Fig 2: Adult Sox30-null testes are abnormally small and spermatids and spermatozoa are not produced. (a) Testes are smaller in Sox30-null males with (b,c,d) empty epididymal tubules compared to wildtype littermates (n = 3). (e,f) Sperm are found in the caput and cauda epididymides of wildtype males, but are absent from Sox30-null epididymides (white arrows, H&E staining). (g) 8-week-old testes show an absence of elongating spermatids and the presence of multinucleated giant cells (black arrows) in Sox30-null testis.

Fig 3: Sox30-null males are infertile with spermatogenesis blocked at the round spermatid stage. (a) PAS staining of 19, 21, 23 and 25 dpp testes reveals that abnormal germ cells (white arrows) are first seen in 21 dpp Sox30-null testes, a timpoint characterized by flattening of the pro-acrosome; most advanced appearance of round spermatids is shown in insert for each genotype and age. At 19 dpp no distinct granules are observed in any of the three genotypes. TEM examination of wildtype samples showing two ultrastructural features characteristic of step 3–4 spermatids: (b) acrosome capping of the nucleus (white arrowhead) and (c) docking of basal body to the nuclear membrane (black arrowhead). (d) These two features were not observed in Sox30-null round spermatids, nor were proacrosomal granules observed at the center of Golgi bodies (black arrow). (e) Nuclei within symplasts do not display any evidence of acrosome formation or tail development.

Fig 4: Expression of Sox30 is specific to the germ cells of the gonads at 13.5 dpc and follows the expression of Stra8 in both fetal ovaries and postnatal testis. (a) qRT-PCR was performed on various tissues of 13.5 dpc embryos, and adult testis, to assess Sox30 expression. Robust Sox30 expression in embryonic tissue is observed only in the gonads, with significantly higher levels detected in the fetal ovary than the fetal testis. Sox30 expression is ~8 fold higher in the adult testis than in the fetal ovary (pools of 4~8 individuals per column; two-tailed unpaired t-test). (b) qRT-PCR analysis of flow-sorted (FACS) populations of fetal gonadal tissues reveals that Sox30 transcript is enriched in the germ cell compartment in both the fetal ovary and testis at 13.5 dpc (cells isolated from pools of 4 to 8 gonad pairs). (c) qRT-PCR analysis in fetal gonads (from 11.5 to 18.5 dpc) demonstrates that the expression of Sox30 in the fetal ovary increases after the expression of Stra8 (pools of 4 to 8 gonad pairs). (d) qRT-PCR analysis demonstrates that expression of Sox30 follows expression of Stra8 in the postnatal testis (each time point represents one individual sample). All error bars represent S.E.M. of technical replicates. Ddx4 is expressed specifically in germ cells and is used as a normalizer for germ cell-specific genes when whole fetal gonadal tissue is analysed. (e) We verified by mRNA in situ hybridisation the previously described germ cell specific expression of Sox30 in the post-natal testis11. Transcripts for Sox30 were detected beginning at mid-pachytema (right panel), peaking in round and elongating spermatids (left panel, arrow). Elongated spermatids have diminished expression and no signal was detected in sperm just prior to spermiation.