Granulosa cells provide elimination of apoptotic oocytes through unconventional autophagy-assisted phagocytosis

Study question: Do human granulosa cells (GCs) ingest and destroy apoptotic oocytes?

Summary answer: Somatic GCs ingest and destroy apoptotic oocytes and other apoptotic substrates through unconventional autophagy-assisted phagocytosis.

What is known already: Most (99%) ovarian germ cells undergo apoptosis through follicular atresia. The mode of cleaning of atretic follicles from the ovary is unclear. Ovarian GCs share striking similarities with testicular Sertoli cells with respect to their origin and function. Somatic Sertoli cells are responsible for the elimination of apoptotic spermatogenic cells through unconventional autophagy-assisted phagocytosis.

Study design, size, duration: Human GCs were tested for the ability to ingest and destroy the apoptotic oocytes and other apoptotic substrates. A systemic study of the main phagocytosis steps has been performed at different time points after loading of apoptotic substrates into the GC.

Participants/materials, setting, methods: Primary cultures of GC retrieved following controlled ovarian stimulation of five women for IVF/ICSI and a human granulosa KGN cell line were incubated with different apoptotic substrates: oocytes which underwent spontaneous apoptosis during the cultivation of immature germ cells for IVF/ICSI; apoptotic KGN cells; and apoptotic membranes from rat retinas. Cultured GC were analyzed for the presence of specific molecular markers characteristic of different steps of phagocytic and autophagy machineries by immunocytochemistry, confocal microscopy, transmission electron microscopy and western blotting, before and after loading with apoptotic substrates.

Main results and the role of chance: Incubation of human GC with apoptotic substrates resulted in their translocation in cell cytoplasm, concomitant with activation of the phagocytosis receptor c-mer proto-oncogene tyrosine kinase MERTK (P < 0.001), clumping of motor molecule myosin II, recruitment of autophagy proteins: autophagy-related protein 5 (ATG5), autophagy-related protein 6 (Beclin1) and the rise of a membrane form of microtubule-associated protein 1 light chain 3 (LC3-II) protein. Ingestion of apoptotic substrates was accompanied by increased expression of the lysosomal protease Cathepsin D (P < 0.001), and a rise of lysosomes in the GCs, as assessed by different techniques. The level of autophagy adaptor, sequestosome 1/p62 (p62) protein remained unchanged.

Large scale data: N/A.

Limitations, reasons for caution: The number of patients described here is limited. Also the dependence of phagocytosis on reproductive hormone status of patients should be analyzed.

Wider implications of the findings: Removal of apoptotic oocytes by surrounding GC seems likely to be a physiological mechanism involved in follicular atresia. Proper functioning of this mechanism may be a new strategy for the treatment of ovarian dysfunctions associated with an imbalance in content of germ cells in the ovaries, such as premature ovarian failure and polycystic ovary syndrome.

Study funding/competing interest(s): The study was funded by Rennes Metropole (AIS 2015) and Agence de BioMédecine. This work was supported by funding from Université de Rennes1, Institut National de la Santé et de la Recherche Médicale (INSERM) and CHU de Rennes. A.B. is funded in part by the program Actions Concertées Interpasteuriennes (ACIP) and a research grant from the European Society of Pediatric Endocrinology. This work is supported by the Agence Nationale de la Recherche Grants ANR-17-CE14-0038 and ANR-10-LABX-73. The authors declare no competing interests.