Treatment of dry age-related macular degeneration

Abstract

The present disclosure provides gene therapy targeting a complement pathway for the treatment of dry age-related macular degeneration.

Description

Cross Reference to Related Applications

[0001] This application claims priority to U.S. Provisional Application No. 63 / 490,736, filed March 16, 2023, and U.S. Provisional Application No. 63 / 607,419, filed December 7, 2023. The disclosure of the above priority applications is incorporated by reference in its entirety SEQUENCE LISTING

[0002] This application contains a Sequence Listing which has been submitted electronically in.XML format and is hereby incorporated by reference in its entirety. The.XML copy, created on March 12, 2024, is named “122548.AR006.xml” and is 189,331 bytes in size. The Sequence Listing contained in the.XML file is part of the specification and is hereby incorporated by reference in its entirety. BACKGROUND

[0003] Age-related macular degeneration (AMD) is the leading cause of vision loss in older adults. There are two types of AMD: dry and wet. Wet AMD, also known as advanced neovascular AMD, is a less common type of AMD that typically results in faster vision loss. On the other hand, dry AMD accounts for 85-90% of AMD cases globally (Schultz et al., Clin Ther. (2021) 43(10): 1792-818).

[0004] Dry AMD is typically accompanied by retinal pigment epithelium (RPE) dysfunction, initially in the macula of the eye, progressing to late stages with RPE cell death, followed by photoreceptor cell death, and ultimately blindness. The hallmark of the disease is the accumulation of drusen in the RPE and activation of the complement pathway. This in turn leads to a strong inflammatory response, geographic atrophy, and RPE cell and photoreceptor cell death, resulting in blindness.

[0005] Human genetic variations in multiple complement factors have been associated with changes in AMD risk and implicate dysregulation of the classical complement pathway and the alternative complement pathway as causal factors in the disease pathogenesis. Cumulative damage to the retina from aging, environmental stress, and other factors triggers inflammation in multiple pathways, including the complement cascade. When regulatory components in these pathways are affected, as with several genetic risk factors for geographic atrophy associated in the complement cascade, chronic inflammation can ultimately lead to retinal cell death characteristic of geographic atrophy / dry AMD. Complement activity and inflammation levels are increased in patients with intermediate AMD and advanced dry AMD with geographic atrophy (GA). GA is an advanced stage of dry AMD referring to areas of cell atrophy and death in the retina, leading to severe loss of bilateral central vision.

[0006] Natural immunity via the complement cascade can clear pathogens or damaged cells via phagocytosis. However, dysregulation of the complement cascade can also cause harmful inflammation. There are three pathways that initiate the complement cascade - the classical pathway, the lectin pathway, and the alternative pathway. The classical pathway is initiated by activation of the Cl complex (Clq, Clr, and Cls) upon binding to IgG or IgM immune complexes, leading to cleavage of C4 and C2, which assemble to form C4b2a, a C3 convertase. The lectin pathway is initiated, for example, by activation of the mannose-binding lectin (MBL) / MBL-associated serine protease (MASP) complex upon binding to oligosaccharides, leading to cleavage of C4 and C2, which assemble to form C4b2a. The alternative pathway is constitutively active at low levels and is initiated by hydrolysis of C3 to C3(H20), which can bind to Factor B (FB), leading to formation of the fluid-phase C3 convertase proenzyme C3(H20)B. This complex is recognized by Factor D (FD) and cleaved to form C3(H20)Bb, the fluid-phase C3 convertase.

[0007] All C3 convertases cleave C3 to an anaphylatoxin C3a and opsonin C3b. Covalently attached C3b mediates phagocytosis of opsonized cells. In addition, C3b, regardless of the initiating pathway, opsonizes through the alternative pathway to amplify the complement response. This amplification triggers activation of the terminal pathway through formation of the C5 convertase, which cleaves C5 to C5a, a potent anaphylatoxin, and C5b, a component of C5b9 or the membrane attack complex (MAC), a large pore complex that can cause cell lysis.

[0008] To date, most management guidelines focus on reducing risk factors and use of dietary supplements (Schutz et al., supra). The first treatment for GA, a C3 inhibitor (SYFOVRE TM ; pegcetacoplan injection), was recently approved, but it requires long-term, frequent intravitreal injections, which limits patient compliance and leads to an increased risk of developing neovascular AMD. Moreover, C3 inhibition does not impede complement effector functions mediated by upstream activation fragments. Another treatment for GA, a C5 inhibitor (IZERVAY TM ; avacincaptad pegol intravitreal solution), was approved by the FDA a few months after SYFOVRE TM was approved, but C5 inhibition has similar drawbacks as C3 inhibition. Therefore, there remains an urgent need to develop an effective one-time therapy for dry AMD. SUMMARY

[0009] The disclosure provides an expression construct comprising a first nucleotide sequence encoding an inhibitor of activated complement subcomponent Cls and a second nucleotide sequence encoding an inhibitor of complement factor Bb; or a pair of expression constructs, one comprising the first nucleotide sequence and the other comprising the second nucleotide sequence. Unless otherwise specified herein, activated Cls is also referred to herein as “Cl s”. Factor Bb is also referred to herein as “FBb” or simply “Bb”.

[0010] In some embodiments, the Cls inhibitor and the Bb inhibitor are each an antibody fragment, optionally wherein the antibody fragment is a single-chain Fv (scFv) or a single-chain Fab (scFab). In some embodiments, the Cls inhibitor is an anti-Cl s antibody fragment comprising heavy chain CDRs (HCDRs) 1-3 in SEQ ID NO: 7 (optionally comprising SEQ ID NOs: 1-3, respectively) and light chain CDRs (LCDRs) 1-3 in SEQ ID NO: 8 (optionally comprising SEQ ID NOs: 4-6, respectively). In some embodiments, the Bb inhibitor is an anti-Bb antibody comprising HCDRs 1-3 in SEQ ID NO: 19 (optionally comprising SEQ ID NOs: 13-15, respectively) and LCDRs 1-3 in SEQ ID NO: 20 (optionally comprising SEQ ID NOs: 16-18, respectively).

[0011] In some embodiments, the Cls inhibitor comprises a heavy chain variable domain (V H ) comprising SEQ ID NO: 7 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto and a light chain variable domain (V L ) comprising SEQ ID NO: 8 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto. In some embodiments, the Bb inhibitor comprises a V H ) comprising SEQ ID NO: 19 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto and a V L .

[0012] In some embodiments, the C1s inhibitor comprises a heavy chain (HC) comprising SEQ ID NO: 10 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto and a light chain (LC) comprising SEQ ID NO: 11 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto. In some embodiments, the Bb inhibitor comprises a HC comprising SEQ ID NO: 22 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto and a LC comprising SEQ ID NO: 23 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0013] In some embodiments, the C1s inhibitor and the Bb inhibitor each comprise one or more charge mutations to facilitate pairing between the heavy chain and the light chain of each inhibitor. In some embodiments, the charge mutations in the C1s inhibitor comprise Q42E and Q292K, where numbering is consistent with SEQ ID NO: 12. In some embodiments, the charge mutations in the Bb inhibitor comprise Q38K and Q288E, optionally further comprising S114A, N137K, and T434E, where numbering is consistent with SEQ ID NO: 24.

[0014] In some embodiments, the C1s inhibitor is an scFv or scFab, where the HC and LC are connected by a peptide linker, optionally where the peptide linker comprises one or more, optionally 2, 3, 4, 5, 6, 7, 8, 9, or 10 G4S (SEQ ID NO: 46) repeat sequences. In some embodiments, the Bb inhibitor is an scFv or scFab, where the HC and LC are connected by a peptide linker, optionally where the peptide linker comprises one or more, optionally 2, 3, 4, 5, 6, 7, 8, 9, or 10 G4S repeat sequences.

[0015] In some embodiments, the expression construct herein comprises a transgene encoding a fusion protein comprising a C1s inhibitor and a Bb inhibitor connected by a peptide linker, optionally where the peptide linker comprises one or more, optionally 2, 3, 4, 5, 6, 7, 8, 9, or 10 G4S repeat sequences. In some embodiments, the transgene is operably linked to a minimal chicken beta-actin (minCBA) promoter.

[0016] In some embodiments, the expression construct herein comprises a bidirectional promoter directing expression of a C1s inhibitor and a Bb inhibitor (as separate molecules), optionally wherein the bidirectional promoter is a pair of CBA promoters placed in opposite directions and separated by a CMV enhancer, further optionally wherein the bidirectional promoter comprises SEQ ID NO: 53 or a nucleotide sequence at least 85% (e.g., at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%) identical thereto.

[0017] In some embodiments, the expression construct expresses a heterodimer comprising (i) a fusion protein comprising a single-chain anti-C1s antibody fragment fused to a HC or LC of an anti-Bb antibody fragment; and (ii) a LC or HC polypeptide of the anti-Bb antibody fragment, wherein the coding sequence of the fusion protein and the coding sequence of the LC or HC polypeptide of the anti-Bb antibody fragment are separated in-frame by a coding sequence of a cleavable peptide, optionally wherein the cleavable peptide comprises a 2A sequence and / or a furin cleavage site, further optionally the expression construct comprises a minCBA promoter.

[0018] In some embodiments, the expression construct expresses a heterodimer comprising (i) a fusion protein comprising a single-chain anti-C1s antibody fragment fused to a HC or LC of an anti-Bb antibody fragment; and (ii) a LC or HC polypeptide of the anti-Bb antibody fragment, wherein the coding sequence of the fusion protein and the coding sequence of the LC or HC polypeptide of the anti-Bb antibody fragment are separated in-frame by a coding sequence of a cleavable peptide, optionally wherein the cleavable peptide comprises a 2A sequence and / or a furin cleavage site, further optionally the expression construct comprises a minCBA promoter.

[0019] In some embodiments, the expression construct encodes a fusion protein comprising, from N-terminus to C-terminus: (i) an anti-Cls scFv, a (G4S)2linker, and an anti-Bb scFv, optionally comprising SEQ ID NO: 55 (with or without a signal peptide) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto; (ii) an anti-Bb scFv, a (G4S)2linker, and an anti-Cls scFv, optionally comprising SEQ ID NO: 57 (with or without a signal peptide) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto; (iii) an anti-Cls scFab, a (G4S)3linker, and an anti-Bb scFab, optionally comprising SEQ ID NO: 26 or 28 (with or without a signal peptide) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto; (iv) an anti-Bb scFab, a (G4S)3linker, and an anti-Cls scFab, optionally comprising SEQ ID NO: 30 or 32 (with or without a signal peptide) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto; (v) an anti-Cls scFab, a (G4S)2linker, and an anti-Bb scFv, optionally comprising SEQ ID NO: 34 or 36 (with or without a signal peptide) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto; or (vi) an anti-Cls scFab, a (G4S)3linker, and an anti-Bb scFv, optionally comprising SEQ ID NO: 59 or 61 (with or without a signal peptide) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0020] In some embodiments, the expression construct encodes an anti-Cls scFab, optionally comprising SEQ ID NO: 12 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto, optionally wherein the amino acid sequence comprises Q42E and Q292K mutations relative to SEQ ID NO: 12; and an anti-Bb scFab, optionally comprising SEQ ID NO: 14 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto, wherein the amino acid sequence comprises Q38K and Q288E, and optionally S114A, N137K, and T434E mutations relative to SEQ ID NO: 14.

[0021] In some embodiments, the expression construct encodes a heterodimer comprising (A) (i) an anti-Cls LC and (ii) a fusion protein comprising an anti-Cls HC fused to an aBb scFab, optionally wherein the expression construct comprises the coding sequence of SEQ ID NO: 39 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto; (B) (i) an anti-Cls LC and (ii) a fusion protein comprising an anti-Cls HC fused to an anti-Bb scFab, optionally wherein the expression construct comprises the coding sequence of SEQ ID NO: 41 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto; (C) (i) a fusion protein comprising an anti-Cls scFab fused to an anti-Bb HC and (ii) an anti-Bb LC, optionally wherein the expression construct comprises the coding sequence of SEQ ID NO: 43 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto; or (D) (i) a fusion protein comprising an anti-Cls scFab fused to an anti-Bb HC and (ii) an anti-Bb LC, optionally wherein the expression construct comprises the coding sequence of SEQ ID NO: 45 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0022] In another aspect, the disclosure provides an isolated nucleic acid comprising a nucleotide sequence selected from SEQ ID NO: 25, 27, 29, 31, 33, 35, 37, 38, 40, 42, 54, 56, 58, 60, 62, 79, or 80 or encoding an amino acid sequence identical to the selected nucleotide sequence.

[0023] In another aspect, the disclosure provides one, two, or more recombinant adeno-associated viruses (rAAV) comprising an expression construct or an isolated nucleic acid herein. In some embodiments, the genome of the rAAV herein comprises an expression construct flanked by AAV2 inverted terminal repeat sequences (ITRs). In some embodiments, the genome comprises SEQ ID NO: 50, 51, or 52; or encodes an amino acid sequence identical to SEQ ID NO: 50, 51, or 52. In some embodiments, the rAAV herein comprises a capsid of AAV2 (optionally wild-type AAV2).

[0024] In an aspect, the disclosure provides a pharmaceutical composition comprising a rAAV herein and a pharmaceutically acceptable carrier.

[0025] In an aspect, the disclosure provides one or more proteins encoded by an expression construct or a rAAV herein.

[0026] In one aspect, the disclosure provides a host cell comprising an expression construct, an isolated nucleic acid, or a rAAV herein.

[0027] In one aspect, the disclosure provides a method for treating dry age-related macular degeneration (AMD) in a patient in need thereof, the method comprising administering an effective amount of a rAAV or pharmaceutical composition herein. In some embodiments, administration is by intravitreal injection. In some embodiments, the patient has geographic atrophy (GA) secondary to dry AMD. In some embodiments, the effective amount is 10 7 to 10 15 , optionally 10 8 to 10 14 , 10 9 to 10 13 , further optionally 2 x 10 9 , 2 x 10 10 , or 2 x 10 11 vector genomes.

[0028] Also provided herein are a recombinant AAV or pharmaceutical composition herein for use in treating dry age-related macular degeneration (AMD) in a patient in need thereof with a therapeutic method herein, and use of a recombinant AAV or pharmaceutical composition herein for the manufacture of a medicament for treating dry age-related macular degeneration (AMD) in a patient in need thereof with a therapeutic method herein.

[0029] In another aspect, the disclosure provides a mammalian promoter comprising a sequence having at least 85%, optionally at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%, or 100% identity to SED ID NO: 83.

[0030] In another aspect, the disclosure provides a bidirectional mammalian promoter comprising a pair of chicken beta-actin promoters placed in opposite directions, separated by a CMV enhancer, optionally wherein the bidirectional mammalian promoter comprises a sequence having at least 85%, optionally at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%, or 100% identity to SED ID NO: 53.

[0031] Other features, objects, and advantages of the application will become apparent from the following detailed description. It should be understood, however, that the detailed description, while indicating embodiments and aspects of the application, is provided by way of illustration only, and is not limiting. Various changes and modifications within the scope of the application will become apparent to those skilled in the art from the detailed description. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1A is a diagram illustrating an exemplary monocistronic construct for expressing linked anti-Cls (aC1s) and anti-Bb (aBb) antibody fragments, e.g., via the illustrated G4S linker. minCBA: minimal chicken beta-actin promoter. scFab: single chain antibody fragment. scFv: single chain antibody variable domain. BGH: bovine growth hormone.

[0033] Figure 1B is a diagram illustrating an exemplary bicistronic construct using a bidirectional promoter (modified minCBA) that allows expression of two separate antibody fragments in opposite directions.

[0034] Figure 1C is a diagram illustrating an exemplary recombinant AAV genome containing an expression cassette for expressing anti-Bb and anti-Cls antibody fragments Figure 1A or Figure 1B . ITR: inverted terminal repeat.

[0035] Figure 2A is a set of diagrams illustrating linked anti-Cls / anti-Bb antibody fragments produced from four exemplary configurations (#5-#8) of a monocistronic construct. Heavy chain variable domain: V H . Light chain variable domain: V L . Heavy chain constant region: C H . Light chain constant region: C L .

[0036] Figure 2B is a pair of diagrams illustrating two exemplary configurations (#9 and #10) of a construct carrying a bidirectional (“BiDir”) promoter driving expression of two separate antibody fragments.

[0037] Figure 2C is a pair of diagrams illustrating exemplary linked anti-Cls / anti-Bb scFab antibody fragments (#11 and #12) with charge mutations (“CM”; Δ) intended to facilitate cognate heavy and light chain pairing. In the figures herein, “Δ” indicates that a charge mutation is present, and does not mean to illustrate the exact location or number of charge mutations in the antibody fragment.

[0038] Figure 2D is a pair of diagrams illustrating exemplary linked anti-Cls / anti-Bb antibody fragments of aC1s scFab – (G4S)2 – aBb scFv with (#14) or without (#13) charge mutations.

[0039] Figure 2Eis a pair of graphs illustrating exemplary linked anti-Cls / anti-Bb antibody fragments with (#16) or without (#15) charge mutations of the aC1s scFab-(G4S)3-aBb scFv.

[0040] Figure 2F is a set of graphs illustrating exemplary linked anti-Cls / anti-Bb antibody fragments containing the self-cleaving peptides F2A or GT2A between the heavy and light chains of the aC1s Fab fragment (#17: aC1s F2A Fab-(G4S)3-aBb scFab; and #18: aC1s GT2A Fab-(G4S)3-aBb scFab) or the heavy and light chains of the aBb Fab fragment (#19: aC1s scFab-(G4S)3-aBb F2A Fab; and #20: aC1s scFab-(G4S)3-aBb GT2A Fab). F2A: a self-cleaving peptide comprising a furin cleavage site linked to a foot-and-mouth disease virus 2A peptide by an SGSG (SEQ ID NO: 81) linker (Fuchs et al., PLOS One [Public Library of Science: One] (2016) doi: 10.1371 / journal.pone.0158009). GT2A: a self-cleaving peptide comprising a furin cleavage site linked to a Lymantria dispar virus 2A peptide by a GSG linker.

[0041] Figure 2G is a pair of graphs illustrating exemplary configurations of constructs carrying bidirectional promoters driving expression of two independent antibody fragments, which differ from constructs #9 and #10 by having charge mutations (#21 and #22).

[0042] Figure 2H is a graph showing, in the case of AAV vector plasmids (including AAV2 ITRs), Figure 2D is a graph of construct #14 (aC1s scFab-(G4S)2-aBb scFv-CM). "aC1s": aC1s. "aBb": aBb.

[0043] Figure 2I is a graph showing, in the case of AAV vector plasmids (including AAV2 ITRs), Figure 2G is a graph of construct #9 (aC1s scFab-bidirectional promoter-aBb scFab) without charge mutations. Figure 2B is a graph of construct #14 (aC1s scFab-(G4S)2-aBb scFv-CM). "aC1s": aC1s. "aBb": aBb.

[0044] Figure 2Jis a graph showing construct aBbscFab-(G4S)3-aC1s scFab-CM (construct #12) in the context of AAV vector plasmid including AAV2 ITR. "aC1s": aC1s. "aBb": aBb. Figure 2C

[0045] Figure 3 is a representative biolayer interferometry (BLI) sensorgram showing that the protein expressed from construct #19 of Figure 2F

[0046] Figure 4A is a graph showing dose-dependent inhibition of complement activation by purified proteins expressed from construct #2 of Figure 2A

[0047] Figure 4B is a graph showing dose-dependent inhibition of complement activation by purified proteins expressed from construct #4 of Figure 2A

[0048] Figure 4C is a graph showing dose-dependent inhibition of complement activation by equimolar mixture of recombinant anti-Bb Fab and anti-C1s Fab and equimolar mixture of purified proteins expressed from constructs #2 and #4 of Figure 2A

[0049] Figure 5 is a set of photographs showing representative vector in situ hybridization of mouse retina 3 weeks post-AAV2 #9 administration. Vector-specific probes targeting the vector genome were used.

[0050] Figure 6 is a set of graphs showing combined inhibition of CP and AP on ARPE19 cells in a CRP-mediated complement activation model of dry AMD. Data shown are mean values of 12 replicates per condition in two independent experiments and standard deviations. "NHS": normal human serum. "CRP": C-reactive protein. **** p < 0.0001.

[0051] Figure 7A and Figure 7B is a graph showing cell-ELISA data depicting complement deposition on induced pluripotent stem cell-derived retinal pigment epithelial cells (iPSC-RPE) in an AMD cellular model. Treatment with anti-Bb and anti-C1s scFabs significantly inhibited complement product C3d Figure 7A ​​​​​) and C5b9 Figure 7B ) deposition on iPSC-RPE. Error bars are standard deviation. **** p < 0.0001.

[0052] Figure 8A and Figure 8B shows immunofluorescent staining of C5b9 on iPSC-RPE. Figure 8A is a set of confocal microscopy images showing C5b9 deposition (red) on iPSC-RPE. Figure 8B is a set of confocal microscopy images showing Figure 8A is a graph showing quantitative analysis of images in

[0053] Figure 9 is a heatmap showing results of ocular examination based on the preclinical ocular toxicology scoring (SPOTS) system. The heatmap shows clinical indicators of control group ocular inflammation and irritation before and after LPS treatment; it shows median severity scores during ophthalmic examination using the SPOTS system. DETAILED DESCRIPTION

[0054] The present disclosure is based on the discovery that dual targeting of the complement classical and alternative pathways can be used to treat ocular diseases associated with dysregulation or over-activation of the complement system in the eye. The present disclosure provides a gene therapy that delivers to the eye of a patient in need an inhibitor of activated complement component 1 subcomponent (aC1s or “C1s” for short herein) and an inhibitor of activated factor B (also known as Bb fragment, FBb, or Bb). The gene therapy can use a viral vector, such as a recombinant adeno-associated virus (AAV, such as AAV2), as a vehicle to deliver a transgene that directly expresses the C1s and Bb inhibitors. In some embodiments, the C1s inhibitor and the Bb inhibitor are antibody fragments, such as single-chain Fab (scFab) or single-chain Fv (scFv). The C1s inhibitor and the Bb inhibitor can be expressed as a single protein, or as two separate proteins.

[0055] In some embodiments, the eye disease to be treated is dry AMD, including associated geographic atrophy. In some embodiments, the patient has dysregulation / excessive activation of the complement system in the RPE choroidal interface. In some embodiments, the present therapy delivers (e.g., intravitreally or subretinally) the present recombinant expression construct (e.g., recombinant AAV2) to retinal ganglion cells (RGCs). Intravitreal delivery of rAAV2 can transduce RGCs in the retina and promote secretion of the inhibitory protein for distribution to the broader retina. For example, rAAV2 can be delivered intravitreally to patients with geographic atrophy (GA) secondary to dry AMD to reduce the growth in size of retinal GA lesions and prevent inevitable vision loss over 12 months. In addition to its benefit as a potential one-time treatment for GA, the presently disclosed gene therapy can have improved efficacy compared to therapeutic approaches that target downstream components in the complement pathway. This is because the current therapy broadly suppresses proximal and terminal mediators of inflammation, phagocytosis, and membrane attack complex-mediated cytolysis.

[0056] Approved or currently developing therapies involve repeated dosing of complement inhibitors (e.g., monthly or every other month). A one-time treatment by intravitreal delivery of a recombinant vector to an outpatient would provide the best of both approaches. Additionally, in other therapies, complement inhibitors can block all complement pathways. In contrast, the present bifunctional complement inhibitor targets the upstream activation step of the complement pathways (AP and CP) that are thought to be drivers of the pathogenesis of dry AMD, rather than targeting the downstream convertases that are common to all three initiating pathways. This approach leaves the Clq and lectin pathways intact to maintain immune surveillance. Moreover, this approach has a superior mode of action because it not only inhibits the membrane attack complex (MAC), but also inhibits the complement amplification loop and terminal events (such as inflammation and opsonization and phagocytosis) mediated by the upstream activation fragments. The present approach can also reduce target-mediated drug disposition (TMDD) because the inhibitor targets activated enzymes, which are typically present at much lower levels compared to intact proenzymes. I. C1s and Bb inhibitors

[0057] The present gene therapy introduces a Cls inhibitor and a Bb inhibitor (linked or unlinked) into the diseased eye of a patient.

[0058] Prior to processing and activation, a human Cls polypeptide can have the amino acid sequence of SEQ ID NO: 65 (UniProt. P09871), with amino acids 1-15 constituting a signal peptide. Upon activation, the Cls polypeptide is cleaved and becomes a disulfide-linked heterodimer, with the heavy chain corresponding to amino acids 16-437 of SEQ ID NO: 65 and the light chain corresponding to amino acids 438-688 of SEQ ID NO: 65. Unless otherwise indicated, a Cls inhibitor herein refers to an inhibitor of this activated form of Cls.

[0059] Prior to processing and activation, a human Factor B polypeptide can have the amino acid sequence of SEQ ID NO: 66 (UniProt. P00751), with amino acids 1-25 constituting a signal peptide. Upon activation, the polypeptide is cleaved into two subcomponents, Factor Ba corresponding to amino acids 26-259 of SEQ ID NO: 66 and Factor Bb corresponding to amino acids 260-764 of SEQ ID NO: 66. Factor Bb is also referred to herein simply as “Bb”.

[0060] The Cls inhibitors and Bb inhibitors herein can be recombinantly linked (e.g., recombinantly expressed as fusion proteins) with or without a peptide linker. When these proteins are introduced into cells via expression vectors, they can also be referred to as “vectorized” proteins (e.g., “vectorized” antibody fragments).

[0061] In some embodiments, the Cls inhibitors and Bb inhibitors are antigen-binding fragments of whole antibodies. A whole “antibody” (Ab) or “immunoglobulin” (Ig) refers to a four- polypeptide chain protein composed of two heavy (H) chains (about 50-70 kDa) and two light (L) chains (about 25 kDa) inter-connected by disulfide bonds. Each heavy chain is comprised of a heavy chain variable domain (V H ) and a heavy chain constant region (C H ). Each light chain is comprised of a light chain variable domain (V L ) and a light chain constant region (C L ). The V H and V L domains can be further subdivided into regions of hypervariability, termed “complementarity determining regions” (CDR), interspersed with regions that are more conserved, termed “framework regions” (FR). Each V H or V L is composed of three CDRs and four FRs, arranged from amino-terminus to carboxy-terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, and FR4. The amino acid Definitions (Lefranc et al., Dev Comp Immunol. (2003) 27(1):55-77; or the definitions of Kabat, Sequences of Proteins of Immunological Interest (National Institutes of Health, Bethesda, MD (1987 and 1991)); Chothia and Lesk, J. Mol. Biol. (1987) 196:901-917; or Chothia et al., Nature (1989) 342:878-83. Additional CDR definition systems include the AbM system and the Martin system (see, e.g., Abhinandan and Martin, Mol Immunol. (2008) 45(14):3832-9).

[0062] The term "antibody fragment," "antigen-binding fragment," or a similar term refers to a portion of an intact antibody that contains the amino acid residues that interact with an antigen and confer specificity and affinity of the fragment for the antigen. An antibody fragment can be a single chain variable fragment (scFv), which is a fusion protein of the V H and V L domains of an antibody, linked by a short peptide linker; a diabody, which is a noncovalent dimer of scFvs (Zapata et al., Protein Eng. (1995) 8(10): 1057-62); or a Fab fragment, which includes a single chain Fab (scFab) fragment. A "Fab" fragment contains the constant domains of the light chain and the first constant domain of the heavy chain (C H1 ). Other non-limiting examples of antigen-binding fragments of antibodies include Fd fragments, Fv fragments, dAb fragments, and minimal recognition units consisting of the amino acid residues that mimic the hypervariable loops of antibodies. In particular embodiments, an antibody fragment is a scFv, Fab, or scFab. A. Anti-Cls scFv and scFab

[0063] In some embodiments, the active Cls inhibitor is an antibody fragment, such as a scFab or scFv derived from the anti-Cl s antibody VH3 / VK2 from WO 2018 / 071676. Antibody fragments derived from variants of this antibody as described in WO 2018 / 071676 or WO 2016 / 164358 and U.S. Patents 10,729,767 and 11,246,926 can also be used herein. In some embodiments, the anti-Cl s (also referred to herein as “aCl s”) scFv or scFab herein comprises CDRs derived from the VH3 / VK2 antibody described above. The CDRs can be defined by any of the well-known systems, including those described above. In some embodiments, the CDRs are defined by the Kabat system, the Chothia system, or the AbM system, as shown in Table A below (SEQ ID NOs are shown in parentheses). Table A Table A

[0064] In some embodiments, the anti-Cl s scFab or scFv comprises heavy chain CDRs (HCDRs) 1-3 comprising SEQ ID NOs: 1-3, respectively, and light chain CDRs (LCDRs) 1-3 comprising SEQ ID NOs: 4-6, respectively.

[0065] In particular embodiments, the anti-Cl s scFv or scFab comprises a V H comprising SEQ ID NO: 7 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto, and a V L In certain embodiments, the anti-Cl s scFv comprises a peptide linker, such as a flexible linker, e.g., a linker comprising (G4S) n (SEQ ID NO: 46) where n = 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10, linking V H and V L In some embodiments, the linker comprises SEQ ID NO: 48 (i.e., n = 3). The V H may be N-terminal or C-terminal to V L In some embodiments, the anti-Cl s scFv comprises SEQ ID NO: 9 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0066] In certain embodiments, the anti-Cls scFab comprises a heavy chain (HC) comprising SEQ ID NO: 10 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto; and a light chain (LC) comprising SEQ ID NO: 11 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto. In further embodiments, the HC and LC are connected by a peptide linker, such as a flexible linker, e.g., a linker comprising (G4S) n (SEQ ID NO: 46) wherein n = 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10, connecting the HC and LC. In some embodiments, the linker comprises SEQ ID NO: 49 (i.e., n = 7). The HC can be N-terminal or C-terminal to the LC. In some embodiments, the aC1s scFab comprises SEQ ID NO: 12 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0067] In some embodiments, the C1s inhibitor is an antibody fragment, such as an scFab or scFv derived from an anti-C1s antibody disclosed in US 2022 / 0380483A1. For example, the C1s inhibitor can comprise the heavy and light chain CDRs or V H HAs of a parent anti-C1s antibody. In some embodiments, the C1s inhibitor comprises the heavy and light chain CDRs or V L . B. Anti-Bb scFv and scFab

[0068] In some embodiments, the Bb inhibitor is an antibody fragment, such as an scFab or scFv derived from an anti-Bb antibody V H 6 / VK7-IgG4v2 from U.S. Patent 11,242,382 and WO 2021 / 216458. Antibody fragments derived from variants of this antibody as described in WO 2021 / 216458 can also be used herein. In some embodiments, the anti-Bb (also referred to herein as “aBb”) scFv or scFab herein comprises CDRs derived from the V H 6 / VK7-IgG4v2 antibody described above. The CDRs can be defined by any of the well-known systems, including those described above. In some embodiments, the CDRs are defined by the Kabat system, the Chothia system, or the AbM system, as shown in Table B below (SEQ ID NOs are shown in parentheses). Table B

[0069] In some embodiments, the anti-Bb scFab or scFv comprises heavy chain CDRs (HCDRs) 1-3 comprising SEQ ID NOS: 13-15, respectively, and light chain CDRs (LCDRs) 1-3 comprising SEQ ID NOS: 16-18, respectively.

[0070] In particular embodiments, the anti-Bb scFv or scFab comprises a V H comprising SEQ ID NO: 19 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto, and a V L In certain embodiments, the anti-Bb scFv comprises a peptide linker, such as a flexible linker, e.g., a linker comprising (G4S) n (SEQ ID NO: 46) wherein n = 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10, linking V H and V L In some embodiments, the linker comprises SEQ ID NO: 48 (i.e., n = 3). V H may be N-terminal or C-terminal to V L In some embodiments, the anti-Bb scFv comprises SEQ ID NO: 21 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0071] In certain embodiments, the anti-Bb scFab comprises a heavy chain (HC) comprising SEQ ID NO: 22 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto, and a light chain (LC) comprising SEQ ID NO: 23 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto. In further embodiments, the HC and LC are linked by a peptide linker, such as a flexible linker, e.g., a linker comprising (G4S) n (SEQ ID NO: 46) wherein n = 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10, linking the HC and LC. In some embodiments, the linker comprises SEQ ID NO: 49 (i.e., n = 7). The HC can be N-terminal or C-terminal to the LC. In some embodiments, the anti-Bb scFab comprises SEQ ID NO: 24 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0072] In some embodiments, the Bb inhibitor is an antibody fragment, such as an scFab or scFv derived from an anti-Bb antibody disclosed in U.S. Patent Nos. 10,131,706; 10,604,563; or 7,964,705. For example, the Bb inhibitor can comprise the heavy and light chain CDRs or V H and V L . C. Anti-Cls / Bb Bispecific Fusion Proteins

[0073] In some embodiments, the C1s inhibitor (e.g., an anti-C1s scFab or scFv) and the Bb inhibitor (e.g., an anti-Bb scFab or scFv) are linked by a peptide linker, such as a flexible linker, e.g., a linker comprising (G4S) n (SEQ ID NO: 46) wherein n = 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10, linking the two inhibitors. In some embodiments, the peptide linker is SEQ ID NO: 47 (n = 2) or 48 (n = 3). The C1s inhibitor can be N-terminal or C-terminal to the Bb inhibitor. The aC1s / aBb fusion protein can have the following exemplary non-limiting configurations (from N-terminal to C-terminal): aC1s scFab - linker - aBb scFab aC1s scFv - linker - aBb scFab aC1s scFab - linker - aBb scFv aC1s scFv - linker - aBb scFv aBb scFab - linker - aC1 scFab aBb scFv - linker - aC1 scFab aBb scFab - linker - aC1 scFv aBb scFv - linker - aC1 scFv wherein the "linker" can be one of the peptide linkers described herein (e.g., a flexible linker described herein), such as (G4S)2(SEQ ID NO: 47) and (G4S)3(SEQ ID NO: 48), and wherein within each configuration, the scFab and / scFv can have heavy and light chains in the order N-Heavy-Light-C or N-Light-Heavy-C.

[0074] To promote homogenous pairing of heavy and light chains within each antigen binding domain of the fusion protein, each antigen binding domain can comprise charge mutations. Charge mutations refer to the substitution of a charge neutral amino acid (e.g., Q) with a positively charged (e.g., K) or negatively charged (e.g., E) amino acid, as well as the substitution of a charged amino acid with an amino acid of opposite charge. To increase pairing of two polypeptide chains, interacting residues on the two chains can be mutated to amino acid residues of opposite charge. Exemplary charge mutations that can aid in homogenous antibody chain pairing are described, e.g., in Tan et al., Biophys J (1998) 75: 1473-82; US2014 / 0242076A1; and WO 2020 / 136566. In some embodiments, • Charge mutations in the aC1s scFv or scFab include Q42E (V L ) and Q292K (V H ) mutations (numbering according to SEQ ID NO: 12); • Charge mutations in the aBb scFv include Q38K (V L ) and Q288E (V H ) (numbering according to SEQ ID NO: 24); and • Charge mutations in the aBb scFab include Q38K (V L ) and Q288E (V H ), and optionally further include S114A (C L ), N137K (C L ), and T434E (C H1 ) (numbering according to SEQ ID NO: 24).

[0075] In some embodiments, the fusion protein has the structure shown in Construct #5 Figure 2A ), wherein the order of components of the fusion protein from N-terminus to C-terminus is aC1s scFv – (G4S)2 – aBb scFv. In particular embodiments, the fusion protein is encoded by SEQ ID NO: 54, or comprises SEQ ID NO: 55 (with or without a signal peptide) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0076] In some embodiments, the fusion protein has the structure shown in Construct #6 Figure 2Athe order of components of the fusion protein, from N- to C-terminus, is aBb scFv - (G4S)2 - aC1s scFv. In particular embodiments, the fusion protein is encoded by SEQ ID NO: 56, or comprises SEQ ID NO: 57 (with or without signal sequence) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0077] In some embodiments, the fusion protein has the structure set forth in Figure 2A the order of components of the fusion protein, from N- to C-terminus, is aC1s scFab - (G4S)3 - aBb scFab. In particular embodiments, the fusion protein is encoded by SEQ ID NO: 25, or comprises SEQ ID NO: 26 (with or without signal sequence) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0078] In some embodiments, the fusion protein has the structure set forth in Figure 2A the order of components of the fusion protein, from N- to C-terminus, is aBb scFab - (G4S)3 - aC1s scFab. In particular embodiments, the fusion protein is encoded by SEQ ID NO: 29, or comprises SEQ ID NO: 30 (with or without signal sequence) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0079] In some embodiments, the fusion protein has the structure set forth in Figure 2C the order of components of the fusion protein, from N- to C-terminus, is aC1s scFab - (G4S)3 - aBb scFab (with CM). In particular embodiments, the fusion protein is encoded by SEQ ID NO: 27, or comprises SEQ ID NO: 28 (with or without signal sequence) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0080] In some embodiments, the fusion protein has the structure set forth in Figure 2Cthe components of the fusion protein, in order from N-terminal to C-terminal, are aBb scFab - (G4S)3 - aC1s scFab (with CM). In particular embodiments, the fusion protein is encoded by SEQ ID NO: 31, or comprises SEQ ID NO: 32 (with or without signal sequence) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0081] In some embodiments, the fusion protein has the structure depicted in Figure 2D the components of the fusion protein, in order from N-terminal to C-terminal, are aC1s scFab - (G4S)2 - aBb scFv. In particular embodiments, the fusion protein is encoded by SEQ ID NO: 33, or comprises SEQ ID NO: 34 (with or without signal sequence) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0082] In some embodiments, the fusion protein has the structure depicted in Figure 2D the components of the fusion protein, in order from N-terminal to C-terminal, are aC1s scFab - (G4S)2 - aBb scFv-CM (#13 with CM in both aC1s and aBb). In particular embodiments, the fusion protein is encoded by SEQ ID NO: 35, or comprises SEQ ID NO: 36 (with or without signal sequence) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0083] In some embodiments, the fusion protein has the structure depicted in Figure 2E the components of the fusion protein, in order from N-terminal to C-terminal, are aC1s scFab - (G4S)3 - aBb scFv. In particular embodiments, the fusion protein is encoded by SEQ ID NO: 58, or comprises SEQ ID NO: 59 (with or without signal sequence) or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto.

[0084] In some embodiments, the fusion protein has the structure depicted in Figure 2E) wherein the order of components of the fusion protein from N-terminus to C-terminus is αBb scFab–(G4S)3–αC1s scFv-CM (with CM). In particular embodiments, the fusion protein is encoded by SEQ ID NO: 60, or comprises the amino acid sequence of SEQ ID NO: 61 (with or without the signal peptide) or at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto. D. Bispecific heterodimers

[0085] In some embodiments, the dual-targeting complement inhibitors are anti-C1s / anti-Bb bispecific heterodimeric proteins. These proteins are encoded by a single open reading frame, but the HC and LC of one of the antibody fragments are cleaved upon intracellular translation and post-translational processing, resulting in two separate polypeptides that fold into two antigen-binding domains. Figure 2F Such configurations are illustrated. In these illustrated configurations, the HC and LC of one of the antibody fragments are connected by a cleavable peptide (e.g., a self-cleaving 2A peptide with or without a protease (e.g., furin) cleavage site). See also the discussion in Section II (“Recombinant expression constructs”).

[0086] In some embodiments, the heterodimer has the structure illustrated in Construct #17 Figure 2F ) wherein the heterodimer consists of (i) an αC1s LC and (ii) a fusion protein comprising an αC1s HC fused to an αBb scFab. In particular embodiments, this heterodimer is encoded by SEQ ID NO: 38, or comprises prior to cleavage the amino acid sequence of SEQ ID NO: 39 or at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto (including or excluding the two signal peptide sequences).

[0087] In some embodiments, the heterodimer has the structure illustrated in Construct #18 Figure 2F ) wherein the heterodimer consists of (i) an αC1s LC and (ii) a fusion protein comprising an αC1s HC fused to an αBb scFab. In particular embodiments, this heterodimer is encoded by SEQ ID NO: 40, or comprises prior to cleavage the amino acid sequence of SEQ ID NO: 41 or at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto (including or excluding the two signal peptide sequences).

[0088] In some embodiments, the heterodimer has the structure illustrated in Construct #19 Figure 2F) wherein the heterodimer consists of (i) a fusion protein comprising an aC1s scFab fused to an aBb HC and (ii) an aBb LC. In particular embodiments, this heterodimer is encoded by SEQ ID NO: 42, or comprises prior to cleavage SEQ ID NO: 43 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto (including or excluding the two signal peptide sequences).

[0089] In some embodiments, the heterodimer has the structure shown in Figure 2F ) wherein the heterodimer consists of (i) a fusion protein comprising an aC1s scFab fused to an aBb HC and (ii) an aBb LC. In particular embodiments, this heterodimer is encoded by SEQ ID NO: 42, or comprises prior to cleavage SEQ ID NO: 43 or an amino acid sequence at least 95% (e.g., at least 96%, 97%, 98%, or 99%) identical thereto (including or excluding the two signal peptide sequences). E. Peptide Linkers

[0090] Peptide linkers connecting antibody fragments and domains of fusion proteins can preferably be flexible linkers in order to allow proper folding, movement, and interaction of the connected domains. In some embodiments, flexible peptide linkers herein comprise primarily small amino acids (e.g., Gly, Ser, or Thr). In some embodiments, peptide linkers herein consist primarily (e.g., more than 50% of residues) of Gly and Ser residues (“GS” linkers). As noted above, such peptide linkers can comprise (G4S)n(SEQ ID NO: 46). By adjusting the copy number “n”, the length of the linker can be adjusted to achieve a desired distance of the connected functional domains. In some embodiments, peptide linkers can comprise additional amino acids such as Thr and Ala to maintain flexibility, and polar amino acids such as Lys and Glu to improve solubility. See, e.g., Chen et al., Adv Drug Deliv Rev. (2013) 65(10): 1357-69. II. Recombinant expression constructs

[0091] The disclosure provides recombinant expression constructs for expressing the C1s / Bb inhibitors herein. These expression constructs have expression cassettes comprising coding sequences for the C1s / Bb inhibitors operably linked to a promoter and a poly(A) signal sequence. The coding sequences can be human codon-optimized to improve expression in human cells. The coding sequences can encode a signal peptide (e.g., from IgG kappa) to support secretion of the protein. The expression cassettes can also comprise additional transcriptional regulatory sequences (e.g., Kozak sequences) and sequences that enhance gene expression or RNA stability (e.g., a WPRE element). A. Configuration of expression constructs 1. Expression constructs encoding a single fusion protein

[0092] In some embodiments, the expression constructs herein are monocistronic and comprise a coding sequence for an aC1s / aBb fusion protein. See, e.g., Figure 1A and Figure 1C For example, the expression construct can be one of numbered constructs #5 through #8 and constructs #11 through #16, the gene products of which are described in the above sections. 2. Expression constructs encoding two separate proteins

[0093] In some embodiments, the expression construct encodes the C1s inhibitor and the Bb inhibitor as two separate proteins. Independent target engagement can eliminate the possibility of steric hindrance.

[0094] For example, the expression construct has two separate expression cassettes, one for the C1s inhibitor (e.g., scFv or scFab) and one for the Bb inhibitor (e.g., scFv or scFab). Each expression cassette has its own transcriptional regulatory sequences, such as a promoter and enhancer.

[0095] In another configuration, the expression construct has a bicistronic expression cassette and a single promoter. The coding sequences for the C1s inhibitor and the Bb inhibitor are transcribed together into one mRNA under the single promoter, and then the RNA sequences of each subtype are translated separately by using an internal ribosome entry site (IRES) in the mRNA. In another approach, the coding sequences for the C1s and Bb inhibitors are separated by a coding sequence for a self-cleaving peptide and / or a protease (e.g., furin) cleavage site, such that translation and subsequent processing of the mRNA transcript produces two separate gene products (C1s inhibitor and Bb inhibitor). An example of a self-cleaving peptide is a 2A peptide, which is a virus-derived peptide typically 18-22 amino acids in length. 2A peptides include T2A, P2A, E2A, and F2A. Translation of the transgene can leave a few amino acid residues from the 2A peptide on one or both gene products. A furin cleavage site can be included to allow removal of the extra amino acid residues.

[0096] In yet another configuration, the bicistronic expression construct comprises a bi-directional promoter that allows separate expression of each inhibitor. See, e.g., the expression construct can be one of the numbered constructs #9, #10, #21, and #22 shown in Figure 2B and Figure 2G (BiDir: bi-directional promoter) • #9: aC1s scFab - bi-directional promoter - aBb scFab, producing separate aC1s scFab and aBb scFab • #10: aBb scFab - bi-directional promoter - aC1s scFab, producing separate aC1s scFab and aBb scFab • #21: aC1s scFab - bi-directional promoter - aBb scFab-CM, producing separate aC1s scFab-CM and aBb scFab-CM • #22: aBb scFab - bi-directional promoter - aC1s scFab-CM, producing separate aC1s scFab-CM and aBb scFab-CM In constructs #21 and #22, both the anti-C1s and anti-Bb scFabs contain a charge mutation (CM) to facilitate homologous pairing of the heavy and light chains within each antibody fragment. 3. Expression construct encoding a heterodimer

[0097] In some embodiments, the expression construct encodes a heterodimer comprising a first single chain antibody fragment (e.g., scFab or scFv) fused to one of the two chains of a second antibody fragment (e.g., Fab), wherein the fusion polypeptide forms a complex with the other chain of the second antibody fragment. The heterodimer is bispecific and can bind to Cls and Bb.

[0098] Exemplary constructs encoding bispecific heterodimer configurations are shown in Figure 2F and listed below: • #17: aC1s F2A Fab - (G4S)3 - aBb scFab, producing a heterodimer consisting of (i) aC1s LC and (ii) a fusion protein comprising aC1s HC fused to aBb scFab • #18: aC1s GT2A Fab - (G4S)3 - aBb scFab, producing a heterodimer consisting of (i) aC1s LC and (ii) a fusion protein comprising aC1s HC fused to aBb scFab • #19: aC1s scFab - (G4S)3 - aBb F2A Fab, producing a heterodimer consisting of (i) a fusion protein comprising aC1s scFab fused to aBb HC and (ii) aBb LC • #20: aC1s scFab - (G4S)3 - aBb GT2A Fab, producing a heterodimer consisting of (i) a fusion protein comprising aC1s scFab fused to aBb HC and (ii) aBb LC In the above constructs, the coding sequence comprising a cleavable peptide (such as F2A and GT2A) results in the production of two separate polypeptides, which are then complexed and folded into a single bispecific heterodimeric protein. F2A and GT2A coding sequences and amino acid sequences are shown in SEQ ID NOs: 38-45. Other coding sequences for cleavable peptides (e.g., those described above) can also be used. 4. Separate expression constructs for Cls inhibitor and Bb inhibitor

[0099] In some embodiments, the Cls inhibitor and Bb inhibitor can be expressed from two separate constructs, e.g., two separate recombinant AAVs, as further described below. The two AAVs can be of the same or different serotypes. B. Transcriptional regulatory sequences

[0100] In the present expression constructs, the coding sequences for the Cls inhibitor and Bb inhibitor are operably linked to transcriptional regulatory sequences such as promoters and enhancers to allow expression of the encoded proteins in the intended target cells.

[0101] In some embodiments, the C1s and Bb inhibitors are produced in a recombinant host cell. In such cases, the promoters and enhancers are those active in the host cell.

[0102] In some embodiments, the C1s and Bb inhibitors are delivered by gene therapy and produced in vivo in the eye of a subject (e.g., a human, non-human primate, or mouse). In such cases, the promoter can be a constitutive or inducible promoter that functions in ocular or retinal cells (e.g., RGCs and RPE cells of the inner and outer nuclear layers, Mueller cells, and photoreceptor cells).

[0103] In some embodiments, the promoter is a minCBA promoter comprising a CMV enhancer, a chicken beta-actin promoter, and an intron sequence. The minCBA promoter can have a sequence that is at least 85% (e.g., at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%) identical or is identical to SED ID NO: 83.

[0104] In some embodiments, the promoter is a bidirectional promoter. The bidirectional promoter can contain, for example, a pair of CBA promoters placed in opposite directions, separated by a CMV enhancer. In particular embodiments, the bidirectional promoter comprises a sequence that is at least 85% (e.g., at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%) identical or is identical to SED ID NO: 53.

[0105] In some embodiments, the expression cassette has a poly(A) signal sequence derived from the bovine growth hormone gene. In particular embodiments, the poly(A) signal sequence comprises a sequence that is at least 85% (e.g., at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%) identical or is identical to the sequence shown in SED ID NO: 51 in italics and underlined in the sequence section below.

[0106] In some embodiments, the expression cassette contains an enhancer, such as a CMV enhancer. In particular embodiments, the CMV enhancer comprises a sequence that is at least 85% (e.g., at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%) identical or is identical to the sequence shown in bold and italics in SED ID NO: 53 in the sequence section below.

[0107] In some embodiments, the expression cassette contains an intron sequence, such as a chimeric intron. The intron sequence can increase the level of transgene expression by facilitating mRNA export out of the nucleus and enhancing mRNA stability. C. Recombinant AAV Expression Vectors

[0108] In some embodiments, a viral vector is used to deliver the vectorized antibody fragment to the eye of a patient. In some embodiments, the expression / delivery vector is a recombinant adeno-associated virus (rAAV) expression vector. The expression constructs herein can be a rAAV genome. In the case of a rAAV genome, the expression cassettes herein can be flanked by a pair of AAV inverted terminal repeat sequences (ITRs), such as AAV2 ITRs. Non-limiting examples of a single- direction, single-cistronic AAV2 recombinant genome are shown in Figure 2H . Non-limiting examples of a bi-directional, double-cistronic AAV2 recombinant genome are shown in Figure 2I .

[0109] An exemplary rAAV genome carrying Construct #9 can have the exemplary nucleotide sequence of SEQ ID NO: 50, or a nucleotide sequence that encodes the same amino acid sequence as SEQ ID NO: 50 and comprises a sequence with at least 50% (e.g., at least 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95%) identity to SEQ ID NO: 50.

[0110] An exemplary rAAV genome carrying Construct #12 can have the exemplary nucleotide sequence of SEQ ID NO: 51, or a nucleotide sequence that encodes the same amino acid sequence as SEQ ID NO: 51 and comprises a sequence with at least 50% (e.g., at least 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95%) identity to SEQ ID NO: 51.

[0111] An exemplary rAAV genome carrying Construct #14 can have the exemplary nucleotide sequence of SEQ ID NO: 52, or a nucleotide sequence that encodes the same amino acid sequence as SEQ ID NO: 52 and comprises a sequence with at least 50% (e.g., at least 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95%) identity to SEQ ID NO: 52.

[0112] The rAAV genome can be constructed by inserting the expression cassettes herein into a rAAV genome from which the major rAAV open reading frames have been excised. Other portions of the rAAV genome can also be deleted, so long as sufficient ITR portions are retained to achieve replication and packaging functions.

[0113] Any suitable AAV serotype can be used. For example, the AAV can be AAV1, AAV2, AAV3, AAV3B, AAV4, AAV5, AAV6, AAV7, AAV8, AAV8.2, AAV9, or AAVrh10, or a pseudotype or serotype that is a mutant, variant, or derivative of one of the AAV serotypes listed herein (i.e., AAVs derived from multiple serotypes). The AAV can be engineered such that its capsid proteins have reduced immunogenicity or enhanced transduction ability in humans or non-human primates.

[0114] In some embodiments, the rAAV herein has an AAV2 capsid. In particular embodiments, the AAV2 capsid is a wild-type AAV2 capsid. In other embodiments, the AAV2 capsid contains mutations that improve the titer and productivity of the rAAV2.

[0115] The viral vectors described herein can be produced using methods known in the art. Any suitable permissive cell or packaging cell can be employed to produce the viral particles. For example, mammalian (e.g., 293 or HeLa) or insect (e.g., Sf9) cells can be used as packaging cell lines. When a recombinant AAV vector is introduced into a host cell that has been infected with a suitable helper virus (or expresses suitable helper functions) and expresses AAV rep and cap gene products (i.e., AAV Rep and capsid proteins), the recombinant AAV vector can be replicated and packaged into infectious viral particles. See, e.g., U.S. Patent 11,261,463. D. Transfection of Host Cells

[0116] When the Cls and Bb inhibitors are delivered directly to a patient, the inhibitors can be produced in recombinant mammalian host cells, such as COS, NS0, 293, HeLa, or CHO cells. Once the vectors are incorporated into the appropriate host, the host is maintained in conditions suitable for high-level expression of the nucleotide sequences and collection and purification of the inhibitors. III. Pharmaceutical compositions and uses

[0117] The disclosure provides pharmaceutical compositions comprising a dual-targeting Cls / Bb inhibitor or a recombinant viral vector (e.g., an AAV vector encoding the inhibitor). The pharmaceutical compositions can comprise a pharmacologically (especially ophthalmologically) acceptable carrier, diluent, and / or excipient. For example, the compositions can comprise a tonicity agent (such as sodium chloride, an amino acid, a sugar, or a combination thereof), a surfactant (such as polysorbate 20 or polysorbate 80), and / or a stabilizer (such as methionine).

[0118] The pharmaceutical compositions can be delivered by intraocular injection, for example, via a temporal corneal limbal injection into the anterior chamber, suprachoroidal injection, intracameral injection, intrasubject injection, subretinal injection, intravitreal injection (e.g., anterior, mid, or posterior vitreous injection).

[0119] The pharmaceutical compositions of the present application can be delivered in a therapeutically effective amount to treat dry AMD and geographic atrophy (GA) secondary to dry AMD. By "therapeutically effective amount" is meant a dosage sufficient to result in the improvement of one or more symptoms of the disease being treated (e.g., growth of GA lesions, retinal lesions, or destruction of retinal layers), and / or slowing of the progression of the disease, for example. The desired result can also include improvement in one or more functional symptoms; for example, the desired result can be a reduction in visual distortion, improvement in central vision, improvement in vision in low light environments, and / or reduction in blurring. By "treat" or "treatment" is meant to improve one or more symptoms of the disease and / or slow the progression of the disease.

[0120] The pharmaceutical compositions of the present application can be delivered in a prophylactically effective amount to prevent the onset of dry AMD or geographic atrophy (GA) secondary to dry AMD. By "prophylactically effective amount" is meant a dosage sufficient to result in the prevention or delay of the onset of dry AMD and / or GA, and / or the prevention or delay of the onset of one or more symptoms of dry AMD and / or GA, for example. The pharmaceutical compositions of the present application can be administered prophylactically to patients at high risk of developing dry AMD, for example, patients with a genetic predisposition.

[0121] In some embodiments, the dose of recombinant AAV (rAAV) injected into the eye is 10 7 to 10 15 vector genomes (vg), for example, 10 8 to 10 14 , 10 9 to 10 13 , or 10 9 to 10 12 vg. In some embodiments, the dose of rAAV is 2 x 10 9 , 2 x 10 10 , or 2 x 10 11 vg.

[0122] In some embodiments, the patient is treated with an anti-inflammatory agent (e.g., a steroid) prior to, during, and / or after rAAV injection to prevent or ameliorate a potential immune response against the rAAV. In some embodiments, the patient can be pretreated with an IgG-degrading enzyme (e.g., IdeS) to reduce pre-existing neutralizing antibodies against the AAV capsid. These immunomodulatory agents can be administered locally or systemically. In some embodiments, the modulatory agent can be administered intraocularly (e.g., intravitreally), orally, intravenously, intramuscularly, or subcutaneously.

[0123] Unless otherwise defined, scientific and technical terms used in connection with the present disclosure shall have the meanings that are commonly understood by those of ordinary skill in the art. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present disclosure, exemplary methods and materials are described below. In case of conflict, the patent specification, including definitions, will control. In addition, unless otherwise specified, a singular term includes pluralities and a plural term includes the singular. Throughout the specification and embodiments, the words “comprise” and “contain” or variations such as “comprises” or “comprising,” will be understood to imply the inclusion of stated integers or groups of integers but not to the exclusion of any other integers or groups of integers. All publications and other references mentioned herein are incorporated by reference in their entirety. Although a number of documents are cited herein, this citation does not constitute an admission that any of these documents forms part of the prior art for any country. As used herein, the term “about” or “approximately,” when applied to one or more

[0124] As used herein, the percent identity of two amino acid sequences (or two nucleic acid sequences) can be determined by, for example, obtained using default parameters (available on the U.S. National Library of Medicine’s National Center for Biotechnology Information website). In some embodiments, the length of a query sequence that is aligned for comparison purposes is at least 30% (e.g., at least 40%, 50%, 60%, 70%, 80%, or 90%) of the length of the reference sequence.

[0125] According to the present disclosure, the back reference in the dependent claims means a shorthand for the direct and explicit disclosure of each and all combinations of the claims to which the back reference indicates. Any compound disclosed herein can be used in any method of treatment disclosed herein, wherein the individual to be treated is as defined anywhere herein.

[0126] To better understand the present application, the following examples are set forth. These examples are for purposes of illustration only and are not to be construed as limiting the scope of the application in any manner. Example Example 1: Carrierized Antibodies and Expression Constructs Thereof

[0127] This example describes the design of bifunctional expression constructs expressing C1s and Bb inhibitors, and the characterization of recombinant proteins produced from these constructs. These constructs have the following features: (i) a single or dual promoter (e.g., minCBA promoter) to drive constitutive transgene expression; (ii) transgenes (e.g., transgenes containing sequences that are human codon-optimized); (iii) different combinations of antibody fragments (e.g., scFab-scFab and scFab-scFv) derived from parental anti-Bb IgG4 antibodies (e.g., V H 6 / VK7-IgG4v2) and parental anti-C1s IgG4 antibodies (e.g., VH3 / VK2 from WO 2018 / 071676); (iv) peptide linkers (e.g., containing G4S repeats) between antibody fragments and between heavy and light chains of each antibody fragment; (v) presence or absence of rationally designed charge mutations (CMs) to promote accurate heavy / light chain pairing; and (vi) polyadenylation sites (e.g., bovine growth hormone (bGH) gene polyadenylation signal). A. Generation of Bifunctional Bi- or Mono-cistronic Constructs

[0128] The bifunctional mono- or bi-cistronic constructs generated herein contain DNA fragments expressing scFv or scFab of constitutive antibody fragments against active C1s and Bb downstream of a ubiquitous minCBA promoter and a poly(A) signal sequence from the bovine growth hormone gene. The entire expression cassette is cloned between sequences from wild-type inverted terminal repeats (ITRs) from AAV serotype 2. Figures 1A-1C A glycine / serine-rich linker (e.g., a linker with G4S repeats) is inserted between the heavy and light chains of each single-chain aC1s and aBb antibody fragment (scFab or scFv) to promote pairing of one pair of V H and V Lproper folding of each antigen binding domain formed. In this study, a linker with seven G4S repeat sequences was used to connect the heavy and light chains of scFab, and a linker with three G4S repeat sequences was used to connect the V H and V L .

[0129] For single-cistronic constructs, exemplary formats are scFab-scFab, scFv-scFab, scFab-scFv, and ScFv-ScFv (see, e.g., Figure 2A , Figure 2D and Figure 2E ). A glycine / serine-rich linker (e.g., a linker with G4S repeat sequences (such as two or three repeat sequences)) was inserted between the two single-chain fragments of the bifunctional fusion protein to achieve flexibility of the bifunctional fusion protein.

[0130] For some single-cistronic constructs, an additional feature is the inclusion of a canonical furin cleavage site (RX(R / K)R) (SEQ ID NO: 82), e.g., in the linkers F2A and GT2A Figure 2F ). Linking the heavy chain (HC) and light chain (LC) genes to a single cassette using 2A peptides will allow improved control over the LC and HC ratio. Insertion of a furin recognition site upstream of the 2A will allow removal of the 2A residues that would otherwise be attached to the HC and / or LC (see, e.g., Figure 2F ).

[0131] For bidirectional bifunctional constructs, a novel bidirectional promoter was designed based on the ubiquitous minimal chicken beta-actin (minCBA) promoter. This promoter supports the simultaneous expression of individual antibody fragments against Factor Cls and Factor Bb. MinCBA contains the CBA promoter and CMV enhancer, but contains a truncated intron sequence. The bidirectional promoter contains a pair of CBA promoters placed in opposite directions and separated by a CMV enhancer (SEQ ID NO: 53). The bidirectional expression construct produces separate anti-Cls and anti-Bb antibody fragments for independent target engagement, which eliminates the possibility of steric hindrance.

[0132] The single-cistronic or double-cistronic expression cassettes were cloned between AAV2 ITR sequences (see, e.g., Figure 1C , Figure 2H and Figure 2I ) for AAV delivery.

[0133] Some experiments used antibody fragments containing charge mutations that promote accurate pairing between the heavy and light chains of each constituent antibody fragment. To generate charge mutations (CMs), specific amino acids were substituted in the variable and / or constant domains of the aCIs and aBb antibody fragments. The following mutated antibody fragments incorporate the following amino acid changes: • aCIs scFab-CM: Q42E and Q292K (numbering consistent with SEQ ID NO: 12) • aBb scFab-CM: Q38K, S114A, N137K, Q288E, and T434E (numbering consistent with SEQ ID NO: 24)

[0134] Exemplary one-way construct configurations are shown in Figure 2A and Figures 2C-2F and are listed below: • #5: aCIs scFv - (G4S)2 - aBb scFv • #6: aBb scFv - (G4S)2 - aCIs scFv • #7: aCIs scFab - (G4S)3 - aBb scFab • #8: aBb scFab - (G4S)3 - aCIs scFab • #11: aCIs scFab - (G4S)3 - aBb scFab-CM (#7 with CMs in both aCIs and aBb) • #12: aBb scFab - (G4S)3 - aCIs scFab-CM (#8 with CMs in both aCIs and aBb) • #13: aCIs scFab - (G4S)2 - aBb scFv • #14: aCIs scFab - (G4S)2 - aBb scFv-CM (#13 with CMs in both aCIs and aBb) • #15: aCIs scFab - (G4S)3 - aBb scFv • #16: aCIs scFab - (G4S)3 - aBb scFv-CM (#15 with CMs in both aCIs and aBb) • #17: aCIs F2A Fab - (G4S)3 - aBb scFab, producing a heterodimer consisting of (i) aCIs LC and (ii) a fusion protein comprising aCIs HC fused to aBb scFab • #18: aC1s GT2A Fab - (G4S)3 - aBb scFab, producing a heterodimer consisting of (i) aC1s LC and (ii) a fusion protein comprising aC1s HC fused to aBb scFab • #19: aC1s scFab - (G4S)3 - aBb F2A Fab, producing a heterodimer consisting of (i) a fusion protein comprising aC1s scFab fused to aBb HC and (ii) aBb LC • #20: aC1s scFab - (G4S)3 - aBb GT2A Fab, producing a heterodimer consisting of (i) a fusion protein comprising aC1s scFab fused to aBb HC and (ii) aBb LC

[0135] Exemplary bidirectional construct configurations are shown in Figure 2B and listed below: • #9: aC1s scFab - bidirectional promoter - aBb scFab, producing separate aC1s scFab and aBb scFab • #10: aBb scFab - bidirectional promoter - aC1s scFab, producing separate aC1s scFab and aBb scFab • #21: aC1s scFab - bidirectional promoter - aBb scFab-CM, producing separate aC1s scFab-CM and aBb scFab-CM • #22: aBb scFab - bidirectional promoter - aC1s scFab-CM, producing separate aC1s scFab-CM and aBb scFab-CM B. Evaluation of B.Bb and C1s binding

[0136] Each DNA construct was transfected into HEK293 cells. Supernatants containing secreted recombinant proteins were harvested and purified on protein L beads. More particularly, the supernatant was incubated with protein L beads for 1 hour at room temperature. The beads were then washed three times with PBS containing 0.05% v / v polysorbate 20. The bead column was then eluted with 0.1 M glycine (pH 2.0) for 10 minutes at room temperature. The eluate was neutralized with 15% v / v 1 M Tris (pH 8.5) and then desalted into PBST by buffer exchange.

[0137] The purity of the recombinant proteins was evaluated on SDS-PAGE (non-reducing and reducing) and mass photometer (mass distribution). In The concentration of the proteins was measured on a Nanodrop® (Thermo Fisher).

[0138] Recombinant proteins were assessed for target engagement and binding affinity to complement Cls enzyme (active Cls or "Cl s" herein) and Factor Bb (Complement Technology, Tyler, TX, USA) using Bio-Layer Interferometry (BLI). EZ-Link TM Sulfo-NHS-LC-LC-Biotin (Thermo Fisher Scientific, Waltham, MA, USA) was used to biotinylate Cls and Bb. Biotinylated Cls or Bb were loaded onto Streptavidin (SA) biosensors (Sartorius, Gottingen, Germany) were then loaded with a range of concentrations of purified protein. To assess dual target engagement, biotinylated active Cls or Bb were loaded onto the sensor followed by loading of purified protein ("first association phase") followed by loading of the non-captured complement target (Bb or active Cls, non-biotinylated; "second association phase"). Assays were performed at 30 °C using PBS containing 0.1% Tween 20 as diluent. Figure 3 ).

[0139] In addition, inhibition of the complement system classical pathway and complement system alternative pathway was assessed using Complement System Classical Pathway and C. Results

[0140] To confirm that the vector-derived antibody fragments were expressed and secreted, supernatants were harvested from HEK293 cells transfected with plasmids encoding the transgenes and antibody fragments containing kappa light chains were enriched from the supernatants using affinity purification with protein L beads. Western blot analysis of the enriched supernatants indicated that all transgenes produced antibody fragments.

[0141] Cell supernatants were assessed for target engagement of the antibody fragments using binding assays. The data indicated that the proteins produced by all expression constructs exhibited dual target engagement to Cls and Bb. The binding affinity of the partially purified bifunctional antibody fragments was within a factor of 2 to 10 of the purified parent anti-Cls and anti-Bb Fabs in all constructs tested.

[0142] Exemplary data are shown in Figure 3 which show that when added in the first "association" phase (for binding to Bb) the proteins produced by construct #19 Figure 2F) An increase in signal was observed when the partially purified antibody fragments were generated. When the second target, aC1s, was added in the second "association" phase, an additional increase in signal was observed Figure 3 ) All antibody fragments generated from the bifunctional constructs tested, except for Construct #5, exhibited similar levels of dual target engagement Figure 2A (see Example 2 below).

[0143] The parent monoclonal antibodies used to design the bifunctional complement inhibitors have previously been shown to inhibit the complement classical pathway (CP; see WO 2016 / 164358) or the alternative pathway (AP; see U.S. Patent 11,242,382), neither of which inhibit the lectin pathway. Using The bifunctional antibody fragments or antibody fragments were evaluated in vitro for their ability to inhibit the activity of both the CP and AP. All bifunctional antibody constructs tested inhibited IgM-stimulated CP activation and LPS-stimulated AP activation (see Example 2 below). The data show that for all constructs tested, the inhibitory activity was within 4-fold of the parent Fab.

[0144] The results demonstrate that the vector-expressed antibody fragments against complement factors Bb and C1s bind to the target complement factors and inhibit activated complement with similar efficiency as the parent individual Fab proteins. These results are unexpected because the parent antibody fragments are Fab generated using recombinant mAB technology methods, i.e., expressed in CHO cells and highly purified, in contrast to the antibody fragments generated from AAV viral pre-plasmids, which are scFab and scFV fragments, and tested as partially purified antibody fragments. Furthermore, for some constructs, the plasmid-derived antibody fragments are monocistronic, so their action is similar to bifunctional antibodies. Despite this difference in design / structure from the parent Fabs, inhibition of each target is largely preserved. Example 2: Functional characterization of anti-C1s and anti-Bb scFabs

[0145] Constructs #2 and #4 were recombinantly expressed and purified to homogeneity as described above and tested in target binding assays as well as serum-based and cell-based functional assays. Direct target binding was measured using surface plasmon resonance (SPR).

[0146] In serum-based The inhibitory activity of the scFabs was tested in enzyme immunoassays. In commercial assay kits, wells of microtiter strips are coated with specific activators of each complement system pathway. In addition, buffers and reagents included in the kits prevent cross-activation of multiple pathways, thus maintaining specificity of pathway activation. The test kit for AP is coated with lipopolysaccharide, while the test kit for CP is coated with human IgM. The final readout is detection of a neoepitope on the C5b9 complex generated by complement pathway activation, measured by colorimetry. The recombinant scFabs were also tested in a modified assay in which microtiter plates are coated with heat aggregated (HAGG) IgG and C3b to allow simultaneous activation of CP and AP; in this assay, the C5b9 complex generated by activation of both pathways is also measured by colorimetry.

[0147] In addition, the recombinant scFabs were tested in an in vitro dry AMD model based on the ARPE19 cell line. In all functional assays, the recombinant scFabs were tested individually and in equimolar mixtures, representing the vector-derived product.

[0148] Table 1 below shows the characterization of the recombinant scFabs and their comparison with the parental scFabs (#2 and #4) and mAbs. Table 1

[0149] These data show that the recombinant scFabs against both Cls and Bb display similar binding and inhibitory properties as their corresponding parental scFabs. Example 3: Properties of exemplary complement inhibitors with charge mutations

[0150] Three expression constructs were selected for further study. The first construct, #14 Figure 2D and Figure 2H contains a single-directional minCBA promoter driving expression of a single transcript encoding an anti-Cl s scFab linked to an anti-Bb scFv by a flexible (G4S)2 linker [aCl s scFab-(G4S)2-aBb scFv], followed by a bGH poly(A) signal. These sequences were human codon-optimized and contain charge mutations to promote accurate strand pairing.

[0151] The second expression construct, construct #12 Figure 2C and Figure 2J), comprising a single-directional minCBA promoter driving expression of a single transcript encoding an anti-Bb scFab linked via a flexible (G4S)3 linker to an anti-Cls scFab [aBbscFab-(G4S)3 - aC1s scFab] followed by a bGH poly(A) signal. These sequences were human codon-optimized and contained charge mutations to promote accurate strand pairing.

[0152] A third expression construct, Construct #9 Figure 2B and Figure 2I ), comprising a bi-directional minCBA promoter driving expression of different transcripts encoding human codon-optimized aBb scFab or aC1s scFab [aC1s scFab + aBb scFab], each followed by a bGH poly(A) signal. In assays performed as described in Example 1, the binding affinities of complement-binding antibody fragments expressed from Construct #9 and #14 for C1s and Bb were within 2 to 6-fold of the purified parent Fabs, while the binding affinities of complement-binding antibody fragments expressed from Construct #12 for C1s and Bb were within about 6 to 7-fold of the purified parent Fabs (Table 2). Table 2 Inhibitors AP K D (Bb, nM) CP K D (C1s, nM) #14 2.5 1.4 #12 6.6 3.6 #9 0.9 1.8 Anti-Bb Fab 0.7 N / A Anti-C1s Fab N / A 0.3

[0153] In assays, the IC 50 values of complement inhibitors derived from Construct #14 were within about 6-fold of the purified anti-C1s Fab (CP inhibition) and the purified anti-Bb Fab (AP inhibition). The IC 50 values of antibody fragments derived from #12 were within about 7-fold of the purified anti-Bb Fab (AP inhibition) and 14-fold of the purified anti-C1s (CP inhibition). The IC 50 values of antibody fragments derived from #9 were within about 3-fold of the purified anti-Bb Fab (AP inhibition) and 25-fold of the purified anti-C1s Fab (CP inhibition) (Table 3). Table 3

[0154] In addition, Constructs #2, #4, #12, and #14 Figure 2A , Figure 2C and Figure 2D ) were expressed and purified to >98% purity using chromatography (referred to as recombinant constructs) and tested head-to-head with the parent anti-C1s and anti-Bb Fabs in assays described in Example 1 to characterize the functional properties of these constructs. Constructs #2 and #4 were selected to represent the two scFabs that would be expressed and secreted from bi-directional vector Construct #9.

[0155] The results of these experiments are summarized in Table 4 below (ND: Not determined). Table 4

[0156] Table 5 below summarizes the in vitro binding and functional inhibition results of the proteins expressed by constructs #2, #4, #5, #6, #7, #8, #9, #11, #12, #13, #14, #15, #16, #17, #18, #19, and #20 from Figures 2A-2F Table 5 NA: Not applicable. ND: Not determined. * Curve not saturated; IC50estimated 50 .

[0157] In addition to direct target binding (BLI) and EIA assays, another functional assay was developed to assess the simultaneous inhibition of CP and AP by these recombinant constructs. In this assay, ELISA plates were coated with HAGG (heat aggregated gamma globulin) and C3b and incubated with 12% C1s depleted serum containing 380 ng / ml zymogen C1s to simultaneously activate CP and AP. The assay conditions were optimized to achieve similar levels of CP and AP activation on the plate. Dose responses of constructs #2 and #4 were tested alone or in equimolar mixtures (representative of the expression conditions from construct #9). An equimolar mixture of the parental anti-C1s Fab and anti-Bb Fab was also tested.

[0158] Under these conditions, constructs #2 and #4 achieved dose dependent but partial inhibition (70-85%; Figure 4A and Figure 4B ). However, when these two constructs were mixed together in equimolar proportions, resulted in >99% inhibition of complement activation, similar to what was observed with the equimolar mixture of the parental Fabs. The observed IC 50 was within 2 to 3 fold of the IC 50 observed with the equimolar mixture of the parental Fabs. Figure 4C ​). See also Table 6, which summarizes the half maximal inhibitory concentration of anti-Cls Fab, anti-Bb Fab, the protein expressed by Construct #2, the protein expressed by Construct #4, or an equimolar mixture of the two, and the maximum inhibition achieved under conditions that simultaneously activate the CP and AP in vitro. Table 6 Construct number IC 50 (nM) Maximum inhibition 2 97.3 75％-80％ 4 633.4 80％-85％ Anti-C1s Fab 68.2 75％-85％ Anti-Bb Fab 462.6 70％-75％ Equimolar mixture of constructs #2 and #4 119.8 >99％ Equimolar mixture of anti-C1s Fab and anti-Bb Fab 51.5 90％-92％ Example 4: In vivo retinal studies in mice

[0159] Based on the above in vitro results, Constructs #9, #12, and #14 were selected for in vivo studies and their ITR plasmid expression cassettes were packaged into AAV2 for delivery to target cells (see, e.g., Figure 2H , Figure 2I and Figure 2J ). This example describes in vivo testing of these vectorized antibody constructs in wild-type mouse retinas to confirm transduction of retinal ganglion cells (RGCs) and secretion of antibody fragments into the vitreous. Antibody fragments secreted into the mouse vitreous were evaluated in in vitro assays to demonstrate target engagement with human complement factors Cls and Bb. Tolerability was assessed by optical coherence tomography (OCT). A. AAV injections

[0160] More particularly, recombinant AAV2 expressing Constructs #9, #12, and #14 flanked by AAV2 ITRs were generated. AAV2 #14, AAV2 #12, and AAV2 #9 were administered to C57BL / 6J mice by intravitreal injection at three doses [10 8 , 10 9 , or 10 10 vector genomes (vg) per eye] and evaluated for retinal transduction, transgene expression, antibody secretion, and tolerability after 3-4 weeks of in vivo exposure. Recombinant AAV2 encoding a secreted VEGF inhibitor was administered at 2 x 10 9 vg per eye in parallel as a positive control. Mice injected with no vector and no transduction served as negative controls. B. Vector transduction

[0161] In quantitative PCR analysis of DNA purified from mouse retinas, vector-derived bGH poly(A) was detected using to quantify vector transduction. The data show that all three vectors successfully transduced the retina, achieving approximately 10 4 -10 5 vg per 500 ng DNA. Levels of transduction from the bifunctional antibody fragment vectors were comparable to those achieved by the positive control. There was a vector dose-dependent increase in transduction for AAV2 #14 (10 10Compared to 10 8 (p = 0.01). Similar results were observed with AAV2#12 (administration of 10...). 9 Several mice with a dose of 1 vg showed relatively low transduction levels; this may be due to technical issues when administering the dose. AAV2#9 contains two copies of bGH polymerase (A), which showed high levels of transduction at all doses.

[0162] Vector transduction and cell targeting in the mouse retina were evaluated using vector-specific probe sets in in situ hybridization (ISH) analysis of fixed, paraffin-embedded eye sections. Each probe set contained 40 pairs of probes approximately 50 bases in length. In eyes treated with each AAV2 vector, vector transduction was primarily detected in retinal ganglion cells (RGCs) and inner nuclear layer (INL) cells, with lower levels of transduction detected in outer nuclear layer (ONL) cells, and minimal transduction detected in retinal pigment epithelium (RPE) cells. Figure 5 ).

[0163] Table 7A summarizes the transduction levels (vector genome / 500 ng genomic DNA) achieved in the mouse retina 3 weeks after intravitreal administration of AAV2#9, AAV2#12, and AAV2#14 (median ± MAD). Table 7A C. Transgenic expression

[0164] Quantitative RT-PCR analysis was performed on RNA purified from mouse retina. Transgenic expression in the retina was measured by determining the bGH polymerase (A) sequence derived from the detection vector. RNA quality was assessed, and samples with an RNA integrity value (RIN) below 6 were excluded from the analysis. Data showed that, after 3 weeks of in vivo exposure, all three AAV vectors produced high levels of transgenic expression (approximately 10-1) in the retina. 5 Up to 10 6 Table 7B summarizes the transgene expression levels achieved in the mouse retina 3 weeks after intravitreal administration of AAV2#9, AAV2#12, and AAV2#14 (bGH transcripts / 500 ng RNA) (median ± MAD). Table 7B

[0165] In all samples, transcript levels correlated with vector genome levels (p = 0.59), and from 10... 9Expression levels were lower in the poorly transduced AAV2#12 retinas of the vg-treated group. AAV2#9 showed a dose-dependent increase in transgene expression (10 10 Compared to 10 8 where p = 0.036; 10 10 Compared to 10 9 where p = 0.0495). D. Antibody expression

[0166] Vector-derived human antibody fragments were detected by immunohistochemistry (IHC) using an anti-human kappa light chain antibody to evaluate expression and distribution of vector-derived complement inhibitors in mouse retinas. Data showed that inhibitors produced by all three vectors were detected in RGCs (retinal ganglion cells) and INL (inner nuclear layer) cells. E. Antibody secretion and target engagement

[0167] To demonstrate that viral vectors produce secreted bifunctional complement inhibitors, inhibitor levels in mouse vitreous humor were evaluated using ELISA methods. Vector-derived complement inhibitors present in mouse vitreous humor were quantified via target engagement capability using C1s and Bb ELISA and purified anti-C1s and anti-Bb scFabs as standards. Table 8A and Table 8B summarize ex vivo dual target engagement results (mean ± SD; ng / mL) of secreted anti-C1s (Table 8A) and anti-Bb (Table 8B) antibody fragments present in mouse vitreous humor at 3 weeks post intravitreal administration of AAV2#9, AAV2#12, and AAV2#14. Table 8A Table 8B

[0168] Overall, C1s and Bb ELISA demonstrated that all three rAAVs resulted in expression and secretion from mouse retinal ganglion cells when delivered intravitreally. Proteins expressed by all three expression vectors can bind to C1s and Bb ex vivo. Overall, data showed that all three expression vectors produced comparable levels of anti-C1s and anti-Bb binding activity in mice. Unexpectedly, retinal ganglion cells can support in vivo production of vectorized antibody fragments that exhibit similar binding properties to the parental antibodies generated in vitro using established recombinant antibody production methods.

[0169] The vitreal levels of bispecific antibody in mice treated with AAV2#14 ranged from about 150 ng / mL to about 900 ng / mL. The vitreal levels of AAV2#12 derived suppressors showed a slight dose-response between treatment groups, increasing from about 80 ng / mL to about 140 ng / mL. The levels of suppressor in the vitreous of mice treated with AAV2#9 increased in a dose-dependent manner, reaching about 1100 ng / mL at the highest dose. In addition to quantifying the levels of suppressor in the vitreous, these data demonstrate ex vivo dual target engagement of vector-derived antibody fragments.

[0170] Vector-derived complement suppressors cannot be evaluated for target engagement and effectiveness in vivo in mice because these suppressors only bind human and non-human primate (NHP) Cls and Bb and do not interact with murine complement factors.

[0171] In mice dosed with the AAV2 positive control (see above), the secretion of VEGF suppressor into the vitreous was measured by ELISA. Two weeks after in vivo exposure, the vitreal levels of VEGF suppressor averaged about 57 ng / mL. Thus, AAV2#14, AAV2#12, and AAV2#9 all produced higher levels of secreted protein than the positive control. F. Tolerability

[0172] Photoreceptor damage can be detected as a thinning of the photoreceptors. Tolerability to the viral vectors was assessed by measuring the thickness of the photoreceptor (PR) layer [outer nuclear layer (ONL) + inner segment / outer segment (IS / OS)] in the retinal optical coherence tomography (OCT) images of mice that were not transduced with a vector and mice that were transduced with a vector. The photoreceptor thickness in the retinas transduced with AAV2#14, AAV2#12, and AAV2#9 was not reduced at any dose (10 8 , 10 9 , or 10 10 vg) compared to the retinas that were not transduced with a vector, indicating that the doses and time points studied in mice had no effect on photoreceptor tolerability. Example 5: Inhibition of complement activation in a cell-based dry AMD model

[0173] C-reactive protein (CRP) is an acute phase reactant protein and also an activator of the classical complement pathway (CP). CRP binds to dying cells and activates the CP, marking these cells for clearance by phagocytes. CRP levels are elevated in inflammatory states. Elevated CRP levels have been shown to be an independent risk factor in the pathogenesis of AMD, and high serum concentrations of CRP are associated with faster progression to advanced disease in AMD and higher severity of vision loss in other retinal diseases such as retinitis pigmentosa (Chen et al., Trans Vis Sci & Techno. (2021) 10(7):7; Molins et al., Front Immunol. (2018) 9:808; and Murakami et al., Acta Ophthalmol. (2018) 96(2):e174-e179). Furthermore, staining of Bruch’s membrane, drusen, and choroidal vessel walls in the eyes of AMD patients has shown elevated levels of CRP, suggesting that complement activation during the disease is at least partially initiated by CRP (Bhutto et al., Br J Ophthalmol. (2011) 95(9):1323-30).

[0174] To reproduce some of these patient features in vitro in a cell-based model, ARPE19 cells (a retinal pigment epithelial (RPE) cell line) were treated with normal human serum (NHS) supplemented with CRP. The extent of complement activation was assessed by monitoring the levels of deposited C3 fragments and C5b9 on the cell surface using an ELISA protocol on cells. The data showed that treatment of ARPE19 cells with NHS supplemented with CRP resulted in increased levels of both C3 fragments and C5b9 on the cells compared to treatment with NHS alone, indicating that the activation of the complement system was stronger in the presence of CRP Figure 6 ) than in its absence. When a complement inhibitor was included in the treatment, the combination inhibition (anti-Cls Fab + anti-Bb Fab) resulted in a stronger reduction of C3 fragments and C5b9 levels than the inhibition levels achieved with anti-Cls Fab (CP) or anti-Bb Fab (AP) alone Figure 6 ). Example 6: New iPSC-derived AMD cell model

[0175] This example describes a new cellular model developed to demonstrate CRP-induced complement activation in AMD. This model measures complement deposition on induced pluripotent stem cell-derived retinal pigment epithelial cells (iPSC-RPE). RPE has many important roles in the eye, responsible for phagocytosis of the outer segment of photoreceptors and the transfer of nutrients from the choroid to the retina among many other important functions. Complement activation on RPE can contribute to inflammation and cell death in AMD. iPSC-RPE were chosen for this model because they maintain the morphology of native RPE and share similar cellular markers. Therefore, measuring complement deposition on the surface of these cells can mimic how certain drug treatments can limit complement activation in the retina during the course of AMD.

[0176] Complement deposition on the surface of iPSC-RPE was measured using a cellular ELISA. iPSC-RPE (FujiFilm Cellular Dynamics, Madison, WI) were grown in fibronectin-coated black / clear bottom 96-well plates. CRP (100 pg / mL) (ImmunoPrecise Antibodies, Utrecht, NL), 10% normal human serum (Complement Technology, Taylor, TX), and complement inhibitors being tested were added to the cell culture media and incubated with the iPSC-RPE overnight. The next day, cells were washed and fixed with 4% paraformaldehyde. After blocking, cells were incubated with anti-C3d or anti-C5b9 HRP-conjugated antibodies (Novus Biologicals, Centennial, CO). QuantaRed TM Enhanced chemifluorescent HRP substrate (ThermoFisher, Waltham, MA) generated a fluorescent signal that was measured using a microplate reader. Data show that treatment with anti-Cls or anti-Bb scFab alone resulted in a significant reduction in C3d and C5b9 deposition on iPSC-RPE; however, the combination of both scFabs maximally reduced complement product deposition Figure 7A and Figure 7B ).

[0177] A similar method was used to fluorescently image C5b9 deposition on iPSC-RPE. In this method, cells were grown on fibronectin-coated 24-well hanging cell culture inserts. Cells were treated with CRP, 10% normal human serum, and complement inhibitors overnight. Z-stack images were captured using a confocal microscope at 40x magnification. For image quantification, three regions of interest (ROIs) were randomly imaged from each sample. The total area of C5b9 within each ROI was calculated and averaged for each sample. The mean value of each group was measured in triplicate and error bars were calculated from the standard deviation of the mean. The fluorescent imaging experiment was repeated three times with three different iPSC-RPE cell lines. Data similarly showed that treatment with a combination of anti-Cls and anti-Bb scFab resulted in a significant reduction of C5b9 (red) staining Figure 8A and Figure 8B ).

[0178] In summary, the results from the iPSC-RPE model suggest that both the classical and alternative pathways can play a role in the pathogenesis of AMD. Blocking each pathway separately resulted in a reduction of complement deposition on RPE cells. However, simultaneous inhibition of both pathways resulted in the greatest reduction of deposition, suggesting that simultaneous inhibition of the classical and alternative pathways can be beneficial for AMD. Example 7: In vivo retinal studies in non-human primates

[0179] This example describes in vivo testing of exemplary vectorized antibody constructs in non-human primates (NHPs) to confirm transduction and transgene expression in the retina. The activity of viral vectors following intravitreal administration in NHPs was evaluated in two studies: (1) a 6-week dose ranging study of AAV2#14 and AAV2#12, and (2) an 8-week single dose study of AAV2#14 and AAV2#9. In each study, NHPs administered with ocular formulation buffer served as controls. A. Study 1

[0180] In the first study, NHPs were administered by intravitreal injection with ocular formulation buffer (N = 2 NHPs) or three doses (2 x 1011vg / eye, based on vector titer determined using an assay to detect BGH poly(A); N = 3 NHPs per treatment group) of AAV2#14 or AAV2#12. Animals were evaluated six weeks after in-life exposure. 9 10 11

[0181] To evaluate vector transduction, vector-specific ​​​Quantification of the vector genome level was performed. TUBB1 was used as a reference gene to confirm comparable DNA input between samples. Data showed that both AAV2#12 and AAV2#14 successfully transduced the NHP retina, resulting in a dose-dependent increase in vector genome levels (dose-response AAV2#14 p = 0.0286, AAV2#12 p = 0.0095). Table 9 below summarizes the transduction levels achieved in the NHP retina at 6 weeks after intravitreal administration (median vector genome / 500 ng genomic DNA). Table 9

[0182] To evaluate transgene expression, transcript-specific methods were used in quantitative RT-PCR analysis of RNA purified from the NHP retina. The levels of vector-derived transgenes were quantified. RNA quality was assessed, and all samples showed an RNA integrity value (RIN) greater than 7.5. One sample was not included in the RNA analysis due to low RNA input. Transcript levels were quantified relative to a double-stranded plasmid DNA standard curve. Data showed that transduction of both AAV2#12 and AAV2#14 resulted in dose-dependent levels of transgene expression in the NHP retina (dose-response AAV2#14 p = 0.0286, AAV2#12 p = 0.0286). Table 10 below summarizes the transcript abundance (median transcripts / 500 ng RNA) in the NHP retina at 6 weeks post-intravitreal administration. Table 10 B. Study 2

[0183] In the second study, ocular preparation buffer was administered to NHPs via intravitreal injection (N = 2 NHPs) or 2 x 10⁻⁶ NHPs. 11 One vg / eye of AAV2#14 or AAV2#9 (N = 3 NHPs / vector treatment group). Vector titer was determined based on the assay of BGH aggregate (A). Animals were evaluated within 8 weeks of in vivo exposure. Due to the presence of serum AAV2 neutralizing antibody (Nab), IgG degrading enzyme (IdeS) was administered intravitreally to all 2 study NHPs 2 days prior to vector administration.

[0184] To evaluate vector transduction, vector-specific methods were used in the quantitative PCR analysis of DNA purified from the NHP retina. Vector genome levels were assayed. For AAV2#9, two different assays were used to assess vector genome levels, one anti-Bb and one anti-Cls arm. Using TUBB as a reference gene, comparable DNA input between samples was confirmed. Despite potential hindrance by pre-existing AAV2 Nab, data showed that both AAV2#14 and AAV2#9 successfully transduced the NHP retina, with AAV2#14 reaching levels of about 9.3 x 10 3 vg and AAV2#9 reaching levels of about 7.6 x 10 4 to about 2.8 x 10 5 vg (median vector genomes / 500 ng of genomic DNA) at 8 weeks post-intravitreal administration.

[0185] To assess transgene expression, transcript-specific Vector-derived transgene levels were assayed. For AAV2#9, anti-Bb and anti-Cls transcripts were expressed independently, and were thus assessed separately. RNA quality was assessed, and all samples showed RNA integrity values (RIN) greater than 7.5. Transcript levels were quantified relative to a standard curve of double-stranded plasmid DNA. AAV2#14 produced an abundance level of about 9.7 x 10 4 transcripts and AAV2#9 produced about 1.6 x 10 6 x 10 5 transcripts (median transcripts / 500 ng of retinal RNA) at 8 weeks post- vivo exposure. C. Persistence studies in NHPs

[0186] The pharmacology and persistence of multiple dose levels of AAV2#9 were evaluated in a 16-week in vivo evaluation study, including a 6-week mid-study necropsy.

[0187] NHPs (cynomolgus monkeys) were administered formulation buffer (180 mM NaCl, 5 mM sodium phosphate, 0.01% PS20, pH 7.4) or multiple dose levels of AAV2#9 (based on vector titer determined by microdroplet digital PCR (ddPCR) using a vector-specific assay targeting the anti-Cls region of AAV2#9) via bilateral intravitreal injection. All NHPs were given prophylactic steroids (1 mg / kg oral prednisolone daily) starting two weeks prior to vector administration and continued throughout the study. Vector-specific C1s and Bb Vector genome levels in NHP retinas were assayed.

[0188] C1s and Bb assays detected comparable vector genome levels in each sample at 6 and 16 weeks time points. At 6 weeks, AAV2#9 transduction resulted in a dose-dependent increase in vector genome levels in the retina. A dose-dependent increase in retinal transduction was also observed at 16 weeks.

[0189] RNAscope TM In ISH analysis, vector biodistribution in NHP eyes was assessed using an AAV2#9 vector-specific probe set (comprising 40 pairs of probes, each spanning approximately 50 bases, designed to detect the sense strand of the vector genome). At 6 and 16 weeks, vector was detected in the retina and irido-ciliary body of eyes administered with AAV2#9. No vector was detected in the optic nerve. In the retina, vector was present in rare cells of the RGC and INL, and was generally present in the foveal and parafoveal regions of the macula.

[0190] In quantitative RT-PCR analysis performed on RNA purified from the right eye, C1s and Bb-specific Assays quantified levels of AAV2#9-derived anti-C1s and anti-Bb transcripts in the NHP retina. Transcript levels were quantified relative to a standard curve of double-stranded plasmid DNA. Transcript levels in the 6- and 16-week cohorts were highly correlated with vector genome levels (Spearman’s r > 0.97). At 6 weeks, AAV2#9 transduction resulted in a dose-dependent increase in transcript levels in the retina. A dose-dependent trend in increasing transcript levels was also observed at 16 weeks.

[0191] To assess the kinetics and persistence of peak scFab expression, aqueous humor was collected at baseline and at weeks 3, 6, 12, and 16. Vitreous humor was collected at necropsy. In aqueous humor collected from some NHPs in the 16-week cohort, scFab levels peaked between 3-6 weeks and persisted to the end of the study at 4 months (day 113). scFab levels in the vitreous humor at 4 months were similar to or higher than levels in the aqueous humor. D. Evaluation of AAV2#9 effectiveness in inhibiting LPS-induced complement activation and ocular inflammation in NHPs

[0192] An acute model of endotoxin-induced inflammation was used to assess the ability of AAV2#9-derived scFab to inhibit complement pathway activation in vivo.

[0193] NHPs (cynomolgus monkeys) were administered formulation buffer (180 mM NaCl, 5 mM sodium phosphate, 0.01% PS20, pH 7.4) or AAV2#9 via bilateral intravitreal injection, followed by bilateral intravitreal administration of lipopolysaccharide (LPS) [0.5 endotoxin units (EU) LPS / eye, from E. coli O111:B4; Sigma-Aldrich L4391] on day 41. NHPs were given prophylactic steroids (1 mg / kg oral prednisolone daily) starting two weeks prior to vector administration and continued daily for four weeks. Prednisolone was tapered off the NHPs prior to LPS administration on day 41. Figure 9

[0194] Free drug levels of AAV2#9-derived scFabs in aqueous and vitreous humor were measured using Bb and C1s target capture ELISAs. Median aqueous and vitreous humor levels of free anti-C1s scFab were approximately 50-100 ng / mL (1-2 nM) and median levels of free anti-Bb scFab reached approximately 100-200 ng / mL (2-4 nM) at the end of the study on day 43. Anti-C1s scFab levels in aqueous and vitreous humor were higher than the equilibrium dissociation constant of anti-C1s scFab for human and cynomolgus monkey C1s (human K D = 0.34 nM; cynomolgus monkey K D = 0.016 nM). Anti-Bb scFab had lower affinity for cynomolgus monkey Bb (K D = 14.8 nM) compared to human Bb (K D = 3.7 nM), and anti-Bb scFab levels reached in aqueous and vitreous humor in this study were lower than the K D of anti-Bb scFab for cynomolgus monkey Bb and were not sufficient to inhibit Bb in the NHP eye.

[0195] To assess complement pathway activation, we used a multiplex ELISA from Quidel to measure activated fragment levels of C4a (classical pathway), Ba (alternative pathway), and sC5b9 (terminal pathway) in aqueous humor. Levels of Ba, C4a, and sC5b9 in the aqueous humor of LPS-treated eyes were increased compared to LPS-untreated control eyes, indicating activation of the alternative, classical, and terminal pathways. C4a and sC5b9 levels were decreased in LPS-treated eyes dosed with AAV2#9 compared to LPS-treated control eyes, indicating inhibition of the classical and terminal pathways. Inhibition of the alternative pathway (Ba) was not detected in AAV2#9-treated eyes, likely due to the lower affinity of anti-Bb scFab for the cynomolgus monkey target.​

[0196] Ocular examinations performed two days after LPS administration detected ocular inflammation in all treatment groups. However, the severity and frequency of clinical indicators of inflammation in eyes treated with AAV2#9 was reduced (scored using the SPOTS system). Sequence SEQ ID NO: 1 - HCDR1 of anti-Cls antibody DDYIH SEQ ID NO: 2 - HCDR2 of anti-Cls antibody RIDPADGHTKYAPKFQV SEQ ID NO: 3 - HCDR3 of anti-Cls antibody YGYGREVFDY SEQ ID NO: 4 - LCDR1 of anti-Cls antibody KASQSVDYDGDSYMN SEQ ID NO: 5 - LCDR2 of anti-Cls antibody DASNLES SEQ ID NO: 6 - LCDR3 of anti-Cls antibody QQSNEDPWT SEQ ID NO: 7 - V of anti-Cls antibody H (Kabat CDRs underlined) SEQ ID NO: 8 - V of anti-Cls antibody L (Kabat CDRs underlined) SEQ ID NO: 9 - aCIs scFv SEQ ID NO: 10 - Heavy chain of anti-Cls Fab SEQ ID NO: 11 - Light chain of anti-Cls Fab SEQ ID NO: 12 - aCIs scFab SEQ ID NO: 13 - HCDR1 of anti-Bb antibody NYAMS SEQ ID NO: 14 - HCDR2 of anti-Bb antibody TISNRGSYTY YPDSVKG SEQ ID NO: 15 - HCDR3 of anti-Bb antibody ERPMDY SEQ ID NO: 16 - LCDR1 of anti-Bb antibody KASQDVGTAVA SEQ ID NO: 17 - LCDR2 of anti-Bb antibody WASTRHT SEQ ID NO: 18 - LCDR3 of anti-Bb antibody HQHSSNPLT SEQ ID NO: 19 - V of anti-Bb antibody H (Kabat CDRs boxed) SEQ ID NO: 20 - V of anti-Bb antibody L (Kabat CDRs boxed) SEQ ID NO: 21 - aBb scFv SEQ ID NO: 22 - Heavy chain of anti-Bb Fab SEQ ID NO: 23 - Light chain of anti-Bb Fab SEQ ID NO: 24 - aBb scFab SEQ ID NO: 25 - [aC1s scFab - (G4S)3 - aBb scFab] nucleic acid sequence (Construct #7, Figure 2A ) SEQ ID NO: 26 - [aC1s scFab - (G4S)3 - aBb scFab] amino acid (signal peptide in bold) (Construct #7, Figure 2A ) SEQ ID NO: 27 - [aC1s scFab - (G4S)3 - aBb scFab-CM] nucleic acid sequence (Construct #11,Figure 2C ) SEQ ID NO:28–[αC1s scFab–(G4S)3–αBb scFab-CM] amino acid sequence (constructor #11, Figure 2C (Signal peptides are bolded; charge mutations are boxed and shown in italics, and the numbers do not include the signal peptides: Q42E and Q292K in αC1s scFab, and Q523K, S599A, N622K, Q773E, and T919E in αBb scFab) SEQ ID NO:29–[αBb scFab–(G4S)3–αC1s scFab] nucleic acid sequence (constructor #8, Figure 2A ) SEQ ID NO:30–[αBb scFab–(G4S)3–αC1s scFab] amino acid sequence (constructor #8, Figure 2A (Signal peptide bolded) SEQ ID NO:31–[αBb scFab–(G4S)3-αC1s scFab-CM] nucleic acid sequence (constructor #12, Figure 2C ) SEQ ID NO:32 [αBb scFab–(G4S)3-αC1s scFab-CM] amino acid sequence (constructor #12, Figure 2C (Signal peptides are bolded; charge mutations are boxed and shown in italics, and the numbers do not include the signal peptides: Q38K, S114A, N137K, Q288E, and T434E in αBb scFab, and Q520E and Q770K in αC1s scFab) SEQ ID NO:33–[αC1s scFab–(G4S)2–αBb scFv] nucleic acid sequence (constructor #13, Figure 2D ) SEQ ID NO: 34 - [aC1s scFab - (G4S)2 - aBb scFv] amino acid sequence (Construct #13, Figure 2D )(signal peptide in bold) SEQ ID NO: 35 - [aC1s scFab - (G4S)2 - aBb scFv-CM] nucleic acid sequence (Construct #14, Figure 2D ) SEQ ID NO: 36 - [aC1s scFab - (G4S)2 - aBb scFv-CM] amino acid sequence (Construct #14, Figure 2D )(signal peptide in bold; charge mutations boxed and shown in italics, numbering does not include signal peptide: Q42E and Q292K in aC1s scFab, and Q519E and Q648K in aBb scFv) SEQ ID NO: 37 - [aC1s scFab - bidirectional promoter - aBb scFab] nucleic acid sequence (Construct #9, Figure 2B ) SEQ ID NO: 38 - [aC1s F2A Fab - (G4S)3 - aBb scFab] nucleic acid sequence (Construct #17, Figure 2F ) SEQ ID NO: 39 - [aC1s F2A Fab - (G4S)3 - aBb scFab] amino acid sequence (Construct #17, Figure 2F )(signal peptide in bold; furin cleavage site underlined; F2A sequence shown in italics) SEQ ID NO: 40 - [aC1s GT2A Fab - (G4S)3 - aBb scFab] nucleic acid sequence (Construct #18, Figure 2F ) SEQ ID NO: 41 - [aCIs GT2A Fab - (G4S)3- aBb scFab] amino acid sequence (Construct #18, Figure 2F )(signal peptide is bold; furin cleavage site is underlined; GT2A sequence is in italics) SEQ ID NO: 42 - [aCIs scFab - (G4S)3- aBb F2A Fab] nucleic acid sequence (Construct #19, Figure 2F ) SEQ ID NO: 43 - [aCIs scFab - (G4S)3- aBb F2A Fab] amino acid sequence (Construct #19, Figure 2F )(signal peptide is bold; furin cleavage site is underlined; F2A sequence is in italics) SEQ ID NO: 44 - [aCIs scFab - (G4S)3- aBb GT2A-Fab] nucleic acid sequence (Construct #20, Figure 2F ) SEQ ID NO: 45 - [aCIs scFab - (G4S)3- aBb GT2A-Fab] amino acid sequence (Construct #20, Figure 2F )(signal peptide is bold; furin cleavage site is underlined; GT2A sequence is in italics) SEQ ID NO: 46 GGGGS SEQ ID NO: 47 GGGGSGGGGS GGGGS SEQ ID NO: 48 GGGGSGGGGS GGGGSGGGGS GGGGSGGGGS GGGGS SEQ ID NO: 49 GGGGSGGGGS GGGGSGGGGS GGGGSGGGGS GGGGS SEQ ID NO: 50 - nucleotide sequence of AAV2#9 Figure 2B andFigure 2I (5' ITR is bold; bGH poly A signal is underlined; reverse complement of aC1sscFab coding sequence is shown in italics; IgG kappa signal coding sequence is shown in italics and underlined; Kozak sequence is boxed; CBA promoter (inverted) is bold and underlined; CBA promoter is boxed and underlined; CMV enhancer is bold and shown in italics; aBb scFab is bold, shown in italics and underlined; and 3' ITR is boxed and shown in italics) SEQ ID NO: 51 - nucleotide sequence of AAV2#12 Figure 2C and Figure 2J (5' ITR is bold; minCBA promoter (which includes a CMV enhancer, CBA promoter, and truncated chimeric intron) is underlined; Kozak sequence is boxed; IgG kappa signal coding sequence is shown in italics; aBb scFab coding sequence is bold and underlined; (G4S)7 linker coding sequence is lower case, bold, and shown in italics; (G4S)3 linker coding sequence is boxed and shown in italics; aC1s scFab coding sequence is bold and shown in italics; bGH poly A is shown in italics and underlined; and 3' ITR is boxed and bold) SEQ ID NO: 52 - nucleotide sequence of AAV2#14 Figure 2D and Figure 2H (3' ITR is bold; minCBA promoter is underlined; Kozak sequence is boxed; IgG kappa signal sequence is shown in italics; aC1s scFab coding sequence is bold and underlined; (G4S)2 linker coding sequence is boxed and shown in italics; (G4S)7 coding sequence is lower case, bold, and shown in italics; aBb scFv coding sequence is bold and shown in italics; bGH poly A signal is shown in italics and underlined; and 5' ITR is boxed and bold) SEQ ID NO: 53 - bidirectional promoter and CMV enhancer (CBA promoter (inverted) is bold and underlined; CMV enhancer is bold and shown in italics; CBA promoter is boxed and underlined) AAGCGAAGCG CGCGGCGGGC G SEQ ID NO:54 - [aC1s scFv - (G4S)2 - aBb scFv] nucleic acid sequence (Construct #5, Figure 2A ) SEQ ID NO:55 - [aC1s scFv - (G4S)2 - aBb scFv] amino acid sequence (Construct #5, Figure 2A ) (signal peptide in bold) SEQ ID NO:56 - [aBb scFv - (G4S)2 - aC1s scFv] nucleic acid sequence (Construct #6, Figure 2A ) SEQ ID NO:57 - [aBb scFv - (G4S)2 - aC1s scFv] amino acid sequence (Construct #6, Figure 2A ) (signal peptide in bold) SEQ ID NO:58 - [aC1s scFab - (G4S)3 - aBb scFv] nucleic acid sequence (Construct #15, Figure 2E ) SEQ ID NO:59 - [aC1s scFab - (G4S)3 - aBb scFv] amino acid sequence (Construct #15, Figure 2E ) (signal peptide in bold) SEQ ID NO:60 - [aC1s scFab - (G4S)3 - aBb scFv-CM] nucleic acid sequence (Construct #16, Figure 2E ) SEQ ID NO:61 - SEQ ID NO:60 - [aC1s scFab - (G4S)3 - aBb scFv-CM] amino acid sequence (Construct #16, Figure 2E ) (signal peptide in bold; charge mutations boxed and shown in italics, numbering does not include signal peptide: Q42E and Q292K in aC1s scFab, and Q524E and Q653K in aBb scFv) SEQ ID NO: 62 - [aC1s scFab - bi-directional promoter - aBb scFab - CM] nucleic acid sequence (construct #21, Figure 2G ) SEQ ID NO: 63 - aC1s scFab - CM set (construct #22, Figure 2G ) (signal sequence in bold; charge mutations in boxes and shown in italics, numbering does not include signal peptide: Q42E and Q292K) SEQ ID NO: 64 - aBb scFab - CM set (construct #22, Figure 2G ) (signal sequence in bold; charge mutations in boxes and shown in italics, numbering does not include signal peptide: Q38K and Q288E, and S114A, N137K and T434E) SEQ ID NO: 65 - human complement C1s amino acid sequence before processing and activation (signal sequence in bold) SEQ ID NO: 66 - human complement factor B before processing and activation (signal peptide in bold) SEQ ID NO: 67 - HCDR1 of anti-C1s antibody GFNIKDDY SEQ ID NO: 68 - HCDR2 of anti-C1s antibody ID PADGHT SEQ ID NO: 69 - HCDR3 of anti-C1s antibody ARYGYGREVFDY SEQ ID NO: 70 - LCDR1 of anti-C1s antibody QSVDYDGDSY SEQ ID NO: 71 - HCDR1 of anti-C1s antibody (Jossia) GFNIKDD SEQ ID NO: 72 - HCDR2 of anti-C1s antibody (Jossia) DPADGH SEQ ID NO: 73 - HCDR1 of anti-Bb antibody GFTF SNYA SEQ ID NO: 74 - HCDR2 of anti-Bb antibody ISNRGS YT SEQ ID NO: 75 - HCDR3 of anti-Bb antibody ARERPMDY SEQ ID NO: 76 - LCDR1 of anti-Bb antibody QDVGTA SEQ ID NO: 77 - HCDR1 of anti-Bb antibody (Josephia) GFTFSNY SEQ ID NO: 78 - HCDR2 of anti-Bb antibody (Josephia) SNRGSY SEQ ID NO: 79 - [aBb scFab - bi-directional promoter - aC1s scFab-CM] nucleic acid sequence (Construct #22, Figure 2G ) SEQ ID NO: 80 [aBb scFab - bi-directional promoter - aC1s scFab] nucleic acid sequence (Construct #10, Figure 2B ) SEQ ID NO: 81 - Peptide linker SGSG SEQ ID NO: 82 - Furin cleavage site RX1X2R, where X1 = any naturally occurring amino acid, and X2 = R or K SEQ ID NO: 83 - minCBA promoter (CMV enhancer underlined; CBA promoter bold and italicized; truncated chimeric intron: bold and underlined)

Claims

1. A single expression construct comprising a first nucleotide sequence encoding an inhibitor of activated complement subcomponent Cls (Cl s inhibitor) and a second nucleotide sequence encoding an inhibitor of complement factor Bb (Bb inhibitor); or a pair of expression constructs, wherein one comprises the first nucleotide sequence and the other comprises the second nucleotide sequence.

2. The expression construct or pair of expression constructs of claim 1, wherein the Cl s inhibitor and the Bb inhibitor are each an antibody fragment, optionally wherein the antibody fragment is a single chain Fv (scFv) or a single chain Fab (scFab).

3. The expression construct or pair of expression constructs of claim 2, wherein (a) the Cl s inhibitor is an anti-Cl s antibody fragment comprising heavy chain CDRs (HCDRs) 1-3 in SEQ ID NO: 7, which optionally comprise SEQ ID NOs: 1-3, respectively, and light chain CDRs (LCDRs) 1-3 in SEQ ID NO: 8, which optionally comprise SEQ ID NOs: 4-6, respectively; and / or (b) the Bb inhibitor is an anti-Bb antibody comprising HCDRs 1-3 in SEQ ID NO: 19, which optionally comprise SEQ ID NOs: 13-15, respectively, and LCDRs 1-3 in SEQ ID NO: 20, which optionally comprise SEQ ID NOs: 16-18, respectively.

4. The expression construct or pair of expression constructs of claim 3, wherein (a) the Cl s inhibitor comprises a heavy chain variable domain (VH) comprising the amino acid sequence of SEQ ID NO: 7 or an amino acid sequence at least 95% identical thereto H ), and a heavy chain variable domain (VH) comprising the amino acid sequence of SEQ ID NO: 7 or an amino acid sequence at least 95% identical to SEQ ID NO:

7. L ) and / or (b) the Bb inhibitor comprises V comprising SEQ ID NO: 19 or an amino acid sequence at least 95% identical thereto H and V comprising SEQ ID NO: 20 or an amino acid sequence at least 95% identical thereto L .

5. The expression construct or pair of expression constructs of claim 3 or 4, wherein (a) the Cl s inhibitor comprises a heavy chain (HC) comprising SEQ ID NO: 10 or an amino acid sequence at least 95% identical thereto, and a light chain (LC) comprising SEQ ID NO: 11 or an amino acid sequence at least 95% identical thereto; and / or (b) the Bb inhibitor comprises a HC comprising SEQ ID NO: 22 or an amino acid sequence at least 95% identical thereto, and a LC comprising SEQ ID NO: 23 or an amino acid sequence at least 95% identical thereto.

6. The expression construct or pair of expression constructs of any one of claims 2-5, wherein the Cl s inhibitor and the Bb inhibitor each comprise one or more charge mutations to promote pairing between the heavy chain and the light chain of each inhibitor.

7. The expression construct or pair of expression constructs of claim 6, wherein (a) the charge mutations in the Cl s inhibitor comprise Q42E and Q292K, wherein the numbering is consistent with SEQ ID NO: 12; and (b) the charge mutations in the Bb inhibitor comprise Q38K and Q288E, optionally further comprising S114A, N137K, and T434E, wherein the numbering is consistent with SEQ ID NO:

24.

8. The one or pair of expression constructs of any one of claims 2-7, wherein the Cls inhibitor is an scFv or scFab, wherein the HC and the LC are connected by a peptide linker, optionally wherein the peptide linker comprises one or more, optionally 2, 3, 4, 5, 6, 7, 8, 9, or 10 G4S (SEQ ID NO: 46) repeat sequences.

9. The one or pair of expression constructs of any one of claims 2-8, wherein the Bb inhibitor is an scFv or scFab, wherein the HC and the LC are connected by a peptide linker, optionally wherein the peptide linker comprises one or more, optionally 2, 3, 4, 5, 6, 7, 8, 9, or 10 G4S (SEQ ID NO: 46) repeat sequences.

10. The single expression construct of any one of claims 1-9, comprising a transgene encoding a fusion protein comprising the Cls inhibitor and the Bb inhibitor connected by a peptide linker, optionally wherein the peptide linker comprises one or more, optionally 2, 3, 4, 5, 6, 7, 8, 9, or 10 G4S (SEQ ID NO: 46) repeat sequences, further optionally wherein the transgene is operably linked to a minimal chicken beta-actin (minCBA) promoter.

11. The single expression construct of any one of claims 1-9, wherein the expression construct comprises a bi-directional promoter that directs expression of the Cls inhibitor and the Bb inhibitor as separate molecules, optionally wherein the bi-directional promoter is a pair of chicken beta-actin (CBA) promoters placed in opposite directions and separated by a CMV enhancer, further optionally wherein the bi-directional promoter comprises SEQ ID NO: 53 or a nucleotide sequence at least 85% identical thereto.

12. The single expression construct of any one of claims 2-9, wherein the expression construct expresses a heterodimer comprising (i) a fusion protein comprising a single chain anti- Cls antibody fragment fused to a HC or LC of an anti-Bb antibody fragment; and (ii) a LC or HC polypeptide of the anti-Bb antibody fragment, wherein the coding sequence of the fusion protein and the coding sequence of the LC or HC polypeptide of the anti-Bb antibody fragment are separated in frame by a coding sequence of a cleavable peptide, optionally wherein the cleavable peptide comprises a 2A sequence and / or a furin cleavage site, further optionally the expression construct comprises a minCBA promoter.

13. The single expression construct of any one of claims 2-9, wherein the expression construct expresses a heterodimer comprising (i) a fusion protein comprising a single chain anti-Bb antibody fragment fused to a HC or LC of an anti-Cl s antibody fragment; and (ii) a LC or HC polypeptide of the anti-Cl s antibody fragment, wherein the coding sequence of the fusion protein and the coding sequence of the LC or HC polypeptide of the anti-Cl s antibody fragment are separated in frame by a coding sequence of a cleavable peptide, optionally wherein the cleavable peptide comprises a 2A sequence and / or a furin cleavage site, further optionally the expression construct comprises a minCBA promoter.

14. The single expression construct of any one of claims 2-10, wherein the expression construct encodes a fusion protein comprising, from N-terminus to C-terminus (i) an anti-Cls scFv, a (G4S)2linker, and an anti-Bb scFv, optionally comprising SEQ ID NO: 55 (with or without signal peptide) or an amino acid sequence at least 95% identical thereto; (ii) an anti-Bb scFv, a (G4S)2linker, and an anti-Cls scFv, optionally comprising SEQ ID NO: 57 (with or without signal peptide) or an amino acid sequence at least 95% identical thereto; (iii) an anti-Cls scFab, a (G4S)3linker, and an anti-Bb scFab, optionally comprising SEQ ID NO: 26 or 28 (with or without signal peptide) or an amino acid sequence at least 95% identical thereto; (iv) an anti-Bb scFab, a (G4S)3linker, and an anti-Cls scFab, optionally comprising SEQ ID NO: 30 or 32 (with or without signal peptide) or an amino acid sequence at least 95% identical thereto; (v) an anti-Cls scFab, a (G4S)2linker, and an anti-Bb scFv, optionally comprising SEQ ID NO: 34 or 36 (with or without signal peptide) or an amino acid sequence at least 95% identical thereto; or (vi) an anti-Cls scFab, a (G4S)3linker, and an anti-Bb scFv, optionally comprising SEQ ID NO: 59 or 61 (with or without signal peptide) or an amino acid sequence at least 95% identical thereto.

15. The expression construct or pair of expression constructs of any one of claims 2-9 and 11, wherein the expression construct or expression constructs encode an anti-Cls scFab, optionally comprising SEQ ID NO: 12 or an amino acid sequence at least 95% identical thereto, optionally wherein the amino acid sequence comprises Q42E and Q292K mutations relative to SEQ ID NO: 12; and an anti-Bb scFab, optionally comprising SEQ ID NO: 14 or an amino acid sequence at least 95% identical thereto, wherein the amino acid sequence comprises Q38K and Q288E, and optionally S114A, N137K, and T434E mutations relative to SEQ ID NO:

14.

16. The single expression construct of any one of claims 2-9, 12, and 13, wherein the expression construct encodes a heterodimer consisting of (A) (i) an anti-Cls LC and (ii) a fusion protein comprising an anti-Cls HC fused to an aBb scFab, optionally wherein the expression construct comprises the coding sequence of SEQ ID NO: 39 or an amino acid sequence at least 95% identical thereto; (B) (i) an anti-Cls LC and (ii) a fusion protein comprising an anti-Cls HC fused to an anti-Bb scFab, optionally wherein the expression construct comprises the coding sequence of SEQ ID NO: 41 or an amino acid sequence at least 95% identical thereto; (C) (i) a fusion protein comprising an anti-Cls scFab fused to an anti-Bb HC and (ii) an anti-Bb LC, optionally wherein the expression construct comprises the coding sequence of SEQ ID NO: 43 or an amino acid sequence at least 95% identical thereto; or (D) (i) a fusion protein comprising an anti-Cls scFab fused to an anti-Bb HC and (ii) an anti-Bb LC, optionally wherein the expression construct comprises the coding sequence of SEQ ID NO: 45 or an amino acid sequence at least 95% identical thereto.

17. An isolated nucleic acid comprising a nucleotide sequence selected from SEQ ID NO: 25, 27, 29, 31, 33, 35, 37, 38, 40, 42, 54, 56, 58, 60, 62, 79, or 80 or encoding one or more amino acid sequences identical to the selected nucleotide sequence.

18. One, two, or more recombinant adeno-associated viruses (rAAV) comprising one or a pair of expression constructs according to any one of claims 1-16 or the isolated nucleic acid of claim 17.

19. The one or more rAAV of claim 18, wherein the genome of the or each rAAV comprises the expression construct flanked by AAV2 inverted terminal repeat (ITR) sequences.

20. The one or more rAAV of claim 19, wherein the genome comprises SEQ ID NO: 50, 51, or 52; or encoding one or more amino acid sequences identical to SEQ ID NO: 50, 51, or 52.

21. The one or more rAAV of any one of claims 18-20, comprising a capsid of AAV2, optionally wild-type AAV2.

22. A pharmaceutical composition comprising the one or more rAAV of any one of claims 18-21 and a pharmaceutically acceptable carrier.

23. One or more proteins encoded by the one or pair of expression constructs or one or more rAAV of any one of claims 1-21.

24. A host cell comprising the one or pair of expression constructs, isolated nucleic acid, or one or more rAAV of any one of claims 1-21.

25. A method for treating dry age-related macular degeneration (AMD) in a patient in need thereof, the method comprising administering an effective amount of the one or more rAAV of any one of claims 18-21 or the pharmaceutical composition of claim 22.

26. The method of claim 25, wherein the administration is by intravitreal injection.

27. The method of claim 25 or 26, wherein the patient has geographic atrophy (GA) secondary to dry AMD.

28. The method of any one of claims 25-27, wherein the effective amount is 10 7 to 10 15 , optionally 10 8 to 10 14 , 10 9 to 10 13 , further optionally 2 x 10 9 , 2 x 10 10 , or 2 x 10 11 vector genomes.

29. The one or more recombinant AAVs of any one of claims 18-21 or the pharmaceutical composition of claim 22 for use in treating dry age-related macular degeneration (AMD) in a patient in need thereof in a method according to any one of claims 25-28.

30. Use of the one or more recombinant AAVs of any one of claims 18-21 or the pharmaceutical composition of claim 22 for the manufacture of a medicament for treating dry age-related macular degeneration (AMD) in a patient in need thereof in a method according to any one of claims 25-28.

31. A mammalian promoter comprising SEQ ID NO: 83 or a sequence having at least 85% identity thereto.

32. A bidirectional mammalian promoter comprising a pair of chicken beta-actin promoters placed in opposite directions, separated by a CMV enhancer, optionally wherein the bidirectional mammalian promoter comprises SEQ ID NO: 53 or a sequence having at least 85% identity thereto.

Images