WO2015038958A1

WO2015038958A1 - Selective recovery

Info

Publication number: WO2015038958A1
Application number: PCT/US2014/055490
Authority: WO
Inventors: Benjamin E. DEVERMAN; Paul E. PATTERSON
Original assignee: California Institute Of Technology
Priority date: 2013-09-13
Filing date: 2014-09-12
Publication date: 2015-03-19
Also published as: US20190292230A1; US9585971B2; EP3044318A1; US20170204144A1; US9957303B2; US11117933B2; EP3044318B1; US20220098243A1; US10202425B2; US20170240885A1; US10934329B2; US20200087353A1; US20180230186A1; US11932668B2; EP3044318A4; US20150079038A1; ES2739288T3; US10301360B2; EP3561062A1; EP3564379A1

Abstract

Provided herein are methods of selective screening. In addition, various targeting proteins and sequences, as well as methods of their use, are also provided.

Description

SELECTIVE RECOVERY

RELATED APPLICATIONS

[0001] The present claims the benefit of U.S. Provisional Applications Ser. No. 61/877506, filed Sept. 13, 2013; Ser. No. 61/983,624, filed April 24, 2014; Ser. No. 62/020658, filed July 3, 2014; and Ser. No. 62/034060, filed August 6, 2014, each of which is hereby incorporated by reference in their entireties.

STATEMENT REGARDING FEDERALLY SPONSORED R&D

[0002] This invention was made with government support under (MH100556 & MH086383 awarded by the National Institutes of Health. The government has certain rights in the invention.

SEQUENCE LISTING IN ELECTRONIC FORMAT

[0003] The present application is being filed along with a Sequence Listing in electronic format. The Sequence Listing is provided as a file entitled SEQUENCELISTING_CALTE103WO.TXT, created on September 11, 2014, which is 98,570 bytes in size. The information in the electronic format of the Sequence Listing is incorporated herein by reference in its entirety.

FIELD OF THE INVENTION

[0004] The present invention relates to methods of selective recovery and targeting proteins and/or methods.

BACKGROUND

[0005] Recombinant adeno-associated viruses (rAAV) are vectors for in vivo gene transfer applications. Several rAAV-based gene therapies are proving to be efficacious, most notably for the treatment of Leber's congenital amaurosis, hemophilia associated with factor IX deficiency and lipoprotein lipase deficiency (Simonelli et al 2010; Nathwani et al 2011; Gaudet et al. 2010). Recently, the first rAAV-based gene therapy, Glybera, was approved by the European Medicines Agency for the treatment of lipoprotein lipase deficiency. rAAVs have also shown success in preclinical models of a large variety of diseases, including Rett syndrome, congenital ALS, Parkinson's, Huntington's disease, Spinal Muscular Atrophy, among others and for the prophylactic delivery of broad neutralizing antibodies against infectious diseases such as HIV and influenza (Garg et al 2013; Valori et al. 2010; Foust et al. 2010; Foust et al 2013; Southwell et al 2009; Balazs et al 2011; and Balazs et al 2013). In addition, rAAVs are also popular vectors for in vivo delivery of transgenes for non-therapeutic scientific studies, such as optogenics.

SUMMARY OF THE INVENTION

[0006] In some embodiments, an AAV vector is provided that comprises an amino acid sequence that comprises at least 4 contiguous amino acids from the sequence TLAVPFK (SEQ ID NO: 1) or KFPVALT (SEQ ID NO: 3).

[0007] In some embodiments, a central nervous system targeting peptide is provided that comprises an amino acid sequence of SEQ ID NO: 1 (or any of the amino acid sequences in FIG. 31).

[0008] In some embodiments, a nucleic acid sequence encoding any four contiguous amino acids in TLAVPFK (SEQ ID NO: 1) or in KFPVALT (SEQ ID NO: 3) is provided (or for any of the sequences from FIG. 31).

[0009] In some embodiments, a method of delivering a nucleic acid sequence to a nervous system is provided. The method can comprise providing a protein comprising TLAVPFK (SEQ ID NO: 1) (or any of the other targeting sequences provided herein, for example, in FIG. 31), wherein the protein is part of a capsid of an AAV, and wherein the AAV comprises a nucleic acid sequence to be delivered to a nervous system; and administering the AAV to the subject.

[0010] In some embodiments, an rAAV genome is provided that comprises at least one inverted terminal repeat configured to allow packaging into a vector and a cap gene.

[0011] In some embodiments a plasmid system is provided that comprises a first plasmid comprising a modified AAV2/9 rep-cap helper plasmid configured such that it eliminates at least one of VP1, VP2, or VP3 expression and a second plasmid comprising a rAAV-cap-in-cis plasmid. [0012] In some embodiments, a method of developing a capsid with a desired characteristic is provided. The method can comprise providing a population of rAAV genomes (of any provided herein), selecting the population by a specific set of criteria, and selecting the rAAV genome that meets the screening criteria.

[0013] In some embodiments, a capsid protein is provided that comprises an amino acid sequence that comprises at least 4 contiguous amino acids from the sequence TLAVPFK (SEQ ID NO: 1) or KFPVALT (SEQ ID NO: 3) (or from any of the sequences in FIG. 31).

[0014] In some embodiments, a library of nucleic acid sequences is provided. The library can comprise a selectable element and one or more recombinase recognition sequences.

[0015] In some embodiments, a method of developing a capsid with a desired characteristic is provided. The method comprises providing a library of plasmids that comprise a capsid gene, and at least one recombinase recognition sequence, configured such that it allows a recombinase-dependent change in a sequence of a plasmid of the library that comprises the capsid gene that is a detectable change. The method can further comprise selecting the population by a specific set of criteria and selecting the rAAV genome that meets the screening criteria.

BRIEF DESCRIPTION OF THE DRAWINGS

[0016] FIG. 1A depicts various embodiments for aspects of various targeting proteins (in this case, examples of AAV capsid proteins), including G2B-13, G2B-26,

THl.1-32 and TH1.1-35.

[0017] FIG. IB depicts some embodiments of selective recovery of capsid proteins.

[0018] FIG. 2A depicts some embodiments of AAV genome manipulation.

[0019] FIG. 2B depicts some embodiments of capsid gene manipulation.

[0020] FIG. 2C depicts a flow chart for some embodiments of selective recovery embodiments.

[0021] FIG. 2D depicts a schematic of AAV genes and their known products. The AAV rep gene makes four protein products shown in black. The cap gene makes three structural capsid proteins (VP1-3) from one reading frame by a combination of alternative splicing and alternative initiator codons. In addition, the capsid gene also encodes an additional protein, assembly-activating protein (AAP), which is expressed from an alternative reading frame.

[0022] FIG. 2E is a schematic for the design of constructs used for some embodiments of the rAAV-based capsid library selection method. A capsid gene is inserted within a recombinant AAV genome flanked by ITRs. The expression and splicing of the AAV capsid gene products is controlled by the AAV5 p41 promoter upstream of the AAV2 rep sequences that contain the splice donor and intron sequences for the capsid gene products. By eliminating most of the rep gene, space (represented by the dotted lines) is available within the rAAV cap-in-cis genome for the insertion of additional elements.

[0023] FIG. 2F is a schematic showing the AAV components of the rep-AAP helper plasmid. Five stop codons were inserted within the capsid VP reading frame to ensure that VP1, VP2 and VP3 expression is eliminated from the rep- AAP helper.

[0024] FIGs. 3A and 3B. A strategy for recombinase-dependent recovery of sequences from transduced target cells. (FIG. 3A) The schematic shows the cap-in-cis rAAV genome. A ubiquitin C promoter fragment can be used to drive expression of an mCherry reporter followed by a synthetic polyA sequence. An AAV capsid gene, controlled by rep regulatory sequences, is followed by a lox71- and lox66-flanked SV40 late polyA signal. The lox66 site is inverted relative to lox71. In this configuration, Cre mediates the inversion (FIG. 3B) of the sequence flanked by the mutant lox sites. After the inversion, incompatible, double mutant lox72 and a loxP site are generated, reducing the efficiency of inversion back to the original state. Using PCR primers represented by the arrows in the schematic, cap sequences can be recovered selectively from genomes that have undergone a Cre-dependent inversion.

[0025] FIGs. 4A-4D. Alternative lox strategies for recombinase dependent recovery. (FIG. 4A) Single inverted loxP or lox71 and lox66 sites can be replaced by double loxP (white triangle) and lox2272 (black triangle) for irreversible recombination. (FIG. 4B) loxP sites or similar sites (lox2272 or other variants) inserted in the same orientation to mediate a deletion also allow selective recombination-dependent amplification. Recombination specific recovery can be achieved by performing the PCR-based recovery with three primers: one 5' of the randomized sequence (black arrow), one reverse primer downstream of the 3' loxP sequence (dark gray arrow) and a second forward primer that binds specifically within the deleted sequence (light gray arrow). This primer out competes the 5 ' (black) forward primer during amplification, reducing amplification of cap sequences from non-recombined sequences. Recovery of non-recombined sequences can also be reduced by digestion with an enzyme that recognizes a site within the sequence deleted by the recombinase. Alternatively, Cre-dependent and -independent products can be separated by size by gel electrophoresis. (FIG. 4C & 4D) Inverted loxP, lox71 and lox66 (shown), or DIO, FLEX sites can be placed in alternative configurations. (FIG. 4C) shows lox sites in an inverted orientation surrounding the rep and cap sequences, which can be inverted in the presence of Cre. (FIG. 4D) Schematic shows the option of flanking the reporter with lox sites. In this embodiment, the reporter is inverted and can be expressed after recombination.

[0026] FIGs. 5A-5C. A split Rep/AAP helper and rAAV-Cap-lox vector produces high titer virus. (FIG. 5A) DNase-resistant AAV genome copies (GCs) produced with the split AAV2/9 rep-AAP and AAV9 cap-in-cis genome (left), the AAV2/9 rep-AAP and a mCherry expressing rAAV genome (no cap - middle) or a control AAV2/9 rep/cap helper with the same AAV2:mCherry genome (right). (FIG. 5B) DNase-resistant viral GCs obtained from larger scale (7-10 150mm plate) preps of libraries with randomized 7-mer sequences replacing AAV9 capsid amino acids 452-8 (left) or inserted after amino acid 588 (right) (n=4/per library + SEM). (FIG. 5C) The PCR fragments containing the capsid sequence variation (black, 452-8 or light gray, 588) libraries are generated and cloned into a rAAV9R-delta-X/Acap- in-cis vector that has been modified to insert unique restriction sites Xbal (X) and Agel (A) flanking the region to be modified.

[0027] FIGs. 6A-6C. Cre-dependent sequence recovery after selection in Cre transgenics or Cre+ cells. (FIG. 6A) The schematic shows an overview of the selection process. In example 3, GFAP-Cre+ mice were injected with AAV virus containing AAV9- cap-in-cis, or the cap libraries with random 7mers at amino acids 452-8 or 588, and PCR products were recovered using primers that selectively amplify sequences from cap-in-cis genomes that have undergone Cre-mediated inversion of the sequence 3' to the cap gene. (FIG. 6B) The image shows an ethidium bromide-stained agarose gel of the PCR products recovered after the second PCR step using primers 1331 and 1312. (FIG. 6C) Recovered PCR products are then cloned into the rAAV9R-delta-X/A-cap-in-cis vector as a first step to generate the next round of capsid virus libraries.

[0028] FIG. 7. The novel AAV variants generate virus with efficiencies similar to AA V9. The graph shows the DNase-resistant viral GCs generated per 150mm dish of near confluent 293 producer cells for AAV9 and the four novel AAV serotypes.

[0029] FIGs. 8A-8C. G2B13 and G2B26 variants mediate enhanced transduction of the brain and spinal cord after IV administration as compared to AAV9. An AAV-CAG-eGFP-2A-ffLUC-WPRE-SV40 pA vector was packaged into AAV9 (left) or the novel variants G2B13 (middle) or G2B26 (right). Iel2 GC of each virus was injected IV into individual 5-week old female wt C57B1/6 mice and the brains of the mice were assessed for GFP expression 6 days later. (FIG. 8A) Panels show native eGFP fluorescence in whole brain. (FIG. 8B) Immunostaining for eGFP expression in the sectioned brains of mice injected with the indicated virus show efficient transduction of multiple cell types including neurons (n) and astrocytes (a). (FIG. 8C) Panels show native eGFP fluorescence in the livers of mice injected with the indicated virus.

[0030] FIGs. 9A-9I. G2B13 and G2B26 variants mediate enhanced transduction of CNS neurons and glia after IV administration as compared to AA V9. A rAAV-CAG-eGFP-2A-ffLUC-WPRE-SV40-pA vector was packaged into G2B13 (FIG. 9A- 9C) or G2B26 (FIG. 9D-9I) and lel2 GC of the indicated virus was injected IV into individual 5-week old female wt C57B1/6 mice. (FIG. 9A-9B, FIG. 9D-9I) Panels show immunostaining for eGFP in the sectioned brains of mice 6 days after they were injected IV with G2B13:CAG-GFP2A-Luc (FIG. 9A-9B) or G2B26:CAG-GFP2A-Luc (FIG. 9D, FIG. 9E, FIG. 9G-9I). Both vectors show transduction of several cell types including neurons (n) and astrocytes (a). (FIG. 9A-9B, FIG. 9D-9E, FIG. 9G-9H) Panels show eGFP immunostaining (upper panels) and NeuN immunostaining (lower panels) from paired image fields. (FIG. 9C and FIG. 9F) Panels show native eGFP fluorescence in the spinal cords of mice injected with G2B13 (FIG. 9C) or G2B26 (FIG. 9F). (FIG. 91) Panel shows eGFP immunostaining in (upper panels) co-localized (arrows) with the glial marker Sox2 (lower panels) from the same image field.

[0031] FIG. 10. Strategy for generating further diversity by combining sequences recovered at multiple sites. Following one or more rounds of selection for novel cap variants at two different sites, the pools of selected variants can be mixed to generate libraries that combine the randomized sequences at two or more sites by overlapping PCR. Using the same strategy, individual clones with novel sequences at 2 more sites can also be combined to generate clones with multiple modifications.

[0032] FIGs. 11A-11E. Cre-dependent sequence recovery after in vivo selection. (FIG. 11 A) The Brain Explorer 2 (Allen Brain Atlas) schematic shows an overview of the selection process. Capsid virus libraries were injected bilaterally into the straita of TH-Cre+ mice (asterisks show approximate injection sites), and capsid sequences were recovered from the substantia nigra (highlighted with a white square). (FIG. 11B) The image shows native mCherry fluorescence from 1 mm slices through the forebrain containing the striatal injection sites. (FIG. 11C) mCherry+ fibers from striatal neurons can be seen in slices from the SNr. The SNr and SNc located dorsal to the SNr were collected for capsid sequence recovery. (FIG. 11D) Panels show ethidium bromide stained PCR products recovered from TH-Cre+ cells of mice injected with AAV virus containing the libraries with random 7mers at amino acids 452-8 or 588 (round 1). (FIG. HE) Cap-in-cis genomes were recovered by a Cre-independent PCR strategy from Cre+ and Cre- mice demonstrating the presence of virus in all samples.

[0033] FIGs. 12A-12H. THl.1-32 and -35 variants exhibit rapid and efficient retrograde transduction of TH+ SNc neurons as well as neurons in additional regions known to project to the striatum. AAV-TH1.1-32:CAG-GFP or AAVTH1.1-35:CAG-GFP were injected into the striatum of adult mice and mice were killed 7 days later for GFP expression analysis. Panels show immunostaining for eGFP (FIG. 12A-B, FIG 12D and FIG. 12F-H) or TH (FIG. 12C and FIG 12E). (FIG. 12A) GFP expression within the striatum surrounding the injection site of the THl.1-35 variant. (FIG 12B-E) Panels show GFP immunostaining (FIG. 12B and FIG. 12D) and TH immunostaining (FIG 12C and FIG. 12E) within the same image field within the SN. Co-localization of GFP and TH+ immunostaining within the same cell is noted with arrows. GFP expression is evident in the SNr and a subpopulation of TH+ neurons in the SNc of a mouse injected with THl.1-35 (FIG 12B-C) or THl.1-32 (FIG 12D-E). (FIG. 12F) GFP immunostaining in the frontal cortex of a mouse injected with the THl.1-35 variant. GFP immunostaining in the thalamus (FIG. 12G) and amygdala (FIG. 12H) of a mouse injected with the THl.1-32 variant. [0034] FIG. 13 depicts a sequence alignment of AAVs and related parvoviruses showing diversity at 7mer insertion/replacement sites. The figure is split into three parts, with the first sheet representing the left-hand side of the figure, the second sheet representing the middle part of the figure, and the third sheet of FIG. 13 representing the right-hand side of the sheet. That is, the names for each of the rows in the first sheet are intended to carry across to the other two sheets (in each row).

[0035] FIG. 14 depicts a structural model of some embodiments of a capsid protein showing the loop regions where targeting sequences can be added or substituted in.

[0036] FIG. 15 depicts some embodiments of the rAAV9R-delta-X/A-cap-in-cis vector.

[0037] FIG. 16 depicts some embodiments of an AAV rep/cap helper plasmid that was modified by inserting a total of 5 stop codons within the cap gene within the VP1, 2 and 3 reading frame (1 stop codon disrupts VP3, 3 disrupt VP2 and all 5 disrupt VP1 - FIG. 2F, SEQ ID NO: 5).

[0038] FIG. 17 depicts some embodiments of a template DNA (pCRII-9R-X/A EK plasmid, SEQ ID NO: 6.)

[0039] FIG. 18 depicts AAV9R-delta-X/A-cap-in-cis, SEQ ID NO: 7, in which the coding region between the Xbal and Agel sites was eliminated to prevent "wt" AAV9R X/A capsid protein production from any undigested vector during library virus production.

[0040] FIG. 19 depicts some embodiments of a sequence of an AAV-PHP.B (AAV-G2B26) capsid VP1 protein.

[0041] FIG. 20 depicts some embodiments of a nucleic acid sequence for an AAV-PHP.B (AAV-G2B26) capsid gene coding sequence.

[0042] FIG. 21 depicts some embodiments in which a nucleic acid sequence for a targeting protein is cloned into an AAV Rep-Cap helper plasmid.

[0043] FIGs. 22A-22F. Recombinase-dependent recovery of AAV capsid sequences from transduced target cells. FIG. 22A is a schematic showing the rAAV-cap- in-cis rAAV genome used for capsid library generation. Cre mediates the inversion of the sequence flanked by the mutant lox sites and PCR primers, represented by half arrows in the schematic, are used to selectively amplify the recombined sequences. (FIG. 22B) The schematic shows the AAV components of the Rep-AAP helper plasmid. Stop codons inserted in the VP reading frame eliminate VPl, VP2 and VP3. (FIG. 22C) DNase-resistant AAV genome copies (GCs) produced with the split AAV2/9 rep-AAP and AAV9 cap-in-cis genome (left) as compared to a control AAV2/9 rep/cap helper with a control AAV-UBC- mCherry genome (middle) or the AAV2/9 rep-AAP and control AAV-UBC-mCherry genome (no cap - right). (FIG. 22D) Representative PCR products showing Cre dependent (top) and Cre independent (bottom) amplification of recovered capsid library sequences from TH-Cre positive or Cre negative mice. (FIG. 22E) The capsid sequence variation libraries at AA452-8 or after AA588 of AAV9 (vertical gradient) are generated by PCR and cloned into a rAAV-Acap-in-cis vector that has been modified to insert unique restriction sites Xbal (x) and Agel (a) flanking the variable region(s). (FIG. 22F) The schematic shows an overview of some embodiments of the selection process.

[0044] FIGs. 23A-23H. AAV-PHP.R2 mediates rapid and efficient retrograde transduction. (FIG. 23A) Capsid libraries were injected into the striatum (left) and tissue from the substantia nigra was collected for capsid sequence recovery 10 days later. (FIG. 23A) Images show native mCherry fluorescence expressed from the cap-in-cis library genomes surrounding the injection sites (left) and mCherry positive axons from striatal neurons in the SNr (right). (FIG. 23B- FIG. 23H) The recovered AAV variant PHP.R2 was used to package ssAAV-CAG-GFP and 7xl0⁹ VG was injected into the striatum. Seven days later, the mice were assessed for GFP expression by immunostaining. The images show GFP positive cells at the striatal injection site (FIG. 23B) or at the indicated brain regions that contain GFP+ cell bodies (FIG 23C, FIG 23E-I). (FIG 23D) shows immunostaining for TH in the SNc. (FIG 23E) shows GFP immunostaining from the same field shown in FIG 23D. Co-localization of GFP and TH immunostaining within the same cell is noted with arrows. Scale bars are 100 um in C, E-H and 20 um in 23D.

[0045] FIGs. 24A-24I. AAV-PHP.B mediates robust transduction of the entire CNS after IV administration. Representative images from mice transduced with lxlO¹² VG of ssAAV-CAG-GFP-2A-Luc packaged in AAV9 or AAV-PHP.B. GFP expression was assessed 3 weeks later by immunostaining (FIG. 24A and FIG. 24C) or native GFP fluorescence (FIG. 24B, FIG. 24D-24G). (FIG. 24A) Images show GFP immunostaining in sagittal brain sections from mice given AAV9 (left), an equivalent dose of AAV-PHP.B (middle) or lxlO¹¹ VG of AAV-PHP.B (right). (FIG. 24B) Representative cortical (left) or striatal (right) 50 um maximum confocal projection images of native eGFP fluorescence from the brains of mice treated as in 24A. (FIG. 24C) Nearly all Calbindin^"1" Purkinje cells (bottom) are GFP⁺ (top) 3 weeks after IV injection of lxlO¹² VG of AAV- PHP.B (FIG. 24D) Representative image of native GFP fluorescence from the lumbar spinal cord. The inset shows an enlargement of the boxed ventral spinal cord area. (FIG. 24E) Confocal maximum projection image of GFP fluorescence from a whole mount retina. (FIG. 24F) Cross section of the retina. (FIG. 24G) CLARITY images of GFP fluorescence from the cortex (left), striatum (right) and ventral spinal cord of a mouse transduced with lxlO¹² VG of AAV-PHP.B. AAV biodistribution in the brain (FIG. 24H) and peripheral organs (FIG. 241) 25 days after injection of lxlO¹¹ VG IV into adult mice. N=3 for AAV-PHP.B and N=4 for AAV9; error bars show standard deviation (s.d.); *p<0.05, ***p<0.001, one-way ANOVA and Bonferroni's multiple comparison test. Scale bars are 1 mm in FIG. 24A and FIG. 24D, 50 um in 24B, and 200 μπι in FIG. 24C. Major tick marks in FIG. 24G are 100 um.

[0046] FIGs. 25A-25J. AAV-PHP.b transduces many CNS neuronal and glial cell types. Representative images show immunostaining for GFP (FIG. 25A-25D, FIG. 25G-25I) or native GFP fluorescence (FIG. 25E, FIG. 25F, and FIG. 25 J). Images are of the brain regions indicated in the panel or striatum (FIG 25E) or hippocampus (FIG. 25H). For FIGs. 25A-25F, the left image shows the antigen immunostaining, while the right image in each pair shows GFP expression. For FIGs. 25G-25J, the top image shows the indicated antigen immunostaining, the middle images show GFP expression and the lower paired images show a higher magnification views of the indicated boxed areas. Mice received lxlO¹¹ VG (FIG. 25A) or lxlO¹² VG (FIG. 25B-25J) at 5-6 weeks of age and were assessed for eGFP expression 3 weeks later. Scale bars are 50 μπι in FIG. 25A, and 20 μπι in FIG. 25B-25J. In all panels, arrows indicate colocalization and asterisks indicate cells that are positive for the indicated antigen but negative for GFP.

[0047] Figure 26. Systemic, low-dose AAV-PHP.B reporter vector labeling can be used together with CLARITY for single cell morphological phenotyping. (FIG 26A) A tiled image of the cortex shows sparse labeling of cells with GFP after IV administration of lxlO¹⁰ VG of rAAV-PHP.B:CAG-GFP. The region highlighted by the box is shown magnified in 3 different orientations (FIG 26B). Two astrocytes can be seen making contact with a blood vessel containing an endothelial cell with a GFP+ nucleus. The astrocyte endfeet can be seen spiraling around the blood vessel. In some embodiments, this can be used to label individual cells, assess the morphology of the cells and see the material's impact on the cells. (FIG. 26C) An image of the striatum shows sparse labeling of cells with GFP after IV administration of lxlO¹⁰ VG of rAAV-PHP.B:CAG-GFP. (FIG. 26D) An individual medium spiny neuron is highlighted using an semi-automated filament tracing method (Imaris, Bitplane software). (FIG. 26E) The same medium spiny neuron highlighted in FIG. 26D is shown isolated along with several closely associated neural processes. In some embodiments, this can be used to sparsely label cells and assess the association of the labeled cells.

[0048] FIGs. 27A-27C Primers used for generating capsid library fragments and Cre-dependent capsid sequence recovery. (FIG. 27A) Schematic shows PCR products as the right hand shaded section with 7AA of randomized sequence (represented by vertical multishaded bars) inserted after amino acid 588 (588i library) or replacing AA452-8 (452-8r library). The primers used to generate these libraries are indicted by name and half arrow. For the generation of the second library, the template was modified to eliminate a naturally occurring Earl restriction site within the capsid gene fragment (xE). In this way, any contamination from amplified wt AAV capsid sequence could be eliminated by digesting the recovered libraries with Earl. (FIG. 27B) Schematic shows the rAAV-Cap-in-cis vector and the primers used to quantify vector genomes (left) and recover sequences that have transduced Cre expressing cells (left). (FIG. 27C) The figure provides the sequences for the primers shown in FIG. 27A and FIG. 27B. Table 0.1 also provides a listing of the sequences:

TABLE 0.1

Primer Purpose Sequence

9CapF Step 1: forward CAGGTCTTCACGGACTCAGACTATCAG SEQ ID NO: 16

CDF Step 1: reversed CAAGTAAAACCTCTACAAATGTGGTAAAATCG SEQ ID NO: 17

by Cre

XF Step 2 forward ACTCATCGACCAATACTTGTACTATCTCTCTAGAAC SEQ ID NO: 18

AR Step 2 reverse GGAAGTATTCCTTGGTTTTGAACCCA SEQ ID NO: 19

TF qPCR forward GGTCGCGGTTCTTGTTTGTGGAT SEQ ID NO: 20

TR qPCR reverse GCACCTTGAAGCGCATGAACTCCT SEQ ID NO: 21

7xNNK 452-8r library CATCGACCAATACTTGTACTATCTCTCTAGAACTATTNNKNNK

generation NNKNNKNNKNNKNNKCAAACGCTAAAATTCAGTGTGGCCGGA SEQ ID NO: 22

7xMNN 588i library GTATTCCTTGGTTTTGAACCCAACCGGTCTGCGCCTGTGCMNN

generation MNNMNNMNNMNNMNNMNNTTGGGCACTCTGGTGGTTTGTG SEQ

ID NO: 23

[0049] FIG. 27D The schematic shows an overview of some embodiments of a process used to introduce 2-site randomization after the first round of selection (combinatorial libraries). This process was used to develop AAV-PHP.R2. Both libraries (452-8r and 588i) were generated by PCR, cloned into the rAAV-Cap-in-cis vector and capsid selection was performed in TH-Cre mice. Sequences from both libraries were recovered and were combined using an overlapping PCR strategy to generate a new library that should contain all possible combinations of the 7mer sequences recovered from the 452- 8r library with all of the 7mer sequences recovered at the 588i in one library. This library was subjected to a second round of selection in TH-Cre mice and recovered variants that showed signs of enrichment were characterized individually.

[0050] FIG. 27E DNase-resistant vector genomes (VGs) obtained from preps of individual variants recovered from GFAP-Cre and TH-Cre selections. Yields are given as vector genome copies per 150 mm dish of producer cells. Error bars show s.d. N=3 for AAV9, PHP.A and PHP.B. N=l for PHP.r. One way analysis of variance (ANOVA).

[0051] FIG. 28A-28E AAV-PHP.A more efficiently and selectively transduces CNS astrocytes. (FIG. 28A) Representative images of GFP immunostaining of brain sections from mice injected as adults with 3xl0ⁿ VG of a ssAAV-CAG-eGFP expressing vector packaged into AAV9 or PHP.A as indicated. (FIG. 28B) Panels show GFP immunostaining (left) and cell nuclei (right) in the cortex of mice that received AAV9 or PHP.A as indicated. AAV biodistribution in the brain (FIG. 28C) and peripheral organs (FIG. 28D) 25 days after injection of lxlO¹¹ VG IV into adult mice. N=4; error bars show standard deviation (s.d.); *p<0.05, **p<0.01, ***p<0.001, one-way ANOVA and Bonferroni's multiple comparison test. (FIG. 28E) Representative images of GFP immunostaining of liver sections from mice injected as adults with 3xl0ⁿ VG of a ssAAV-CAG-eGFP expressing vector packaged into AAV9 or PHP.A as indicated. Scale bar is 50 μπι in 28B.

[0052] FIG. 29. Rapid cellular level tropism characterization with whole animal tissue clearing. Images show native GFP fluorescence in the indicated organs of mice 3 weeks after IV injection of 1x10 12 AAV9 or AAV-PHP.B as indicated. The indicated organs were rendered optically transparent using the PARS-based CLARITY (a whole body tissue clearing method - Yang et al. 2014). Confocal Z-stack images were reconstructed into three-dimensional images using Imaris software (Bitplane).

[0053] FIG. 30 depicts a set of embodiments for amino acid and nucleic acid of

AAV9.

[0054] FIG 31 Depicts a set of embodiments for additional targeting proteins. Any of the targeting protein embodiments provided in FIG. 31 can be swapped out for any of the embodiments involving any other particular targeting protein embodiment described herein. Similarly, any of the nucleic acids provided in FIG. 31 can similarly be swapped out for any of the particular nucleic acids provided herein. The figure depicts the most highly enriched sequences recovered from the second round of selection for AAV variants that transduce GFAP-Cre+ astrocytes following intravenous administration.

DETAILED DESCRIPTION OF EMBODIMENTS

[0055] rAAVs have reinvigorated the field of gene therapy and facilitate the gene transfer critical for a wide variety of basic science studies. Several characteristics make rAAVs attractive as gene delivery vehicles: (i) they provide long-term transgene expression, (ii) they are not associated with any known human disease, (iii) they elicit relatively weak immune responses, (iv) they are capable of transducing a variety of dividing and non- dividing cell types and (v) the rAAV genome can be packaged into a variety of capsids, or protein coat of the virus, which have different transduction characteristics and tissue tropisms. Despite these advantages, the use of AAV for many applications is limited by the lack of capsid serotypes that can efficiently transduce certain difficult cell types and by the lack of serotypes that can efficiently and selectively target a desired cell type/organ after systemic delivery.

[0056] Using directed evolution to improve AAV capsid characteristics. One approach that has been used to develop rAAVs with improved tissue/cell type targeting is to perform directed evolution on the AAV capsid gene. Typically this is done by making a library of replication competent AAVs that are modified to introduce random mutations into the AAV cap gene, which codes for the capsid proteins that determines the tissue tropism. The AAV capsid virus library is then injected in an animal or delivered to cells in culture. After a certain time, capsid sequences that are present in the cells/tissue of interest are recovered. These recovered sequences are then used to generate a new pool of viruses and then the process is repeated. Through repeated rounds of selection/sequence recovery, sequences that generate capsids that function better (i.e., those repeatedly pass the selection process) will be enriched. The capsids that exhibit an improved ability to transduce the target can then be recovered and assessed as individual clones or mutated further and subjected to additional rounds of selection.

[0057] Directed evolution has been used to generate AAVs that evade neutralizing antibodies (Maheshri et al 2006) and better target glioma cells (McGuire et al. 2010), airway epithelium (Excoffon et al. 2009) and photoreceptors in the retina after intravitreal injection (Dalkara et al 2013). In addition, using a human/mouse chimeric liver model, Lisowski et al. developed a rAAV that specifically and efficiently targeted the human hepatocytes (2013).

[0058] Some of the embodiments herein described provide methods for the enrichment and selective recovery of sequences with desirable traits from libraries of sequence variants using a recombination-dependent recovery strategy. This method is widely applicable for the selective enrichment of sequences from randomized libraries that mediate an increased contact between the nucleic acid containing the randomized sequence and a recombinase that recognizes a specific sequence or sequences present on the same nucleic acid as the randomized sequence. The recombinase can be expressed in response to desired stimuli, in a desired subcellular compartment or expressed in a specific target population of cells in vitro or in vivo.

[0059] As an example of the use of some embodiments, an option for selectively recovering adenoassociated virus (AAV) capsid sequences that encode capsid proteins that more efficiently and/or selectively transduce specific Cre recombinase (Cre) expressing target cell populations has been provided herein. Cre recognition sites (loxP or variants of loxP sites) can be inserted into an rAAV genome adjacent or flanking to the capsid gene. In this way, when the capsid gene enters the nucleus of a Cre expressing cell and is converted to dsDNA, Cre can induce a recombination event between the lox sites within the rAAV genome resulting on an inversion or deletion (depending on their relative orientations) of the sequence flanked by the lox sites. Using a recovery strategy that is dependent on the recombination event, capsid sequences that encode capsids that direct the rAAV genome to the nucleus of Cre+ cells can be enriched through one or more rounds of selection. Capsid gene directed evolution is only one example application of this technology. In some embodiments, the method can be adapted for the selection of any other coding or non-coding sequences with desirable traits within an AAV genome or any sequences within other viruses or non- viral nucleic acids that alter the interaction with the recombinase.

[0060] The following sections provide a brief set of definitions for the various terms and then various embodiments that have been produced through these screening methods. Following this, detailed variants and embodiments of the screening method are provided, as well as examples thereof.

Definitions:

[0061] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention as claimed. In this application, the use of the singular includes the plural unless specifically stated otherwise. In this application, the use of "or" means "and/or" unless stated otherwise. Furthermore, the use of the term "including", as well as other forms, such as "includes" and "included", is not limiting. Also, terms such as "element" or "component" encompass both elements and components comprising one unit and elements and components that comprise more than one subunit unless specifically stated otherwise. Also, the use of the term "portion" can include part of a moiety or the entire moiety.

[0062] The section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described. All documents, or portions of documents, cited in this application, including but not limited to patents, patent applications, articles, books, and treatises, are hereby expressly incorporated by reference in their entirety for any purpose. As utilized in accordance with the present disclosure, the following terms, unless otherwise indicated, shall be understood to have the following meanings:

[0063] A "plasmid" is as a nucleic acid that can be used to replicate recombinant DNA sequences within a host organism. The sequence can be preferably double stranded DNA. [0064] The term "recombinase recognition sequence" or "recombinase recognition site" refers to a sequence of nucleic acid that is recognizable by a recombinase and can serve as the substrate for a recombination event catalyzed by said recombinase. The sequence can be double stranded DNA.

[0065] The term "virus genome" refers to a nucleic acid sequence that is flanked by cis acting nucleic acid sequences that mediate the packaging of the nucleic acid into a viral capsid. For AAVs and parvoviruses, for example it is known that the "inverted terminal repeats" (ITRs) that are located at the 5' and 3' end of the viral genome have this function and that the ITRs can mediate the packaging of heterologous, for example, non-wt virus genomes, into a viral capsid.

[0066] The term "element" refers to a separate or distinct part of something, for example, a nucleic acid sequence with a separate function within a longer nucleic acid sequence.

[0067] The term "rAAV" refers to a "recombinant AAV". Recombinant AAV refers to an AAV genome in which part or all of the rep and cap genes have been replaced with heterologous sequences.

[0068] The term "AAV" or "adeno-associated virus" refers to a Dependoparvovirus within the Parvoviridae genus of viruses. Herein, AAV can refer to an AAV derived from a naturally occurring "wild-type" virus, an AAV derived from a rAAV genome packaged into a capsid derived from capsid proteins encoded by a naturally occurring cap gene and/or a rAAV genome packaged into a capsid derived from capsid proteins encoded by a non- natural capsid cap gene, for example, AAV-PHP.B.

[0069] The term "rep-cap helper plasmid" refers to a plasmid that provides the viral rep and cap gene functions. This plasmid can be useful for the production of AAVs from rAAV genomes lacking functional rep and/or the capsid gene sequences.

[0070] The term "vector" is defined as a vehicle for carrying or transferring a nucleic acid. Examples of vectors include plasmids and viruses.

[0071] The term "cap gene" refers to the nucleic acid sequences that encode capsid proteins that form, or contribute to the formation of, the capsid, or protein shell, of the virus. In the case of AAV, the capsid protein may be VP1, VP2, or VP3. For other parvoviruses, the names and numbers of the capsid proteins can differ. [0072] The term "rep gene" refers to the nucleic acid sequences that encode the non-structural proteins (rep78, rep68, rep52 and rep40) required for the replication and production of virus.

[0073] A "library" may be in the form of a multiplicity of linear nucleic acids, plasmids, viral particles or viral vectors. A library will include at least two linear nucleic acids.

[0074] When the inserted nucleic acid sequences are randomly generated, N=A,C,G or T; K=G or T; M=A or C.

[0075] Unless specified otherwise, the left-hand end of any single-stranded polynucleotide sequence discussed herein is the 5' end; the left-hand direction of double- stranded polynucleotide sequences is referred to as the 5 ' direction.

[0076] As used herein, "operably linked" means that the components to which the term is applied are in a relationship that allows them to carry out their inherent functions under suitable conditions. For example, a control sequence in a vector that is "operably linked" to a protein coding sequence is ligated thereto so that expression of the protein coding sequence is achieved under conditions compatible with the transcriptional activity of the control sequences.

[0077] The term "host cell" means a cell that has been transformed, or is capable of being transformed, with a nucleic acid sequence and thereby expresses a gene of interest. The term includes the progeny of the parent cell, whether or not the progeny is identical in morphology or in genetic make-up to the original parent cell, so long as the gene of interest is present.

[0078] The term "naturally occurring" as used herein refers to materials which are found in nature or a form of the materials that is found in nature.

[0079] The term "treat" and "treatment" includes therapeutic treatments, prophylactic treatments, and applications in which one reduces the risk that a subject will develop a disorder or other risk factor. Treatment does not require the complete curing of a disorder and encompasses embodiments in which one reduces symptoms or underlying risk factors. [0080] The term "prevent" does not require the 100% elimination of the possibility of an event. Rather, it denotes that the likelihood of the occurrence of the event has been reduced in the presence of the compound or method.

[0081] Standard techniques can be used for recombinant DNA, oligonucleotide synthesis, and tissue culture and transformation (e.g., electroporation, lipofection). Enzymatic reactions and purification techniques can be performed according to manufacturer's specifications or as commonly accomplished in the art or as described herein. The foregoing techniques and procedures can be generally performed according to conventional methods well known in the art and as described in various general and more specific references that are cited and discussed throughout the present specification. See, e.g., Sambrook et ai, Molecular Cloning: A Laboratory Manual (2d ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1989)), which is incorporated herein by reference for any purpose. Unless specific definitions are provided, the nomenclatures utilized in connection with, and the laboratory procedures and techniques of, analytical chemistry, synthetic organic chemistry, and medicinal and pharmaceutical chemistry described herein are those well known and commonly used in the art. Standard techniques can be used for chemical syntheses, chemical analyses, pharmaceutical preparation, formulation, and delivery, and treatment of patients

[0082] As described herein, SEQ ID NO: 4 - rAAV-cap-in-cis plasmid may also be referred to as: AAV-cap-in-cis, rAAV-Cap-in-cis vector, rAAV-CAP-in-cis genome, rAAV-Cap-in-cis construct, rAAV9-cap-in-cis, rAAV9R-X/A-cap-in-cis, or cap-in-cis, rAAV mCherry-cap-lox71/66 genome, rAAV-CAP-in-cis-lox, AAV9 cap-in-cis genome, rAAV-cap-in-cis rAAV genome, or the cap-in-cis rAAV genome.

[0083] As described herein, SEQ ID NO:7 (for example)- rAAV-delta-cap-in-cis may also be referred to as: rAAV9R-delta-X/A-cap-in-cis, rAAV9R-delta-X/A-cap-in-cis vector, rAAV-Acap-in-cis vector, rAAV-cap-in-cis acceptor vector, cap-in-cis acceptor construct, rAAV9R-delta-X/A-cap-in-cis acceptor construct, rAAV-cap-in-cis library acceptor, or AAV9R-delta-X/A-cap-in-cis.

[0084] As described herein, SEQ ID NO: 5 (for example) - AAV Rep-AAP helper may also be referred to as the Rep-AAP, rep-AAP helper and REP-AAP helper, AAV REP- AAP helper, AAV2/9 rep-AAP, or AAV2/9 REP-AAP helper plasmid. [0085] As described herein, SEQ ID NO: 6 (for example)- pCRII-9R-X/A EK plasmid or pCRII-9Cap-xE are interchangeable terms.

[0086] As described herein, AAV-PHP.B denotes the same thing as AAV-PHP.b, which denotes the same things as AAV-G2B26.

[0087] As described herein, AAV-PHP.A denotes the same things as AAV-

PHP.a.

[0088] As described herein, AAV-PHP.R2 denotes the same thing as AAV- PHP.r, which denotes the same thing as AAV-THl.1-35.

[0089] As described herein, 1253 is also referred to as 9CapF; 1316 is also referred to as CDF; 1331 is also referred to as XF; 1312 is also referred to as AR; 1287 is also referred to as 7xNNK; and 1286 is also referred to as 7xMNN.

[0090] The term "central nervous system" or "CNS" as used herein refers to the art recognized use of the term. The CNS includes the brain, optic nerves, cranial nerves, and spinal cord. The CNS also includes the cerebrospinal fluid, which fills the ventricles of the brain and the central canal of the spinal cord.

[0091] Systemic administration of vectors including a capsid protein that includes a targeting protein of SEQ ID NO: l of the are particularly suitable for delivering exogenous DNA sequences encoding polypeptides, proteins, or non-coding DNA, RNA, or oligonucleotides to, for example, cells of the CNS of subjects afflicted by a CNS disease.

Targeting Sequence:

[0092] In some embodiments, a central nervous system targeting peptide is provided. In some embodiments, the peptide comprises an amino acid sequence of SEQ ID NO: 1. In some embodiments, the peptide is further conjugated to a nanoparticle, a second molecule, a viral capsid protein, or inserted between amino acids 588 and 589 of AAV9 (SEQ ID NO: 2, FIG. 30).

[0093] In some embodiments, the central nervous system targeting peptide includes 4 or more amino acids of residues that overlap with residues 585 to 595 within SEQ ID NO: 8.

[0094] In some embodiments, the central nervous system targeting peptide includes 4 or more contiguous amino acids of SEQ ID NO: 1 (or any of the sequences in FIG. 31). In some embodiments, the central nervous system targeting peptide comprises 4-7 amino acids of SEQ ID NO: 1 (or any of the sequences in FIG. 31). In some embodiments, the central nervous system targeting peptide comprises 4-6 amino acids of SEQ ID NO: 1 (or any of the sequences in FIG. 31). In some embodiments, the central nervous system targeting peptide includes one or more of the following options: TLAVPFK (SEQ ID NO: 1); LAVPFK (SEQ ID NO: 31); AVPFK(SEQ ID NO: 32); VPFK(SEQ ID NO: 33); TLAVPF(SEQ ID NO: 34); TLA VP (SEQ ID NO: 35); or TLAV(SEQ ID NO: 36). In some embodiments, the targeting peptide can consist of, consist essentially of, or comprise one or more of the sequences in FIG. 31. In some embodiments, 2 or fewer amino acids can be altered within TLAVPFK (or for any of the sequences within FIG. 31). In some embodiments, one amino acid can be altered within TLAVPFK (or for any of the sequences within FIG. 31). In some embodiments, the alteration is a conservative alteration (within any of the targeting peptides provided herein). In some embodiments, the alteration is a deletion or insertion of one or two amino acids (within any of the targeting peptides provided herein). In some embodiments, the amino acid can include a non-natural amino acid. In some embodiments, the central nervous system targeting peptide sequence can be one or more of: SQTLA, QTLAV, TLA VP, LAVPK, AVPKA, VPKAQ. In some embodiments, the targeting peptide can be at least 75% identical to one or more of the above sequences, for example, at least 80% identical.

[0095] In some embodiments, the central nervous system targeting peptide sequence can be inverted, such as in KFPVALT (SEQ ID NO: 3) (similarly, any of the sequences in FIG. 31 can also be inverted). In such embodiments, 4 or more contiguous amino acids can be employed. For example, the sequence can comprise and/or consist of KFPV, FPVA, PVAL, VALT, etc. In some embodiments, the targeting peptide can be at least 75% identical to one or more of the above sequences, for example, at least 80%.

[0096] In some embodiments, the central nervous system targeting peptide comprises an amino acid sequence that comprises at least 4 contiguous amino acids from the sequence TLAVPFK (SEQ ID NO: 1) or KFPVALT (SEQ ID NO: 3) or any of the sequences in FIG. 31. In some embodiments, the amino acid sequence results in an increase in CNS cell transduction by the AAV. In some embodiments, the amino acid sequence is part of a capsid protein of the AAV vector. In some embodiments, the sequence TLAVPFK (SEQ ID NO: 1 ; (or any of the sequences in FIG. 31)) is inserted between AA588-589 of an AAV sequence of the vector (SEQ ID NO: 2). In some embodiments, the sequence TLAVPFK (SEQ ID NO: 1; or or any of the sequences in FIG. 31) is inserted between AA586-592 of an AAV sequence of the vector (SEQ ID NO: 2). In some embodiments, the sequence TLAVPFK (SEQ ID NO: 1 ; or any of the sequences in FIG. 31) further comprises at least two of amino acids 587, 588, 589, or 590 of SEQ ID NO: 2. In some embodiments, the targeting peptide can be at least 75% identical to one or more of the above sequences.

[0097] In some embodiments, the central nervous system targeting peptide comprises, consists, or consists essentially of any one or more of the above sequences. In some embodiments, the central nervous system targeting peptide is inserted into a longer peptide, as described herein.

[0098] In some embodiments, the targeting peptide is part of an AAV, as described herein. In some embodiments, the targeting peptide is part of an AAV9. In some embodiments, the targeting peptide can be linked to any molecule that should be targeted as desired. In some embodiments, the targeting peptide can be linked, without limitation, to a recombinant protein, antibody, a cell, a diagnostic, a therapeutic, a nanomolecule, etc.

[0099] In some embodiments, the targeting sequence is an amino acid sequence. In some embodiments, the targeting sequence is a nucleic acid sequence. In some embodiments, the targeting sequence is a capsid protein comprising an amino acid sequence that comprises at least 4 contiguous amino acids from the sequence TLAVPFK (SEQ ID NO: 1) and/or KFPVALT (SEQ ID NO: 3) and/or any of the sequences in FIG. 31.

[0100] Some embodiments of options of targeting sequences, as outlined in the examples below, are provided in FIG. 1A, which includes G2B-13, G2B-26, THl.1-32, and THl.1-35, by way of example.

[0101] In some embodiments, the targeting protein can be inserted into any desired section of a protein. In some embodiments, the targeting protein can be inserted into a capsid protein. In some embodiments, the targeting protein is inserted on a surface of the desired protein. In some embodiments, the targeting protein is inserted into the primary sequence of the protein. In some embodiments, the targeting protein is linked to the protein. In some embodiments, the targeting protein is covalently linked to the protein. In some embodiments, the targeting protein is inserted into an unstructured loop of the desired protein. In some embodiments, the unstructured loop can be one identified via a structural model of the protein.

[0102] In some embodiments, the unstructured loop can be one identified by sequence comparisons, such as shown in FIG. 13. FIG. 13 shows an alignment of VP1 capsid amino sequence from AAV and related parvoviruses aligned to AAV9. Sequence identity is shown as a dot. The AAs that differ from AAV9 are indicated. The numbering is based on AAV9 VP1. Only AA 418-624 are shown, although such an alignment can be done by one of skill in the art for any desired section of protein. Shaded vertical bars of different length represent the relative conservation at each AA. Longer bars indicate greater conservation. Horizontal shaded bars indicate sites of the unstructured loops into which the targeting protein can be inserted.

[0103] In some embodiments, the location of insertion of the targeting protein into the desired protein can be achieved by a structural model. An example of such a structural model is shown in FIG. 14. FIG. 14 depicts a structural model highlighting surface loops randomized by targeted sequence insertion. The insert (which is a blow up) depicts a ribbon diagram of AAV9 surface model constructed in PyMol from the AAV9 Protein Data Bank file 3uxl.pdb. The capsid surface is shown in gray and the loop regions chosen for sequence insertion are highlighted by shading (AA586-592) and (AA452-458). Other regions of sequence insertion or replacement can be identified from within regions that are not highly conserved. Additional examples include the regions of AAV9 between AA262- 269, AA464-473, AA491-495, AA546-557 and AA659-668 or the homologous regions of other the capsid proteins from other AAVs or parvoviruses.

[0104] In some embodiments, the capsid protein can comprise or consist of the sequence shown in FIG. 19, SEQ ID NO: 8. The underlined amino acid is a K to R mutation that was made to provide a unique Xbal restriction site, this can be optional. For references, SEQ ID NO: 1 (TLAVPFK) is in bold text. Any of the other targeting peptides provided herein (for example, in FIG. 31) can also be inserted in place of SEQ ID NO: 1. FIG. 20 depicts some embodiments of a nucleic acid sequence for an AAV-PHP.B (AAV-G2B26) capsid gene coding sequence). The recovered nucleic acid sequence encoding SEQ ID NO: 1 (TLAVPFK) is in bold and underlined text. The mutations introduced to insert or remove restriction sites are highlighted with double underlined italicized text. Vectors

[0105] In some embodiments, a viral vector can include one or more of the noted targeting sequences (for example, any of the central nervous system targeting peptides noted herein or any peptide provided by the screening methods provided herein). In some embodiments, an AAV vector can be provided that comprises a sequence TLAVPFK (SEQ ID NO: 1) (or any of the other targeting proteins provided herein, including those in FIG. 31).

[0106] In some embodiments, one or more targeting sequences can be employed in a single system. For example one can employ one or more targeting sequences and also modify other sites to reduce the recognition of the AAVs by the pre-existing antibodies present in the host, such as a human. In some embodiments, the AAV vector can include a capsid, which influences the tropism/targeting, speed of expression and possible immune response. The vector can also include the rAAV, which genome carries the transgene/therapeutic aspects (e.g., sequences) along with regulatory sequences. In some embodiments, the vector can include the targeting sequence within/on a substrate that is or transports the desired molecule (therapeutic molecule, diagnostic molecule, etc.).

[0107] In some embodiments, any one or more of the targeting sequences provided herein can be incorporated into a vector.

[0108] In some embodiments, the sequence TLAVPFK (SEQ ID NO: 1) (or any of the other targeting proteins provided herein, including those in FIG. 31) results in an increase in CNS cell transduction from a virus containing the vector. In some embodiments, the increase is at least 2, 10, 20, 30, 40, 50, 60, 70, 80, 90, or at least 100 fold more than transduction without the targeting sequence. In some embodiments, there is a 40-90 fold increase in transduction of the CNS, as compared with AAV9 transduction.

[0109] In some embodiments, the sequence TLAVPFK (SEQ ID NO: 1) (or any of the other targeting proteins provided herein, including those in FIG. 31) is part of a capsid protein of the AAV vector. In some embodiments, the sequence TLAVPFK (SEQ ID NO: 1) (or any of the other targeting proteins provided herein, including those in FIG. 31) is inserted between AA588-589 of an AAV sequence of the vector (SEQ ID NO: 2). In some embodiments, the sequence TLAVPFK (SEQ ID NO: l) (or any of the other targeting proteins provided herein, including those in FIG. 31) is inserted within AA452-458 of an AAV sequence of the vector (SEQ ID NO: 2). In some embodiments, the sequence TLAVPFK (SEQ ID NO: l) (or any of the other targeting proteins provided herein, including those in FIG. 31) is inserted within AA491-495 of an AAV sequence of the vector (SEQ ID NO: 2). In some embodiments, the sequence TLAVPFK (SEQ ID NO: l) (or any of the other targeting proteins provided herein, including those in FIG. 31) is inserted within AA546-557 of an AAV sequence of the vector (SEQ ID NO: 2). In some embodiments, any of the targeting sequences (or combination thereof) in FIG. 31 can be used and/or substituted for any of the embodiments provided herein regarding SEQ ID NO: 1. Thus, for example, one or more of the sequences within FIG. 31 can be inserted between AA588-589 of an AAV sequence of the vector (SEQ ID NO: 2). In some embodiments, the targeting sequence can be one or more of: SVSKPFL (SEQ ID NO: 28); FTLTTPK (SEQ ID NO: 29); or MNATKNV (SEQ ID NO: 30). FIG. 31 depicts some of the most highly enriched sequences recovered from the second round of selection for AAV variants that transduce GFAP-Cre+ astrocytes following intravenous administration.

[0110] In some embodiments, the targeting sequence that is part of the vector can comprise any four contiguous AAs within AAV9 VP1 AA585-598 of SEQ ID NO: 8.

[0111] While the numbering is not identical between serotypes, the exact insertion site is not critical. In some embodiments, the targeting sequence is inserted within the unstructured (see FIG. 14) and poorly conserved (see alignment, FIG. 13) surface exposed loops. In some embodiments, the insertion of the targeting sequence can be achieved within other AAV capsids by inserting the targeting sequence within the homologous unstructured loops of other AAV sequences.

[0112] In some embodiments, an rAAV genome is provided. The genome can comprise at least one inverted terminal repeat configured to allow packaging into a vector and a cap gene. In some embodiments, it can further include a sequence within a rep gene required for expression and splicing of the cap gene. In some embodiments, the genome can further include a sequence capable of expressing VP3.

[0113] In some embodiments, the only protein that is expressed is VP3 (the smallest of the capsid structural proteins that makes up most of the assembled capsid - the assembled capsid is composed of 60 units of VP proteins, -50 of which are VP3). In some embodiments, VP3 expression alone is adequate to allow the method of screening to be adequate.

[0114] In some embodiments, the system for screening involves placing the selectable element, (which in some embodiments can be the AAV cap gene into the AAV genome) together with one or more recombinase recognition sites (loxP or mutant loxP sites are preferred, but others could be used). In some embodiments, the AAV genome can be defined by a nucleic acid comprising at least one inverted terminal repeat.

[0115] In some embodiments, the rAAV genome further comprises a mCherry reporter cassette comprising a ubiquitin C gene, a mCherry cDNA, and a minimal synthetic polyA sequence.

[0116] In some embodiments, the genome further comprises cre-dependent switch comprising: a polyA sequence and a pair of inverted loxP sites flanking the polyA sequence. In some embodiments, the polyA sequence is downstream of the cap gene. In some embodiments, the pair of inverted loxP sites comprises lox71 and lox66. In some embodiments, the genome contains only those sequences within the rep gene required for expression and splicing of the cap gene product.

[0117] In some embodiments, AAV-PHP.B delivers genes efficiently to one or more organs including, but not limited to the central nervous system, liver, muscle, heart, lungs, stomach, adrenal gland, adipose and intestine.

[0118] In some embodiments, a capsid library is provided that comprises AAV genomes that contain both the full rep and cap sequence that have been modified so as to not prevent the replication of the virus under conditions in which it could normally replicate (co- infection of a mammalian cell along with a helper virus such as adenovirus). A pseudo wt genome can be one that has an engineered cap gene within a "wt" AAV genome.

[0119] In some embodiments, the capsid library is made within a "pseudo-wild type" AAV genome containing the viral replication gene (rep) and capsid gene (cap) flanked by inverted terminal repeats (ITRs). In some embodiments, the capsid library is not made within a "pseudo-wild type" AAV genome containing the viral replication gene (rep) and capsid gene (cap) flanked by inverted terminal repeats (ITRs). [0120] In some embodiments, the rAAV genome contains the cap gene and only those sequences within the rep gene required for the expression and splicing of the cap gene products (FIG. 22B).

[0121] In some embodiments, a capsid gene recombinase recognition sequence is provided with inverted terminal repeats flanking these sequences.

[0122] In some embodiments, the system could be used to develop capsids that exhibit enhanced targeting of specific cells/organs, select for capsids that evade immunity, select for genomes that are more at homologous recombination, select for genome elements that increase the efficiency of conversion of the single stranded AAV genome to a double stranded DNA genome within a cell and/or select for genome elements that increase the conversion of AAV genome to a persistent, circularized form within the cell.

Nucleic Acid Sequences

[0123] In some embodiments, a nucleic acid sequence encoding any of the targeting sequences provided herein is provided. In some embodiments, the nucleic acid sequence is AAGTTTCCTGTGGCGTTGACT FOR SEQ ID NO 3). ACT TTG GCG GTG CCT TTT AAG (SEQ ID NO:49) for a sequence encoding the AA sequence of SEQ ID NO: 1. In some embodiments, the nucleic acid sequence is one that will hybridize to this sequence under stringent conditions. In some embodiments, the nucleic acid sequence includes a nucleic acid sequence that encodes for SEQ ID NOS: 1 and/or 3 and the sequence is part of a larger nucleic acid sequence. In some embodiments, any one or more of the sequences from FIG. 31 can provide the noted nucleic acid sequence (that is, any nucleic acid sequence that encodes for any of these sequences can be provided). In some embodiments, the nucleic acid sequence is one that will hybridize to any of the sequences within FIG. 31 (or the sequences that encode the amino acid sequences) under stringent conditions. In some embodiments, the nucleic acid sequence includes a nucleic acid sequence that encodes for any of the sequences within FIG. 31 and the sequence is part of a larger nucleic acid sequence. In some embodiments, the nucleic acid sequence is one or more of: AGTGTGAGTAAGCCTTTTTTG (SEQ ID NO: 24); TTTACGTTGACGACGCCTAAG (SEQ ID NO: 26); or ATGAATGCTACGAAGAATGTG (SEQ ID NO: 27). [0124] In some embodiments, the nucleic acid sequence that encodes for SEQ ID NO: 1 is inserted between a sequence encoding for amino acids 588 and 589 of AAV9 (SEQ ID NO: 2).

[0125] In some embodiments, a nucleic acid sequence encoding any four contiguous amino acids in TLAVPFK (SEQ ID NO: 1) or in KFPVALT (SEQ ID NO: 3) is provided. In some embodiments, a nucleic acid sequence encoding any five contiguous amino acids in TLAVPFK (SEQ ID NO: 1) or in KFPVALT (SEQ ID NO: 3) is provided. In some embodiments, a nucleic acid sequence encoding any six contiguous amino acids in TLAVPFK (SEQ ID NO: 1) or in KFPVALT (SEQ ID NO: 3) (or any of the other targeting proteins provided herein, including those in FIG. 31) is provided.

[0126] In some embodiments, the nucleic acid sequence is inserted between a sequence encoding for amino acids 588 and 589 of AAV9 (SEQ ID NO: 2).

[0127] In some embodiments, a plasmid system is provided. The plasmid can include a first plasmid comprising a modified AAV2/9 rep-cap helper plasmid comprising at least one in frame stop codon within its VP1, VP2 and VP3 reading frame. The stop codon is positioned to disrupt VP expression without altering the amino acid sequence of the assembly activating protein. The plasmid system can further include a second plasmid comprising a rAAV-cap-in-cis plasmid.

[0128] In some embodiments, the method does not involve expressing single VP proteins from heterologous plasmids to generate "mosaic" capsids made from VP proteins encoded by different plasmids.

[0129] In some embodiments, a library of nucleic acid sequences is provided. The library can comprise a selectable element and one or more recombinase recognition sequences. In some embodiments, the nucleic acid sequences and one or more recombinase recognition sequences are incorporated within a virus genome. In some embodiments, the viral genome is an AAV genome. In some embodiments, the selectable element encodes an AAV capsid. In some embodiments, the selectable element is a genetic element that increases conversion to dsDNA. In some embodiments, the selectable element increases the efficiency of homologous recombination between the element and the endogenous genome. In some embodiments, the recombinase recognition sequences are comprised of one or more loxP sites. In some embodiments, the loxP site is a lox71 site and an inverted lox66 site. [0130] In some embodiments, the gene encoding the targeting protein and/or the capsid can be cloned into an AAV Rep-Cap helper plasmid in place of the existing capsid gene. When introduced together into producer cells, this plasmid can be used to package an rAAV genome into the targeting protein and/or capsid. Producer cells can be any cell type possessing the genes necessary to promote AAV genome replication, capsid assembly and packaging. Preferred producer cells are 293 cells, or derivatives, HELA cells or insect cells together with helper virus or a second plasmid encoding the helper virus genes known to promote rAAV genome replication. In some embodiments, an AAV rep-cap helper sequence can be modified to introduce a tetracycline-inducible expression system in between the rep and the cap gene to increase capsid expression and virus production. In some embodiments, a tetracycline transactivator cDNA, poly adenylation sequence, tetracycline responsive element and AAV5 p41 promoter and AAV2 splicing regulatory elements contained within the AAV2 rep gene are inserted between the rep gene and the gene encoding the capsid or targeting protein. Use of this inducible rep-cap plasmid when making rAAV provides 1.5-2- fold more virus than the AAV2/9 rep-cap plasmid. Some embodiments of such a nucleic acid cloned into a plasmid are depicted in FIG. 21, SEQ ID NO: 10. The cap gene sequence is underlined in FIG. 21. Uppercase letters indicate sites where the capsid sequence differs from AAV9.

Methods of Use

[0131] In some embodiments, a method of delivering a nucleic acid sequence to a nervous system (or other desired system) is provided. The method can include providing a protein comprising any one or more of the targeting sequences provided herein. The protein can be part of a capsid of an AAV. The AAV can comprise a nucleic acid sequence to be delivered to a nervous system. One can then administer the AAV to the subject.

[0132] In some embodiments, the nucleic acid sequence to be delivered to a nervous system comprises one or more sequences that would be of some use or benefit to the nervous system and/or the local of delivery or surrounding tissue or environment. In some embodiments, it can be a nucleic acid that encodes a trophic factor, a growth factor, or other soluble factors that might be released from the transduced cells and affect the survival or function of that cell and/or surrounding cells. In some embodiments, it can be a cDNA that restores protein function to humans or animals harboring a genetic mutation(s) in that gene. In some embodiments, it can be a cDNA that encodes a protein that can be used to control or alter the activity or state of a cell. In some embodiments, it can be a cDNA that encodes a protein or a nucleic acid used for assessing the state of a cell. In some embodiments, it can be a cDNA and/or associated RNA for performing genomic engineering. In some embodiments, it can be a sequence for genome editing via homologous recombination. In some embodiments, it can be a DNA sequence encoding a therapeutic RNA. In some embodiments, it can be a shRNA or an artificial miRNA delivery system. In some embodiments, it can be a DNA sequence that influences the splicing of an endogenous gene.

[0133] In some embodiments, the resulting targeting molecules can be employed in methods and/or therapies relating to in vivo gene transfer applications to long-lived cell populations. In some embodiments, these can be applied to any rAAV-based gene therapy, including, for example: spinal muscular atrophy (SMA), amyotrophic lateral sclerosis (ALS), Parkinson's disease, Pompe disease, Huntington's disease, Alzheimer's disease, Battens disease, lysosomal storage disorders, glioblastoma multiforme, Rett syndrome, Leber's congenital amaurosis, chronic pain, stroke, spinal cord injury, traumatic brain injury and lysosomal storage disorders. In addition, rAAVs can also be employed for in vivo delivery of transgenes for non-therapeutic scientific studies such as optogenetics, gene overexpression, gene knock-down with shRNA or miRNAs, modulation of endogenous miRNAs using miRNA sponges or decoys, recombinase delivery for conditional gene deletion, conditional (recombinase-dependent) expression, or gene editing with CRISPRs, TALENs, and zinc finger nucleases.

[0134] Provided herein are methods for treating and/or preventing Huntington's disease using the methods and compositions described herein. The method of treating and/or preventing Huntington's disease can include identifying the subject(s), providing a vector for delivery of a polynucleotide to the nervous system of the subject as provided herein, administering the vector in an effective dose to the subject thereby treating and/or preventing Huntington's disease in the subject. In some embodiments, the methods for treating a subject with Huntington's disease involve compositions where the vector delivers the polynucleotide composition comprising a Zinc finger protein (ZFP) engineered to represses the transcription of the Huntingtin (HTT) gene. In some embodiments, the ZFP selectively represses the transcription of the HTT gene allele responsible for causing the Huntington's disease in the subject by binding to the CAG repeat region of the HTT gene in a CAG repeat length- dependent manner. In some embodiments, the ZNFTR selectively represses transcription of both alleles of the HTT gene.

[0135] In some embodiments, the therapeutic item to be administered to the subject comprises a short hairpin RNA (shRNA) or microRNA (miRNA) that knocks down Huntingtin expression by inducing the selective degradation of, or inhibiting translation from, RNA molecules transcribed from the disease causing HTT allele by binding to the CAG repeat. In some embodiments, the therapeutic item to be administered to the subject comprises a short hairpin RNA (shRNA) or microRNA (miRNA) that knocks down Huntingtin expression by inducing the degradation of, or inhibiting translation from, RNA molecules transcribed from one or both alleles of the HTT gene. In some embodiments, the therapeutic item to be administered to the subject comprises a short hairpin RNA (shRNA) or microRNA (miRNA) that knocks down Huntingtin expression by inducing the selective degradation of, or inhibiting translation from, RNA molecules transcribed from the disease causing HTT allele through the selective recognition of one or more nucleotide polymorphisms present within the disease causing allele. The nucleotide polymorphisms can be used by one skilled in the art to differentiate between the normal and disease causing allele.

[0136] In some embodiments, the therapeutic item to be administered to the subject comprises a polynucleotide that encodes an RNA or protein that alters the splicing or production of the HTT RNA. In some embodiments, the therapeutic item to be administered to the subject comprises a polynucleotide that encodes one or more polypeptides and/or RNAs for genome editing using a Transcription activator-like effector nuclease (TALEN), zinc finger nuclease or clustered regularly interspaced short palindromic repeats - cas9 gene (CRISPR/Cap9) system engineered by one skilled in the art to induce a DNA nick or double- stranded DNA break within or adjacent to the HTT gene to cause an alteration in the HTT gene sequence. In some embodiments, the therapeutic item to be administered to the subject comprises a polynucleotide encoding a polypeptide that binds to a polypeptide from the HTT gene, alters the conformation of a polypeptide from the HTT gene or alters the assembly of a polypeptide from the HTT gene into aggregates or alters the half-life of a polypeptide from the HTT gene. In some embodiments, the therapeutic item to be administered to the subject comprises a polynucleotide that encodes a RNA or polypeptide that causes or prevents a post-transcriptional modification of a polypeptide from the HTT gene. In some embodiments, the therapeutic item to be administered to the subject comprises a polynucleotide that encodes a polypeptide from a chaperone protein known to those skilled in the art to influence the conformation and/or stability of a polypeptide from the HTT gene.

[0137] In some embodiments, the therapeutic item to be administered to the subject comprises regulatory elements known to one skilled in the art to influence the expression of the RNA and/or protein products encoded by the polynucleotide within desired cells of the subject.

[0138] In some embodiments, the therapeutic item to be administered to the subject comprises a therapeutic item applicable for any disease or disorder of choice. In some embodiments, this can include compositions for treating and/or preventing Alzheimers disease using the methods and compositions described herein, for example, ApoE2 or ApoE3 for Alzheimer's disease; SMN for the treatment of SMA; frataxin delivery for the treatment of Friedreich's ataxia; and/or shRNA or miRNA for the treatment of ALS.

[0139] In some embodiments, the therapeutic item for delivery is a protein (encodes a protein) or RNA based strategy for reducing synuclein aggregation for the treatment of Parkinson's. For example delivering a polynucleotide that encodes a synuclein variant that is resistant to aggregation and thus disrupts the aggregation of the endogenous synuclein.

[0140] In some embodiments, a transgene encoding a trophic factor for the treatment of AD, PD, ALS, SMA, HD can be the therapeutic item involved. In some embodiments, a trophic factor can be employed and can include, for example, BDNF, GDNF, NGF, LIF, and/or CNTF.

[0141] Dosages of a viral vector can depend primarily on factors such as the condition being treated, the age, weight and health of the patient, and may thus vary among patients. For example, a therapeutically effective human dosage of the viral vector is generally in the range of from about 0.1 ml to about 100 ml of solution containing concentrations of from about lxlO⁹ to lxlO¹⁶ genomes virus vector. A preferred human dosage can be about lxl0¹³ to lxlO¹⁶ AAV genomes. The dosage will be adjusted to balance the therapeutic benefit against any side effects and such dosages may vary depending upon the therapeutic application for which the recombinant vector is employed. The levels of expression of the transgene can be monitored to determine the frequency of dosage resulting from the vector of the invention.

[0142] In some embodiments, the polynucleotides vector also includes regulatory control elements known to one skilled in the art to influence the expression of the RNA and/or protein products encoded by the polynucleotide within desired cells of the subject.

[0143] In some embodiments, functionally, expression of the polynucleotide is at least in part controllable by the operably linked regulatory elements such that the element(s) modulates transcription of the polynucleotide, transport, processing and stability of the RNA encoded by the polynucleotide and, as appropriate, translation of the transcript. A specific example of an expression control element is a promoter, which is usually located 5' of the transcribed sequence. Another example of an expression control element is an enhancer, which can be located 5' or 3' of the transcribed sequence, or within the transcribed sequence. Another example of a regulatory element is a recognition sequence for a microRNA. Another example of a regulatory element is an intron and the splice donor and splice acceptor sequences that regulate the splicing of said intron. Another example of a regulatory element is a transcription termination signal and/or a polyadenylation sequences.

[0144] Expression control elements and promoters include those active in a particular tissue or cell type, referred to herein as a "tissue-specific expression control elements/promoters." Tissue-specific expression control elements are typically active in specific cell or tissue (for example in the liver, brain, central nervous system, spinal cord, eye, retina or lung). Expression control elements are typically active in these cells, tissues or organs because they are recognized by transcriptional activator proteins, or other regulators of transcription, that are unique to a specific cell, tissue or organ type.

[0145] Expression control elements also include ubiquitous or promiscuous promoters/enhancers which are capable of driving expression of a polynucleotide in many different cell types. Such elements include, but are not limited to the cytomegalovirus (CMV) immediate early promoter/enhancer sequences, the Rous sarcoma virus (RSV) promoter/enhancer sequences and the other viral promoters/enhancers active in a variety of mammalian cell types; promoter/enhancer sequences from ubiquitously or promiscuously expressed mammalian genes including, but not limited to, beta actin, ubiquitin or EFlalpha; or synthetic elements that are not present in nature.

[0146] Expression control elements also can confer expression in a manner that is regulatable, that is, a signal or stimuli increases or decreases expression of the operably linked polynucleotide. A regulatable element that increases expression of the operably linked polynucleotide in response to a signal or stimuli is also referred to as an "inducible element" (that is, it is induced by a signal). Particular examples include, but are not limited to, a hormone (for example, steroid) inducible promoter. A regulatable element that decreases expression of the operably linked polynucleotide in response to a signal or stimuli is referred to as a "repressible element" (that is, the signal decreases expression such that when the signal, is removed or absent, expression is increased). Typically, the amount of increase or decrease conferred by such elements is proportional to the amount of signal or stimuli present; the greater the amount of signal or stimuli, the greater the increase or decrease in expression.

[0147] Any one or more of the above aspects can be included within any of the vectors provided herein, in combination with any targeting protein.

Method of Selection

[0148] Current directed evolution protocols used to enhance AAV capsids have several shortcomings. The first is that it is difficult to design an in vivo screen that specifically recovers sequences from the target cell of interest when that target cell is one of many cell types in a complex organ. Typically, after the virus is administered in vivo, the tissue of interest is collected, and virus DNA is recovered from the DNA of the entire tissue, or region of tissue. Recently, Dalkara et al. reported the use of FACS sorting the target cells (photoreceptors) as a means to selectively recover capsid sequences present in those cells. But this method is labor intensive and costly, especially for sorting from large volumes of dissociated tissues. In addition, this additional sorting effort does not overcome the other major limitation of selecting for AAV capsid sequences: all capsid sequences present within the cell/tissue are recovered regardless of whether or not these viruses functionally transduced any cells. In other words, sequences from viruses stuck on the cell surface or viruses that entered the cell bound to a receptor that trafficked to an intracellular compartment not compatible with AAV unpackaging and transduction are recovered by these screens along with sequences that encoded capsids that successfully induced transgene expression in the target cell population. Therefore, non-functional capsids are also enriched by typical selection methods. Finally, most current methods also require the use of libraries made from replication competent AAV, which is a potential biosafety concern, especially if the virus will be introduced in animal facilities where there are primates since these viruses could replicate in animals carrying helper viruses. Herein is described an AAV capsid library screening platform that overcomes one or more of each of these limitations.

[0149] Successful production of an AAV capsid variant library depends upon each variant cap gene being packaged by the particular capsid proteins it encodes. Therefore, it is useful that the cap gene is present in cis (within the AAV genome). However, it is not essential that the non-structural rep genes be present in cis. Herein is disclosed a replication incompetent rAAV genome expressing the cap gene and in place of much of the rep sequence, several recombinant elements have been added that provide a way to selectively recover only those capsid sequences that have functionally transduced the target cell population of interest without the need for target cell isolation.

Selective recovery of AAV capsid sequences from specific Cre+ cell populations

[0150] In some embodiments, the approach incorporates a Cre recombinase- dependent switch that uses PCR (polymerase chain reaction) to selectively recover capsid sequences that have transduced Cre+ target cells. This can be accomplished by inserting mutant Cre recognition sites (lox66 and lox71 in a head-to-head orientation) into the rAAV genome around a sequence adjacent to the cap gene (FIG. IB). Cre recombination results in an inversion of the sequence flanked by the lox66 and lox71 sites, and one can then use a PCR recovery strategy that only amplifies the cap gene sequence from rAAV genomes after Cre-mediated inversion of the cap gene adjacent sequence (one PCR primer binds the invertible sequence and one binds the cap gene). Mutant loxP sites (lox66 and lox71) can be chosen so that the inversion would be less reversible (Alberts et al. 1995). FIG. IB shows an embodiments of a rAAV plasmid that has been developed.

[0151] In some embodiments, the method takes advantage of the large number of Cre transgenic mice that have been (and can be) developed. These lines express Cre under the control of cell specific promoters such that Cre is present only in a subpopulation of cells within a given organ. Hundreds of Cre transgenic lines are available from commercial vendors and academic sources, and custom lines can be generated.

[0152] In some embodiments, one can apply this for developing capsids that more efficiently transduce astrocytes in the central nervous system after IV virus administration. Transgenic mGFAP-Cre mice are available that express Cre specifically within astrocytes and neural stem cells (NSCs) in the adult brain and spinal cord. Using the Cre-dependent sequence recovery strategy, one can deliver rAAV capsid libraries in vivo, collect DNA from the entire brain and spinal cord and recover rAAV capsid sequences specifically from astrocytes and NSCs.

[0153] Another advantage of some embodiments of this approach is that this Cre dependent strategy only recovers those sequences that have transduced the target cell. AAV is a single stranded DNA virus, and its genome must enter the nucleus and be converted to double stranded DNA (dsDNA) for functional transduction. Since Cre only recombines dsDNA, only those capsid sequences that have trafficked properly to the cell nucleus and have been converted to dsDNA will be recovered.

Inclusion of a reporter gene cassette to facilitate cell sorting

[0154] For cases where Cre+ transgenics are not available, one can also incorporate a reporter cassette driven by a ubiquitous promoter/enhancer to facilitate sorting of transduced/transgene expressing cells from within a mixed population. This second option is more labor intensive than the Cre-based strategy as it requires generating single cell suspensions and FACS or magnetic bead/antibody-based sorting. But the reporter method is also powerful in that it can be combined with sorting for specific target cell populations using antibodies to known surface markers or with GFP transgenics to limit recovery to a particular population. And like the Cre strategy above, it will only lead to the recovery of sequences that are present in transduced cells. The reporter also facilitates following the transduction characteristics of the pooled library during screening (useful for both the Cre- and reporter- dependent methods).

[0155] The technology described herein can be used in conjunction with any transgenic line expressing Cre in the target cell type of interest to develop AAV capsids that more efficiently transduce that target cell population. Applications include, but are not limited to, developing capsids that are more efficient at transducing specific cell types in any organ after IV AAV administration, targeting specific populations of neurons, improving interneuronal transport, targeting tumor cells, hematopoetic stem cells, insulin producing beta cells, lung epithelium, etc. The method is not limited to any one virus delivery method. The vector may be delivered via any route including, but not limited to, oral, intravenous, intraarticular, intracardiac, intramuscular, intradermal, topical, intranasal, intraparitoneal, rectal, sublingual, subcutaneous, epidural, intracerebral, intracerebroventricular, intrathecal, intravitreal or subretinal administrations. The system can also be used to develop viruses that better cross specific barriers (blood brain barrier, gut epithelium, placenta, etc.). The method can also be used in vitro to develop capsids that are better at achieving nuclear entry and second strand synthesis (conversion to dsDNA).

[0156] In addition, this system is not limited to AAV9. Any starting AAV capsid (naturally occurring or modified variants) can be incorporated into this rAAV-cap vector, mutagenized by standard methods to create the capsid library and then screened with this Cre-dependent recovery strategy. Preferred AAV capsids include AAV1 , AAV2, AAV3, AAV3b, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAV10, human isolates (for example, hu.31 and hu.32), rhesus isolates (for example, rh.8 and rh.lO), AAVs or related parvoviruses from other primates, mammals and non-mammalian species. Furthermore, this method is not limited to any one commonly used capsid mutagenesis strategy. Any method can be used to generate the library diversity, including but not limited to capsid domain shuffling, random sequence insertion and random mutagenesis by error prone PCR. Finally, the vector has been designed to be modular making it simple to replace various elements such as the reporter cassette or capsid sequence to further customize the screening options.

[0157] To make this system possible, an option for incorporating the AAV capsid gene within the rAAV genome while providing the other non- structural AAV gene products in trans from a helper plasmid was developed. This was not necessarily straightforward since the cap gene codes for 4 proteins (the capsid proteins VP1, VP2, VP3 and the assembly activating protein (AAP)) using a combination of alternative splicing, alternative start codons and alternative reading frames. Maintaining proper regulation of the expression of these proteins is relevant for efficient virus production. It was found that the capsid proteins could be expressed from a rAAV genome when a fragment of the 3' end of the rep gene, which contains critical promoter/enhancer and splicing signals, is included (FIG. 2A). Retaining only the 3' end of the rep gene together with the capsid gene left enough space within the rAAV genome to incorporate the Cre invertible polyA sequence downstream of the cap gene as well as an mCherry (red fluorescent protein) reporter cassette.

[0158] To insure that the capsid is made entirely from the capsid gene encoded within the rAAA capsid library genome, an AAV helper plasmid that would provide the AAV non-structural proteins but not any capsid protein expression (typically rAAV are produced by supplying both rep and cap genes in trans from the AAV helper plasmid) was developed. Using a AAV2/9 RepCap plasmid vector core as a starting point, 5 stop codons were inserted within the cap gene near the translation initiation sites for the three capsid proteins VP1, VP2 and VP3 (FIG. 2B). This effectively eliminated rAAV production unless the VP 1-3 capsid proteins are provided elsewhere (such as in cis on the AAV genome cap library construct described above).

[0159] In some embodiments, a method of developing a capsid with a desired characteristic is provided. The method can comprise providing a population of rAAV genomes provided herein. The method can further involve screening the population by a specific set of criteria. The method can further involve selecting the rAVV genome that meets the screening criteria.

[0160] In some embodiments, some of the methods of screening provided herein provide at least one of the following advantages. First, in some embodiments, the method makes use of the growing library of Cre-transgenics to provide selective pressure for capsids that more efficiently transduce genetically defined cell populations (for example, see cre .j ax, gensat, creportal, connectivity .brain-map (MGI) (all ".org"). Second, in some embodiments, since Cre only recombines double stranded DNA (dsDNA) and the AAV genome is single stranded, only those capsid sequences that mediate the proper intracellular trafficking and conversion of the packaged genome to a persistent dsDNA form will be recovered. Therefore, such an approach can provide additional selective pressure for functional capsids.

[0161] As depicted in FIG. 2C, additional sequence variants can be selected based on their ability to mediated an increased association of the nucleic acid carrying the sequence library and recombinase site with the recombinase for similar or different desired applications. In some embodiments, the process can start by generating a library of DNA sequence variants (1A). In some embodiments, this can include 10^Λ2 if not more sequences (for example 10^Λ6 or more). Within the same nucleic acid, one can also incorporate one or more recombinase recognition sequences (IB). A strategy is then designed for recombination-dependent sequence recovery/amplification of the sequence variants. (IB). This can involve one or more recombinase recognition sites. One can then combine the library and recombination sites to transfer library DNA fragments into a vector with recombinase recognition sites (2). One can then deliver the library (for example, in vivo and/or in vitro). The recombinase (REC+) expression is restricted to one or more target cell populations or compartments (3). One can then apply a selected selective pressure to the system such that one can recover/amplify sequences based on the presence or absence of the recombinase-mediated recombination events on the nucleic acids comprising the library variants (4). This process can be repeated if necessary, transferring the recovered or amplified selected variants back into the library acceptor vector (5B) for 1 to 5, or more rounds of selection. One can then obtain and characterize the variant sequences (5A) by various methods, such as Sanger sequencing or next generation sequencing. Finally, one can then characterize the function of any or all of the individual variants.

[0162] A method for selectively recovering capsid sequences that have transduced specific target cell populations within complex tissue samples. The AAV genome has two genes-rep, which encodes 4 nonstructural proteins relevant for replication (rep78, rep68, rep52 and rep40) and cap, which encodes three proteins (VP1, VP2, and VP3) that form the shell, or capsid, of the virus (FIG. 2D). In addition, the cap gene also encodes an accessory protein Assembly-activating protein (AAP) that is required for capsid assembly. Capsid directed evolution methods make use of replication competent AAV so that the capsid gene is present in cis (that is, within the viral genome). However, successful production of an AAV capsid variant library depends only upon each variant cap gene being packaged by the particular capsid proteins it encodes. Therefore, while it is useful that the cap gene is present in cis, it is not essential that the nonstructural replication (rep) genes be present in cis. With this in mind, a replication-incompetent, rAAV genome expressing the cap gene and only those regions of the rep gene necessary for expression and splicing of the capsid gene (FIG. 2D) has been developed. In place of the remaining rep sequences, several recombinant elements that provide a means to selectively recover only those capsid sequences that have functionally transduced the target cell population of interest without the need for target cell isolation have been incorporated (FIG. 3A). To ensure that the proteins encoded by the cap are properly expressed, the splicing donor and acceptor sequences (and all intervening sequences) present within the AAV2 rep gene upstream of the AAV cap gene within the recombinant genome (FIG. 2E) were incorporated. A p41 promoter fragment from AAV5 was used to drive translation from the capsid gene (SEQ ID NO: 4, FIG. 15, depicting the entire plasmid: the plasmid backbone, AAV ITRs, UBC-mCherry-syn-pA, AAV5 p41 promoter-AAV2 rep splicing seq, AAV9 cap, lox71-SV40 polyA-lox66-ITR). To provide rep and AAP helper function, an AAV rep/cap helper plasmid was modified by inserting a total of 5 stop codons within the cap gene within the VPl, 2 and 3 reading frame (1 stop codon disrupts VP3, 3 disrupt VP2 and all 5 disrupt VPl - FIG. 2F, SEQ ID NO: 5, FIG. 16). These stop codons were designed such that they did not disrupt the coding sequence of the AAP protein, which is encoded within an alternative reading frame.

[0163] Selective recovery of AAV capsid sequences from specific Cre+ cell populations. To facilitate selective recovery of only those capsid sequences that encode the capsid protein that mediate transduction of a specific target cell population, a system was designed to take advantage of the large number of Cre transgenic mice, rats or other Cre transgenic organisms that have been (and can be) developed. These lines express Cre under the control of cell-specific promoters such that Cre is present only in a subpopulation of cells within a given organ. This approach incorporates a Cre recombinase-dependent "switch" that provides a means to selectively recover capsid sequences that have transduced Cre+ target cells. Cre recombination results in an inversion or deletion (depending on the configuration the lox sites used - see FIGs. 3 and 4) of a sequence within the AAV genome, and a PCR- based recovery strategy was used that only amplifies the cap gene sequence from rAAV genomes that have undergone a Cre-mediated inversion event. This strategy can also be adapted to select for AAV capsids that target cells that had previously been made to express Cre through non transgenic means (e.g., prior transduction with a Cre expressing virus), which could be useful for selection in larger species where Cre transgenes are not available.

[0164] An advantage of some of these embodiments is that this recombinase- dependent strategy only recovers those sequences that have transduced the target cell. AAV is a single stranded DNA virus, and its genome must be converted to double-stranded DNA (dsDNA) for functional transduction. Since Cre only recombines dsDNA, only those capsid sequences that have trafficked properly and have been converted to dsDNA will be recovered. This increases the selective pressure applied, which we anticipate will reduce the number of selection rounds that are necessary to develop viruses with improved properties.

[0165] This application is not limited to using Cre-lox as a recombinase/target site system. Other embodiments can include recombinases/integrases including, for example, Flp, phiC31 or Bxbl. The method can also be adapted for use with recombination- dependent, non-PCR-based recovery strategies. Furthermore, a recombinant AAV genome lacking most of the rep sequences was used to provide space for the lox switch and a reporter cassette, a cre-dependent switch could alternatively be inserted within a "replication competent" AAV genome in such a manner that it did not disrupt virus gene expression and packaging.

[0166] Recent efforts to use rAAV as a vehicle for gene therapy hold promise for its applicability as a treatment for human diseases based on genetic defects. rAAV vectors provide long-term expression of introduced genes from an episomal genome, although integration of the rAAV genome into the host chromosomes has been noted (Kaeppel 2013). An additional advantage of rAAV is its ability to perform this function in both dividing and non-dividing cell types including hepatocytes, neurons and skeletal myocytes. rAAV has been used successfully as a gene therapy vehicle to enable expression of erythropoietin in skeletal muscle of mice (Kessler et al., 1996), tyrosine hydroxylase and aromatic amino acid decarboxylase in the CNS in monkey models of Parkinson disease (Kaplitt et al., 1994) and Factor IX in skeletal muscle and liver in animal models of hemophilia. At the clinical level, the rAAV vector has been used in human clinical trials to deliver the cftr gene to cystic fibrosis patients, the Factor IX gene to hemophilia patients (Flotte, et al., 1998, Wagner et al, 1998).

[0167] Recombinant AAV is produced in vitro by introduction of gene constructs into cells known as producer cells. Some systems for production of rAAV employ three fundamental elements: 1) a gene cassette containing the gene of interest, 2) a gene cassette containing AAV rep and cap genes and 3) a source of "helper" virus genes. [0168] The first gene cassette is constructed with the gene of interest flanked by inverted terminal repeats (ITRs) from AAV. ITRs function to direct the packaging of the gene of interest into the the AAV virion. The second gene cassette contains rep and cap, AAV genes encoding proteins needed for replication and packaging of rAAV. The rep gene encodes four proteins (Rep 78, 68, 52 and 40) required for DNA replication. The cap genes encode three structural proteins (VP1, VP2, and VP3) that make up the virus capsid.

[0169] The third element is relevant because AAV-2 does not replicate on its own. Helper functions are protein products from helper DNA viruses that create a cellular environment conducive to efficient replication and packaging of rAAV. Adenovirus (Ad) has been used almost exclusively to provide helper functions for rAAV. The gene products provided by Ad are encoded by the genes El a, Elb, E2a, E4orf6, and Va.

[0170] Production of rAAV vectors for gene therapy can be carried out in vitro, using suitable producer cell lines such as 293 and HeLa. One strategy for delivering all of the required elements for rAAV production utilizes two plasmids and a helper virus. This method relies on transfection of the producer cells with plasmids containing gene cassettes encoding the necessary gene products, as well as infection of the cells with Ad to provide the helper functions. This system employs plasmids with two different gene cassettes. The first is a proviral plasmid encoding the recombinant DNA to be packaged as rAAV. The second is a plasmid encoding the rep and cap genes. To introduce these various elements into the cells, the cells are infected with Ad as well as transfected with the two plasmids. Alternatively, in more recent protocols, the Ad infection step can be replaced by transfection with an adenovirus "helper plasmid" containing the VA, E2A and E4 genes. As provided herein, the rep and cap arrangements can be in trans for the screening aspects.

[0171] While Ad has been used conventionally as the helper virus for rAAV production, it is known that other DNA viruses, such as Herpes simplex virus type 1 (HSV- 1) can be used as well. The minimal set of HSV-1 genes required for AAV-2 replication and packaging has been identified, and includes the early genes UL5, UL8, UL52 and UL29. These genes encode components of the HSV-1 core replication machinery, i.e., the helicase, primase, primase accessory proteins, and the single-stranded DNA binding protein. This rAAV helper property of HSV-1 has been utilized in the design and construction of a recombinant Herpes virus vector capable of providing helper virus gene products needed for rAAV production.

[0172] The following examples are presented as exemplary embodiments only, and are not intended to be limiting on the scope of the claims. In addition, there are further sections of various embodiments provided between some of the various examples below, as appropriate, and as indicated by the text and spacing of the document.

EXAMPLE 1

[0173] The split rep-AAP and rAAV-cap-in-cis constructs generate high titer rAAV. To test whether the split rep-AAP helper and rAAV-cap-in-cis system generates rAAV virus, a triple transfection of 293T (ATCC) cells was performed with the rep-AAP helper, rAAV mCherry-cap-lox71/66 genome and the adenoviral helper construct pHelper. Plasmids were transfected at a ratio of 2: 1:4 (0.263 ug total DNA/cm2 of plated cell surface area), respectively using linear polyethylenimine (PEI) as the transfection reagent with a N:P ratio of 25. With these constructs, one was able to generate recombinant virus with an efficiency that was equivalent to that observed with the standard AAV2/9 rep/cap helper, a rAAV2 genome expressing mCherry only and pHelper (FIG. 5A). In contrast, when the rep- AAP helper and pHelper were used together with an rAAV genome encoding mCherry, but not an AAV cap, little to no virus was generated. This confirms that capsid expression in cis was required for rAAV production.

[0174] Generating the capsid libraries: introducing short randomized sequences into surface loops of AAV9. Several strategies can be used to introduce sequence diversity into the cap gene. Examples include, but are not limited to (i) error prone PCR, which introduces mutations at a controllable rate throughout a region of the cap gene amplified by PCR, (ii) capsid domain shuffling, where libraries are generated through recombination events between fragmented capsid sequences generated from a panel of different capsid serotypes and (iii) targeted sequence modification at specific sites using primers with mixed bases, which generates stretches of randomized sequences at specific sites within the capsid. Each of these methods has advantages and disadvantages. In some embodiments, one can use targeted sequence modification strategy to replace or insert random sequences of seven amino acids (21 nucleotides) into two different surface loops. EXAMPLE 2

[0175] To generate libraries of AAV capsid variants, seven amino acids of randomized sequence was introduced into the AAV9 capsid. In one library, (452-8r) AA452- 8, (VP1 counting) was replaced by randomized sequence. In a second library, (588i) seven AA of randomized sequence was inserted after AA588 in the AAV9 capsid. Using this targeted randomization strategy the sites can be randomized together in the same library or randomized sequentially after selection at an individual site.

[0176] The library fragments were generated by PCR. AA452-458 of AAV9 were replaced with 7 random amino acids through the use of a primer containing a stretch of 21 hand-mixed bases (7 x NNK, Primer 1287). Primer 1312 was used as a reverse primer. For the 588i library, a stretch of 7AA was inserted after AA588 using a primer containing a stretch of 21 hand-mixed bases (7 x MNN, primer 1286). Primer 1331 was used as a forward primer. The PCR conditions reactions were performed using 200nM of each primer, 0.1-0.5 ng of template DNA (pCRII-9R-X/A EK plasmid, SEQ ID NO: 6, FIG. 17), 200um dNTPs, 0.5ul Q5 Hot Start, High-Fidelity DNA Polymerase (NEB), lOul 5X buffer and lOul GC enhancer provided by the manufacturer. The template plasmid contained a fragment of the AAV9 capsid gene that has been modified to have two unique restriction sites (Xbal and Agel) flanking the region that was varied (this region creates an overlap with the rAAV9R- X/A-cap-in-cis acceptor plasmid cut with the same enzymes, see FIG. 5C). In addition, the PCR template fragment was further modified to eliminate a naturally occurring Earl restriction site within the capsid gene fragment and insert a Kpnl site. The modification to remove the Earl restriction site provides a way to eliminate any "wild-type" AAV capsid vector sequence contamination from the libraries that might arise during cloning by digesting the libraries with the Earl enzyme. The Earl digestion step may not be necessary if care is taken to eliminate the possibility of wt AAV capsid sequence carry over/amplification. The insertion of the Xbal site caused a K449R mutation, but the other mutations introduced into the AAV9 sequence are silent.

[0177] To facilitate cloning of the PCR fragments comprising the capsid library sequences into a recombinant AAV genome, the rAAV-cap-in-cis plasmid was modified to introduce the same two unique restriction sites, Xbal and Agel, within the capsid sequence flanking the region that will be replaced by the PCR-based libraries (FIG. 5C). In addition, the coding region between the Xbal and Agel sites was eliminated to prevent "wt" AAV9R X/A capsid protein production from any undigested vector during library virus production (AAV9R-delta-X/A-cap-in-cis, SEQ ID NO: 7, FIG. 18).

[0178] To assemble the PCR library products into the acceptor vector, the PCR products can be digested with Xbal and Agel restriction enzymes and then ligated into the cap-in-cis acceptor construct cut with the same enzymes. Alternatively, the PCR products and the rAAV-cap-in-cis acceptor vector can be assembled using the Gibson Assembly method (Gibson et al., 2009). Enzymatic assembly of DNA molecules up to several hundred kilobases. Nature Methods, 6(5), 343-345. doi: 10.1038/nmeth.l318). In the examples presented here, the Gibson Assembly method was used to consistently assemble over 100 ng of Plasmid Safe DNase-resistant circular DNA from an assembly reaction made from 400 ng of Xbal and Agel digested, alkaline phosphatase treated AAV9R-delta-X/A-cap-in-cis vector and 67 ng of library PCR product with 30 ul of 2X Gibson Assembly Master Mix (NEB) in a total volume of 60 ul. The reactions were run at 50C for 120 minutes.

[0179] Following Gibson assembly, reaction products were digested with a Plasmid Safe (PS) DNase as directed (Epicenter), which digests linearized but not circularized DNA molecules. The assembly reactions were incubated with 1 ul (10U) of PS DNase in a reaction containing 2ul ATP and 7ul of the reaction buffer supplied by the manufacturer (Epicentre) at 37C for 20 minutes followed by a heat inactivation step at 70C for 30 minutes. This reaction typically yielded over lOOng of assembled plasmid (as defined by the measured amount of product remaining after the PS DNase reaction (measured by Qubit dsDNA Broad Range kit from Invitrogen). lOOng is sufficient to transfect 10 150mm dishes at lOng/dish. It was useful to transfect this amount of the rAAV-cap-in-cis library plasmid to minimize number of packaging cells that were transfected with multiple copies of the rAAV-cap-in-cis plasmid, which could cause the generation of mosaic capsids. Mosaic capsids (those having a capsid shell composed of more than one capsid protein variant) would only carry one capsid variant genome. Therefore, not all of the amino acids within the capsids would be encoded by the capsid gene within the packaged genome By directly transfecting the assembled DNA, rather than first transforming it into competent cells and amplifying it in bacteria, it was possible to transfect the packaging cells with a maximally diverse library (theoretically >lel0 unique sequences). Transfection of 293 cells for capsid library virus production.

[0180] 7-10 150mm dishes of near confluent 293T cells that had been seeded 16- 30 hour prior to transfection were typically transfected. In addition to the lOng of rAAV-cap- in-cis library vector, 5.7 ug of pucl8, 11.4 ug of Rep-AAP helper and 22.82 ug of pHelper (per dish) were co-transfected using PEI at a N:P ratio of 25 (see Grieger et al 2006). The transfection mix was made in phosphate buffered saline (PBS) and was incubated at room temperature for 10 minutes and then added drop wise into the media. 12-18 hours after transfection, the media on the transfected cells was exchanged for fresh DMEM supplemented with 5% FBS, IX Pen/Strep and IX non-essential amino acid mix (Invitrogen). This media was then collected 48 hours after transfection, and replaced with fresh media. At 60 hours post transfection the media and cells were collected. Virus present in the media was concentrated by precipitation by adding poly(ethylene glycol) and sodium chloride to 8% and 0.5M, respectively. The cell pellets were resuspended in 10 mM Tris, 2mM MgCl₂ and the viruses were released from the cells by 3 freeze-thaw cycles (alternating between a bath made from 100% ethanol and dry ice and a 37C water bath. After the final thaw at 37C, the lysates were treated with 50U of Benzonase for 1 hour at 37C. The virus precipitated from the media was then collected by centrifugation at 4000xg for 30 minutes at 4C. The pelleted virus from the media was resuspended in the same Tris-MgCl₂ buffer as above and then combined with the cell lysate viral stock. At this time, deoxycholine (DOC) was added to 0.5% and the virus stock was incubated at 37C for an additional 30 minutes. The virus stock was then adjusted to 500mM NaCl and incubated for a further 30 minutes before the lysate was cleared by spinning at 4000xg for 15 minutes at 25C. After spinning, the cleared viral stock lysates were purified over iodixanol (Optiprep, Sigma) step gradients (15%, 25%, 40% and 60% as described by Ayuso et al 2010). Viruses were then sterile filtered and dialyzed with Amicon Ultra 100K Centrifugal filters as directed (Invitrogen) and concentrated in PBS. Virus titers were determined by measuring the number of DNasel- resistant genome copies (GCs) using qPCR and a linearized plasmid as a control (Gray et al 2011).

[0181] The virus production was halted at 60 hours post-transfection to reduce the likelihood of secondary transduction of the producer cells by the rAAV-cap-in-cis virus that is released into the medium. [0182] Secondary transduction of cells that were successfully transfected with all of the plasmids necessary for virus production may lead to the generation of viruses from more than one capsid sequence (i.e., mosaics). Effort can be taken to minimize mosaic virus production to ensure that each capsid gene is packaged only into the physical capsid variant that it encodes.

Discussion of additional embodiments and Further Examples

[0183] Based on the results from the initial examples presented below, it is expected that this system can be used in conjunction with any transgenic line expressing a recombinase in the target cell type of interest to develop AAV capsids that more efficiently transduce that target cell population. Applications include, but are not limited to, developing capsids that are more efficient at transducing specific cell types in any organ after IV AAV administration, targeting specific populations of neurons after intraparenchymal brain injections, improving neuronal transport (anterograde or retrograde), targeting tumor cells, hematopoetic stem cells, insulin producing beta cells, lung epithelium, skeletal or cardiac muscle. Thus, the selection methods provided herein can be applied for one or more these aspects.

[0184] In addition, the approach can be used to select for viruses that target human cells in human/mouse chimeric animals (the human cells would be made to express Cre prior to in vivo delivery). This last example can be useful in the successful development of efficient vectors for gene therapy applications as there is evidence that the AAV serotypes that function best in animal models may not always function with the same efficiencies in humans (Lisowski et al 2013). Therefore, it can be advantageous to select for viruses that most efficiently transduce the target human cell population in the in vivo context of an animal model.

[0185] In addition, the method can be used in conjunction with any virus delivery method (e.g., intravenously, SC, IP, intramuscular, intranasal, i.c.v, intrathecal, oral or intracranial/intraparenchymal brain injection). In some embodiments, the vector can be delivered via any route including, but not limited to: oral, intravenous, intraarticular, intracardiac, intramuscular, intradermal, topical, intranasal, intraparitoneal, rectal, sublingual, subcutaneous, epidural, intracerebral, intracerebroventricular, intrathecal, intravitreal or subretinal administrations.

[0186] Although the examples herein have used AAV9 as a starting point, any naturally occurring or previously engineered AAV capsid could also be used as a starting point for selection using this approach. Furthermore, this method could also be useful for identifying other coding or non-coding sequences within an AAV or other viral genome that influenced transduction of recombinase expressing cells. Preferred examples include selecting for sequences within the AAV genome that increase conversion of the viral genome to dsDNA, increase the persistence of viral genomes by facilitating recombination or circularization, increase the efficiency of integration of the viral genome into a favored site(s) in the cellular genome or sequences that influence gene expression in the target cell population.

[0187] In some embodiments, provided herein is the use of CREATE (Cre Rtfcombinase-based AAV Targeted Evolution), a novel platform for the selective recovery of capsid sequences that transduce Cre⁺ target cell populations. Using CREATE, it was possible to develop several new AAV capsid variants with useful properties, including one, AAV- PHP.R2, that mediates efficient retrograde transduction within the brain as early as seven days post administration, and a second variant, AAV-PHP.B that crosses the adult mouse blood brain barrier (BBB) and transduces a variety of CNS neural cell types with an efficiency that is at least 40-fold greater than AAV9, the current standard for systemic delivery. In addition, whole animal tissue clearing using PARS-based CLARITY (Yang et al., 2014b) as a more rapid method for assessing serotype tropism at the cellular level and as a method to study individual cell morphology in the brain when combined with low-dose systemic AAV-PHP.B delivery is provided. Used together, transduction mapping in intact tissues and the Cre-based capsid selection method presented provide a novel platform that should facilitate further custom virus development.

EXAMPLE 3:

IN VIVO SELECTION USING GFAP-CRE TRANSGENIC ANIMALS [0188] AAV capsids were developed that more efficiently transduce cells within the CNS of adult mice after IV injection. For this purpose, transgenic mGFAP-Cre mice were used that express Cre specifically within astrocytes and neural stem cells (NSCs) in the adult brain and spinal cord. The capsid libraries (1.2el l GC) were delivered intravenously (IV) to mice through the retro-orbital sinus. 7-8 days later, the DNA was collected from the entire brain and spinal cord and recovered rAAV capsid sequences from GFAP-Cre+ cells (FIG. 6). Vector DNA was recovered from one hemisphere of the brain and half of the spinal cord using 4 ml of Trizol (Invitrogen). The manufacture's protocol was followed, and the aqueous, RNA-containing fraction was precipitated with isopropanol and subjected to three washes in 70% ethanol made with water (all water used for PCR recovery of capsid sequences in this protocol is treated with UV using a UV light box for 10-15 minutes prior to use). The precipitated material was then resuspended in lOmM Tris pH8.0. In addition to RNA, this fraction also contains a significant fraction of the viral genome as well as some mitochondrial DNA. To eliminate the RNA, which reduced the efficiency of the PCR-based recovery of capsid sequences, the samples were treated with 1 ul of RNase (Qiagen) overnight. Alternative strategies for selective recovery of viral genomes away from the animal's genomic DNA could also be used, e.g., the HIRT extraction protocol (Hirt 1967), sized-based gel-purification, sequences specific capture/hybridization methods or selective digestion of the mouse genomic DNA by PS DNase following digestion with a restriction enzyme that does not cut the rAAV-cap-in-cis genome.

[0189] Capsid sequence recovery was performed in a Cre-dependent manner using primers 1253 + 1316. PCR conditions were 20-28 cycles with 95C 20 sec/60C for 20 sec/72C for 30 sec using Q5 Hot Start High-fidelity DNA Polymerase. The PCR product was then diluted 1: 10 and then used as a template for a second, PCR reaction that generated the X to A fragment (using primers 1331 + 1312) that was cloned back into the rAAV9R- deltaX/A-cap-in-cis acceptor construct to generate the next round of library virus using the same methods describe above. 1 ul of the Gibson Assembly reactions was then diluted 1: 10 and transformed into Sure2 competent cells (Agilent) as directed by the manufacturer. At least 10 colonies/library were picked 12-16 hours later, DNA was isolated by miniprep kit (Qiagen) and the clones were sequenced. Alternatively, DNA from the clones can be amplified by PCR using primers 1253 and 1312 and sequenced directly, eliminating the need to perform mini plasmid DNA preps.

[0190] After the first round of selection all of the clones sequenced for both libraries were unique. Therefore, a second selection round was performed to further enrich for the most potent sequences. The assembled rAAV-cap-in-cis library regenerated after the first round of selection was used to generate a second round of virus which was then injected into a second batch of GFAP-Cre+ mice as described above. After the second round, two sequences, G2B13 and G2B26 showed evidence of enrichment (Table 2).

TABLE 2

Enriched Sequences from GFAP-Cre in vivo selection

Mouse r, , Selection

Delivery . site 7mer DNA sequence(s) 7mer AA sequence(s) total line ' rounds

clones

GFAP CAGTCGTCGCAGACGCC QSSQTPR (SEQ ID NO:

G2B-13 452-8

Cre TAGG (SEQ ID NO: 48) 54) ^{18 /0} GFAP ACTTTGGCGGTGCCTTTT

G2B-26 588 TLAVPFK (SEQ ID NO: 1) 27% Cre AAG (SEQ ID NO: 49)

ATTCTGGGGACTGGTAC ILGTGTS (452-8) (SEQ ID

TH1.1- -,_u„ intracrainial , , 452-8 + TTCG (SEQ ID NO: 50) NO: 55) 18% n TH-Cre , , . , . 1+1

32 (striatum) 588 ACGCGGACTAATCCTGA TRTNPEA (588) (SEQ ID 9%

GGCT (SEQ ID NO: 51) NO: 56)

ATTCTGGGGACTGGTAC ILGTGTS (452-8) (SEQ ID

TH1.1- -,_u„ intracrainial , , 452-8 + TTCG (SEQ ID NO: 52) NO: 57) 18% n_c TH-Cre , , . , . 1+1

35 (striatum) 588 AATG G G G G G ACTAGTAG NGGTSSS (588) (SEQ ID 36%

TTCT (SEQ ID NO: 53) NO: 58)

[0191] To test the variants recovered, the sequences were cut with BsiWI and Agel and ligated into an AAV2/9R-X/A rep/cap helper (AAV2/9 rep/cap helper modified with the AAV9R-X/A capsid sequence from rAAV-cap-in-cis plasmid) also cut with BsiWI and Agel and transformed into DH5alpha competent cells (NEB). Plasmid DNA was purified using an Endofree Plasmid Maxi Kit (Qiagen). The resulting rep/cap plasmids carrying the novel variant sequences, or AAV2/9 rep/cap as a control, were then used to package a rAAV genome containing a dual eGFP-2A-luciferase reporter cassette driven by a ubiquitous CAG promoter (rAAV-CAG-eGFP-2A-Luc-WPRE-SV40pA). The novel capsids packaged the genome with efficiencies comparable with AAV9 (FIG. 7). Iel2 GC of each vector was injected IV into individual adult female C57B1/6 mice. Six days later, the mice were perfused with 4% paraformaldehyde in lOOmM phosphate buffer and the brains were examined for eGFP fluorescence.

[0192] Remarkably, transduction by the G2B26 variant was efficient enough that the native eGFP fluorescence throughout the intact brain could be seen with a IX objective on an epifluorescence microscope (FIG. 8A). At this same exposure setting, little to no eGFP fluorescence is evident in the brain from the mouse injected with AAV9. In sections prepared from a brain from a mouse injected with G2B26, transduction of neurons and glia in all regions examined in the brain and spinal cord were seen (FIGs. 8 and 9). In certain thalamic nuclei, over 90% of the NeuN+ cell bodies expressed GFP (FIG. 9G). Transduction of motor neurons in the ventral spinal cord was also robust (FIG. 9F). Numerous Sox2+ glia expressed GFP (FIG. 91). The G2B13 variant also transduced astrocytes and neurons more efficiently than AAV9, but the effect was not as dramatic as compared to the transduction by G2B26 (FIGs. 8 and 9A-C). The G2B13 variant showed strong transduction of fiber tracts in the dorsal brain stem (FIG. 9B) and spinal cord (FIG. 9C) as well as robust liver transduction (FIG. 8C). It also appears that the G2B26 variant provides more rapid onset of expression in the CNS than AAV9. Transduction by AAV9 transduction at six days post- injection was weak. Stronger expression was observed per cell and more eGFP expressing cells with the same dose of AAV9 at 21 days post injection.

[0193] Since rapid unpackaging has been proposed to be an important component of transduction efficiency and viral genome persistence (Wang et al. 2007), recovering capsid sequences soon after injection (in this case 7-8 days) may be an important component of a successful selection.

[0194] In the example above, the number of cycles can be determined empirically with the optimal number of cycles being within a range that yields more product from samples taken from Cre+ cells/animals than from samples lacking Cre+ cells. If the PCR reaction is allowed to continue past this optimal range by performing too many cycles, products may be recovered even from Cre negative samples. It can be desirable to avoid doing too many cycles.

EXAMPLE 4: IN VIVO SELECTION FOR IMPROVED RETROGRADE TRANSDUCTION

USING TH-CRE ANIMALS

[0195] In a second test of the Cre-dependent selection platform, it was asked whether one could generate novel AAV serotypes that lead to more efficient transduction of TH+ neurons in the substantia nigra (SN) compact part after virus injection into the striatum, a structure that receives axons from TH neurons. This selection scheme is designed to develop AAVs that are capable of rapid retrograde transport and transduction of TH+ and non-TH+ cells.

[0196] The same libraries were used as initially generated for the Example 3 selection and injected 0.6ul of the virus bilaterally into the striata of adult TH-Cre+ male mice using the stereotaxic coordinates 0.7mm rostral, 2.0mm lateral and 3.0mm ventral from Bregma. 10 days later, the region containing the SN was collected and isolated virus DNA from the tissue as described above. For these dissections, the mCherry reporter expressed from the rAAV-cap-in-cis genome aided in the identification of the SN (FIG. IOC) and the confirmation that the virus libraries injections had targeted the desired areas (FIG. 10B). Virus DNA was obtained from the SN-containing tissue sample using Trizol (Invitrogen) as described above and the same Cre-dependent PCR strategy was used to selectively recover those capsid sequences that led to the transduction of TH+ neurons (FIG. 10D). Using primers that amplify capsid sequences from all genomes regardless of recombination status (1253 + 1267) demonstrate that the viral sequences were also present in the Cre- controls (FIG. 10D, lower panel). Sequences recovered through the Cre-recombination dependent strategy were cloned back into the rAAV-cap-in-cis library acceptor as described above in Example 3.

[0197] Colonies were picked for sequencing. After the first round of selection, all of the tested sequences were unique, so a second round of selection was performed as described above. In addition to continuing with the libraries modified at the two individual sites, combinatorial libraries were also made by mixing all of the sequences recovered at the 452-8 replacement site with the sequences recovered at the 588 insertion site using the PCR strategy outlined in FIG. 10. Capsid virus libraries from the recovered sequences were prepared, selected again in TH-Cre mice and recovered as described above. [0198] After the second selection round in the combinatorial library, several sequences at both randomization sites showed evidence of enrichment. Several novel capsid sequences were selected to test as individual variants. The sequences were cloned into an AAV2/9R-X/A rep/cap helper using unique BsiWI and Agel sites present in both vectors. The resulting rep/cap plasmids carrying the novel variant sequences, or AAV2/9 rep/cap as a control, were then used to package a single stranded (ss) rAAV-CAG-GFP-W-pA genome. The novel capsids packaged the genome with efficiencies comparable with AAV9 (FIG. 7). 7e9 VGs (in 0.5ul) of each virus was injected individually, and bilaterally, into adult C57B1/6 mice using the same stereotaxic coordinates described above. 7 days later, mice were given an overdose of Euthasol and killed by cardiac perfusion with 4% PFA as described above. At this time, there were few if any GFP+/TH+ neurons in the SNc of mice that received an injection of the control virus, AAV9:CAG-GFP-W-SV40pA. In contrast, there were numerous GFP+/TH+ neurons present in the mice given injections of the same rAAV genome packaged into the novel clones THl.1-32 (FIG 12D) or TH 1.1 -35 (FIG 23D)

[0199] A listing of the primers used in examples 3 and 4 is provided in Table 1 below:

TABLE 1

Primer Purpose Sequence

Cre-dependent

1253 CAGGTCTTCACGGACTCAGACTATCAG (SEQ ID NO: 16)

amplification, forward

θΑΑθθΰΰΤΑΑΤΑΰποτΑΰΑΰΑΰΑΤΑΰΤΑθΑΑΰΤΑπΰΰτοΰΑΤΰΑΰΤΰ

1254 9R-X/A delta, reverse

(SEQ ID NO: 37)

0Τ0Τ0ΤΑ0ΑΑ0ΤΑπΑ0000π000π0ΑΑΑΑ00ΑΑ00ΑΑΤΑ0π0 (SEQ

1255 9R-X/A delta, forward

ID NO: 38)

Library recovery, non-

1267 GTCCAAACTCATCAATGTATCTTATCATGTCTG (SEQ ID NO: 39)

recombined, reverse

1280 VP1 stop, reverse GAGTCAATCTGGAAGTTAACCATCGGCA (SEQ ID NO: 40)

1281 VP1 stop, forward GATGGTTAACTTCCAGATTGACTCG (SEQ ID NO: 41)

1283 VP2 stop, reverse GACTACTCTACAGGCCTCTTCTATCCAG (SEQ ID NO: 42)

1284 VP2 stop, forward GATAGAAGAGGCCTGTAGAGTAGTCTCC (SEQ ID NO: 43)

1285 VP3 stop, reverse ΟΑΤΟΟΟΟΑΟΟΤΤΑΰπΑΤΙΌΤ ΐΌ (SEQ ID NO: 44)

1286 VP3 stop, forward GACAATAACTAAGGTGCCGATGGAGTGG (SEQ ID NO: 45)

Site 588

GTATTCCTTGG I I I I GAACCCAACCGGTCTGCGCCTGTGCMNNMNNMN

1286 randomization,

NMNNMNNMNNMNNTTGGGCACTCTGGTGGTTTGTG (SEQ ID NO: 23) reverse

Site 452-8

ΟΑΤΟΟΑΟΟΑΑΤΑθπΰΤΑΟΤΑΤΟΤΟΤΟΤΑΰΑΑΟΤΑΤΤΝΝΚΝΝΚΝΝΚΝΝΚΝ

1287 randomization,

NKNNKNNKCAAACGCTAAAATTCAGTGTGGCCGGA (SEQ ID NO: 22) forward Primer Purpose Sequence

Site 452-8, reverse

1312 and X/A fragment GGAAGTATTCCTTGGTTTTGAACCCA (SEQ ID NO: 19)

generation, reverse

Library recovery, Cre-

1316 dependent forward CAAGTAAAACCTCTACAAATGTGGTAAAATCG (SEQ ID NO: 17)

(reversed by '

recombination)

Site 588, forward and

1331 X/A fragment ACTCATCGACCAATA GTACTATCTCTCTAGAAC (SEQ ID NO: 18) generation, forward

Combinatorial library

generation, Earl to GTCTCTGCCGGTACOTGmGCCAAAAATTAAAGATCCA (SEQ ID NO:

Kpnl mutation

insertion, Rev

Combinatorial library

generation, Earl to GCAAACAAGGTACCGGCAGAGACAACGTGGATGCGGACA (SEQ ID NO: Kpnl mutation 47)

insertion, For

[0200] By using the platform for selection provided herein, it was possible to developed several capsids that provide enhanced, widespread gene expression in the CNS.

[0201] Notably, one capsid (AAV-PHP.R2) was capable of rapid, retrograde transport within CNS neurons after intracerebral injection, while another capsid (AAV- PHP.B) transduced cells throughout central nervous systems with 40-90-fold greater efficiency than AAV9 when delivered systemically. AAV-PHP.B transduces both neurons and glia and is therefore well suited for gene transfer to global CNS neural cell types including neurons, astrocytes and oligodendrocytes.

[0202] Given the large collection of cell type-specific Cre transgenic lines, the present capsid-selection platform is a valuable resource for customizing gene delivery vectors for biomedical applications.

EXAMPLE 5

[0203] Within the rAAV cap-in-cis recombinant genome, two elements were introduced to facilitate the selection. The first is an mCherry reporter cassette, having a 398 base pair promoter fragment from the ubiquitin C gene (UBC), the 711 bp mCherry cDNA, and a 118 bp 3 ' untranslated region containing a 51 bp synthetic poly adenylation (polyA) sequence (Levitt et al., 1989). The second, and more relevant element is a Cre-dependent "switch", having a pair of inverted, modified loxP sites (lox71 and lox66) (Araki et al., 1997) flanking a SV40 polyA sequence downstream of the cap gene. This floxed element created a Cre-invertible sequence that allows for the selective PCR amplification and recovery of only those cap sequences contained within the AAV genomes that have transduced Cre⁺ cells (FIG. 22A).

[0204] To provide rep, but not cap, gene function in trans, an AAV2/9 REP-CAP helper plasmid was modified by inserting five in frame stop codons within the reading frame for the capsid proteins, VP1, VP2 and VP3. These stop codons were designed to disrupt capsid protein expression, but not alter the amino acid sequence of the assembly activating protein (AAP), which is expressed from an alternative reading frame within the cap gene (FIG. 22B) (Sonntag et al., 2010). In this way, the modified REP-AAP helper plasmid continues to provide all of the AAV gene products in trans, save for capsid protein expression. To test whether this split rAAV-CAP-in-cis-lox and REP-AAP helper system efficiently generates rAAV, a triple transfection of HEK 293T cells was performed with the rAAV-CAP-in-cis genome, the REP-AAP helper, and the adenoviral helper plasmid, pHelper. Importantly, with these plasmids, it was possible to generate recombinant virus with an efficiency that was equivalent, if not greater than, that observed when an AAV2/9 REP-CAP helper was used to package a rAAV genome encoding mCherry (AAV-UBC- mCherry) (FIG. 22C).

[0205] In contrast, when the AAV REP-AAP helper was used to package AAV- UBC-mCherry, lacking the cap gene in cis, little to no virus was generated, confirming that capsid protein expression from the rAAV-CAP-in-cis-lox vector was required for rAAV production. Used together, the rAAV-CAP-in-cis-lox and AAV REP-AAP helper provided a novel platform, which is here below termed CREATE, for selective capsid sequence recovery from genetically defined populations of cells within complex tissue samples.

EXAMPLE 6

[0206] Two AAV9-based capsid libraries were generated by PCR using a mixed base randomization strategy. One library was made by inserting 7 amino acids of randomized sequence between AA588-9 (VP1 position) of the AAV9 capsid and another with 7 amino acids of randomized sequence replacing AA452-8 of AAV9. The cloning strategy was designed such that the recoverable PCR product would contain only the stretch of amino acids spanning the variable regions (sequences between AA450 and AA592), which encompasses a significant portion of the surface exposed amino acids, while the rest of the capsid sequence within the backbone vector remains unmodified. Library fragments were then cloned into the rAAV-delta-cap-in-cis vector and assembled products were directly transfected into packaging cells to produce virus, bypassing the primary bottleneck of library diversification, bacterial transformation. With this approach, the library diversity is limited by the number of transfected cells, rather than the number of bacterial transformants resulting in an estimated diversity of Ixl0⁷-lxl0⁸ unique sequences. Using this approach, it was possible to achieve yields of 5-lOxlO¹¹ VGs.

[0207] Vectors that mediate efficient retrograde transduction of neurons, i.e., the uptake of vector by axons and transport back to the nucleus, are desired for neuronal circuit tracing and intersectional approaches for circuit-specific gene expression, and may also have uses for clinical gene delivery. While viruses such as recombinant rabies and herpes simplex virus (HSV), exhibit highly efficient retrograde transduction and are useful for short-term circuit tracing studies, their long-term toxicity precludes their use for longitudinal experiments or experiments where their impact on cellular health would cofound (e.g. optogenetics, aging, neurodegeneration studies). For long-lasting gene expression, AAVs capable of efficient retrograde transduction would be highly valuable as they would allow the extensive tool-set available in the rAAV genome format to be applied to applications requiring retrograde transduction (NIH Brain Initative Working Group, 2013).

EXAMPLE 7

IN VIVO SELECTION FOR AAV VARIANTS WITH ENHANCED RETROGRADE TRANSDUCTION IN THE RODENT CNS

[0208] Several AAV serotypes have been shown to mediate retrograde transduction of neurons in the CNS with varying efficiencies (Aschauer et al., 2013; Castle et al., 2014a; Castle et al., 2014b; Cearley and Wolfe, 2007; Hutson et al., 2012; Low et al., 2013; Salegio et al., 2013; Samaranch et al., 2012). To develop AAV capsids with improved retrograde transduction, an in vivo selection for capsids that transduced TH⁺ dopaminergic neurons in the substantia nigra via retrograde transport from their axons within striatum (Smith and Bolam, 1990) was set up. The AAV-CAP-in-cis-lox 452-8r and 588i libraries were separately injected into the striata of adult TH-Cre⁺ mice. 10 days later the tissue surrounding the substantia nigra (SN) (FIG. 23A) was dissected and isolated viral DNA. For these dissections, the mCherry reporter expressed from the AAV-cap-in-cis library vectors aided in the identification of the SN as the SN pars reticulata (SNr) was easily identified from the mCherry^"1" axons that project to the SNr from the striatum. mCherry expression in the striatum confirmed that the virus library injections had been properly targeted (FIG. 23A). After the first round of selection, 10 clones from each library were sequenced and it was found that all of the tested sequences were unique, so a second round of selection was performed. To further diversify the libraries after the initial round of enrichment, combinatorial libraries were made by mixing all of the sequences recovered from the 452-8r library with all of the sequences recovered from the 588i library by PCR (see FIG. 27D). Viral capsid libraries from the combinatorial library were prepared and selected again in TH- Cre mice as described above. After the second selection round, several sequences at both randomization sites showed evidence of enrichment.

[0209] The most highly enriched variant, PHP.R2, was further characterized by testing it individually (see Table 3 for sequence information and enrichment data).

TABLE 3

Variant Selection Route Rounds Site{s) 7mer DNA sequence AA seq. m

PHP.R2 TH i.e. 1+1 452-8r ATTCTG G G G ACTG GTACTTCG ILGTGTS (SEQ ID NO: 55) 18%

588i (SEQ ID NO: 50) NGGTSSS (SEQ ID NO: 58) 36%

AATGGGGGGACTAGTAGTTCT

(SEQ ID NO: 53)

PHP.A GFAP i.v. 2 588i TATACTTTGTCGCAG G GTTG G YTLSQGW (SEQ ID NO: 60) 40%

(SEQ ID NO: 59)

PHP.B GFAP i.v. 2 588i ACTTTGGCGGTGCC I I I I AAG TLAVPFK (SEQ ID NO: 1 ) 27%

(SEQ ID NO: 49) [0210] Table 3 lists the AAV-PHP variants, the Cre transgenic line used to perform the library selection, the route of administration and the number of selection rounds used to enrich for the improved variants. 1+1 refers to one round of selection of the two separate libraries and then an additional round of selection of the combinatorial library. The site within AAV9 that was modified in each recovered variant is listed as is the 7mer DNA sequence(s) and amino acid sequence(s) (AA seq.) that are modified in each capsid variant. The number of occurrences of the enriched sequence as a percentage of the total number of clones sequenced is also given.

[0211] AAV-PHP.R2 was used to package a single stranded (ss) AAV-CAG-GFP genome and injected it into the striatum of adult mice. Notably, after only 7 days, robust GFP expression was observed at the striatal injection site (FIG. 23B) as well as at sites distant to the injection that are known to send projections to the striatum including the SNc (FIG. 23C and 23D), the cortex (FIG. 23E and 23F), the thalamus (FIG. 23G) and the amygdala (FIG. 23H). These results demonstrate that AAV-PHP.R2 can provide rapid retrograde transduction of several distributed neuronal populations.

EXAMPLE 8

IN VIVO SELECTION FOR AAV VARIANTS CAPABLE OF WIDESPREAD CNS TRANSDUCTION FOLLOWING SYSTEMIC ADMINISTRATION

[0212] The present example examines the development of AAV capsids that more efficiently transduce cells throughout the CNS. Several AAVs, most notably AAV9, rh.lO and rh.8, transduce CNS neurons and glia after neonatal or adult systemic, intravenous delivery (Duque et al., 2009; Foust et al., 2009; Gray et al., 2011 ; Samaranch et al., 2011 ; Yang et al., 2014a). While systemic rAAV administration with these serotypes is capable of widespread CNS delivery, the transduction efficiency is significantly reduced compared to that achievable in other organs such as liver, heart or skeletal muscle (Pulicherla et al., 2011). The present example demonstrates the use of the CREATE platform to develop capsids that more efficiently transduce the CNS globally. This was done given the important roles astrocytes play in the pathogenesis of neurodegenerative disease, together with the baseline tropism of AAV9 for astrocytes. [0213] The AA452-8r and ΑΑ588Ϊ capsid libraries described above were delivered into transgenic mGFAP-Cre mice that express Cre from the mouse GFAP promoter, which is expressed within astrocytes and neural stem cells (NSCs) in the adult brain and spinal cord (Garcia et al., 2004). lxlO¹¹ VG of each capsid library were injected into separate adult GFAP-Cre positive mice and GFAP-Cre negative mice as controls. Seven days later, virus DNA from the brains and spinal cords and recovered capsid sequences from viral genomes that had undergone Cre-mediated recombination by PCR were isolated. The recovered fragments were cloned back into the rAAV-CAP-in-cis-lox acceptor vector, and clones from each library were picked at random for sequencing. As observed in the first round of the TH-Cre selection, all of the tested sequences recovered from both libraries after the first round were unique. After the second round, a single sequence, designate as AAV- PHP.B, was identified from the 588i library and showed signs of enrichment.

[0214] To assess the AAV-PHP.B variant individually, this capsid or AAV9 was used to package a dual-GFP and firefly Luciferase reporter vector, ssAAV-CAG-GFP-2A- Luc. AAV-PHP.B packaged the recombinant genome with an efficiency similar to AAV9 (FIGs. 27A-27E). Next, lxlO¹² VG of ssAAV-CAG-GFP-2A-Luc packaged into AAV- PHP.B or AAV9 was delivered into adult mice by IV injection and assessed transduction by GFP expression three weeks later. Remarkably, this variant transduced the entire CNS with high efficiency as indicated by immunostaining for GFP (FIG.24A and 24C) and analysis of native eGFP fluorescence in several brain regions (FIGs. 24B, 24D, and 24G), the spinal cord (FIG. 24D and 24G) and retina (FIG. 24E-24F). Native GFP fluorescence remained dramatically increased over AAV9 even when 10-fold less AAV-PHP.B was delivered (FIG. 25A-right and 25D). In stark contrast with AAV9, which sparsely labels neurons and glia, individual transduced astrocytes were difficult to discern in mice that received lxlO¹² VG AAV-PHP.B, but could be seen in animals that received 10-fold less virus (FIGs. 24A-24B and FIG. 25A). AAV-PHP.B also transduced cerebellar Purkinje cells with strikingly high efficiency as demonstrated by co-localization of GFP and Calbindin immunostaining (FIG. 24C). Taking advantage of recent advances in CLARITY-based tissue clearing (Yang et al., 2014b), the native eGFP fluorescence was imaged through several hundred micron thick sections of tissue from the cortex, striatum and ventral spinal cord. These 3D renderings further demonstrate the efficiency of transduction by the AAV-PHP.B variant and confirm tissue clearing (Chung and Deisseroth, 2013; Tomer et al., 2014; Yang et al., 2014b) as a means of assessing the three-dimensional distribution of transduced cells within the brain (FIG. 24G).

[0215] To quantify CNS transduction by this variant as compared to AAV9, the number of viral genomes present in several brain regions 25 days post injection were measured. Brain and spinal cord transduction by AAV-PHP.B was between 40 and 92-fold more efficient than AAV9, depending on the region examined (FIG. 24H), while outside of the CNS, the AAV-PHP.B vector transduced several peripheral organs less efficiently than AAV9 (FIG. 241). Remarkably, in all regions other than the cerebellum, the number of viral genomes detected in the CNS in mice treated with AAV-PHP.B was similar that observed in the liver and greater than that seen in the other peripheral organs examined. In stark contrast, after AAV9 transduction, the number of AAV genomes detected within any of the CNS regions examined was at least 120-fold less than the number found in the liver. Therefore, while the tropism of AAV-PHP.B was not CNS specific, the enhanced transduction exhibited by this vector was CNS specific. In an initial selection using CREATE in GFAP-Cre mice an AAV9-based variant, AAV-PHP.A, was identified that exhibited both more efficient astrocyte transduction as well as reduced tropism for several peripheral organs (FIG. 28A- 28E). Based on these results, AAV-PHP.B appears applicable for non-invasive, CNS-wide gene transfer in the adult.

[0216] AAV9 preferentially transduces astrocytes when delivered systemically to adult animals, but it also transduces neurons in several regions (FIG.24A and FIG. 25A, Foust et al 2009, and Yang et al 2014). To examine the cell types transduced by AAV- PHP.B, the colocalization of GFP expression with proteins expressed in specific cell populations was analyzed. AAV-PHP.B transduced GFAP⁺ astrocytes (FIG. 25A), CC1⁺ oligodendrocytes (FIG. 25B), NeuN⁺ neurons (FIG. 25C and FIG. 25D) but not ΙΒΑΓ microglia (FIG. 25E).

[0217] Several cell types of clinical importance are also targeted with high efficiency including TH⁺ dopaminergic neurons in the SNc (FIG. 25F), spinal motor neurons (FIG. 34D and FIG. 24H) and striatal medium spiny neurons (FIG. 25D). In addition, several interneuron populations were also transduced (FIG. 25G-25J), although strong eGFP fluorescence was rarely found to colocalize with cells with Calretinin staining (FIG. 25 J). [0218] In sum, adult IV administration of AAV-PHP.B can be used to target, with high efficiency, numerous CNS cell types of scientific and clinical interest.

Tissue clearing for serotype tropism characterization and 3D cell phenotyping

[0219] Because CLARITY allows for the 3D imaging of cells in thick sections or intact tissue (Chung et al 2013; Yang et al. 2014) and AAV-PHP.B transduces numerous glial and neuronal cell types in the brain, whether CLARITY could be used together with a low dose of AAV-PHP.B to provide random, Golgi-like labeling of neural cells in the CNS was examined. To evaluate this approach, lxlO¹⁰ VG of AAV-PHP.B expressing GFP was delivered into adult mice by IV injection. These mice were perfusion cleared and native GFP fluorescence in the brains of the mice was imaged. Individual neurons, astrocytes and endothelial cells were visible and could be imaged through at least 400 um of cleared tissue (FIG. 26D).

[0220] This approach can be useful for studying the morphology of individual cells in normal and diseased states. This approach can be used to co-express a reporter along with any of the following examples of genetic elements to investigate the effects of said genetic element on cell morphology or connectivity in vivo: a gene encoding a protein of interest; Cre, or another recombinase, for conditional gene modification in transgenic animals harboring a floxed target allele(s); conditional, floxed, alleles to transgenic animals made to express Cre in a defined target cell population; a gene knockdown cassettes containing a suitable promoter and shRNA or miRNA, or an endogenous miRNA sponge or decoy. Given the ease of adjusting the labeling/gene modification frequency by modulating the amount of virus administered, this vector could also be used to address questions related to cell autonomy by generating genetic mosaics.

[0221] Whole animal tissue clearing using PARS-based CLARITY may also be useful for the assessment of AAV tropism at a cellular level. This is typically a labor- intensive process that requires processing, mounting and imaging individual thin (1-100 micron) sections of tissue from each organ. The potential whole animal tissue clearing to reduce this burden was explored. lxlO 1¹2" VG of ssAAV-CAG-GFP-2A-Luc packaged into AAV-PHP.B or AAV9 was delivered into adult mice by IV injection. Three weeks later, all of the tissue in the mice was cleared using the PARS-based CLARITY method described in Yang et al. (2014) and used confocal imaging and 3D image reconstruction (Imaris software, Bitplane) to assess native GFP expression as a reporter of vector transduction. In several organs, including skeletal muscle, lung, pancreas, and the liver, the mice that received AAV- PHP.B showed a reduction in the expression of GFP as compared to the mice that received an equivalent dose of AAV9 (FIG 29). The reduced GFP expression in several peripheral organs observed with AAV-PHP.B as compared with AAV9 appears consistent with the number of VGs detected for each vector in these same peripheral organs (FIG. 241). Note the GFP positive nerve fibers present in the muscle and pancreas of mice injected with AAV9 and AAV-PHP.B.

[0222] rAAV labeling combined with CLARITY will be a useful approach for studying the 3D morphologies of peripheral nerves.

[0223] The following aspects apply to the experiments outlined in Examples 7 and 8 above:

Mice

[0224] 5-week-old female C57B1/6 mice were purchased from the Jackson Labs (Maine). GFAP-Cre mice expressing Cre under the control of the mouse GFAP promoter (Garcia et al., 2004) and TH-Cre mice (Savitt et al., 2005) were from the Jackson Labs.. In vivo selection was performed in adult mice of either sex.

Plasmids

[0225] The rAAV-cap-in-cis-lox plasmid contains the following elements cloned into a vector containing AAV2 ITRs (Balazs et al. 2011). An mCherry expression cassette (398bp fragment of the human UBC gene upstream of the mCherry reporter followed by a 48bp synthetic polyA sequence - (Levitt et al., 1989) followed by the AAV9 capsid cassette. Expression of the AAV9 capsid gene was placed under the control of the p41 promoter sequence from AAV5 (1680-1974 of GenBank AF085716.1; Qiu, Nayak Pintel 2002 and Farris and Pintel 2008) and splicing sequences taken from the rep gene from the AAV9 packaging plasmid (U. Penn). A SV40 polyA sequence flanked by inverted lox71 and lox66 sites was placed downstream of the AAV9 capsid. The rAAV-cap-in-cis-lox plasmid was modified to introduce two unique restriction sites, Xbal and Agel, within the capsid sequence flanking the region that was replaced by the PCR fragment. The insertion of the Xbal site caused a K449R mutation and the mutations required to insert the Agel site were silent.

[0226] In the second iteration of the library construction (used in the second GFAP-Cre capsid selection that yielded PHP.B), two modifications were made to reduce contamination of the libraries by AAV9 or the starting AAV9R X/A capsid. First, the coding region between the Xbal and Agel sites was eliminated in the plasmid used for the capsid library cloning (rAAV-Acap-in-cis acceptor) to eliminate any potential carryover of undigested plasmid. Second, the PCR fragment covering the capsid library variable region between the Xbal and Agel sites was modified to remove a unique Earl restriction site (xE) within this region of AAV9 and insert a unique Kpnl site. The modified xE fragment was TA cloned into pCRII to generate pCRII-9Cap-xE, which served as the template for our later library PCR fragments. Eliminating the Earl site provided a secondary precaution allowing for the digestion of any contaminating AAV9 sequences recovered by PCR. It was not necessary to use this digestion step as taking standard PCR precautions including UV treating reagents and pipettors and using the rAAV-ACap-in-cis acceptor for cloning the libraries was sufficient to prevent contamination from AAV9 or AAV9R X/A.

[0227] The AAV2/9 REP-AAP helper plasmid was constructed by introducing 5 stop codons into the coding sequence of the VP reading frame of the AAV9 gene at AAs 6, 10, 142, 148, 216 (VP1 numbering). The stop codon at AA216 was designed such that it did not disrupt the coding sequence of the AAP protein, which is encoded within an alternative reading frame.

Library Generation

[0228] The 452-8r and 588i library fragments were generated by PCR using Q5 Hot Start, High-Fidelity DNA Polymerase (NEB). A schematic showing the approximate primer binding sites and the primer sequences are given in FIGs. 27A and 27C, respectively. To facilitate cloning of the PCR fragments comprising the capsid library sequences into a rAAV genome, the rAAV-cap-in-cis plasmid was modified to introduce two unique restriction sites, Xbal and Agel, within the capsid sequence flanking the region that was replaced by the PCR fragment. The insertion of the Xbal site caused a K449R mutation and the mutations required to insert the Agel site were silent. To prevent contamination of the libraries by "wild type" AAV9R X/A capsid, the coding region between the Xbal and Agel sites was eliminated from the rAAV-cap-in-cis plasmid to create rAAV-Cap-in-cis.

[0229] To generate the rAAV based library, the PCR products containing the library and the Xbal and Agel digested cap-in-cis acceptor vector were assembled using Gibson Assembly. The reaction products were then treated with PS DNase (Epicentre) to eliminate any unassembled fragments. This reaction typically yielded over lOOng of assembled plasmid (as defined by the amount of DNA remaining after a PS DNase digestion step). lOOng is sufficient to transfect 10 150mm dishes at 10 ng/dish.

Virus Production And Purification

[0230] Recombinant AAVs were generated by triple transfection of 293T cells using PEL Viral particles were harvested from the media at 72 h post transfection and from the cells and media at 120 h. Virus present in the media was concentrated by precipitation with 8% poly(ethylene glycol) and 500 mM sodium chloride (Ayuso et al 2010) and then the precipitated virus was added to the lysates prepared from the collected cells. The viruses were purified over iodixanol (Optiprep, Sigma) step gradients (15%, 25%, 40% and 60% as described by Zolotukhin et al 1999). Viruses were concentrated and formulated in PBS. Virus titers were determined by measuring the number of DNasel-resistant vector genome copies (VGs) using qPCR and the linearized genome plasmid as a control (Gray et al 2011).

[0231] For capsid library virus generation, two modifications were made to the virus production protocol to reduce the production of mosaic capsids that could arise from the presence of multiple capsid sequences in the same cell. First, only 10 ng per dish of AAV-Cap9-in-cis library vector per dish was transfected to insure that the vast majority of transfected cells only received one capsid variant sequence. Second, the virus was collected earlier (48 h and 60 hours, instead of 72 h and 120 h as above) to minimize the secondary transduction of the producer cells with the rAAV library virus released into the medium.

In Vivo Selection

[0232] For the selections in GFAP-Cre mice, lxlO¹¹ vg of the capsid libraries were injected IV (retro-orbital route) into adult Cre+ mice. Seven or eight days post- injection, mice were euthanized and the brain and spinal cord were collected. Vector DNA was recovered from one hemisphere of the brain and half of the spinal cord using 4-5 ml of Trizol (Invitrogen). For the selections in TH-Cre mice, 8xl0⁹ vg of each capsid was injected by intracranially using the stereotaxic coordinates 0.7mm rostral, 2.0mm lateral and 3.0mm ventral from bregma. 10 days later, the region containing the substantia nigra was collected and the tissue was homogenized in 1 ml of Trizol. For virus DNA isolation, the manufacture's RNA extraction protocol was followed (the upper aqueous, RNA-containing fraction collected). In addition to RNA, it was found that this fraction also contains a significant portion of the viral genome as well as some mitochondrial DNA. RNA was eliminated by treating the samples with 1 ul of RNase (Qiagen) at 37C overnight. The Cre recombination-dependent PCR strategy involved a two-step amplification strategy (FIG. 27A-27E). Sequence recovery was first performed in a Cre-dependent manner using the primers 9CAPF and CDF (FIG. 27A-27E). PCR was performed for 20-26 cycles of 95C for 20 sec, 60C for 20 sec and 72C for 30 sec using Q5 Hot Start High-fidelity DNA Polymerase. The PCR product was then diluted 1: 10-1 :100 and then used as a template for a second, PCR reaction using primer XF and AR that generated a shorter fragment that was cloned back into the rAAV-delta-cap-in-cis acceptor construct as described above, lul of the Gibson Assembly reactions was then diluted 1:10 and transformed into Sure2 competent cells (Agilent) as directed by the manufacturer to generate individual clones for sequencing.

[0233] Clones that showed evidence of enrichment were cut with BsiWI and Agel and ligated into a custom 2/9R-X/A rep/cap helper also cut with BsiWI and Agel and then transformed into DH5alpha competent cells (NEB). The resulting rep/cap plasmids carrying the novel variant sequences, or AAV2/9 rep/cap as a control, were then used to package a rAAV genome containing a dual eGFP-2A-luciferase reporter cassette driven by a ubiquitous CAG promoter (rAAV-CAG-eGFP-2A-Luc-WPRE-SV40pA) for the IV injected variants (AAV-PHP.A and AAV-PHP.B) or a similar vector lacking the Luc gene (rAAV-CAG- eGFP-WPRE-SV40pA) for the intracranial injections (AAV-PHP.R2).

Tissue Preparation And Immunostaining

[0234] Mice were anesthetized with Nembutal and transcardially perfused first with 0.1 M phosphate buffer (PB), pH 7.4 and then with freshly prepared 4% paraformaldehyde in PB. Brains were postfixed overnight and then sectioned by vibratome or cryoprotected and sectioned by cryostat. Immunostaining was performed on the floating sections by diluting primary and secondary antibodies in PBS containing 10% goat or donkey serum 0.5% Triton X-100 or no detergent (GAD67 staining). Primary antibodies used were rabbit anti-GFP (1: 1000; Invitrogen), chicken anti-GFP (1:1000; Abeam), mouse anti-CCl (1 :200; Calbiochem), rabbit anti-GFAP (1:1000; Dako), mouse anti-NeuN (l:500;Millipore), rabbit anti Ibal (1:500; Biocare Medical), mouse anti-Calbindin (l:200;Sigma), rabbit anti-Calretinin (l :1000;Chemicon), mouse anti-GAD67 (l: 1000;Millipore), mouse anti-Parvalbumin (1 :1000). Primary antibodies incubations were performed for 16-24 hours at room temperature. The sections washed and incubated with secondary antibodies conjugated to Alexa 568 (1:1000; Invitrogen) for 2-16 hours.

Tissue Clearing

[0235] Mice were perfused via peristaltic pump through the left ventricle with phosphate buffer (PB) followed by an initial perfusion with 60-80 mL of 4% PFA in PB at a flow rate of 14 mL per minute. The flow rate was then reduced to 2-3 mL per minute and continued for 2 hours at room temperature. The mice were then placed in individual custom- built perfusion chambers and perfused with 200 mL of recycling 4% acrylamide in PB at the same flow rate at room temperature overnight followed by a 2-hour perfusion flush with PB to remove residual polymers/monomers in the vasculature. The polymerization process was initiated placing the chambers in a 42°C water bath and delivering, by perfusion at the same flow rate, 200 mL of recycling, degassed 0.25% VA-044 initiator in PB. After polymerization was complete, the mice were perfused with a clearing solution of 8% SDS in 0.1M PB, pH 7.5 for 7 days. The SDS containing solution was changed two times and then flushed by the perfusion of roughly 2 L of nonrecirculating PB overnight. Tissue samples cleared of lipids were incubated in RIMS solution (Yang et al. 2014) until imaging (at least one week for optimal transparency of unsectioned mouse brain tissue). The samples were then mounted in RIMS and enclosed with a coverglass on a slide using an appropriate thickness spacer (iSpacer, SunJin Lab Co.). Images were taken with a Zeiss LSM 780 single- photon microscope. 3 dimensional image reconstructions were performed using Imaris imaging software (Bitplane). Vector Biodistribution

[0236] Mice were injected IV with lxlO¹¹ VG of a rAAV-CAG-GFP2A-Luc- WPRE-SV40-pA vector packaged into the indicated capsids. 25 days later, the mice were euthanized and tissues and indicated brain regions were collected and frozen at -80C. DNA was isolated using Qiagen DNeasy Blood and Tissue kit. Vector genomes were detected using PCR primers that bind to the WPRE element and were normalized to mouse genomes using primers specific to the mouse glucagon gene. Absolute quantification was performed by comparing unknown samples to serial dilutions of standards of known concentration.

EXAMPLE 9

METHOD OF TREATMENT EMPLOYING TARGETING PROTEINS

[0237] A subject having a disorder that can be treated by the application of a nucleic acid to be expressed within a subject is identified. The subject is then administered a first amount of a vector that includes the polynucleotide to be expressed. The polynucleotide encodes for a therapeutic protein. The vector will include a capsid protein that includes a targeting protein section that is SEQ ID NO: 1, so as to allow proper targeting of the protein to be expressed to the appropriate system within the subject. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the protein to be expressed is expressed within the subject in the appropriate system.

EXAMPLE 10

METHOD OF TREATMENT OF HUNTINGTON'S DISEASE

[0238] A subject having Huntington's disease is identified. The subject is then administered a first amount of a vector that includes the polynucleotide to be expressed. The polynucleotide encodes for a therapeutic protein. The vector will include a capsid protein that includes a targeting protein section that is SEQ ID NO: 1, so as to allow proper targeting of the protein to be expressed to the nervous system within the subject. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the protein to be expressed is expressed within the subject in the nervous system. EXAMPLE 11

METHOD OF TREATMENT OF HUNTINGTON'S DISEASE

[0239] A subject having Huntington's disease is identified. The subject is then administered a first amount of a vector that includes a polynucleotide that encodes for a small non-coding RNA (small hairpin RNA (shRNA) or microRNA (miRNA)) configured to reduce expression of the Huntingtin protein by its sequence). The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1, so as to allow proper targeting of the said polynucleotide to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the small non-coding RNA is expressed the subject in the nervous system.

EXAMPLE 12

METHOD OF TREATMENT

[0240] A subject having Huntington's disease is identified. The subject is then systemically administered a first amount of a vector that includes a polynucleotide that encodes for a Zinc finger protein (ZFP) engineered to represses the transcription of the Huntingtin (HTT) gene. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the ZFP to the nervous system, among other organs. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the ZFP is expressed the subject in the nervous system.

EXAMPLE 13

METHOD OF TREATMENT

[0241] A subject having Huntington's disease is identified. The subject is then systemically administered a first amount of a vector that includes a polynucleotide that encodes for a small non-coding RNA (small hairpin RNA (shRNA) or microRNA (miRNA)) designed by one skilled in the art to reduce expression of the Huntingtin protein. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the polynucleotide to the nervous system, among other organs. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the small non-coding RNA is expressed the subject in the nervous system.

EXAMPLE 14

METHOD OF TREATMENT

[0242] A subject having Alzheimer's disease is identified. The subject is then administered a first amount of a vector that includes a polynucleotide that encodes for an anti-Abeta antibodies or antibody fragments. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the antibody or antibody fragment to be expressed to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the antibody or antibody fragment is expressed the subject in the nervous system.

EXAMPLE 15

METHOD OF TREATMENT

[0243] A subject having Alzheimer's disease is identified. The subject is then administered a first amount of a vector that includes a polynucleotide that encodes for an apolipoprotein E (ApoE) protein, preferably the human apoE polypeptide apoE2 or modified variant of apoE2. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the antibody or antibody fragment to be expressed to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the ApoE protein is expressed the subject in the nervous system.

EXAMPLE 16

METHOD OF TREATMENT [0244] A subject having spinal muscular atrophy (SMA) is identified. The subject is then administered a first amount of a vector that includes a polynucleotide that encodes for a survival motor neuron 1 (SMN1) polypeptide. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the SMN protein to be expressed to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the SMN protein is expressed the subject in the nervous system.

EXAMPLE 17

METHOD OF TREATMENT

[0245] A subject having Friedreich's ataxia is identified. The subject is then systemically administered a first amount of a vector that includes a polynucleotide that encodes for a frataxin protein. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the frataxin protein to be expressed to the nervous system and heart, among other organs. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the frataxin protein is expressed the subject in the nervous system and heart.

EXAMPLE 18

METHOD OF TREATMENT

[0246] A subject having Pompe disease is identified. The subject is then systemically administered a first amount of a vector that includes a polynucleotide that encodes for an acid alpha-glucosidase (GAA) protein. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the GAA protein to be expressed to the nervous system and heart, among other organs. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the GAA protein is expressed the subject in the nervous system and heart. EXAMPLE 19

METHOD OF TREATMENT

[0247] A subject having Late Infantile neuronal ceroid lipofuscinosis (LINCL) is identified. The subject is then systemically administered a first amount of a vector that includes a CLN2 polynucleotide that encodes for the tripeptidyl peptidase 1 protein. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the tripeptidyl peptidase 1 protein to be expressed to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the tripeptidyl peptidase 1 protein is expressed the subject in the nervous system.

EXAMPLE 20

METHOD OF TREATMENT

[0248] A subject having the Juvenile NCL form of Batten disease is identified. The subject is then systemically administered a first amount of a vector that includes a CLN3 polynucleotide that encodes for the battenin protein. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the battenin protein to be expressed to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the battenin protein is expressed the subject in the nervous system.

EXAMPLE 21

METHOD OF TREATMENT

[0249] A subject having Canavan disease is identified. The subject is then systemically administered a first amount of a vector that includes an ASPA polynucleotide that encodes for the aspartoacylase protein. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the aspartoacylase protein to be expressed to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the aspartoacylase protein is expressed the subject in the nervous system.

EXAMPLE 22

METHOD OF TREATMENT

[0250] A subject having Parkinson's disease is identified. The subject is then systemically administered a first amount of one or more vectors that each includes one or more polynucleotide(s) that encode an enzyme(s) necessary for the increased production of dopamine from non-dopaminergic cells. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of said enzyme(s) to be expressed to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the enzyme(s) is expressed the subject in the nervous system.

EXAMPLE 23

METHOD OF TREATMENT

[0251] A subject having Parkinson's disease is identified. The subject is then systemically administered a first amount of a vector that includes a polynucleotide that encode a modified, aggregation-resistant form of alpha-synuclein protein that reduces the aggregation of endogenous alpha-synuclein. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the aggregation-resistant alpha-synuclein protein to be expressed to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the protein is expressed the subject in the nervous system. EXAMPLE 24

METHOD OF TREATMENT

[0252] A subject having amyotrophic lateral sclerosis or frontal dementia caused by a mutation in C90RF72 is identified. The subject is then administered a first amount of a vector that includes a polynucleotide that encodes a non-coding RNA(s) that reduce nuclear RNA foci caused by the hexanucleotide expansion (GGGGCC) in the subjects cells. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the RNA(s) to be expressed to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the RNA(s) is expressed the subject in the nervous system.

EXAMPLE 25

METHOD OF TREATMENT

[0253] A subject having multiple sclerosis is identified. The subject is then systemically administered a first amount of a vector that includes a polynucleotide that encode a trophic or immunomodulatory factor, for example leukemia inhibitory factor (LIF) or ciliary eurotrophic factor (CNTF). The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the said factor to be expressed to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the factor is expressed the subject in the nervous system.

EXAMPLE 26

METHOD OF TREATMENT

[0254] A subject having amyotrophic lateral sclerosis caused by SOD1 mutation is identified. The subject is then administered a first amount of a vector that includes a polynucleotide that encodes for a small non-coding RNA (small hairpin RNA (shRNA) or microRNA (miRNA)) designed by one skilled in the art to reduce expression of mutant SOD1 protein. The vector will include a capsid protein that includes a targeting protein of SEQ ID NO: 1 or any of the targeting proteins in FIG. 31, so as to allow proper targeting of the small non-coding RNA to be expressed to the nervous system. If needed, the subject is administered a second or third dose of the vector, until a therapeutically effective amount of the small non-coding RNA is expressed the subject in the nervous system.

Additional embodiments

[0255] In some embodiments, provided herein is a CREATE - Cre Recombinase- based AAV Targeted Evolution platform.

[0256] In some embodiments, provided herein is an AAV-PHP.B, which allows Z?road gene delivery to CNS neurons and glia via the vasculature.

[0257] In some embodiments, provided herein is an AAV-PHP.R2, which allows Rapid Retrograde transduction in the CNS.

Incorporation by Reference

[0258] All references cited herein, including patents, patent applications, papers, text books, and the like, and the references cited therein, to the extent that they are not already, are hereby incorporated herein by reference in their entirety. To the extent that any of the definitions or terms provided in the references incorporated by reference differ from the terms and discussion provided herein, the present terms and definitions control.

Equivalents

[0259] The foregoing written specification is considered to be sufficient to enable one skilled in the art to practice the invention. The foregoing description and examples detail certain preferred embodiments of the invention and describe the best mode contemplated by the inventors. It will be appreciated, however, that no matter how detailed the foregoing may appear in text, the invention may be practiced in many ways and the invention should be construed in accordance with the appended claims and any equivalents thereof.

Claims

WHAT IS CLAIMED IS:

1. An AAV vector comprising an amino acid sequence that comprises at least 4 contiguous amino acids from the sequence TLAVPFK (SEQ ID NO: 1) or KFPVALT (SEQ ID NO: 3).

2. The AAV vector of claim 1, wherein the amino acid sequence results in an increase in CNS cell transduction by the AAV.

3. The AAV vector of claim 1, wherein the amino acid sequence is part of a capsid protein of the AAV vector.

4. The AAV vector of claim 1, wherein the sequence TLAVPFK (SEQ ID NO: 1) is inserted between AA588-589 of an AAV sequence of the vector (SEQ ID NO: 2).

5. The AAV vector of claim 1, wherein the sequence TLAVPFK (SEQ ID NO: 1) is inserted between AA586-592 of an AAV sequence of the vector (SEQ ID NO: 2).

6. The AAV vector of claim 1, wherein the sequence TLAVPFK (SEQ ID NO: 1) further comprises at least two of amino acids 587, 588, 589, or 590 of SEQ ID NO: 2.

7. A central nervous system targeting peptide, the peptide comprising an amino acid sequence of SEQ ID NO: 1.

8. The peptide of claim 7, further conjugated to a nanoparticle, a second molecule, or a viral capsid protein.

9. The targeting peptide of claim 7, wherein the targeting peptide is part of an

AAV.

10. The targeting peptide of claim 9, wherein the targeting peptide is part of an

AAV9.

11. A nucleic acid sequence encoding any four contiguous amino acids in TLAVPFK (SEQ ID NO: 1) or in KFPVALT (SEQ ID NO: 3).

12. The nucleic acid sequence of claim 11, wherein the nucleic acid sequence is inserted between a sequence encoding for amino acids 588 and 589 of AAV9 (SEQ ID NO: 2).

13. A method of delivering a nucleic acid sequence to a nervous system, the method comprising:

providing a protein comprising TLAVPFK (SEQ ID NO: 1), wherein the protein is part of a capsid of an AAV, and wherein the AAV comprises a nucleic acid sequence to be delivered to a nervous system; and

administering the AAV to the subject.

14. The method of claim 13, wherein the nucleic acid sequence to be delivered to a nervous system comprises one or more of:

a) a trophic factor, a growth factor, or other soluble factors that might be released from the transduced cells and affect the survival or function of that cell and/or surrounding cells;

b) a cDNA that restores protein function to humans or animals harboring a genetic mutation(s) in that gene;

c) a cDNA that encodes a protein that can be used to control or alter the activity or state of a cell;

d) a cDNA that encodes a protein or a nucleic acid used for assessing the state of a cell;

e) a cDNA and/or associated guide RNA for performing genomic engineering;

f) a sequence for genome editing via homologous recombination;

g) a DNA sequence encoding a therapeutic RNA;

h) a shRNA or an artificial miRNA delivery system

i) a DNA sequence that influences the splicing of an endogenous gene.

15. An rAAV genome comprising:

at least one inverted terminal repeat configured to allow packaging into a vector; and

a cap gene.

16. The rAAV genome of claim 15, further comprising a sequence within a rep gene required for expression and splicing of the cap gene.

17. The rAAV genome of claim 15, further comprising a sequence capable of expressing at least one of: a) VPl, VP2 and VP3, b) VPl, c) VP2 d) VP3, e) VPl and VP2, f) VP2 and VP3, or g) VPl and VP3.

18. The rAAV genome of claim 15, further comprising a cre-dependent switch comprising:

a polyA sequence; and

a pair of inverted loxP sites flanking the polyA sequence.

19. The rAAV genome of claim 18, wherein the polyA sequence is downstream of the cap gene.

20. The rAAV genome of claim 18, wherein the pair of inverted loxP sites comprises lox71 and lox66.

21. The rAAV genome of claim 15, wherein the genome contains only those sequences within the rep gene required for expression and splicing of the cap gene product.

22. A plasmid system comprising:

a first plasmid comprising a modified AAV2/9 rep-cap helper plasmid configured such that it eliminates at least one of VPl, VP2, or VP3 expression; and a second plasmid comprising a rAAV-cap-in-cis plasmid.

23. A method of developing a capsid with a desired characteristic, the method comprising:

providing a population of rAAV genomes of claim 15;

selecting the population by a specific set of criteria; and

selecting the rAAV genome that meets the screening criteria.

24. A capsid protein comprising an amino acid sequence that comprises at least 4 contiguous amino acids from the sequence TLAVPFK (SEQ ID NO: 1) or KFPVALT (SEQ ID NO: 3).

25. A library of nucleic acid sequences comprising a selectable element and one or more recombinase recognition sequences.

26. The library of nucleic acid sequences of claim 25, wherein the nucleic acid sequences and one or more recombinase recognition sequences are incorporated within a virus genome.

27. A method of developing a capsid with a desired characteristic, the method comprising:

providing a library of plasmids that comprise:

a capsid gene; and

at least one recombinase recognition sequence, configured such that it allows a recombinase-dependent change in a sequence of a plasmid of the library that comprises the capsid gene that is a detectable change;

selecting the population by a specific set of criteria; and

selecting the rAAV genome that meets the screening criteria.