In vitro system completely restores oogenesis in congenitally infertile mice

Abstract STUDY QUESTION Can in vitro systems, combined with transient gene expression or factor supplementation, completely restore fertility in congenitally infertile mice? SUMMARY ANSWER Transient expression of Kitl via adeno-associated virus (AAV) vectors or supplementation with recombinant KITL in KitlSl-t/KitlSl-t mice—a model of congenital infertility caused by a mutation in the Kitl locus—resulted in the production of mature oocytes and the birth of healthy, fertile offspring. WHAT IS KNOWN ALREADY Although in vivo gene delivery has enabled offspring production in infertile mouse models, low efficiency, unpredictability of parturition timing, inflammatory risk, possible viral genome integration, and lack of real-time oogenesis observation remain major concerns. Despite the potential of in vitro oogenesis as an alternative, complete functional restoration of gene deficiency has not been reported. STUDY DESIGN, SIZE, DURATION AAV-mCherry was applied to wild-type mouse ovaries, and expression levels were compared across 15 serotypes (2.5 × 1011 viral genomes/ml; N = 4–12; 4-day infection, 20-day culture) to identify optimal AAV serotypes for ovarian gene delivery. The effects of AAV-Kitl infection (six doses; N = 3–5) and recombinant KITL supplementation (four doses; N = 5) on oocyte growth were evaluated in KitlSl-t/KitlSl-t mouse ovaries. On culture day 17 or 18, secondary follicles were isolated and cultured for an additional 16 days to evaluate oocyte competence for maturation, fertilization, and full-term development. Offspring were delivered 52–53 days after treatment initiation. PARTICIPANTS/MATERIALS, SETTING, METHODS Ovaries from KitlSl-t/KitlSl-t mice were dissociated into single cells and reaggregated in U-bottom wells with media containing AAV8-Kitl, AAV9-Kitl, or recombinant KITL. Reconstituted ovaries were cultured on insert membranes, thereby allowing primordial follicles to develop into secondary follicles. Isolated secondary follicles were further cultured to the antral stage, and cumulus–oocyte complexes were subjected to IVM and IVF. The resulting embryos were transferred to foster mothers. Finally, the offspring were subjected to PCR screening for AAV sequences and fertility tests. MAIN RESULTS AND THE ROLE OF CHANCE AAV8, AAV9, AAVrh.10, and AAVrh.32.33 induced significantly higher levels of mCherry expression in wild-type mouse ovaries than 10 of the 15 AAV evaluated serotypes in vitro (P < 0.05). AAV8-Kitl promoted primordial follicle activation in a dose-dependent manner in KitlSl-t/KitlSl-t mouse ovaries, with the highest number of secondary follicles (80 per reconstituted ovary) obtained at 1.0 × 1011 vg/ml (P < 0.05). In contrast, AAV9-Kitl required 2.5- to 10-fold higher titers to achieve comparable levels of secondary follicle formation. Contrastingly, no secondary follicles were formed in KitlSl-t/KitlSl-t mouse ovaries following mock treatment. Furthermore, supplementation with 200 ng/ml recombinant KITL supported secondary follicle formation at levels comparable to those in the wild-type mouse ovaries. More than 10% of fertilized oocytes developed to full term, regardless of the treatment method. AAV DNA was not detected in the genomes of the 47 offspring, and all tested female mice exhibited normal fertility. LARGE SCALE DATA N/A. LIMITATIONS, REASONS FOR CAUTION To date, complete in vitro oogenesis has only been achieved in mice; its applicability to other species, including humans, remains unverified. WIDER IMPLICATIONS OF THE FINDINGS This study establishes a novel and controllable in vitro platform to compensate for gene function through transient gene expression or factor supplementation, without permanent genomic modification. This approach provides a powerful framework for the dissection of gene functions during oogenesis, modeling of reproductive disorders, and development of fertility restoration strategies in both clinical and conservation contexts. STUDY FUNDING/COMPETING INTEREST(S) This study was supported by KAKENHI (grant numbers 18H05547, 23K27088, and 25H01353). The authors declare no competing interests.