Directed Evolution-Engineered Muscle-Tropic AAVs for Feline Anti-Mullerian Hormone (AMH)-Based Contraception

Philippe Godin¹; Natalie Sicher¹; Lindsey M. Vansandt²; Marie C. Meinsohn¹; Julie L. Barnes²; Christina Bunner²; Josephine Cleverdon¹; Rokib Hasan³; Noah Davidsohn³; Guangping Gao⁴; Dan Wang⁴; William F. Swanson²; Thomas Conlon⁵; David Pepin¹

1. Pediatric Surgical Research Laboratories, Massachusetts General Hospital, Department of Surgery, Harvard Medical School, Boston, United States;

2. Center for Conservation and Research of Endangered Wildlife (CREW), Cincinnati Zoo & Botanical Garden, Cincinnati, United States;

3. Rejuvenate Bio, San Diego, United States;

4. Horae Gene Therapy Center, University of Massachusetts Chan Medical School, Worcester, United States;

5. The Michelson Found Animals Foundation Inc, Michelson Philanthropies, Los Angeles, United States.

Abstract Text:

The uncontrolled reproduction of free-roaming domestic cats amplifies their welfare challenges and the ecological pressure they exert on wildlife populations. Because of logistic and economic constraints, surgical sterilization alone cannot scale to control the reproduction of the hundreds of millions of stray and feral cats worldwide. In female mammals, anti-Müllerian hormone (AMH) is produced by granulosa cells of growing follicles and involved in the inhibition of primordial follicle activation and pre-antral follicle growth. We have previously reported that the adeno-associated virus serotype 9 (AAV9)-mediated delivery of an AMH transgene could suppress folliculogenesis and induce long-term contraception in adult female mice and domestic cats. Functional transduction of long-lived skeletal muscle cells is key for long-term transgene expression. AAV9 exhibits a broad tissue biodistribution and a high liver tropism, which results in only a fraction of the administered dose to transduce the muscle compartment and can cause liver damage when administered at high doses. The objective of this study was to evaluate the safety and efficacy of alternative AAV capsid variants designed to enhance muscle cell transduction and deliver a feline AMH transgene in adult female mice and domestic cats. We used directed evolution to generate new AAV capsid variants that specifically target cat skeletal muscle cells. Briefly, we produced a library of AAVs generated by capsid DNA shuffling, which was administered intravenously to two cats, and muscle biopsies were sequenced to identify enriched variants. We then compared the intramuscular injection of 1- or 5x10¹² viral genomes per kilogram (vg/kg) of four AAVs generated by directed evolution (AAV.CM1/4/9/10), to naturally occurring AAV serotypes (AAV6/8), or muscle-enhanced/liver-detargeted AAVs (AAV.MYO, AAV9.HR), and our first generation AAV9 capsid. Six-week-old female mice (N=5 per group) were followed for five- (AAV9-AMH and AAV.CM1/4/9/10-AMH) or up to twelve months post-injection (AAV9-AMH and remaining vectors). Observation of injection sites and general health, and necropsy, hematology and serum chemistry at endpoint revealed that all vectors maintained an excellent safety profile. Importantly, heart function was not impacted in mice that received muscle-targeted variants as evidenced by normal blood levels of two cardiac biomarkers. All mice except those that received AAV.CM1 had supraphysiological levels of AMH, and significantly reduced ovarian weight (One-way ANOVA: P < 0.0001 for all) and antral follicles numbers (One-way ANOVA: from P=0.0001 to P=0.0122 for all) when compared to mice that received an empty-vector control injection. These findings suggest these AAVs can induce potent ovarian suppression and anovulation. Nonetheless, only the AAV.CM9/10 groups had average circulating AMH levels comparable to AAV9. To evaluate the safety and ability of CM4/9/10-AMH vectors to induce supraphysiological levels of AMH in the target species, we administered a single intramuscular injection of 2x10¹² vg/kg of each vector (AAV9-AMH for control) to adult female domestic cats (N=2 per group). Six months after treatment administration, we did not observe any abnormalities at injection sites, on hematology and serum chemistry, and on physical and cardiac examinations. Average circulating AMH levels in the blood of each group of cats were comparable to those in the AAV9-AMH control group, suggesting that CM4/9/10 could be viable candidate capsids for long-term persistence of muscle expression to achieve lifetime contraception in female domestic cats.