeably herein. Sequence identity is herein defined as a relationship between two or more amino acid (polypeptide or protein) sequences or two or more nucleic acid (polynucleotide) sequences, as determined by comparing the sequences. In the art, "identity" and “similarity” also means the degree of sequence relatedness between amino acid or nucleic acid sequences, as the case may be, as determined by the match between strings of such sequences. "Identity" and "similarity" can be readily calculated by known methods. Sequence identity” and “sequence similarity” can be determined by alignment of two peptide or two nucleotide sequences using global or local alignment algorithms, depending on the length of the two sequences. Sequences of similar lengths are preferably aligned using global alignment algorithms (e.g. Needleman Wunsch) which align the sequences optimally over the entire length, while sequences of substantially different lengths are preferably aligned using local alignment algorithms (e.g. Smith Waterman). Sequences may then be referred to as "substantially identical” or “essentially similar” when they (when optimally aligned by for example the programs GAP or BESTFIT using default parameters) share at least a certain minimal percentage of sequence identity (as defined below). GAP uses the Needleman and Wunsch global alignment algorithm to align two sequences over their entire length (full length), maximizing the number of matches and minimizing the number of gaps. A global alignment is suitably used to determine sequence identity when the two sequences have similar lengths. Generally, the GAP default parameters are used, with a gap creation penalty = 50 (nucleotides) I 8 (proteins) and gap extension penalty = 3 (nucleotides) / 2 (proteins). For nucleotides the default scoring matrix used is nwsgapdna and for proteins the default scoring matrix is Blosum62 (Henikoff & Henikoff, 1992, PNAS 89, 915-919). Sequence alignments and scores for percentage sequence identity may be determined using computer programs, such as the GCG Wisconsin Package, Version 10.3, available from Accelrys Inc., 9685 Scranton Road, San Diego, CA 92121-3752 USA, or using open source software, such as the program “needle” (using the global Needleman Wunsch algorithm) or “water” (using the local Smith Waterman algorithm) in EmbossWIN version 2.10.0, using the same parameters as for GAP above, or using the default settings (both for ‘needle’ and for ‘water’ and both for protein and for DNA alignments, the default Gap opening penalty is 10.0 and the default gap extension penalty is 0.5; default scoring matrices are Blossum62 for proteins and DNAFull for DNA). When sequences have a substantially different overall length, local alignments, such as those using the Smith Waterman algorithm, are preferred.

Alternatively, percentage similarity or identity may be determined by searching against public databases, using algorithms such as FASTA, BLAST, etc. Thus, the nucleic acid and protein sequences of the present invention can further be used as a “query sequence” to perform a search against public databases to, for example, identify other family members or related sequences. Such searches can be performed using the BLASTn and BLASTx programs (version 2.0) of Altschul, et al. (1990) J. Mol. Biol. 215:403 — 10. BLAST nucleotide searches can be performed with the NBLAST program, score = 100, wordlength = 12 to obtain nucleotide sequences homologous to oxidoreductase nucleic acid molecules of the invention. BLAST protein searches can be performed with the BLASTx program, score = 50, wordlength = 3 to obtain amino acid sequences homologous to protein molecules of the invention. To obtain gapped alignments for comparison purposes, Gapped BLAST can be utilized as described in Altschul et al., (1997) Nucleic Acids Res. 25(17): 3389-3402. When utilizing BLAST and Gapped BLAST programs, the default parameters of the respective programs (e.g., BLASTx and BLASTn) can be used. See the homepage of the National Center for Biotechnology Information at http://www.ncbi.nlm.nih.gov/. As used herein, the term "selectively hybridizing", “hybridizes selectively” and similar terms are intended to describe conditions for hybridization and washing under which nucleotide sequences at least 66%, at least 70%, at least 75%, at least 80%, more preferably at least 85%, even more preferably at least 90%, more preferably at least 95%, more preferably at least 98% or more preferably at least 99% homologous to each other typically remain hybridized to each other. That is to say, such hybridizing sequences may share at least 45%, at least 50%, at least 55%, at least 60%, at least 65, at least 70%, at least 75%, at least 80%, more preferably at least 85%, even more preferably at least 90%, more preferably at least 95%, more preferably at least 98% or more preferably at least 99% sequence identity.

A preferred, non-limiting example of such hybridization conditions is hybridization in 6X sodium chloride/sodium citrate (SSC) at about 45°C, followed by one or more washes in 1 X SSC, 0.1 % SDS at about 50°C, preferably at about 55°C, preferably at about 60°C and even more preferably at about 65°C.

Highly stringent conditions include, for example, hybridization at about 68°C in 5x SSC/5x Denhardt's solution I 1.0% SDS and washing in 0.2x SSC/0.1 % SDS at room temperature. Alternatively, washing may be performed at 42°C.

The skilled artisan will know which conditions to apply for stringent and highly stringent hybridization conditions. Additional guidance regarding such conditions is readily available in the art, for example, in Sambrook et al., 1989, Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Press, N.Y.; and Ausubel et al. (eds.), Sambrook and Russell (2001) "Molecular Cloning: A Laboratory Manual (3^rd edition), Cold Spring Harbor Laboratory, Cold Spring Harbor Laboratory Press, New York 1995, Current Protocols in Molecular Biology, (John Wiley & Sons, N.Y.).

Of course, a polynucleotide which hybridizes only to a poly A sequence (such as the 3' terminal poly(A) tract of mRNAs), or to a complementary stretch of T (or U) resides, would not be included in a polynucleotide of the invention used to specifically hybridize to a portion of a nucleic acid of the invention, since such a polynucleotide would hybridize to any nucleic acid molecule containing a poly (A) stretch or the complement thereof (e.g., practically any double-stranded cDNA clone).

A "nucleic acid construct" or "nucleic acid vector" is herein understood to mean a man-made nucleic acid molecule resulting from the use of recombinant DNA technology. The term "nucleic acid construct" therefore does not include naturally occurring nucleic acid molecules although a nucleic acid construct may comprise (parts of) naturally occurring nucleic acid molecules. A "vector" is a nucleic acid construct (typically DNA or RNA) that serves to transfer an exogenous nucleic acid sequence (i.e. DNA or RNA) into a host cell. A vector is preferably maintained in the host by at least one of autonomous replication and integration into the host cell’s genome. The terms "expression vector" or “expression construct" refer to nucleotide sequences that are capable of affecting expression of a gene in host cells or host organisms compatible with such sequences. These expression vectors typically include at least one “expression cassette” that is the functional unit capable of affecting expression of a sequence encoding a product to be expressed and wherein the coding sequence is operably linked to the appropriate expression control sequences, which at least comprises a suitable transcription regulatory sequence and optionally, 3' transcription termination signals. Additional factors necessary or helpful in affecting expression may also be present, such as expression enhancer elements. The expression vector will be introduced into a suitable host cell and be able to affect expression of the coding sequence in an in vitro cell culture of the host cell. A preferred expression vector will be suitable for expression of viral proteins and/or nucleic acids, particularly recombinant parvoviral proteins and/or nucleic acids, such as baculoviral vectors for expression of parvoviral proteins and/or nucleic acids in insect cells.

A "parvoviral vector" is defined as a recombinantly produced parvovirus or parvoviral particle that comprises a polynucleotide to be delivered into a host cell, either in vivo, ex vivo or in vitro. An adeno-associated virus (AAV) vector is an example of a parvoviral vector. Herein, a parvoviral or AAV vector refers to the polynucleotide comprising part of the parvoviral genome, usually at least one ITR, and a transgene, which polynucleotide is preferably packaged in a parvoviral or AAV capsid.

As used herein, the term "promoter" or "transcription regulatory sequence" refers to a nucleic acid fragment that functions to control the transcription of one or more coding sequences, and is located upstream with respect to the direction of transcription of the transcription initiation site of the coding sequence, and is structurally identified by the presence of a binding site for DNA- dependent RNA polymerase, transcription initiation sites and any other DNA sequences, including, but not limited to transcription factor binding sites, repressor and activator protein binding sites, and any other sequences of nucleotides known to one of skill in the art to act directly or indirectly to regulate the amount of transcription from the promoter. A "constitutive" promoter is a promoter that is active in most tissues under most physiological and developmental conditions. An "inducible" promoter is a promoter that is physiologically or developmentally regulated, e.g. by the application of a chemical inducer or biological entity.

The term "reporter" may be used interchangeably with marker, although it is mainly used to refer to visible markers, such as green fluorescent protein (GFP) or luciferase.

The terms "protein" or "polypeptide" are used interchangeably and refer to molecules consisting of a chain of amino acids, without reference to a specific mode of action, size, 3- dimensional structure or origin.

The term "gene" means a DNA fragment comprising a region (transcribed region), which is transcribed into an RNA molecule (e.g. an mRNA) in a cell, operably linked to suitable regulatory regions (e.g. a promoter). A gene will usually comprise several operably linked fragments, such as a promoter, a 5' leader sequence, a coding region and a 3'-nontranslated sequence (3'-end) comprising a polyadenylation site. "Expression of a gene" refers to the process wherein a DNA region which is operably linked to appropriate regulatory regions, particularly a promoter, is transcribed into an RNA, which is biologically active, i.e. which is capable of being translated into a biologically active protein or peptide.

The term "homologous" when used to indicate the relation between a given (recombinant) nucleic acid or polypeptide molecule and a given host organism or host cell, is understood to mean that in nature the nucleic acid or polypeptide molecule is produced by a host cell or organisms of the same species, preferably of the same variety or strain. If homologous to a host cell, a nucleic acid sequence encoding a polypeptide will typically (but not necessarily) be operably linked to another (heterologous) promoter sequence and, if applicable, another (heterologous) secretory signal sequence and/or terminator sequence than in its natural environment. It is understood that the regulatory sequences, signal sequences, terminator sequences, etc. may also be homologous to the host cell. In this context, the use of only "homologous" sequence elements allows the construction of "self-cloned" genetically modified organisms (GMO's) (self-cloning is defined herein as in European Directive 98/81/EC Annex II). When used to indicate the relatedness of two nucleic acid sequences the term "homologous" means that one single-stranded nucleic acid sequence may hybridize to a complementary single-stranded nucleic acid sequence. The degree of hybridization may depend on a number of factors including the amount of identity between the sequences and the hybridization conditions such as temperature and salt concentration as discussed later.

The terms "heterologous" and "exogenous" when used with respect to a nucleic acid (DNA or RNA) or protein refers to a nucleic acid or protein that does not occur naturally as part of the organism, cell, genome or DNA or RNA sequence in which it is present, or that is found in a cell or location or locations in the genome or DNA or RNA sequence that differ from that in which it is found in nature. Heterologous and exogenous nucleic acids or proteins are not endogenous to the cell into which they are introduced but have been obtained from another cell or are synthetically or recombinantly produced. Generally, though not necessarily, such nucleic acids encode proteins, i.e. exogenous proteins, that are not normally produced by the cell in which the DNA is transcribed or expressed. Similarly, exogenous RNA encodes for proteins not normally expressed in the cell in which the exogenous RNA is present. Heterologous/exogenous nucleic acids and proteins may also be referred to as foreign nucleic acids or proteins. Any nucleic acid or protein that one of skill in the art would recognize as foreign to the cell in which it is expressed is herein encompassed by the term heterologous or exogenous nucleic acid or protein. The terms heterologous and exogenous also apply to non-natural combinations of nucleic acid or amino acid sequences, i.e. combinations where at least two of the combined sequences are foreign with respect to each other.

As used herein, the term "non-naturally occurring" when used in reference to an organism means that the organism has at least one genetic alternation that is not normally found in a naturally occurring strain of the referenced species, including wild-type strains of the referenced species. Genetic alterations include, for example, modifications introducing expressible nucleic acids encoding proteins or enzymes, other nucleic acid additions, nucleic acid deletions, nucleic acid substitutions, or other functional disruption of the organism's genetic material. Such modifications include, for example, coding regions and functional fragments thereof for heterologous or homologous polypeptides for the referenced species. Additional modifications include, for example, non-coding regulatory regions in which the modifications alter expression of a gene or operon. Genetic modifications to nucleic acid molecules encoding enzymes, or functional fragments thereof, can confer a biochemical reaction capability or a metabolic pathway capability to the non-naturally occurring organism that is altered from its naturally occurring state. As used herein, the term “operably linked” refers to a linkage of polynucleotide (or polypeptide) elements in a functional relationship. A nucleic acid is “operably linked” when it is placed into a functional relationship with another nucleic acid sequence. For instance, a transcription regulatory sequence is operably linked to a coding sequence if it affects the transcription of the coding sequence. Operably linked means that the DNA sequences being linked are typically contiguous and, where necessary to join two protein encoding regions, contiguous and in reading frame.

An expression control sequence is "operably linked" to a nucleotide sequence when the expression control sequence controls and regulates the transcription and/or the translation of the nucleotide sequence. Thus, an expression control sequence can include promoters, enhancers, internal ribosome entry sites (IRES), transcription terminators, a start codon in front of a proteinencoding gene, splicing signal for introns, and stop codons.

The term "expression control sequence" is intended to include, at a minimum, a sequence whose presence is designed to influence expression, and can also include additional advantageous components. For example, leader sequences and fusion partner sequences are expression control sequences. The term can also include the design of the nucleic acid sequence such that undesirable, potential initiation codons in and out of frame, are removed from the sequence. It can also include the design of the nucleic acid sequence such that undesirable potential splice sites are removed. It includes sequences or polyadenylation sequences (pA) which direct the addition of a polyA tail, i.e., a string of adenine residues at the 3'-end of a mRNA, sequences referred to as polyA sequences. It also can be designed to enhance mRNA stability. Expression control sequences which affect the transcription and translation stability, e.g., promoters, as well as sequences which affect the translation, e.g., Kozak sequences, are known in insect cells. Expression control sequences can be of such nature as to modulate the nucleotide sequence to which it is operably linked such that lower expression levels or higher expression levels are achieved.

The term "library" as used herein refers to a collection of elements or members that differ from one another in at least one aspect. For example, a library of nucleic acid molecules or a library of nucleic acid constructs is a collection of at least two nucleic acid molecules or at least two nucleic acid constructs that differ from one another by at least one nucleotide. Likewise, a library of AAV capsid variants is a collection of at least two AAV capsid variants (virions) that differ from one another by at least one nucleotide and/or at least one amino acid in their capsid proteins.

The term "tropism" as used herein refers to the preferential targeting by a virus (e.g., an AAV) of cells of a particular host species or of particular cell types within a host species. For example, a virus that can infect cells of the heart, lung, liver, and muscle has a broader (i.e., increased) tropism relative to a virus that can infect only lung and muscle cells. Tropism can also include the dependence of a virus on particular types of cell surface molecules of the host. For example, some viruses can infect only cells with surface glycosaminoglycans, while other viruses can infect only cells with sialic acid (such dependencies can be tested using various cells lines deficient in particular classes of molecules as potential host cells for viral infection). In some cases, the tropism of a virus describes the virus's relative preferences. For example, a first virus may be able to infect all cell types but is much more successful in infecting those cells with surface glycosaminoglycans. A second virus can be considered to have a similar (or identical) tropism as the first virus if the second virus also prefers the same characteristics (e.g., the second virus is also more successful in infecting those cells with surface glycosaminoglycans), even if the absolute transduction efficiencies are not similar. For example, the second virus might be more efficient than the first virus at infecting every given cell type tested, but if the relative preferences are similar (or 5 identical), the second virus can still be considered to have a similar (or identical) tropism as the first virus. In some embodiments, the tropism of a virion comprising a subject variant AAV capsid protein is not altered relative to a naturally occurring virion. In some embodiments, the tropism of a virion comprising a subject variant AAV capsid protein is expanded (i.e., broadened) relative to a naturally occurring virion. In some embodiments, the tropism of a virion comprising a subject variant AAV capsid protein is reduced relative to a naturally occurring virion.

Any reference to nucleotide or amino acid sequences accessible in public sequence databases herein refers to the version of the sequence entry as available on the filing date of this document.

Detailed description of the invention

The present inventors have surprisingly found that a functionally defective phenotype of AAV9 capsid variants can be rescued by introducing additional modifications into the amino acid sequences of the functionally defective capsid that promote CNS tropism to produce novel capsid variants that combine enhanced CNS targeting with liver de-targeting features.

Adeno-associated virus capsid protein variants

In a first aspect, there is provided an adeno-associated virus (AAV) capsid protein variant. In one embodiment, the AAV capsid protein is a variant of an AAV serotype 9 (AAV9) or Clade F AAV capsid protein. A “variant” of an AAV9 or Clade F AAV capsid protein is herein understood as an AAV capsid protein that comprises at least one amino acid difference (mutation) as compared to a wild type AAV9 or Clade F AAV capsid protein. The at least one amino acid difference can be a substitution, insertion or deletion of an amino acid in the amino acid sequence of the wild type capsid protein.

A wild type AAV9 capsid protein is herein understood to comprise, consist of, or to be comprised within the amino acid sequence of SEQ ID NO: 2 (AAV9 VP1 capsid protein; GenBank® Database Accession No. AY530579.1 ; GenBank® Database Accession No. AAS99264.1 ). A wild type Clade F AAV capsid protein is herein understood as a capsid protein of a Clade F AAV isolate as defined by Gao et al. (2004, J. Virol. 78: 6381-6388), and which, in addition to AAV9, includes the hu.31 (encoded by GenBank® Database Accession No. JA400111 .1) and hu.32 (encoded by GenBank® Database Accession No. JA400112.1). In one embodiment, a wild type Clade F AAV capsid protein is herein understood as a capsid protein of an AAV isolate that comprises, consists of, or is comprised within an amino acid sequence with more than 87.5, 88, 89, 90, 91 , 92, 93, 94, 95, 96, 97, 98 or 99% sequence identity to SEQ ID NO: 2. The designation of all amino acid positions in the AAV capsid proteins and variants thereof as used herein is with respect to AAV9 VP1 capsid subunit numbering of SEQ ID NO: 2. Corresponding positions in amino acid sequences other than SEQ ID NO: 2 are identified using a sequence alignment, preferably using algorithms and settings as set out above.

Thus, in one embodiment, there is provided an AAV9 or Clade F AAV capsid protein variant comprising a mutation in one or more amino acids, which mutation(s) confer to the capsid protein at least one of: i) a phenotype of decreased liver transduction as compared to a control capsid protein not having the mutation(s); and, ii) a phenotype of decreased overall transduction as compared to a control capsid protein not having the mutation(s), and wherein the capsid protein further comprises an amino acid sequence that confers increased CNS transduction as compared to a control capsid protein not having the amino acid sequence.

In one embodiment, the phenotypes of decreased liver transduction, decreased overall transduction and increased CNS transduction of a capsid protein variant as compared to a control capsid protein are phenotypes achieved upon systemic administration of an recombinant AAV (rAAV) vector comprising the capsid protein variant.

The phenotypes of decreased liver transduction, decreased overall transduction and increased CNS transduction of a capsid protein variant as compared to a control capsid protein can be determined by administering an rAAV vector comprising the capsid protein variant to a nonhuman animal and determining the transduction profile (i.e. distribution) of the rAAV vector over various tissues of interest in the animal, and preferably comparing the transduction profile rAAV vector comprising the capsid protein variant with the transduction profile of a control rAAV vector, preferably under otherwise identical conditions. In one embodiment, the transduction profile is determined upon administration is systemic administration. In one embodiment, the transduction profile is determined 2, 3, 4, 5 or 6 weeks post-administration. In one embodiment, the non-human animal is a mouse or other research animal, such as, but not limited to, a rabbit, a rat, a monkey or a non-human primate.

As will be understood, the tissues of interest in determining the transduction profile, depend on the desired tropism and de-targeting of the capsid protein variant. For improved CNS tropism and liver de-targeting, the tissues of interest will at least include CNS (brain) and liver but can include further major organs such as heart, lung, and skeletal muscle. For a more specific determination of the CNS transduction profile various sections of the brain can be examined separately, such as frontal brain, caudal brain and cerebellum.

Methods for determining transduction profiles are known in the art per se and are for example described in Pulicherla et al. (2011 , supra), Hanlon et al. (2019, supra), Nonnenmacher et al., (2020, supra), Goersten et al. (2022, supra), herein below and in the Examples herein.

Thus in one embodiment, a capsid protein variant as described herein has at least one mutation that confers a phenotype of decreased liver transduction as compared to a control capsid protein not having the mutation(s). A phenotype of decreased liver transduction is herein understood as a phenotype that displays a significantly decrease in liver transduction (at least more than 2, 3, 5, 10, 20, 50, 100, 200 or 500 fold) but showing modest-to-no change or decrease (twofold or lesser) in other tissue types, as compared to a control capsid protein not having the mutation (s).

In one embodiment, a capsid protein variant as described herein has at least one mutation that confers a phenotype of decreased overall transduction as compared to a control capsid protein not having the mutation(s). A phenotype of decreased overall transduction is herein understood as a phenotype that displays a significantly decrease in transduction of multiple organs/tissue types (at least more than 2, 3, 5, 10, 20, 50, 100, 200 or 500 fold) as compared to a control capsid protein not having the mutation(s).

In one embodiment, a capsid protein variant as described herein comprises an amino acid sequence that confers increased CNS transduction as compared to a control capsid protein not having the amino acid sequence. A phenotype of increased CNS transduction is herein understood as a phenotype that displays a significantly increased transduction CNS tissues (at least more than 2, 3, 5, 10, 20, 50, 100, 200 or 500 fold) as compared to a control capsid protein not having the amino acid sequence. In one embodiment, a capsid protein variant as described herein comprises an amino acid sequence that confers increased transduction of at least one of frontal brain, caudal brain and cerebellum. In a preferred embodiment, the phenotype of increased CNS transduction is herein understood as a phenotype that displays a significantly increased transduction CNS tissues (at least more than 2, 3, 5, 10, 20, 50, 100, 200 or 500 fold), but showing modest-to-no change or increase (twofold or lesser) in other tissue types, as compared to a control capsid protein not having the amino acid sequence.

In one embodiment, a capsid protein variant as described herein comprises at least one mutation that confers a phenotype of decreased liver transduction as compared to a control capsid protein not having the mutation(s), wherein the at least one mutation comprises one or more mutations of amino acids in amino acid regions 452-458, 498-504, 590-595 and/or 582-587. In one embodiment, the at least one mutation comprises at least one mutation selected from the group consisting of mutations at positions W595, Q592, W503, N498, E500, and the amino acid sequence DGAATKN at positions 452 - 458. In one embodiment, the capsid protein variant comprises a W595C mutation, a Q592L mutation, a W503R mutation, a N498Y mutation, a E500D mutation or any combination thereof. It is thus understood that the at least one mutation results in a phenotype of decreased liver transduction as compared to a control not having the mutation(s).

In one embodiment, a capsid protein variant as described herein comprises at least one mutation that confers a phenotype of of decreased overall transduction as compared to a control capsid protein not having the mutation(s), wherein the capsid protein variant comprises a mutation at T138, S414, G453, K557, T568, T582 and Q590 or any combination thereof. In one embodiment, the capsid protein variant comprises mutations at T568 and Q590, or at T138, S414, G453, K557 and T582. In one embodiment, the capsid protein variant comprises the T568P and Q590L mutations, or the T138A, S414N, G453D, K557E and T582I mutations. In one embodiment the capsid protein variant comprises mutations at S414 and G453, or at S414, G453, K557 and T582. It is thus understood that the at least one mutation results in a phenotype of overall transduction as compared to a control not having the mutation(s). In one embodiment, a capsid protein variant as described herein comprises an amino acid sequence that confers increased CNS transduction, wherein the amino acid sequence comprises or consists of an amino acid sequence selected from the group consisting of STTLYSP, FVVGQSY, DGTLAVPFK and WPTSYDA, and amino acid sequences differing by no more than four, three, two or one amino acid(s) therefrom. In one embodiment, the amino acid sequence that confers increased CNS transduction comprises or consists of an amino acid sequence selected from the group consisting of FVVGQSY, FVVAQSY, FVWQSY, FVVLQSY, FVVIQSY, FVVNQSY, FVVSQSY, FVVEQSY, FVVQQSY, FVVCQSY, FVVTQSY, FVVPQSY, FVVQSY, FVGVQSY, FVVQGSY, FVQGVSY, FSVGQVY and FSQGVVY. In a preferred embodiment, the amino acid sequence that confers increased CNS transduction comprises or consists of an amino acid sequence selected from the group consisting of FVVGQSY, FVVAQSY, FVVPQSY, FVGVQSY and FVVQGSY. In one embodiment, the amino acid sequence that confers increased CNS transduction is inserted in a hypervariable region of the AAV capsid protein sequence, preferably in a hypervariable region selected from Loop IV or Loop VIII.

In one embodiment, a capsid protein variant as described herein is a capsid protein variant comprising the T138A, S414N, G453D, K557E and T582I mutations and the amino acid sequence FVVGQSY inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589.

In one embodiment, a capsid protein variant as described herein is a capsid protein variant comprising the S414N, G453D, K557E and T582I mutations and the amino acid sequence is selected from the group FVVGQSY, FVVAQSY, and FVVQGSY inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589. In one embodiment, a capsid protein variant as described herein is a capsid protein variant comprising the S414N, G453D, K557E and T582I mutations and the amino acid sequence FVVGQSY inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589. In an alternative embodiment, a capsid protein variant as described herein is a capsid protein variant comprising the S414N, G453D, K557E and T582I mutations and the amino acid sequence is selected from the group FVGVQSY and FVVPQSY inserted in Loop VIII, preferably inserted between amino acid positions 587 and 588.

In one embodiment, a capsid protein variant as described herein is a capsid protein variant comprising the T138A, S414N, G453D, K557E and T582I mutations and the amino acid sequence WPTSYDA inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589.

In one embodiment, a capsid protein variant as described herein is a capsid protein variant comprising the S414N, G453D, K557E and T582I mutations and the amino acid sequence WPTSYDA inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589.

In one embodiment, a capsid protein variant as described herein is a capsid protein variant comprising the amino acid sequence DGAATKN at positions 452 - 458 and the amino acid sequence FVVGQSY inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589.

In one embodiment, a capsid protein variant as described herein is a capsid protein variant comprising the amino acid sequence DGAATKN at positions 452 - 458 and the amino acid sequence WPTSYDA inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589.

In one embodiment, a capsid protein variant as described herein is a capsid protein variant comprising, comprised in or consisting of an amino acid sequence selected from the group consisting of SEQ ID NO.’s: 6, 7, 9 and 10. In one embodiment, a capsid protein variant as described herein is a capsid protein variant comprising, comprised in or consisting of an amino acid sequence selected from the group consisting of SEQ ID NO.’s: 6, 7, 9, 10 and 96.

In one embodiment, a capsid protein variant as described herein is a variant at least one of a VP1 , VP2 and VP3 capsid protein. It will be understood by those skilled in the art that the modifications described herein if inserted into the AAV cap gene may result in modifications in the VP1 , VP2 and/or VP3 capsid subunits, depending on their location in the VP1 amino acid sequence. However, in one embodiment, the capsid subunits can be expressed independently to achieve modification in only one or two of the capsid subunits, for example only in VP1 , VP2, VP3, VP1 and VP2, VP1 and VP3, or VP2 and VP3.

Nucleic acid molecules, host cells and methods for producing an AAV vector

In a second aspect, there is provided a nucleic acid encoding an AAV capsid protein variant as described herein. In one embodiment, the nucleic acid molecule is an expression construct for expression of the capsid protein(s). Preferably, the expression construct is a construct for expression of the capsid protein(s) in a host cell that is suitable for the production of an AAV, such as a mammalian or insect cell line as further defined below.

Thus, in one embodiment, the expression construct for expression of the capsid protein(s) is an insect cell-compatible vector or a mammalian cell-compatible vector. An "mammalian cellcompatible vector” is understood to be a nucleic acid molecule capable of productive transformation or transfection of a mammalian cell or cell line. Mammalian cell-compatible vectors are well-known in the art. An “insect cell-compatible vector” is understood to be a nucleic acid molecule capable of productive transformation or transfection of an insect or insect cell. Exemplary insect cellcompatible vectors include plasmids, linear nucleic acid molecules, and recombinant viruses, such as baculoviruses. Any vector can be employed as long as it is insect cell-compatible. The mammalian or insect cell-compatible vector may integrate into the cell’s genome but the presence of the vector in the cell need not be permanent and transient episomal vectors are also included. The vectors can be introduced by any means known, for example by chemical treatment of the cells, electroporation, or infection.

In one embodiment, the vector is a baculovirus, a viral vector, or a plasmid. In a more preferred embodiment, the insect cell-compatible vector is a baculovirus, i.e. the nucleic acid construct is a baculovirus-expression vector (BEV). It is well-known that baculovirus-expression vectors are particularly suitable for the transfer of nucleic acids to insect cells and methods for their use are described for example in: Summers and Smith, 1986, “A Manual of Methods for Baculovirus Vectors and Insect Culture Procedures”, Texas Agricultural Experimental Station Bull. No. 7555, College Station, Tex.; Luckow, 1991 , In Prokop et al., “Cloning and Expression of Heterologous Genes in Insect Cells with Baculovirus Vectors' Recombinant DNA Technology and Applications”, 97-152; King and Possee, 1992, “The baculovirus expression system”, Chapman and Hall, United Kingdom; O'Reilly, Miller, and Luckow, 1992, “Baculovirus Expression Vectors: A Laboratory Manual”, New York; Freeman and Richardson, 1995, “Baculovirus Expression Protocols”, Methods in Molecular Biology, volume 39; US 4,745,051 ; US2003148506; and WO 03/074714.

In one embodiment, there is provided at least one expression construct comprising separate expression cassettes for each of the VP1 , VP2 and VP3 capsid protein variants as described herein. In one embodiment, there is provided at least one expression construct comprising a separate expression cassette for a VP1 capsid protein variant as described herein and a separate expression cassette for the VP2 and VP3 protein variants as described herein. The various expression cassettes for the different capsid protein variants as described herein can be present together on a single expression construct/nucleic acid molecule, or one or more of the expression cassettes for the different capsid proteins can be present on two or three separate expression constructs/nucleic acid molecules, each comprising one of the expression cassettes for the different capsid proteins. As will be understood, the use of separate expression cassettes for one or more of the different capsid proteins allows to produce AAV vectors with an above-described modification/mutation in only some of capsid proteins, e.g. only in VP1 and not in VP2 and VP3 or vice versa, or it allows to produce AAV vectors with different above-described modifications/mutations in the different capsid proteins. Expression constructs for separate expression of the various capsid proteins in mammalian cells are e.g. disclosed in Judd et al. (Mol Ther Nucleic Acids. 2012; 1 : e54). Expression constructs for separate expression of the various capsid proteins in insect cells are disclosed in WO2022/253955.

In another embodiment, there is provided an expression construct comprising a single expression cassette for expression of all three of the VP1 , VP2 and VP3 capsid protein variants as described herein, preferably from a single coding sequence. Expression constructs for expression of all three of the VP1 , VP2 and VP3 capsid proteins from a single expression cassette in mammalian cells are e.g. disclosed in Clark et al. (1995, Hum. Gene Ther. 6, 1329-134), Gao et al. (1998, Hum. Gene Ther. 9, 2353-2362), Inoue and Russell (1998, J. Virol. 72, 7024-7031), Grimm et al. (1998, Hum. Gene Ther. 9, 2745-2760) and Xiao et al. (1998, J. Virol. 72, 2224-2232). Expression constructs for expression of all three of the VP1 , VP2 and VP3 capsid proteins from a single expression cassette in insect cells are e.g. disclosed in Urabe et al. (2002, Hum. Gene Ther. 13:1935-1943), WG2007/046703, WO2015/137802 and WO2019/016349. As will be understood, expression of all three of the VP1 , VP2 and VP3 capsid protein variants as described herein, from a single coding sequence (in a single expression cassette), allows to produce AAV vectors comprising mutations/modifications as described herein in all three of its VP1 , VP2 and VP3 capsid proteins.

In a third aspect, there is provided a host cell comprising a nucleic acid molecule or expression construct for expression of the capsid protein variants as described herein. The host cell, preferably is a host cell that is suitable for the production of AAV vectors. Accordingly the host cell is a host cell that is amenable to in vitro culture, preferably at large scale. Host cell that are suitable for the production of AAV vectors are well-known in the art and will typically be a mammalian or an insect cell line. Mammalian cell lines for producing AAV vectors are selected from among any mammalian species, including, without limitation, cells such as A549, WEHI, 3T3, 10T1/2, BHK, MDCK, COS 1 , COS 7, BSC 1 , BSC 40, BMT 10, VERO, WI38, HeLa, a HEK 293 cell (which express functional adenoviral E1), Saos, C2C12, L cells, HT1080, HepG2 and primary fibroblast, hepatocyte and myoblast cells derived from mammals including human, monkey, mouse, rat, rabbit, and hamster. The selection of the mammalian species providing the cells is not a limitation of this disclosure; nor is the type of mammalian cell, i.e., fibroblast, hepatocyte, tumor cell. Mammalian cell lines for producing AAV vectors in particular include a broad range of HEK293 cell lines, of which the HEK293T cell line is preferred.

Insect cell lines for producing AAV vectors can be any cell line that is suitable for the production of heterologous proteins. Preferably the insect cell allows for replication of baculoviral vectors and can be maintained in culture, more preferably in suspended culture. In a preferred embodiment, the insect cell allows for replication of recombinant parvoviral vectors, including rAAV vectors. For example, the cell line used can be from Spodoptera frugiperda, Drosophila, or mosquito, e.g., Aedes albopictus derived cell lines. Preferred insect cells or cell lines are cells from the insect species which are susceptible to baculovirus infection, including e.g. S2 (CRL-1963, ATCC), Se301 , SelZD2109, SeUCRI , Sf9, Sf900+, Sf21 , BTI-TN-5B1-4, MG-1 , Tn368, HzAml , Ha2302, Hz2E5, High Five (Invitrogen, CA, USA) and expresSF+® (US 6,103,526; Protein Sciences Corp., CT, USA).

In one embodiment, the host cell comprising the expression constructs) for expression of the capsid protein variant(s) as described herein, comprises further nucleotide sequences for the expression of an AAV vector. Such further nucleotide sequences typically include expression constructs for expression of AAV rep proteins in the host cell in question. Such further nucleotide sequences in addition usually include a nucleic acid construct comprising a transgene that is flanked by at least one AAV ITR sequence.

Thus, a nucleic acid construct comprising a transgene that is flanked by at least one AAV ITR sequence preferably becomes incorporated into the genome of a recombinant parvoviral (rAAV) vector when produced in a suitable host cell that is expressing AAV rep and cap gene products. In one embodiment, the nucleotide sequence comprising the transgene is flanked by two AAV ITR nucleotide sequences, whereby the nucleotide sequence comprising the transgene is located in between the two AAV ITR nucleotide sequences. In one embodiment, the nucleotide sequence comprising the transgene can be incorporated into the rAAV vector produced in the insect cell if it is located between two regular ITRs, or is located on either side of an ITR engineered with two D regions. Thus, in a preferred embodiment, there is provided a nucleic acid construct comprising two AAV ITR nucleotide sequences and wherein the nucleotide sequence comprising the transgene is located between the two AAV ITR nucleotide sequences.

Typically, the nucleic acid construct, including ITRs and nucleotide sequence comprising the expression cassette for an AAV capsid variant and the enhancer element operably linked thereto, is 5,000 nucleotides (nt) or less in length in order not to exceed the maximum AAV packaging limit of 5.5 kbp. However, other embodiments, wherein oversized nucleic acid constructs are used, e.g. more than 5,000 nt in length, or even more than the maximum AAV packaging limit of 5.5 kbp do still allow the generation of rAAV vectors and are therefore not excluded.

AAV sequences that may be used as described herein for the production of a recombinant AAV virion, i.e. an AAV vector, in insect cells can be derived from the genome of any AAV serotype. Generally, the AAV serotypes have genomic sequences of significant homology at the amino acid and the nucleic acid levels, provide an identical set of genetic functions, and produce virions which are essentially physically and functionally equivalent, and replicate and assemble by practically identical mechanisms. For the genomic sequence of the various AAV serotypes and an overview of the genomic similarities see e.g. GenBank Accession number U89790; GenBank Accession number J01901 ; GenBank Accession number AF043303; GenBank Accession number AF085716; Chlorini et al. (1997, J. Vir. 71 : 6823-33); Srivastava et al. (1983, J. Vir. 45:555-64); Chlorini et al. (1999, J. Vir. 73:1309-1319); Rutledge et al. (1998, J. Vir. 72:309-319); and Wu et al. (2000, J. Vir. 74: 8635-47). Any AAV serotype can be used as source of AAV nucleotide sequences for use in the context of the present invention. Preferably the AAV ITR sequences for use in the context of the present invention are derived from AAV1 , AAV2, AAV4 and/or AAV7. Likewise, the Rep (Rep78/68 and Rep52/40) coding sequences are preferably derived from AAV1 , AAV2, AAV4 and/or AAV7.

AAV Rep and ITR sequences are particularly conserved among most serotypes. The Rep78 proteins of various AAV serotypes are e.g. more than 89% identical and the total nucleotide sequence identity at the genome level between AAV2, AAV3A, AAV3B, and AAV6 is around 82% (Bantel-Schaal et al., 1999, J. Virol., 73(2):939-947). Moreover, the Rep sequences and ITRs of many AAV serotypes are known to efficiently cross-complement (i.e., functionally substitute) corresponding sequences from other serotypes in production of AAV particles in mammalian cells. US2003148506 reports that AAV Rep and ITR sequences also efficiently cross-complement other AAV Rep and ITR sequences in insect cells. Modified "AAV" sequences also can be used in this context, e.g. forthe production of rAAV vectors in insect cells. Such modified sequences e.g. include sequences having at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, or more nucleotide and/or amino acid sequence identity (e.g., a sequence having about 75-99% nucleotide sequence identity) to an AAV1 , AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAV10, AAV11 , AAV12 or AAV13 ITR or Rep can be used in place of wild-type AAV ITR or Rep sequences.

In a fourth aspect, there is provided a method for producing an AAV vector virion as described herein, comprising capsid protein variant as described herein. The method preferably comprising the steps of: a) culturing a host cell as herein defined above under conditions such that the AAV vector is produced; and, b) optionally, one or more of recovery, purification and formulation of the AAV vector.

The AAV in the supernatant can be recovered and/or purified using suitable techniques which are known to those of skill in the art. For example, monolith columns (e.g., in ion exchange, affinity or IMAC mode), chromatography (e.g., capture chromatography, fixed method chromatography, and expanded bed chromatography), centrifugation, filtration and precipitation, can be used for purification and concentration. These methods may be used alone or in combination. In one embodiment, capture chromatography methods, including column-based or membrane-based systems, are utilized in combination with filtration and precipitation. Suitable precipitation methods, e.g., utilizing polyethylene glycol (PEG) 8000 and NH3SO4, can be readily selected by one of skill in the art. Thereafter, the precipitate can be treated with benzonase and purified using suitable techniques. In addition, recovery may preferably comprises the step of affinity-purification of the (virions comprising the) recombinant parvoviral (rAAV) vector using an anti-AAV antibody, preferably an immobilised antibody. The anti-AAV antibody preferably is a monoclonal antibody. A particularly suitable antibody is a single chain camelid antibody or a fragment thereof as e.g. obtainable from camels or llamas (see e.g. Muyldermans, 2001 , Biotechnol. 74: 277-302). The antibody for affinity-purification of rAAV preferably is an antibody that specifically binds an epitope on an AAV capsid protein, whereby preferably the epitope is an epitope that is present on capsid protein of more than one AAV serotype. E.g. the antibody may be raised or selected on the basis of specific binding to AAV2 capsid but at the same time also it may also specifically bind to AAV9 of Clade F capsids.

In general, suitable methods for producing an AAV vector virion as described herein in mammalian or insect host cells, and means therefore (such as expression constructs for expression of AAV rep proteins), are described, for mammalian cells in: Clark et al. (1995, Hum. Gene Ther. 6, 1329-134), Gao et al. (1998, Hum. Gene Ther. 9, 2353-2362), Inoue and Russell (1998, J. Virol. 72, 7024-7031), Grimm et al. (1998, Hum. Gene Ther. 9, 2745-2760), Xiao et al. (1998, J. Virol. 72, 2224-2232) and Judd et al. (Mol Ther Nucleic Acids. 2012; 1 : e54), and for insect cells in: Urabe et al. (2002, Hum. Gene Ther. 13:1935-1943), WG2007/046703, WG2007/148971 , WG2009/014445, WG2009/104964, WO2011/122950, WO2013/036118, WO2015/137802, WO2019/016349 and in co-pending applications EP21177449.2, PCT/EP2021/058794 and PCT/EP2021/058798, all of which are incorporated herein in their entirety.

A recombinant AAV vector virion

In a fifth aspect, there is provided a recombinant AAV (rAAV) vector virion comprising at least one AAV capsid protein variant as described herein.

An "rAAV vector", an "rAAV virion" or an "rAAV vector virion" is defined herein as a recombinantly produced AAV or AAV particle that comprises a polynucleotide to be delivered into a host cell, either in vivo, ex vivo or in vitro. Herein, an “AAV vector construct” or an “AAV provector” refers to the polynucleotide comprising the viral genome or part thereof, usually at least one ITR, and a transgene.

In one embodiment, an rAAV vector virion comprising at least one AAV capsid protein variant as described herein, does not comprise at least one of a wild type VP1 , VP2, and VP3 capsid proteins. Hence, an rAAV vector virion as described can comprise one or at most two of the VP1 , VP2, and VP3 capsid proteins in wild type form, but not all three. In one embodiment, an rAAV vector virion comprising at least one AAV capsid protein variant as described herein comprises a nucleic acid molecule encapsidated by the at least one capsid protein variant. Hence, an rAAV vector virion as described herein consists of a capsid comprising one or more of the VP1 , VP2, and VP3 AAV capsid protein variant as described herein, which capsid comprises (or “’’packages”) a nucleic acid molecule. In one embodiment, the nucleic acid molecule comprises a (nucleotide sequence encoding a) transgene flanked by at least one AAV inverted terminal repeat (ITR). In one embodiment, the nucleic acid molecule comprises a transgene flanked by two ITRs, one on each side of the transgene(s). Such nucleic acid molecules, i.e. AAV provectors, can be replicated and packaged into infectious viral particles when present in a suitable host cell that is expressing AAV rep and cap gene products (i.e. AAV Rep and Cap proteins).

Transgenes

The nucleotide sequence comprising the transgene as defined herein above may thus comprise a nucleotide sequence encoding a gene product of interest (for expression in the mammalian cell) or encoding a nucleotide sequence targeting a gene of interest (for silencing said gene of interest in a mammalian cell), and may be located such that it will be incorporated into an recombinant parvoviral (rAAV) vector replicated in the insect cell. It is understood that a particularly preferred mammalian cell in which the "gene product of interest" is to be expressed or silenced, is a human cell. Any nucleotide sequence can be incorporated for later expression in a mammalian cell transfected with the recombinant parvoviral (rAAV) vector produced as provided herein. The nucleotide sequence may e.g. encode a protein or it may express an RNAi agent, i.e. an RNA molecule that is capable of RNA interference such as, e.g. an shRNA (short hairpinRNA) or an siRNA (short interfering RNA). "siRNA" means a small interfering RNA that is a short-length doublestranded RNA that are not toxic in mammalian cells (Elbashir et al., 2001 , Nature 411 : 494-98; Caplen et al., 2001 , Proc. Natl. Acad. Sci. USA 98: 9742-47). In a preferred embodiment, the nucleotide sequence comprising the transgene may comprise two coding nucleotide sequences, each encoding one gene product of interest for expression in a mammalian cell. Each of the two nucleotide sequences encoding a product of interest is located such that it will be incorporated into a recombinant parvoviral (rAAV) vector replicated in the insect cell.

The product of interest for expression in a mammalian cell may be a therapeutic gene product. A therapeutic gene product can be a polypeptide, or an RNA molecule (si/sh/miRNA), or other gene product that, when expressed in a target cell, provides a desired therapeutic effect. A desired therapeutic effect can for example be the ablation of an undesired activity (e.g. VEGF), the complementation of a genetic defect, the silencing of genes that cause disease, the restoration of a deficiency in an enzymatic activity or any other disease-modifying effect. Examples of therapeutic polypeptide gene products include, but are not limited to growth factors, factors that form part of the coagulation cascade, enzymes, lipoproteins, cytokines, neurotrophic factors, hormones and therapeutic immunoglobulins and variants thereof. Examples of therapeutic RNA molecule products include miRNAs effective in silencing diseases, including but not limited to polyglutamine diseases, dyslipidaemia or amyotrophic lateral sclerosis (ALS).

The diseases that can be treated using a recombinant parvoviral (rAAV) vector produced as described herein are not particularly limited, other than generally having a genetic cause or basis. For example, the disease that may be treated with the disclosed vectors may include, but are not limited to, acute intermittent porphyria (AIP), age-related macular degeneration, Alzheimer’s disease, arthritis, Batten disease, Canavan disease, Citrullinemia type 1 , Crigler Najjar, congestive heart failure, cystic fibrosis, Duchene muscular dystrophy, dyslipidemia, glycogen storage disease type I (GSD-I), hemophilia A, hemophilia B, hereditary emphysema, homozygous familial hypercholesterolemia (HoFH), Huntington’s disease (HD), Leber’s congenital amaurosis, methylmalonic academia, ornithine transcarbamylase deficiency (OTC), Parkinson’s disease, phenylketonuria (PKU), spinal muscular atrophy, paralysis, Wilson disease, epilepsy, Pompe disease, amyotrophic lateral sclerosis (ALS), Tay-Sachs disease, hyperoxaluria (PH-1), spinocerebellar ataxia type 1 (SCA-1), SCA-2, SCA-3, micro-dystrophin, Gaucher’s types II or III, arrhythmogenic right ventricular cardiomyopathy (ARVC), Fabry disease, familial Mediterranean fever (FMF), proprionic acidemia, fragile X syndrome, Rett syndrome, Niemann-Pick disease and Krabbe disease. As AAV vectors preferred herein have CNS tropism, preferred diseases that can be treated using such AAV vectors produced as provided herein are diseases, preferably genetic diseases, of the central nervous system and/or (genetic) diseases that can be treated by targeting the AAV vector to the CNS.

Examples of therapeutic gene products to be expressed include antibodies, N-acetyl-alpha- glucosaminidase, (NaGLU), Treg167, Treg289, EPO, IGF, IFN, GDNF, FOXP3, Factor VIII, Factor IX, insulin, Aromatic L-Amino Acid Decarboxylase (AADC), ApoE2, Frataxin, survival motor neuron (SMN) protein, glucocerebrosidase, N-sulfoglucosamine sulfohydrolase, iduronate 2-sulfatase, alpha-L-iduronidase, palmitoyl-protein thioesterase 1 , tripeptidyl peptidase 1 , battenin, CLN5, CLN6 (linclin), MFSD8, CLN8, aspartoacylase (ASPA), progranulin (GRN), MeCP2, beta-galactosidase (GLBI) and/or gigaxonin (GAN). As preferred AAV vectors as provided herein have CNS tropism, preferred therapeutic gene products are those that are useful in the treatment of genetic diseases of the central nervous system.

Examples of endogenous genes the expression of which is to be inhibited and/or modified for a therapeutic effect by targeting with an RNAi agent include: superoxide dismutase 1 (SOD1), chromosome 9 open reading frame 72 (C9ORF72), TAR DNA binding protein (TARDBP), ataxin-1 (ATXN1), ataxin-2 (ATXN2), ataxin-3 (ATXN3), huntingtin (HTT), amyloid precursor protein (APP), apolipoprotein E (ApoE), microtubule-associated protein tau (MAPT), alpha-synuclein (SNCA), voltagegated sodium channel alpha subunit 9 (SCN9A), and/or voltage-gated sodium channel alpha subunit 10 (SCN1 OA). As preferred AAV vectors as provided herein have CNS tropism, preferred endogenous genes the expression of which is to be inhibited and/or modified for a therapeutic effect by targeting with an RNAi agent include those endogenous genes that play a role in genetic diseases of the central nervous system. Alternatively, or in addition as another gene product, the nucleotide sequence comprising the transgene as defined herein above may further comprise a nucleotide sequence encoding a polypeptide that serves as a selection marker protein to assess cell transformation and expression. Suitable marker proteins for this purpose are e.g. the fluorescent protein GFP, and the selectable marker genes HSV thymidine kinase (for selection on HAT medium), bacterial hygromycin B phosphotransferase (for selection on hygromycin B), Tn5 aminoglycoside phosphotransferase (for selection on G418), and dihydrofolate reductase (DHFR) (for selection on methotrexate), CD20, the low affinity nerve growth factor gene. Sources for obtaining these marker genes and methods for their use are provided in Sambrook and Russel, supra. Furthermore, the nucleotide sequence comprising the transgene as defined herein above may comprise a further nucleotide sequence encoding a polypeptide that may serve as a fail-safe mechanism that allows to cure a subject from cells transduced with the recombinant parvoviral (rAAV) vector as described herein, if deemed necessary. Such a nucleotide sequence, often referred to as a suicide gene, encodes a protein that is capable of converting a prodrug into a toxic substance that is capable of killing the transgenic cells in which the protein is expressed. Suitable examples of such suicide genes include e.g. the E.coli cytosine deaminase gene or one of the thymidine kinase genes from Herpes Simplex Virus, Cytomegalovirus and Varicella-Zoster virus, in which case ganciclovir may be used as prodrug to kill the transgenic cells in the subject (see e.g. Clair et al., 1987, Antimicrob. Agents Chemother. 31 : 844-849).

The nucleotide sequence comprising a transgene as defined herein above for expression in a mammalian cell, further preferably comprises at least one mammalian cell-compatible expression control sequence, e.g. a promoter, that is/are operably linked to the sequence coding for the gene product of interest. Many such promoters are known in the art (see Sambrook and Russel, 2001 , supra). Constitutive promoters that are broadly expressed in many cell-types, such as the CMV, CAG and PGK promoters, may be used. However, more preferred will be promoters that are inducible, tissue-specific, cell-type-specific, or cell cycle-specific. For example, for liver-specific expression (as disclosed in PCT/EP2019/081743) a promoter may be selected from an a1 -antitrypsin promoter, a thyroid hormone-binding globulin promoter, an albumin promoter, LPS (thyroxine-binding globin) promoter, HCR-ApoCII hybrid promoter, HCR-hAAT hybrid promoter and an apolipoprotein E promoter, LP1 , HLP, minimal TTR promoter, FVIII promoter, hyperon enhancer, ealb-hAAT. Other examples include the E2F promoter for tumor-selective, and, in particular, neurological cell tumor-selective expression (Parr et al., 1997, Nat. Med. 3:1145-9) or the IL-2 promoter for use in mononuclear blood cells (Hagenbaugh et al., 1997, J Exp Med; 185: 2101-10). For example, for neuron-specific expression a promoter may be selected from a neuron-specific enolase (NSE) promoter, platelet-derived growth factor (PDGF) promoter, platelet-derived growth factor B-chain (PDGF-(3) promoter, synapsin or synapsin-1 (Syn or Syn-1) promoter, methyl-CpG binding protein 2 (MeCP2) promoter, Ca^+/calmodulin-dependent protein kinase II (CaMKII) promoter, metabotropic glutamate receptor 2 (mGluR2) promoter, neurofilament light (NFL) or heavy (NFH) promoter, p-globin minigene np2 promoter, preproenkephalin (PPE) promoter, enkephalin (Enk) promoter and excitatory amino acid transporter 2 (EAAT2) promoter. For astrocyte-specific expression a promoter may be selected from glial fibrillary acidic protein (GFAP) and EAAT2 promoters. For oligodendrocyte-specific expression the myelin basic protein (MBP) promoter a promoter may be selected. A particularly preferred promoterfor expression oftransgene in the peripheral and/or central nervous system is the CBh promoter (Gray et al., 2011 , Hum. Gene Ther. 22:1143-1153).

In one embodiment, in case transgene/therapeutic gene product is or includes a (small) RNA molecule, such as an siRNA, shRNA, miRNA, crRNA or a guide RNA), the promoter is an RNA polymerase III promoter such as a promoterfrom a U6 snRNA gene, preferably a primate or human U6 promoter.

Various modifications of the nucleotide sequences as defined above, including e.g. the wildtype parvoviral sequences, for proper expression in insect cells is achieved by application of well- known genetic engineering techniques such as described e.g. in Sambrook and Russell (2001) "Molecular Cloning: A Laboratory Manual (3rd edition), Cold Spring Harbor Laboratory, Cold Spring Harbor Laboratory Press, New York. Various further modifications of coding regions are known to the skilled artisan which could increase yield of the encode proteins. These modifications are within the scope of the present disclosure.

Compositions

In a sixth aspect, there is provided a composition comprising an AAV vector virion as described herein, i.e. an AAV vector virion comprising at least one AAV capsid protein variant as described herein.

In one embodiment, there is provided a composition comprising an AAV vector virion as described herein and a suitable excipient, such as a buffer, stabilizer, antioxidant etc. In one particular embodiment these compositions are used to transduce cells in vitro or ex vivo, in which case the excipients will need to be compatible with cell culture.

In other preferred embodiments the compositions are used for treatment of (human) subjects. For that purpose there is provided a pharmaceutical composition comprising an AAV vector virion as described herein and at least one pharmaceutically acceptable carrier. In the case of AAV gene delivery vehicles a pharmaceutical composition typically comprise physiological buffers, such as e.g. PBS, comprising further stabilizing agents such as e.g. sucrose. Such compositions are compatible with and suitable and intended for use in subsequent intravenous, intrathecal, intraparenchymal, intravitreal, subretinal administration or for use in organ-targeted vascular delivery such as intraportal or intracoronary delivery or isolated limb perfusion.

Uses

In a seventh aspect, there is provided the use of an AAV vector virion as described herein, or a composition comprising the AAV vector virion.

In one embodiment, there is provided an AAV vector virion as described herein, or a composition comprising the AAV vector virion for use as a medicament. In one embodiment, the AAV vector virion as described herein, or the composition comprising the AAV vector virion, is for use (as a medicament) in the treatment of a condition of the central nervous system.

In an eighth aspect, there is provided an AAV vector virion as described herein, or a composition comprising the AAV vector virion, for use (as a medicament) in gene therapy. In one embodiment, there is provided a method of gene therapy comprising the step of administering an effective amount of an AAV vector virion as described herein, or a composition comprising the AAV vector virion, to a subject in need of gene therapy.

In one embodiment, the gene therapy is for the treatment of a disease defined herein above (e.g. a disease or condition that can be treated by gene therapy of the central nervous system), preferably using a therapeutic gene as herein indicated above and/or using an RNAi agent for inhibiting or modifying the expression of an endogenous gene as indicated herein above.

Methods for identifying an AAV capsid variant with a desired characteristic

In an nineth aspect, there is provided a method for identifying an AAV capsid variant with a desired characteristic. In one embodiment, the method comprises: a) providing a library comprising a plurality of individually produced AAV capsid variants, wherein each member in the library differs by at least one amino acid from amino acid sequences of AAV capsid variants of other members in the library, and wherein each member in the library comprises a DNA construct, wherein the DNA construct comprises: i) a unique molecular identifier (UMI) that is unique for the member in the library;ii) a reporter gene operably linked to a promoter for driving expression in a mammalian cell; and, iii) at least one AAV ITR, b) contacting the library with a culture of cells, an organoid or a tissue, or administering the library to a non-human animal; c) allowing AAV capsid variants in the library to transduce the cells, the organoid, the tissue or cells in the animal; and, d) identifying at least one AAV capsid variant that has transduced at least one desired cell-type in the cells, the organoid, the tissue or in the animal in b), as an AAV capsid variant with the desired characteristic by sequencing its UMI, and optionally recovery of the AAV capsid variant with the desired characteristic from the desired cell.

Thus, in the method, each member in the library differs by at least one amino acid from amino acid sequences of AAV capsid variants of other members in the library. In one embodiment, an amino acid sequence of an AAV capsid variant of a member in the library comprises a mutation (e.g., insertion, deletion and/or substitution) of one or more amino acids. In one embodiment, an amino acid sequence of an AAV capsid variant of a member in the library comprises at least one amino acid difference (mutation) as compared to a wild type AAV capsid protein. A wild type AAV capsid protein is herein understood as an AAV capsid protein having the amino acid sequence of a naturally occurring AAV isolate, e.g., a clinical isolate. In one embodiment, an amino acid sequence of an AAV capsid variant of a member in the library comprises at least one amino acid difference (mutation) as compared to a wild type AAV capsid protein, in a VR region of a surface loop of the capsid protein variant, including in at least one of VR-I, VR-II, VR-III, VR-IV, VR-V, VR-VI, VR-VII, VR-VIII and VR-IX, of which VR-IV and VR-VIII are preferred and VR-VIII is more preferred. In one embodiment, an amino acid sequence of an AAV capsid variant of a member in the library comprises at least one amino acid difference (mutation) as compared to a wild type AAV capsid protein comprises an insertion of a peptide. The inserted peptide can comprise 2, 3, 4, 5, 6, 7, 8, 9, 10, 1 1 , 12, 13, 14, 15 or more amino acids.

In one embodiment, the above described at least one amino acid difference (mutation) as compared to a wild type AAV capsid protein is introduced into the AAV capsid proteins VP1 , VP2 or VP3, or in two of the capsid proteins in any combination, or in all three. In some embodiments, the mutations are introduced into VP1. In some embodiments, the mutations are introduced into VP2. In some embodiments, the mutations are introduced into VP3. In some embodiments, the mutations are introduced into VP1 and VP2. In some embodiments, the mutations are introduced into VP1 and VP3. In some embodiments, the mutations are introduced into VP2 and VP3. In some embodiments, the mutations are introduced into VP1 , VP2, and VP3.

In one embodiment, the above described at least one amino acid difference (mutation) as compared to a wild type AAV capsid protein is a single mutation (e.g., an insertion, a deletion and/or a substitution of an amino acid ) introduced at a single site in a capsid protein’s amino acid sequence, while in other embodiments, more than 1 , 2, 3, 4, 5, 6, 7, 8, 9 10, 15, 20, 25, 30, 40, 50, 100 or more (including any number between 1 and 100 or more) mutations (e.g., insertions, deletions and/or substitutions) are introduced in the amino acid sequence of a capsid protein.

In one embodiment, an amino acid sequence of an AAV capsid variant of a member in the library comprises at least one amino acid difference (mutation) as compared to a wild type AAV capsid protein, whereby wild type AAV capsid protein can be a capsid protein of a serotype selected from the group consisting of AAV1 , AAV2, AAV3A, AAV3B, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAV10, AAV1 1 , AAV12, AAV13 and AAVrhl 0. In a further embodiment, the wild type AAV capsid protein can be a capsid protein of a Clade A, Clade B, Clade C, Clade D, Clade E or Clade F AAV isolate as defined by Gao et al. (2004, J. Virol. 78: 6381-6388), exemplary sequences of which are disclosed in W02005/033321 and are available from GenBank® Database with Accession No.’s AY530553 to AY530629.

In the method for identifying an AAV capsid variant with a desired characteristic as described herein, each member in the library comprises a DNA construct, wherein the DNA construct comprises: i) a unique molecular identifier (UMI) that is unique for the member in the library; ii) a reporter gene operably linked to a promoter for driving expression in a mammalian cell; and, iii) at least one AAV ITR.

Thus, in one embodiment, the nucleic acid construct comprises a reporter gene operably linked to a promoter for driving expression in a mammalian cell. As understood herein, the reporter gene preferably encodes a protein that, when expressed in a mammalian cell produces a colorimetric, fluorescent, or luminescent signal that will allow cells that contain and express the reporter gene to be identified from surrounding cells that do not contain the reporter gene. In one embodiment, the reporter gene encodes a protein that produces a colorimetric signal, such as e.g. genes encoding secreted alkaline phosphatase (SEAP) or p-galactosidase. In one embodiment, the reporter gene encodes a protein that produces a fluorescent signal, preferred examples of which include EGFP, mCherry, mCloverS, mRubyS, mApple, iRFP, tdTomato, mVenus, YFP and RFP. In one embodiment, the reporter gene encodes a protein that produces a luminescent signal, preferred examples of which include genes encoding Photinus pyralis (firefly) luciferase and Renilla reniformis (sea pansy) luciferase and nanoluciferase. In one embodiment of the nucleic acid construct, the promoter operably linked to the reporter gene for driving its expression in a mammalian cell is a constitutive promoter. In other embodiments, the promoter operably linked to the reporter gene is cell type and/or tissue specific promoter. As will be understood by the skilled person, the cell type and/or tissue specificity of the promoter will depend on the desired cell type and/or tissue-tropism of the AAV capsid variant to be obtained. In some embodiments, the cell type and/or tissue specific promoter for driving expression of the reporter gene in a mammalian cell is a promoter selected from the group consisting of human synapsin promoter (hSynl), transthyretin promoter (TTR), cytokeratin 18, cytokeratin 19, unc-45 myosin chaperon B (unc45b) promoter, cardiac troponin T (cTnT) promoter, glial fibrillary acidic protein (GFAP) promoter, myelin basic protein (MBP) promoter, and methyl-CpG-binding protein 2 (Mecp2) promoter. In a preferred embodiment, the cell type and/or tissue specific promoter is human synapsin promoter (hSynl).

In one embodiment, the nucleic acid construct comprises an unique molecular identifier (UMI), also referred to as “barcode”. A UMI is a substantially unique sequence or barcode, preferably fully unique, that is specific for a nucleic acid molecule, i.e. unique for each nucleic construct used in the library. The UMI may have random, pseudo-random or partially random, or non-random nucleotide sequences. A UMI can be used to uniquely identify the originating molecule from which a sequencing read is derived. For example, reads of amplified nucleic acid molecules can be collapsed into a single consensus sequence from each originating nucleic acid molecule. As indicated above, the UMI may be fully or substantially unique. Fully unique is to be understood herein as that every nucleic acid molecule or construct provided herein comprises a unique tag that differs from all the other tags comprised in further nucleic acid molecules/nucleic acid constructs used herein. Substantially unique is to be understood herein in that each nucleic acid molecule or construct provided herein comprises a random UMI, but a low percentage of these nucleic acid molecules/nucleic acid constructs may comprise the same UMI. Preferably, substantially unique molecular identifiers are used in case the chances of tagging the exact same molecule comprising the same sequence with the same UMI is negligible. Preferably, the UMI is fully unique in relation to a specific sequence of the nucleic acid molecule or construct, e.g. fully unique in relation to a specific capsid variant packaging the nucleic acid construct comprising the specific UMI. The UMI preferably has a sufficient length to ensure this uniqueness. An identifier sequence may range in length from about 2 to 100 nucleotide bases or more, and preferably has a length between about 4-18 nucleotide bases, usually between about 8-12 nucleotide bases. Preferably the identifier sequence does not contain two or more consecutive identical bases. Furthermore, there is preferably a difference between individual identifier sequences of at least two, preferably at least three bases.

In one embodiment, the nucleic acid construct comprises at least one AAV ITR as herein described above. In the nucleic acid construct the reporter gene operably linked to a promoter for driving expression in a mammalian cell and the UMI are preferably flanked by at least one AAV ITR sequence, more preferably they are present in between the two AAV ITR sequences, such that the reporter gene linked to its promoter and the UMI are incorporated into the genome of a rAAV vector produced in a suitable host cell that is expressing AAV rep and cap gene products.

The method for identifying an AAV capsid variant with a desired characteristic as described herein, further comprises a step (b), wherein the library is contacted with a culture of cells, an organoid or a tissue, or wherein the library is administered to a non-human animal. In one embodiment, the culture of cells, an organoid or a tissue, or the non-human animal comprise at least one desired cell-type, e.g. cell-type for which an AAV capsid variant with improved tropism is to be identified. In one embodiment, the culture of cells, an organoid or a tissue, or the non-human animal can further comprise at least one non-desired cell-type, e.g. cell-type for which the AAV capsid variant to be identified has reduced transduction efficiency. In one embodiment, the library is systemically administered to a non-human animal, e.g. using an intravenous route of administration.

In a next step (c) of the method for identifying an AAV capsid variant with a desired characteristic as described herein, the AAV capsid variants in the library are allowed to transduce the cells, the organoid, the tissue or cells in the animal. Thus, in one embodiment of step (c), sufficient time is allowed to elapse for the AAV capsid variants in the library are allowed to transduce the (desired) cells, the organoid, the tissue or cells in the animal and to express the reporter gene.

In a next step (d) of the method for identifying an AAV capsid variant with a desired characteristic as described herein, at least one AAV capsid variant that has transduced at least one cell of the desired cell-type in the cells, the organoid, the tissue or in the animal is identified. In one embodiment, the at least one AAV capsid variant is identified by sequencing its UMI. In one embodiment, the at least one AAV capsid variant is identified as an AAV capsid variant with the desired characteristic. In one embodiment, step (d) further comprises recovery of the AAV capsid variant with the desired characteristic from a cell of the desired cell-type.

In one embodiment, step (d) comprises detection of transduction of a cell of the desired celltype by detecting expression of the reporter gene in at least one cell of the desired cell-type.

In one embodiment of the method for identifying an AAV capsid variant with a desired characteristic as described herein, the desired cell-type, e.g. cell-type for which an AAV capsid variant with improved tropism is to be identified, is a cell of an organ or tissue selected from liver, skeletal muscle, heart, diaphragm muscle, kidney, brain, stomach, intestines, skin, endothelial cells and lungs. In one embodiment of the method for identifying an AAV capsid variant with a desired characteristic as described herein, an AAV capsid variant with a desired characteristic to be identified has a reduced transduction efficiency of a non-desired cell-type, which non-desired celltype can be a cell of an organ or tissue selected from liver, skeletal muscle, heart, diaphragm muscle, kidney, brain, stomach, intestines, skin, endothelial cells and lungs.

In one embodiment, the method for identifying an AAV capsid variant with a desired characteristic as described herein, is a method wherein in step d) in at least one cell of the desired cell-type, for at least two members in the library, at least one of the mRNA expressed from and the genome copy-number of the least two members is quantified and identified by sequencing its UMI, and wherein the member with at least one of the highest mRNA expression level and genome copynumber is identified as the AAV capsid variant with the desired characteristic. In one embodiment of the method, in step d), in cells of more than one cell-types including the at least desired cell-type, for at least two members in the library, at least one of the mRNA expressed from and the genome copy-number of the least two members is quantified and identified by sequencing its UMI, and wherein the member with the most desired distribution over the more than one cell-types is identified as the AAV capsid variant with the desired characteristic. In one embodiment, the cells of more than one cell-types also include at least a non-desired cell-type.

The present invention has been described above with reference to a number of exemplary embodiments as shown in the drawings. Modifications and alternative implementations of some parts or elements are possible, and are included in the scope of protection as defined in the appended claims.

Description of the figures

Figure 1. Barcoded genome used for AAV production in HEK293T cells. GFP and nanoluciferase reporters, followed by a 12-nucleotide barcode before the poly adenylation signal and under the control of a CBh promoter.

Figure 2. AAV titers for novel variants (black bars) remains comparable to those of know variants (grey bars), as measured by qPCR specific for GFP reporter.

Figure 3. SDS Page gel run with purified AAV and known capsid batches. The VP123 ratio of the capsids produced display the natural stoichiometry of ~1 :1 :10 VP1 :VP2:VP3 for AAV.

Figure 4. Composition of the final AAV capsid library injected into the mice. The capsid library consists of 26 different capsids variants, which are not equally represented. Data labels focuses on the proportion (in percentage) of each of the novel capsids and known capsids in the capsid library.

Figure 5. Fold enrichment of novel and known capsids relative to AAV9, in the frontal brain, caudal brain and cerebellum of C57BL6 mice. Bars show the average of 4 mice.

Figure 6. Fold enrichment of novel and known capsids relative to AAV9, in the frontal brain, caudal brain and cerebellum of BALBc mice. Bars show the average of 4 mice.

Figure 7. Fold enrichment of novel and known capsids relative to AAV9, in the liver of both C57BL6 and BALBc mice. Vector genomes and transcripts in harvested tissues. Vector genomes in the cerebrum

(A), cerebellum (B) and liver (C), and vector transcripts in the cerebrum (D), cerebellum (E) and liver (F). Dotted horizontal line indicated the lowest limit of quantification (LLOQ) of the assay. Each bar represents the mean and standard deviation (SD) of 6 mice, and each dot represents the value of each mouse.

Nanoluciferase (nLuc) protein levels in cerebrum, cerebellum and liver tissues. Each bar represents the mean and standard deviation (SD) of 6 mice, and each dot represents the value of each mouse.

Fold change nano-luciferase relative fluorescent units (RFU) in cells overexpressing

Ly6C1 receptor relative to cells overexpressing the negative control receptor TMEM30a.

In-vivo validation of xBBB capsid experiment design. vDNA concentrations in liver tissue of control capsid (AAV9) and the two BBB crossing capsids after 4 weeks in-life. vDNA concentrations in cerebellum of control capsid (AAV9) and the two BBB crossing capsids.

Brain enrichment after normalization to the AAV-F library input, and relative to AAV9.

Horizontal line indicates no enrichment and corresponds to AAV9 wild-type capsid

Examples Introduction

Developing novel adeno-associated virus (AAV) capsids that can efficiently and specifically transduce cells in the central nervous system (CNS) through systemic administration is one of the most intriguing challenges in the field of AAV-based gene therapy. The presence of the blood-brain barrier (BBB) limits the entry of large molecules and viruses from the bloodstream into the CNS. Additionally, the broad tissue tropism of first-generation AAV results in high-level vector sequestration and transduction of peripheral off-target organs, such as the liver, upon systemic delivery. This not only raises safety concerns, including immune activation and liver toxicity, but it also reduces the amount of AAV vectors that are successfully delivered to the CNS, resulting in a narrow therapeutic window. Up to date, several novel AAV capsids with either improved CNS targeting or improved liver detargeting features have been engineered and discovered. These novel capsids candidates result from in vivo or in vitro screening of a collection of capsids, known as capsid library, and using next-generation sequencing (NGS) as experimental readout to retrieve the most enriched capsid candidates in the tissue of interest. The initial capsid library can be generated by a variety of methods, such as random mutagenesis, peptide insertion/display in one or more of the exposed loops of the capsid gene, or science-based rational design.

In the first example, we investigated whether the novel capsids (AAV.CAP-B10-9P31 , AAV.CAP-B10-F, 9.47-9P31 and 9.47-AAV-F) result in successful assembly of vector particles carrying the barcoded genome. In addition, in the second example we provide in vivo evidence of the enhanced CNS targeting and liver de-targeting features of the novel capsids compared to known variants harvested from the literature.

1 . Example 1 : AAV production of novel capsids

1.1 Methods

1. 1. 1 Construction of novel capsid variants

We produced 26 AAV capsids, which were screened in vivo in parallel in a multiplexed ‘race’ to determine which capsid would transduce the brain most efficiently when administered systemically. As references, our capsid library comprised wild-type variants, known variants harvested from the literature which contain either random mutations or peptide insertions that provide them either I iver-d etargeting or CNS targeting properties. The library further included novel capsids that result of combining known peptides or point mutations. Each capsid was individually produced in HEK293T cells using triple plasmid transfection method. As a result, all capsids carry a self-complementary (sc) barcoded genome which consists of two reporter genes, green fluorescent protein, GFP, and nanoluciferase, nLuc, and a 12 nucleotide barcode, unique for each capsid (Table 1) after the nLuc stop codon and before the poly A tail, under the control of a CBh promoter (Figure 1). After production, capsids were pooled and the resulting capsid library was intravenously (IV) administered to 7-week-old mice of strains C57BL6 and BALBc via tail vein injection at a dose of 1 ,4e11 vg/mouse (5.4e9 vg per capsid/mouse; vg = vector genome). Two weeks after vector delivery frontal brain, caudal brain, cerebellum and liver (as peripheral organ representative) were harvested. The barcoded reporter transgene was amplified by PCR and the resulting amplicons analyzed by next-generation sequencing (NGS) to assess the level of enrichment of each CNS capsid in each tissue. Table 1 . Barcode sequence (12 nucleotides) of each of the AAV capsids pooled in the capsid library. Novel capsids are marked in bold.

1.1.2 Production of novel capsid variants in HEK293T cells AAV viral vectors were produced in HEK293T cells using the triple plasmid transfection standard method. In brief, HEK293T were maintained in Dulbecco’s modified Eagle’s medium with GlutaMAX and 10% fetal bovine serum. Polyethylenimine hydrochloride transfection (PEImax®) transfection was performed using 16.6 pg adenovirus helper plasmid (pHelper), 16.6 pg barcoded transgene plasmid (pITR) and 16.6 pg of pRepCap plasmid containing AAV2 replicase (Rep) and each of the tested capsid genes (Genewiz).

Cells were lysed 72 hours after transfection with lysis buffer (1.5 M NaCI, 0.5 M Tris-HCI, 1 mM MgCh, 10% Triton X-100, pH 8.5) at 28 °C for 1 hour, followed by Benzonase treatment for 1 hour at 37 °C. Cell debris was removed by centrifugation for 15 minutes at 1900 x g after which the supernatant (crude lysate) containing the viral particles was purified with a batch binding protocol using POROS™ CaptureSelect™ AAVX Affinity Resin (ThermoFisher) for 2 hours. After two hours the affinity resin was washed with 0.2 M HPO4, pH 7.5 buffer and bound vectors were eluted by the addition of 0.2 M Glycine, pH 2.5. The pH of the eluted vectors was immediately neutralized by the addition of 0.5 M Tris-HCI, pH 8.5. Purified AAV vectors batches were stored at -20 °C.

1. 1.3 Analysis of novel capsid variants by QPCR and SDS PAGE gel electrophoresis

Titers of purified vector batches were determined using qPCR with primers and probes binding to the GFP transgene. AAVs were treated with DNAse at 37 °C to degrade exogenous DNA. AAV DNA was then released from the particles by a short (30min) heat treatment (37 °C) in the presence of 1 M NaOH. Next the alkaline environment was neutralized by the addition of an equal volume of 1 M HCI. Neutralized DNA was diluted 10 x in WFI 16 ng/pl PolyA, after which the samples were used in a qPCR with primers specific for the GFP transgene; forward AGCAAAGACCCCAACGAGAA, reverse GCGGCGGTCACGAACTC and probe Fam- CGCGATCACATGGTCCTGCT-mgb.

VP protein composition of purified vectors were determined by SDS page gel electrophoresis. Briefly, 15 pl of purified vectors was mixed with 5 pl 4 x Laemmli loading buffer (Biorad) supplemented with B-mercaptoethanol (Biorad). Following protein denaturation for 5 min at 95 °C, material was loaded on a Stain free polyacrylamide gel (Biorad). Next, the samples were electrophoretically separated for 35 minutes at 200 Volts. Following electrophoresis stain (tryptophan based) was developed under UV-light for 5 minutes after which VP proteins were then visualized under UV light on a Chemidoc imaging system (Biorad).

1.2 Results

The novel capsids were successfully assembly of vector particles carrying the barcoded genome. After production and purification AAV titers were determined with a qPCR specific for the GFP transgene. For all novel capsids, and using the same transfection method and same amount of plasmid DNA per dish, we found genome titers similar to those of the previously described capsids, ranging between 2.2e11 and 7.1 e11 vector genomes (vg) per mL of AAV capsid batchbound (BB) material (Figure 2). To investigate if the novel AAV capsids resulted in the correct VP stoichiometry we performed SDS-PAGE gel electrophoresis on batch binding purified material (Figure 3). VP1 :2:3 ratio of the resulting capsids has a natural ratio of approximately 1 :1 :10, reflective of an infective AAV capsid. 2. Example 2: Composition of AAV capsid library and in vivo analysis of its tissue distribution in mice upon intravenous administration

2.1 Methods

2. 1. 1 Composition of AAV capsid library, quality control and formulation

26 individual library members, as indicated in Table 1 , were individually produced as described in Example 1 .1 .2. Cells were lysed 72 hours after transfection with lysis buffer, followed by Benzonase and cell debris was removed by centrifugation as described.

After centrifugation, an aliquot was taken from each capsid crude lysed bulk (CLB) for titering (qPCR) and the remaining CLB was pooled and purified as a library using POROS™ CaptureSelect™ AAVX Affinity Resin (ThermoFisher) in an affinity chromatography setting ( KTA System). After purification, ultrafiltration/diafiltration (UD/DF) was performed in a hollow fiber set-up and buffer was exchanged to PCR-0.001 % Pluronic.

The titer of the final library in formulation buffer was determined using qPCR with primers and probes binding to the GFP transgene as described in Example 1 .1 .3.

2.1 .2 In vivo study, tissue processing and amplicon NGS and analysis

The final library was intravenously (IV) administered to 7-week-old mice of strains C57BL6 and BALBc via tail vein injection at a dose of 1 ,4e11 vg/mouse. Two weeks after vector delivery frontal brain, caudal brain, cerebellum and liver (as peripheral organ representative) were harvested. Tissue samples were pulverized and DNA and RNA was isolated using AllPrep DNA/RNA Mini (Qiagen). Total RNA was reverse transcribed using Maxima H minus reverse transcriptase (Thermo Fisher) primed with OligodT(is). Additionally, viral DNA was also isolated from the original AAV capsid library that was injected into the mice following the Purelink viral DNA/RNA mini kit (ThermoFisher) protocol. The barcoded reporter transgene was amplified from DNA, cDNA and original library by PCR for 40 cycles using Q5® Hot Start High-Fidelity DNA polymerase (NEB) with specific primers (forward CGCCGAACATGATCGACTATT, reverse CCCTTGGACGAGACTGAAC). The NEBNext® Ultra II DNA Library Prep kit (Illumina) was used to process the resulting amplicons. The quality and yield of the libraries was measured with the Fragment Analyzer and were further clustered in a NovaSeq6000 flow cell according to manufacturer’s protocols (Illumina). Each sample was sequenced at paired-end 151 bp read length (PE150) with a minimum of 10 million paired-end reads (GenomeScan). Known 12-nucleotide barcode sequences were found and counted from the sequencing reads using a barcode identification and python counting script downloaded from Github (https://qithub.com/JonasWeinmann/AAV-barcode-detection-and-normalization) adapted for our barcode sequences (Table 1), length (12 nucleotides) and flanking sequences (BCV_left="GTGGTCTTCTCA" and BCV_right="TTACGCCAGAAT"). The script was run on a Linux virtual machine running Ubuntu 20.04.3 LTS. The script outputs a text file indicating the barcode identity and the number or sequencing reads aligned to it. For each sample, the number of reads aligned to each barcode was divided by the total number of reads aligned to barcodes in that given sample and, to further normalize forthe capsid variability in the original capsid library that was injected, the proportion of each barcode in each tissue was divided by the proportion of each barcode in the original library. The resulting number represents the fold enrichment of each capsid in each of the tissues, relative to the library that was injected. Finally, all numbers were scaled to 1 to represent the normalized proportion of each capsid in a given tissue.

2.2 Results

Prior to injection into mice, the final purified AAV capsid library mice was titered using primers and a probe targeting the GFP transgene, and resulted to be 7.7e11 vg/mL. Additionally, the DNA packaged by the capsids in the library was extracted, the barcode sequence was amplified by PCR and the resulting amplicon was sequenced using NGS to evaluate the contribution of each of the capsids in the initial library (Figure 4), which was then used to normalize the expression levels of each capsid in each tissue.

Both types of mice strains (BALBc and C57BL6) received 175 pL of the capsid library via intravenous injection in the caudal vein, which resulted in a dose of 1 ,4e11 vg/mouse. Two weeks after library injection liver and brain tissues (divided in frontal and caudal brain and cerebellum) were harvested and following DNA and RNA isolation, cDNA synthesis and PCR to amplify the barcode present in the transgene, the amplicon was sequenced using NGS and the sequencing data was analyzed to evaluate the enrichment and the proportion of each of the capsids in each tissue, relative to the library that was injected.

Figures 5 and 6 show the fold transcriptional enrichment, compared to AAV9, of all novel and known capsids in either C57BL6 (Figure 5) or BALBc strain (Figure 6), while Figure 7 shows the liver fold enrichment relative to AAV9 for both strains.

As expected from previous results published in the literature, AAV-F targets all brain regions of both mice strains, displaying a 33 to 52-fold increase transduction compared to AAV9 in C57BL6 mice (Figure 5) and 136 to 219-fold increase in BALBc (Figure 6), depending on the brain region. However, AAV-F liver transduction remains virtually the same as AAV9 for both mice strains (Figure 7). When building the F peptide on top of the transduction defective 9.47 capsid (which, according to our NGS results, does not transduce the brain at the delivered dose in any of the mice strains) brain transduction is still higher than AAV9 in both mice strains, particularly in BALBc caudal brain (47-fold increase, Figure 6), while liver transduction is decreased 24-fold in C57BL6 and 55-fold in BALBc, compared to AAV9 (Figure 7). On the other hand, when building the F peptide on top of the known AAV.CAP-B10 capsid variant, which has lower liver transduction than AAV9 in both strains (Figure 7) and, according to our results at the delivered dose, higher brain transduction than AAV9 only in the C57BL6 mice strain (up to 24-fold higher, Figure 5), the brain and liver transduction remains similar to AAV.CAP-B10 in C57BL6 strain, but the brain transduction is rescued in the BALBc strain, with up to 61-fold higher transcription than AAV9 (Figure 6) while keeping the liver detargeting (Figure 7).

Regarding the 9P31 peptide-display capsid, as expected from published results, brain transcription enhanced compared to AAV9, particularly in the frontal brain, up to 104-fold in C57BL6 and up to 291 -fold in BALBc (Figures 5 and 6 respectively), while detargeting the liver (Figure 7). When building this 9P31 peptide on top ofthe transduction defective 9.47 capsid, brain transduction is still higher than AAV9 in both strains (again, particularly in the frontal brain), while liver transcription is very strongly reduced compared to AAV9, particularly in BALBc (Figure 7). On the other hand, when building the 9P31 peptide on top of the known AAV.CAP-B10 capsid variant, brain transduction is slightly increased in C57BL6 strain compared to the 9P31 capsid variant alone, however this effects are only observed in the caudal brain and cerebellum of BALBc strain (Figure 6). Regarding liver transduction of the AAV.CAP-B10-9P31 variant, this is reduced by up to 32-fold compared to AAV9, but remains similar to the parental capsid 9P31 , but higher than the parental capsid AAV.CAP-B10, showing no synergistic effect of both capsid features regarding liver detargeting (Figure 7).

In summary, we have produced and tested in vivo four novel capsids (9.47-F, 9.47-9P31 , AAV.CAP-B10-F and AAV.CAP-B10-9P31) and compared their brain and liver performance with the parental known capsids (9.47, AAV.CAP-B10, AAV-F and 9P31) and the golden standard AAV9, to investigate whether the synergistic effects of the capsid features (peptide insertions, random mutations) result in novel capsids that have the ability to transduce the brain better than AAV9 while keeping the liver detargeting feature (Table 2).

Table 2: List of mutations in the novel and known capsid sequence (VP1 notation). Underlined letters indicate amino acid changes compared to the wild type AAV9 (sequence 1)

Our results show that the novel capsids 9.47-F and 9.47-9P31 transduce the brain more efficiently than AAV9 although less than the parental capsids AAV-F and 9P31 . However, both capsids strongly detarget the liver compared to the parental capsids (thanks to the random mutations in the 9.47 backbone) and therefore potentially decreases liver toxicity and increases the therapeutic window. The novel capsids AAV.CAP-B10-F and AAV.CAP-B10-9P31 transduce the brain better than AAV9 in both mice strains, although less than the parental capsids AAV-F in both mice strains. Regarding liver detargeting, capsid 9P31 already showed strong liver detargeting compared to AAV9, rendering AAV.CAP-B10-9P31 with very little phenotypic advantage since liver detargeting is similar for both capsids. However, AAV.CAP-B10-F does keep the liver detargeting feature due to the AAV.CAP-B10 backbone. Additionally, AAV.CAP-B10 was only able to transduce the brain in the C57BL6 mouse strain, the mice strain where it was previously discovered, but not in BALBc according to our results, and the addition of the F peptide was able to rescue the phenotype in this mouse strain.

3. Example 3: Single-plex validation of 9.47-F candidate in vivo

3.1 Methods

3. 1. 1 Production of AA V capsid variants

AAV viral vectors AAV9, AAV-F and 9.47-F were produced in HEK293T cells using the triple plasmid transfection standard method. 72 hours after transfection the virus was harvested and purified by capture with Poros CaptureSelect AAV9 resin (ThermoFisher), further concentrated with Amicon ULTRA 15, Ultracel PL Membrane, 100 kDa, prepared in formulation buffer PBS-0.001 % Pluronic and filtered. AAVs were tittered by qPCR using ITR specific PCR primer. The AAV purity was checked by separating the viral proteins in 10% SDS-PAGE, and the integrity of the packed vector genomes was checked on 0.8% agarose. Additionally, endotoxin levels in the final product were determined using Endosafe-nestgen-PTS Spectrometer (Charles River laboratories). The sequence of the transgene to be expressed was the same for all three capsids (SEQ ID NO: 70) while the sequence for AAV9, AAV-F and 9.47-F capsids proteins are SEQ ID NO:2, SEQ ID NO:3 and SEQ ID NO:6, respectively.

3. 1.2 In vivo study in BALBc mice and tissue harvesting

Each of the three AAV capsids was intravenously administered (IV) to 7-week-old mice of BALBc strain via tail vein injection at a dose of 5e12 vg/mouse, with 6 mice per capsid group. Four weeks after vector delivery the mice were anesthetized with isoflurane 5%, perfused with saline solution, decapitated and whole brain and liver were sampled. Additionally, the whole brain was further divided into two parts: the cerebellum and the rest of the brain (cerebrum). Finally, all samples were snap frozen until further molecular analysis.

3.1.3 Processing of samples for molecular biology readout

Frozen tissue samples were pulverized with CryoPrep System CP02 (Covaris) and lysed with RLT Lysis Buffer in the Tissue Lyser II device (Qiagen). For protein analysis, total protein was measured from homogenized tissue using the Bradford assay, and upon total protein normalization of the samples, nanoLuc protein was quantified using the Nano-Gio® Luciferase Assay System (Promega). Additionally, DNA and RNA were simultaneously isolated from each sample using the Allprep DNA/RNA 96 kit (Qiagen). Upon nucleic acid isolation, 100 ug of total RNA were reverse transcribed using Maxima First Strand cDNA Synthesis kit for RT-qPCR with dsDNase (ThermoFisher Scientific) primed with Random Hexamers.

Quantification of the transgene (SEQ ID NO: 74) from either DNA or cDNA templates was performed with TaqMan Fast Universal PCR MasterMix2X (ThermoFisher Scientific). Forthe vector genomes DNA (vDNA) quantification, primers/probe used were targeting the CMV enhancer sequence (Fw: AGTAACGCCAATAGGGACTTTC, Rev: GGCGTACTTGGCATATGATACA, Probe: TTACGGTAAACTGCCCACTTGGCA; SEQ ID Nos. 75 to 77), while for the quantification of vector transcripts from cDNA samples, primers/probe used were targeting the NLuc transgene (Fw: GGAGGTGTGTCCAGTTTGTT, Rev: ATGTCGATCTTCAGCCCATTT, Probe:

ATCCAAAGGATTGTCCTGAGCGG; SEQ ID Nos. 78 to 80).

3.2 Results

In this example we investigated whether the 9.47-F candidate successfully transduces the brain while keeping the liver detargeting feature. Vector genome analysis in the brain and cerebellum show that both AAV-F and 9.47-F capsids deliver more than 10-fold more vectors to these tissues compared to AAV9 (Figure 8A-B), while analysis of the liver shows that, while vector genomes are similar for both AAV9 and AAV-F, levels are decreased by more than 500-fold in the liver of mice injected with 9.47-F capsids (Figure 8C). Vector transcript analysis in the brain, cerebellum and liver (Figure 8D-F) are in line with the vector genome results, showing increased functional transduction of capsids AAV-F and 9.47-F in both brain regions (particularly in the cerebellum), while decreased transduction in the liver in mice injected with 9.47-F.

Additionally, luciferase (nLuc) protein levels in the forebrain and cerebellum show that both AAV-F and 9.47-F capsids have higher nLuc readout than AAV9, particularly in the cerebellum (Figure 9), while nLuc protein levels in the liver are decreased in 9.47-F capsid at least by 30-fold compared to AAV9.

Overall, in this example we have evaluated in a single-plex manner the in vivo behavior of the AAV9-derived 9.47-F candidate with point mutations S414N, G453D, K557E and T582I and with 7mer FWGQSY peptide inserted between Q588 and A589. We can conclude that 9.47-F shows increase biodistribution and functional transduction in both the cerebellum and cerebrum parts compared to AAV9 wild-type, due to the presence of the 7mer peptide, while keeping the liver detargeting feature derived from the four point mutations.

4 Example 4: In vitro Ly6C1 binding assay of AAV-F capsid variants

4.1 Methods

4.1.1 Production of A A V-F variants

Each of the AAV-F capsid variants (including AAV-F) was produced individually in HEK293T cells using the triple plasmid transfection standard method. 72 hours after transfection the virus was harvested and purified by capture with POROS™ CaptureSelect™ AAVX Affinity Resin (ThermoFisher). The sequence of the transgene was the same for all capsids and is indicated in SEQ ID NO: 83. In brief, each capsid carries a self-complementary (sc) genome which consists of two reporter genes (GRP and nLuc). Upon production, AAVs were tittered using primers and probe targeting GFP (Fw primer: AGCAAAGACCCCAACGAGAA, Rev primer: GCGGCGGTCACGAACTC, probe: CGCGATCACATGGTCCTGCT; SEQ ID Nos. 11 to 13) and viral proteins were evaluated in 4-10% SDS-PAGE gel. Based on QC results, each of the 28 candidates that successfully result in AAV capsid assembly were used in an in vitro assay in order to evaluate their ability to interact with Ly6c1 receptor. 4. 1.2 Ly6C1 in vitro binding assay

In orderto evaluate the interaction of each ofthe AAV-F capsid variants to the Ly6c1 receptor, HEK293T cells were seeded in 48 well-plates (1 e5 cells/well) and reverse transfected with either a mouse Ly6C1 overexpression plasmid (SEQ ID NO:81) or with human TMEM30a plasmid (SEQ ID NO:82) a non-related transmembrane protein that acts as negative control. 48 hours after reverse transfection, cells were transduced with either of the AAV-F variants. 48 hours after transduction cells were lysed and the nanoluciferase activity derived from the capsid transgene was measured for each well using Nano-Gio® Luciferase Assay System kit (Promega). Finally, the fold change in nLuc readout between Ly6C1 and TMEM30a receptor was calculated to evaluate the enrichment of each capsid in Ly6C1 -overexpressing cells.

4.2 Results

This example we have produced 28 AAV-F variants (Table 3) and explored in vitro their ability to bind the Ly6C1 receptor.

Table 3. Sequence of the alternative AAV-F peptide insertions and location of the insertion.

Numbering is based on AAV9 viral protein 1 (VP1) sequence.

Upon cell transfection to overexpress the Ly6C1 or negative control receptor (TMEM30a), nLuc activity derived from transduction was measured for each of the transfected cells. Fold change was calculated by dividing the nLuc RLU readout of the Ly6C1 receptor relative to the same activity derived from cells overexpressing the TMEM30a. Based on these results (Figure 10), only the AIF1 peptide (FVVAQSY inserted between amino acids Q588-A589) showed enrichment in Ly6C1 relative to TMEM30a receptor at levels similar to those found in AAV-F capsid.

Based on this results we can conclude that the AAV-AIF1 maintains the capability of interacting with the Ly6C1 receptor in vivo. In order to prove this in vivo while maintaining the liver detargeting feature of 9.47-AAV, we performed an in vivo study to evaluate whether 9.47-AAV-AIF1 (SEQ ID NO:84) maintains the ability to cross the blood brain barrier upon systemic administration while keeping the liver detargeting feature. Additionally, a modification of the 9.47-AAV backbone excluding two of the mutations (from S414N, G453D, K557E, T582I 4-point mutations to only S414N, G453D 2-point mutations) and including the AIF1 peptide, 9.47ND-AAV-AIF1 (SEQ ID NO:85), was also tested in vivo.

5 Example 5: In-vivo validation of capsid

5.1 Methods

5. 1. 1 DNA/RNA isolation from in vivo samples

For DNA and RNA isolation, AllPrep Mini Kit from Qiagen (Catalog no. 80204) was used in all experiments described, following manufacturer’s protocol. RLT Plus lysis buffer was applied to pulverized animal tissue. The tissue suspension was then applied to Lysing Matrix D tubes with Matrix D beads (MP Biomedicals 116913100) and homogenized using TissueLyser system (Qiagen). The DNA and RNA were then isolated from this extract by applying it first to the DNA binding column from All Prep DNA/RNA mini kit. The flow-through from the DNA column was then added to the RNA binding column. After centrifugation- and washing steps, DNA and RNA was eluted from the corresponding columns. This way, DNA and RNA was isolated from the same sample. DNA and RNA quantity and integrity were determined by Nanodrop.

5.1.2 Vector DNA quantification from in vivo samples

Vector genome copies were quantified by using TaqMan qPCR assay (Thermo Fisher Scientific) with primers and probes against the CMV enhancer region of the vector, and primers and probes against mouse p-actin for housekeeping gene control (SEQ ID NOs 74 to 76 and 89 to 91). The quantification (GC/ug DNA) was done using a linear plasmid to generate a standard curve with a dilution range of this plasmid. This standard curve was used to calculate the vector DNA copy number form the DNA isolation from in vivo tissue samples. 5.2 Results

This in-vivo experiment further evaluates the efficacy of blood-brain barrier (BBB) crossing capsids by comparing them to a control capsid (AAV9). Our study employed a vector encoding for two proteins expressed from two open reading frames (ORFs). The investigation was conducted using male wild-type C57BL/6J mice, with a total of n=6 animals injected intravenously with 2.2E11 gc/animal for each vector. Following a 4-week duration in vivo, the mice were euthanized, and liver and cerebellum tissue was harvested and processed to assess vector DNA (vDNA) levels. This experimental design (Figure 11) aimed to elucidate the transduction efficiency and potential tissuespecific effects of the BBB crossing capsids.

The vDNA concentrations in liver tissue were compared between the control capsid (AAV9) and two BBB crossing capsids (9.47-AAV-AIF1 and 9.47ND-AAV-AIF1 , also referred to as 4mut 9.47-AAV-AIF-1 and 2mut 9.47ND-AAV-AIF-1 , respectively; SEQ ID NOs. 84 and 85). Results, as presented in Figure 12, revealed significantly higher vDNA levels (measured in ug/ml) in the control AAV9 capsid compared to both 9.47-AAV-AIF1 and 9.47ND-AAV-AIF1 (unpaired t-test, p=0.03425 and p=0.03502 respectively).

The vDNA concentrations in cerebellum tissue were also compared among the control capsid (AAV9) and the two BBB crossing capsids (9.47-AAV-AIF1 and 9.47ND-AAV-AIF1) as presented in Figure 13. It was observed that 9.47-AAV-AIF1 exhibited significantly higher vDNA concentrations compared to the control AAV9 capsid (unpaired t-test, p=0.04748). Additionally, AIF1 showed a trend towards increased vDNA levels compared to the control capsid, although this trend did not reach statistical significance.

The findings of this study demonstrate that the control AAV9 capsid resulted in significantly higher vDNA concentrations in liver tissue compared to the BBB crossing capsids 9.47-AAV-AIF1 and 9.47ND-AAV-AIF1. This suggests a potential advantage of the BBB crossing capsids in terms of reducing non-targeted transduction in the liver, which could be beneficial for achieving targeted gene delivery to the brain

The study also indicated that 9.47-AAV-AIF1 resulted in significantly higher vDNA concentrations in cerebellum tissue compared to the control AAV9 capsid. Conversely, 9.47ND- AAV-AIF1 exhibited a trend towards increased vDNA levels compared to the control, albeit not statistically significant. These findings suggest differential effects of BBB crossing capsids on vDNA distribution in the cerebellum.

6 Example 6: In vivo study of AAV-F variants

6.1 Methods

6. 1. 1 Production of AIF variants

Each of the AAV-F capsid variants (including AAV-F) was produced in HEK293T cells using the triple plasmid transfection standard method and purified with POROS™ CaptureSelect™ AAVX Affinity Resin (ThermoFisher). Each AAV-F variant carries a self-complementary (sc) barcoded genome (plasmid backbone SEQ ID NO: 86) consisting on two reporter genes (GFP, nLuc) followed by a 12 nucleotide barcode (unique for each capsid , Table 4) and under the control of the human synapsin promoter (hSynl). Upon titration of each individual variant (qPCR), all variants were mixed to produce a capsid cocktail which was further buffer exchanged to PBS- 0.001% Pluronic using D-Tube Dialyzer Mega MWCO 6-8 kDa (Merk) and concentrated using Amicon® Ultra Centrifugal Filter with 100 kDa MWCO (Sigma). The final cocktail was tittered, endotoxin tested and verified for potency in vitro in neuroblastoma-derived (SH-SY5Y) cells prior to the in vivo study in mice. Additionally, viral DNA was also isolated from the original AAV capsid cocktail that was injected into the mice following the Purelink viral DNA/RNA mini kit (ThermoFisher) protocol.

Table 4. Barcode sequence (12 nucleotides) of each of the AAV-F variants pooled in the cocktail for in vivo testing. 6. 7.2 In vivo study in BALBc mice and tissue harvesting

The capsid cocktail was intravenously administered (IV) to 7-week-old mice of BALBc strain via tail vein injection at a dose of 2.2e11 vg/mouse, with 5 mice per capsid group. Four weeks after vector delivery the mice were anesthetized with isoflurane 5%, perfused with saline solution, decapitated and whole brain and liver were sampled. All samples were snap-frozen until further molecular analysis.

6. 7.3 Sample processing, amplicon NGS and analysis

Frozen tissue samples were pulverized with CryoPrep System CP02 (Covaris) and lysed with RLT Lysis Buffer in the Tissue Lyser II device (Qiagen). DNA and RNA were simultaneously isolated from each sample using the Allprep DNA/RNA 96 kit (Qiagen). Upon nucleic acid isolation, 100 ug of total RNA were reverse transcribed using Maxima First Strand cDNA Synthesis kit for RT-qPCR with dsDNase (ThermoFisher Scientific) primed with Random Hexamers. Quantification of the transgene (SEQ ID NO: 74) from either DNA or cDNA templates was performed with TaqMan Fast Universal PCR MasterMix 2X (ThermoFisher Scientific). For the vector genomes DNA (vDNA) quantification, primers/probe used were targeting the CMV enhancer sequence (Fw: AGTAACGCCAATAGGGACTTTC, Rev: GGCGTACTTGGCATATGATACA, Probe:

TTACGGTAAACTGCCCACTTGGCA; sEQ ID NOs. 75 to 77), while for the quantification of vector transcripts from cDNA samples, primers/probe used were targeting the NLuc transgene (Fw: GGAGGTGTGTCCAGTTTGTT, Rev: ATGTCGATCTTCAGCCCATTT, Probe:

ATCCAAAGGATTGTCCTGAGCGGT; sEQ ID NOs. 78 to 80). Additionally, the barcode from the transgene was PCR amplified from brain cDNA and from the input AAV capsid cocktail administered to the mice using Q5® Hot Start High-Fidelity DNA polymerase (NEB) with forward primer CGCCGAACATGATCGACTATT and reverse primer CCCTTGGACGAGACTGAAC for 35 cycles (SEQ ID Nos. 87 and 88). The TruSeq Nano DNA library preparation kit (Illumina) was used to generate the sequencing libraries which were sequenced in a NovaSeq6000 flow cell according to manufacturer’s protocols (Illumina). Each sample was sequenced at paired-end 151 bp read length (PE150) with a minimum of 10 million paired-end reads (Macrogen Europe). Known 12-nucleotide barcode sequences were found and counted from the sequencing reads using a barcode identification and python counting script downloaded from Github (https://qithub.com/JonasWeinmann/AAV-barcode-detection-and-normalization) adapted for our barcode sequences (Table 3), length (12 nucleotides) and flanking sequences (BCV_left="GTGGTCTTCTCA" and BCV_right="TTACGCCAGAAT"). The script was run on a Linux virtual machine running Ubuntu 20.04.3 LTS. The script outputs a text file indicating the barcode identify and the number or sequencing reads aligned to it. For each sample, the number of reads aligned to each barcode was divided by the total number of reads aligned to barcodes in that given sample and, to further normalize forthe capsid variability in the original capsid library that was injected, the proportion of each barcode in each tissue was divided by the proportion of each barcode in the original library. The resulting number represents the fold enrichment of each capsid in each of the tissues, relative to the library that was injected. Finally, all numbers were scaled to the values of AAV in order to visualize the fold enrichment relative to AAV9.

6.2 Results

A capsid cocktail of AAV-F variants carrying a barcoded genome with GFP and nLuc transgenes under the control of the neuron-specific hSynl promoter was IV administered into BALBc mice in order to evaluate the transduction level of each capsid in the brain, separate to in vitro ability to bind the Ly6C1 receptor, and biodistribution into the liver. This was carried out because AAV9, which has brain expression, is known not to bind the Ly6C1 receptor. The barcoded genome was amplified from the administered AAV cocktail, and the resulting amplicon was sequenced using NGS to evaluate the contribution of each capsid in the initial library, which was then used to normalize the expression levels of each of the capsids in the mouse brain (cDNA). Finally, all capsids were normalized to wild type AAV9, which is expressed in the brain, to calculate the brain enrichment compared to the input and relative to AAV9 (Figure 14).

By analyzing the brain enrichment score in the brain (Figure 14) we can conclude that 10 of the AAV-F variants are more enriched in the mouse brain than the original AAV-F capsid, and based on Example 4 this 10 variants are not mLy6C1 receptor binders according to our results. This suggests that other receptor is being used for crossing the blood-brain-barrier in mouse.

Based on this results, the 3 best-producing capsids (SEQ ID Nos. 92 to 94) of those 10 top candidates are produced at large scale in Sf+ cells using transient transfection method followed by baculovirus infection for replication, and they are used in a singleplex mouse study, confirming their BBB-crossing capabilities.

Claims

1 . An adeno-associated virus (AAV) serotype 9 (AAV9) or Clade F AAV capsid protein variant comprising a mutation in one or more amino acids, which mutation(s) confer to the capsid protein variant at least one of: i) a phenotype of decreased liver transduction as compared to a control capsid protein not having the mutation(s); and, ii) a phenotype of decreased overall transduction as compared to a control capsid protein not having the mutation(s), and wherein the capsid protein variant further comprises an amino acid sequence that confers increased CNS transduction as compared to a control capsid protein not having the amino acid sequence.

2. An AAV capsid protein variant according to claim 1 , wherein at least one of: i) the capsid protein variant comprises a mutation in one or more amino acids in amino acid regions 452-458, 498-504, 590-595 and/or 582-587, and wherein the mutation(s) result in a phenotype of decreased liver transduction as compared to a control not having the mutation (s); ii) the capsid protein variant comprises a mutation at T138, S414, G453, K557, T568, T582 and Q590 or any combination thereof, and wherein the mutation(s) result in a phenotype of decreased overall transduction as compared to a control capsid protein not having the mutation(s); and, iii) the amino acid sequence that confers increased CNS transduction comprises an amino acid sequence selected from the group consisting of STTLYSP, FVVGQSY, DGTLAVPFK and WPTSYDA, and sequences differing by no more than four, three, two or one amino acid(s) therefrom.

3. An AAV capsid protein variant according to claim 1 or 2, wherein at least one of: i) the capsid protein variant comprises at least one mutation selected from the group consisting of mutations at positions W595, Q592, W503, N498, E500, and the amino acid sequence DGAATKN at positions 452 - 458, and wherein the mutation(s) result in a phenotype of decreased liver transduction as compared to a control not having the mutation(s); ii) the capsid protein variant comprises mutations at both T568 and Q590, or at both S414 and G453, or all of T138, S414, G453, K557 and T582, or at all of S414, G453, K557 and T582, and wherein the mutation(s) result in a phenotype decreased overall transduction as compared to a control capsid protein not having the mutation(s); and, iii) the amino acid sequence that confers increased CNS transduction comprises an amino acid sequence selected from the group consisting of FVVGQSY, FVVAQSY, FVVVQSY, FVVLQSY, FVVIQSY, FVVNQSY, FVVSQSY, FVVEQSY, FVVQQSY, FVVCQSY, FVVTQSY, FVVPQSY, FVVQSY, FVGVQSY, FVVQGSY, FVQGVSY, FSVGQVY and FSQGVVY, preferably selected from the group consisting of FVVGQSY, FVVAQSY, FVVPQSY, FVGVQSY and FVVQGSY.

4. An AAV capsid protein variant according to any one of claims 1 to 3, wherein at least one of: i) the capsid protein variant comprises a W595C mutation, a Q592L mutation, a W503R mutation, a N498Y mutation, a E500D mutation or any combination thereof, wherein the mutation(s) result in a phenotype of decreased liver transduction as compared to a control not having the mutation (s); ii) the capsid protein variant comprises the T568P and Q590L mutations, or the S414 and G453D mutations, or the T138A, S414N, G453D, K557E and T582I mutations or the S414N, G453D, K557E and T582I mutations; and, iii) the inserted amino acid sequence is inserted in a hypervariable region of the AAV capsid protein sequence, preferably in a hypervariable region selected from Loop IV or Loop VIII.

5. An AAV capsid protein variant according to any one of claims 1 to 3, wherein the capsid protein variant comprises: a) the T138A, S414N, G453D, K557E and T582I mutations and the amino acid sequence FVVGQSY or FVVAQSY inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589; b) the S414N, G453D, K557E and T582I mutations and the amino acid sequence FVVGQSY or FVVAQSY inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589; c) the T138A, S414N, G453D, K557E and T582I mutations and the amino acid sequence WPTSYDA inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589; d) the S414N, G453D, K557E and T582I mutations and the amino acid sequence WPTSYDA inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589; e) the amino acid sequence DGAATKN at positions 452 - 458 and the amino acid sequence FVVGQSY or FVVAQSY inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589; or, f) the amino acid sequence DGAATKN at positions 452 - 458 and the amino acid sequence WPTSYDA inserted in Loop VIII, preferably inserted between amino acid positions 588 and 589.

6. An AAV capsid protein variant according to any one of the preceding claims, wherein the AAV capsid protein variant is comprised in an amino acid sequence selected from the group consisting of SEQ ID NO.’s: 6, 7, 9, 10 96.

7. An AAV capsid protein variant according to any one of the preceding claims, wherein the AAV capsid protein variant is a variant of at least one of a VP1 , VP2 and VP3 capsid protein.

8. A nucleic acid encoding an AAV capsid protein variant according to any one of claims 1 to 7.

9. A recombinant AAV (rAAV) vector virion comprising at least one AAV capsid protein variant according to any one of claims 1 to 7, and preferably not comprising at least one of a wild type VP1 , VP2, and VP3 capsid protein.

10. An rAAV vector virion according to claim 9, wherein the rAAV vector virion comprises a nucleic acid molecule encapsidated by the at least one capsid protein variant, and wherein the nucleic acid molecule comprises a transgene flanked by at least one AAV inverted terminal repeat (ITR).

11. A composition comprising an AAV vector virion according to claim 9 or 10, wherein preferably the composition is a pharmaceutical composition comprising the AAV vector virion and at least one pharmaceutically acceptable carrier.

12. An AAV vector virion according to according to claim 9 or 10, or a composition according to claim 11 , for use as a medicament, wherein preferably the medicament is used in the treatment of a condition of the central nervous system.

13. A method for identifying an AAV capsid variant with a desired characteristic, the method comprising: a) providing a library comprising a plurality of individually produced AAV capsid variants, wherein each member in the library differs by at least one amino acid from amino acid sequences of AAV capsid variants of other members in the library, and wherein each member in the library comprises a DNA construct, wherein the DNA construct comprises: i) a unique molecular identifier (UMI) that is unique for the member in the library; ii) a reporter gene operably linked to a promoter for driving expression in a mammalian cell; and, iii) at least one AAV ITR, b) contacting the library with a culture of cells, an organoid or a tissue, or administering the library to a non-human animal; c) allowing AAV capsid variants in the library to transduce the cells, the organoid, the tissue or cells in the animal; and, d) identifying at least one AAV capsid variant that has transduced at least one cell of a desired cell-type in the cells, the organoid, the tissue or in the animal in b), as an AAV capsid variant with the desired characteristic by sequencing its UMI, and optionally recovery of the AAV capsid variant with the desired characteristic from a cell of the desired cell-type.

14. A method according to claim 13, wherein step d) comprises detection of transduction of a cell of the desired cell-type by detecting expression of the reporter gene in at least one cell of the desired cell-type.

15. A method according to claim 13 or 14, wherein in step d) in at least one cell of the desired cell-type, for at least two members in the library, at least one of the mRNA expressed from and the genome copy-number of the least two members is quantified and identified by sequencing its UMI, and wherein the member with at least one of the highest mRNA expression level and genome copy-number is identified as the AAV capsid variant with the desired characteristic.

16. A method according to claim 15, wherein in cells of more than one cell-types including the at least desired cell-type, for at least two members in the library, at least one of the mRNA expressed from and the genome copy-number of the least two members is quantified and identified by sequencing its UMI, and wherein the member with the most desired distribution over the more than one cell-types is identified as the AAV capsid variant with the desired characteristic.