Follistatin regulates germ cell nest breakdown and primordial follicle formation

Abstract

Follistatin (FST) is an antagonist of activin and related TGFβ superfamily members that has important reproductive actions as well as critical regulatory functions in other tissues and systems. FST is produced as three protein isoforms that differ in their biochemical properties and in their localization within the body. We created FST288-only mice that only express the short FST288 isoform and previously reported that females are subfertile, but have an excess of primordial follicles on postnatal day (PND) 8.5 that undergo accelerated demise in adults. We have now examined germ cell nest breakdown and primordial follicle formation in the critical PND 0.5-8.5 period to test the hypothesis that the excess primordial follicles derive from increased proliferation and decreased apoptosis during germ cell nest breakdown. Using double immunofluorescence microscopy we found that there is virtually no germ cell proliferation after birth in wild-type or FST288-only females. However, the entire process of germ cell nest breakdown was extended in time (through at least PND 8.5) and apoptosis was significantly reduced in FST288-only females. In addition, FST288-only females are born with more germ cells within the nests. Thus, the excess primordial follicles in FST288-only mice derive from a greater number of germ cells at birth as well as a reduced rate of apoptosis during nest breakdown. These results also demonstrate that FST is critical for normal regulation of germ cell nest breakdown and that loss of the FST303 and/or FST315 isoforms leads to excess primordial follicles with accelerated demise, resulting in premature cessation of ovarian function.

Fig. 1.

Extended germ cell nest breakdown in FST288-only females. From four to six ovaries were used to count the follicle numbers in each day of each genotype. A, H and E staining of PND 5.5 WT ovary. The outer cortex contains a number of primordial follicles and oocytes still in the nest as well as primary and secondary follicles in the medulla. B, H and E stain of FST288-only ovary showing the number of germ cells remaining within nests is greater in FST288-only ovaries compared with WT. C, GCNA1 staining of PND 5.5 WT ovary confirming germ cell identity of cells within nests. D, GNCA1 staining of FST288-only ovary showing greater number of germ cells within nests compared with WT. E, Total number of oocytes remaining within the nest at PND 0.5–8.5 in both WT and FST288-only mice. Note that FST288-only ovaries have significantly more germ cells at PND 0.5 suggesting increased proliferation before birth. F, Same data as E but now expressed as percent of total germ cells. G, Total number of oocytes in each ovary regardless of stage at PND 0.5, 3.5, 5.5, 8.5, 42, and 98. Loss of germ cells is reduced in FST288-only ovaries compared with WT (20% vs. 40% respectively) during germ cell nest breakdown (up to PND 8.5), but this rate is greater at later ages leading to premature exhaustion of the primordial follicle pool before 1 year of age (22). Scale bar, 100 μm (A–D). *, P < 0.05.

Fig. 2.

Proliferation, apoptosis, and FST in neonatal ovaries of WT and FST288-only mice. Four or five ovaries from each genotype at each age analyzed were used to detect proliferation or apoptosis. A, Proliferation in PND 1.5 ovary is shown by double immunofluorescence for BrdU (red) and GCNA-1 (green). No double-stained cells were detected, indicating that all proliferation was in somatic cells. B, Same staining in PND 1.5 FST288-only ovary. No double-stained cells were detected. C, Double immunofluorescence for BrdU (red), Vasa (green), and DAPI (blue) in 1.5-d FST288-only mouse. D, Enlargement of white box from C showing example of one of two double-stained germ cells observed of 7,500 examined. E, Apoptosis in germ cells of PND 0.5 WT ovary labeled for cleaved PARP-1 (green), GCNA-1 (red), and DAPI (blue). F, Same staining in 0.5-d FST288-only ovary showing reduced germ cell apoptosis relative to WT. G, Same staining in 5.5-d WT ovary, no apoptotic germ cells observed. H, Same staining in 5.5-d FST288-only ovary showing presence of apoptotic germ cells. I, Immunohistochemical staining for FST in PND 3.5 WT ovary. FST staining is prominent in oocyte cytoplasm, some stromal cells, some granulosa cells, and in ovarian surface epithelia. Intense stain is observed in some oocyte nuclei, but not all. Inset shows that staining disappears if FST antibody is preincubated with FST protein. J, FST immunohistochemistry in PND 3.5 FST288-only ovary. The staining pattern is similar to WT but less intense, although some oocyte nuclei are still intensely stained. Scale bar, 100 μm (A–C and E–J) and 10 μm (D).

Fig. 3.

Quantitation of apoptosis staining in germ cells. A, Number of apoptotic cells as a fraction of total germ cells in PND 0.5 ovaries. Apoptosis in WT ovary is significantly greater than in FST288-only ovary. B, Number of apoptotic cells as a fraction of total germ cells in PND 5.5 ovaries. At this age, FST288-only ovaries have more apoptotic germ cells than WT. C, Apoptotic germ cells as a percent of total germ cells for PND 0.5–8.5 showing that apoptosis in WT ovaries is high but drops quickly and is almost undetectable by PND 5.5, whereas apoptosis in FST288-only ovaries starts lower but remains relatively constant through PND 5.5 and then decreases to nearly undetectable by PND 8.5. *, P < 0.05.