Recombinant adeno-associated virus (rAAV) vectors are widely used for in vivo gene therapy, yet their potential to integrate into the host genome raises concerns about insertional mutagenesis and oncogenic risk, particularly in the liver where vector exposure is highest. To address this, we analyzed rAAV integration patterns in primary human hepatocytes xenografted into FRG mouse livers and in hepatocytes from cynomolgus macaques following systemic rAAV administration. High-resolution integration site mapping yielded approximately 1.5 million and 1.3 million unambiguously mapped sites in human and macaque genomes, respectively. Both datasets revealed a bias toward integration within transcriptionally active genes and regions of open chromatin, consistent with previous reports, but no particular preference for genes implicated in hepatocellular carcinoma was observed. While numerous common integration sites (CISs) were identified, their distribution differed between species. Notably, a CIS was observed at the AAVS1 locus in human hepatocytes, raising the possibility of Rep-mediated integration. These findings highlight the need for continued monitoring of integration events in clinical settings. Overall, the data support a low oncogenic risk profile for the evaluated vector while reinforcing the value of direct human liver integration analyses to refine risk assessment and guide the development of safer gene therapy platforms.

Here, we report highly efficient functional repair of the ornithine transcarbamylase (OTC) locus in mutant mouse and human hepatocytes in vivo using a dual adeno-associated virus system delivering CRISPR-Cas9 editing reagents and a promoterless donor for targeted integration. The approach was mutation agnostic and targeted intronic sequences to prevent inadvertent inactivation of hypomorphic alleles. Notably, in a murine model, we corrected the metabolic defect and simultaneously achieved liver-wide restoration of physiological metabolic zonation of Otc expression by capturing native cis-acting regulatory elements. The effectiveness of this approach was confirmed using a universally configured therapeutic cassette in patient-derived primary human hepatocytes in vivo. These data provide a powerful template to guide further optimization of this approach and, given the high editing efficacy required for phenotypic effect in OTC deficiency, have broader relevance to other liver disease phenotypes.

Complement activation contributes to kidney damage in many types of glomerulonephritis and complement inhibition therapy is approved for IgA nephropathy and C3 glomerulopathy. However, inhibition is not specific to the kidney resulting in unnecessary systemic complement inhibition and increased infection risk. To develop an effective inhibitor of glomerular complement we combined complement factor H-related protein 5 (FHR51-9), which binds to glomerular complement C3, with the complement regulatory domains of the key negative regulator of C3 activation complement factor H (FH1-5). One week after adeno-associated virus (AAV) mediated expression of the FHR51-9FH1-5 fusion protein in factor H (FH)-deficient mice, glomerular C3b/iC3b/C3c was significantly reduced and properdin resolved completely compared to controls. There was no change to circulating C3 levels and FHR51-9FH1-5 was detected in glomeruli in association with C3d. Six and twenty weeks after AAV8-FHR51-9FH1-5 treatment in hFH-FHR5mut mice (a mouse model of CFHR5 nephropathy) glomerular C3b/iC3b/C3c, C3d, and C5 were significantly reduced and properdin resolved completely compared to controls. Pre-administration of AAV-FHR51-9FH1-5 also ameliorated abnormal glomerular C3b/iC3b/C3c, C3d, C5, and properdin in a triggered CFHR5 nephropathy model. In vitro FHR51-9FH1-5 showed dose-dependent binding to surface-immobilized C3, C3b, iC3b, and C3d; factor I cofactor activity; and reduced C3a generation in an alternative pathway convertase assay. Taken together, the FHR51-9FH1-5 protein reduced glomerular C3 in experimental models of C3 glomerulopathy driven by either FH deficiency or mutated FHR5. These preclinical data indicate that FHR51-9FH1-5 protein represents a novel treatment strategy for complement-mediated kidney disease.

The selection of an appropriate promoter is important to the design and optimisation of adeno-associated viral (AAV) vector-based cardiac gene therapies. The expression cassette design can impact efficacy and safety of the vector. This study is the first to use a novel AAV barcode-seq method for the simultaneous evaluation of a panel of cardiac-specific promoters in a high-throughput manner. Functional analyses of our cardiac promoter kit packaged in three different capsids were performed using neonatal rat ventricular myocytes (NRVM), human iPSC-derived cardiomyocytes (hiPSC-CMs), HuH7 hepatocellular carcinoma cells, as well as mouse, rat, sheep and pig models. The cardiac troponin T (cTnT) promoter showed the most promise overall as a cardiac-specific promoter across all cardiac models tested. The results validate the barcode-seq technique as a powerful and versatile approach that enables high-throughput, quantitative analysis of various expression cassettes in commonly used models of cardiac gene therapy.

C3 glomerulopathy (C3G) and atypical haemolytic uraemic syndrome (aHUS) are rare kidney disorders characterized by the dysregulation of the alternative pathway (AP). Unlike aHUS, current treatment options for C3G are limited. Factor H (FH) plays a key role in regulating the AP, and mice lacking FH develop a disease similar to C3G. We previously developed mouse homodimeric minimal FH (mHDM-FH) as a gene therapy using an adeno-associated virus (AAV) vector incorporating a liver-specific promoter to rebalance the AP activation. We previously noted that a single dose of AAV-mHDM-FH (1 × 1011) restored complement regulation in Cfh−/− mice by 21 days post inoculation, however this was associated with a strong anti-HDM-FH response, which was markedly reduced in Cfh+/− mice.In the current study, we focussed on long-term evaluation of AAV-mHDM-FH in both Cfh+/− mice and C3N/N mice (the C3 D1115N gain-of-function mouse model of aHUS). At six months post-inoculation, C3N/N mice treated with AAV-mHDM-FH showed significantly improved survival compared to AAV-GFP controls. However, the therapeutic effect was partially limited due to reduced binding affinity of mHDM-FH to the mutant C3 D1115N protein. These findings indicate that while AAV-mHDM-FH retains some functionality in C3 mutation-driven disease, its efficacy may not match therapies targeting the terminal complement pathway.In Cfh+/− mice, no signs of disease or safety concerns were observed over the six-month period. Importantly, increased serum levels of both C3 and FH were detected, suggesting improved control of the amplification loop in these mice. AAV transgene expression remained detectable in the liver at six months, an encouraging result for long-term efficacy.Across both experimental time points, anti-AAV antibody responses were detected in Cfh+/− mice. These responses peaked at 21 days post-inoculation but were present at relatively low levels after 6 months. This pattern aligns with observations in patients receiving AAV-based gene therapies. Despite the presence of anti-AAV antibodies, the 6 month data showed transgene expression and no safety concerns, leaving the extent to which these antibodies impact therapeutic efficacy still uncertain. Looking ahead, we aim to enhance long-term vector expression by exploring the use of alternative promoters tailored to different cellular targets. References: 1. Yang Y, Denton H, Davis O R, Smith-Jackson K, Kerr H, Herbert A P, Barlow P N, Pickering M C and Marchbank K J An Engineered Complement Factor H Construct for Treatment of C3 Glomerulopathy J Am Soc Nephrol 2018 29(6):1649–1661.

Monoclonal antibodies are useful tools to dissect the neutralizing antibody response against the adeno-associated virus (AAV) capsids that are used as gene therapy delivery vectors. The presence of pre-existing neutralizing antibodies in large portions of the human population poses a significant challenge for AAV-mediated gene therapy, primarily targeting the capsid leading to vector inactivation and loss of treatment efficacy. This study structurally characterizes the interactions of 21 human-derived neutralizing antibodies from three patients treated with the AAV9 vector, Zolgensma®, utilizing high-resolution cryo-electron microscopy. The antibodies bound to the 2-fold depression or the 3-fold protrusions do not conform to the icosahedral symmetry of the capsid, thus requiring localized reconstructions. These complex structures provide unprecedented details of the mAbs binding interfaces, with many antibodies inducing structural perturbations of the capsid upon binding. Key surface capsid amino acid residues were identified facilitating the design of capsid variants with antibody escape phenotypes. These AAV9 capsid variants have the potential to expand the patient cohort to include those that were previously excluded due to their pre-existing neutralizing antibodies against the wtAAV9 capsid, and the possibly of further treatment to those requiring redosing.

Type 1 diabetes (T1D) is caused by the autoimmune destruction of the pancreatic insulin-producing β cells. This study investigated a novel gene therapy approach to prevent disease development by replacing pancreatic β cell function with that from transdifferentiated liver cells. A clinically applicable third-generation lentiviral vector was used to deliver a cocktail of β cell transcription factors (Pdx1, NeuroD1, and MafA) to the portal vein of 5- to 6-week-old non-obese diabetic (NOD) mice. At the experimental endpoint (30 weeks), 100% of the NOD mice that received the lentiviral vector expressing the three β cell transcription factors were normoglycemic. Additionally, intraperitoneal glucose tolerance tests revealed that treated NOD mice could normalize blood glucose concentrations as efficiently as non-diabetic control animals. RT-PCR detected a range of pancreatic markers, such as somatostatin (Sst), Glut2, and most important, mouse insulin (INS1 and INS2), which was also found to be stored in the liver. Liver function tests remained normal. Collectively, these data show that expression of these β cell transcription factors led to partial pancreatic transdifferentiation and halted the development of hyperglycemia and abnormal glucose tolerance, which are the hallmarks of T1D. Thus, this approach holds substantial promise as a potential prophylactic strategy.

Extracellular DNA (ecDNA) released from injured and dying cells powerfully induces injurious inflammation. In this study we show ecDNA renal presence in patients and experimental mice with myeloperoxidase anti-neutrophil cytoplasmic antibody-associated glomerulonephritis (MPO-ANCA GN). Twice daily administration of intravenous DNase I (ivDNase I) in two models of anti-MPO GN was effective at reducing glomerular deposition of ecDNA, histological injury, leukocyte infiltration and NETosis. Comprehensive investigation into DNase I modes of action revealed the enzyme reduced lymph node DC numbers and their activation status, resulting in decreased frequency of MPO-specific CD4 effector T cells (IFN-γ, and IL17A producing), reductions in dermal anti-MPO delayed type hypersensitivity responses and increased frequency of MPO-specific T regulatory cells. Renal expression of inflammatory chemokines were also decreased. To overcome the translational obstacle of the short half-life of DNase I (<5 hours), we tested an adeno-associated viral vector encoding DNase I in one of the models. Along with the endpoint changes described above, a single vector treatment also enhanced therapeutic benefit as seen by reductions in MPO-ANCA and albuminuria. These results indicate ecDNA is a potent driver of anti-MPO GN and that DNase I is a potential therapeutic that can be delivered using gene technology. ### Competing Interest Statement The authors have declared no competing interest.

Developing clinically predictive model systems for evaluating gene transfer and gene editing technologies has become increasingly important in the era of personalized medicine. Liver-directed gene therapies present a unique challenge due to the complexity of the human liver. In this work, we describe the application of whole human liver explants in an ex situ normothermic perfusion system to evaluate a set of fourteen natural and bioengineered adeno-associated viral (AAV) vectors directly in human liver, in the presence and absence of neutralizing human sera. Under non-neutralizing conditions, the recently developed AAV variants, AAV-SYD12 and AAV-LK03, emerged as the most functional variants in terms of cellular uptake and transgene expression. However, when assessed in the presence of human plasma containing anti-AAV neutralizing antibodies (NAbs), vectors of human origin, specifically those derived from AAV2/AAV3b, were extensively neutralized, whereas AAV8- derived variants performed efficiently. This study demonstrates the potential of using normothermic liver perfusion as a model for early-stage testing of liver-focused gene therapies. The results offer preliminary insights that could help inform the development of more effective translational strategies.

Monoclonal antibodies (mAbs) are useful tools to dissect the neutralizing antibody response against the adeno-associated virus (AAV) capsids used as gene therapy delivery vectors. This study structurally characterizes the interactions of 21 human-derived antibodies from patients treated with the AAV9 vector, Zolgensma ® , utilizing high-resolution cryo-electron microscopy. The majority of the bound antibodies do not conform to the icosahedral symmetry of the capsid, thus requiring localized reconstructions. These complex structures provide unprecedented details of the mAbs binding interfaces, with some antibodies inducing structural perturbations of the capsid upon binding. Key surface capsid amino acid residues were identified facilitating the design of capsid variants with an antibody escape phenotype, with the potential to expand the patient cohort treatable with AAV9 vectors to include those that were previously excluded due to their pre-existing neutralizing antibodies, and possibly also to those requiring redosing.