SRY (sex-determining region on the Y chromosome), the master regulator of male development in
mammals, has been studied extensively for more than 17 years. The precise mechanism by which
SRY protein triggers the chain of events that culminates in testis development, remains unclear.
SRY is the founding member of the SOX gene family, and like other SOX factors it may function
both as an architectural and modular transcription factor. Deciphering the code of SRY is very
complex since in vitro assays, along with available genetic evidence, suggest that SRY may
potentially work as a transcriptional activator or repressor. This complexity has increased with
evidence describing SRY, and other SOX factors, as RNA binding proteins involved in premessenger
splicing. Compelling evidence shows that the function of Sry, instead of the robust gene
one might expect as the pillar of male sexual development, hangs by a thin thread. This situation
has profound biological, medical and evolutionary implications. What is clear is that SRY protein
mediates, directly or indirectly, the up-regulation of SOX9 specifically in the male gonad, and that
SOX9 protein in turn, is sufficient to trigger Sertoli-cell differentiation and testis formation.
This thesis comprises a series of analyses of SRY and SOX gene activities, associated with their
regulation of testis formation in the mouse, and performed at various levels of the SRY-SOX9 axis.
The first of these analyses tested whether a previously described nuclear SRY-interacting protein,
KRAB-O, is necessary for SRY’s function as the testis-determining factor. The analysis of the
SRY-KRAB interaction turned out to be unexpectedly complex because KRAB-O is a member of a
large gene family, and, as reported herein, many genes of the KRAB-O/Rsl cluster are expressed in
the developing mouse gonads at the time of sex determination. Using RNAi-based knockdown
experiments, it was found that Sox9 expression was negatively affected by the loss of function of
KRAB genes in a cellular system, suggesting that SRY might require KRAB proteins to initiate
male sex determination. However, transgenic KRAB knockdown mice did not show a gonadal
phenotype, possibly due to redundancy between members of the gene cluster.
The second analysis tested the hypothesis that the action of any SOX protein, rather than SRY
per se, might be sufficient to trigger male sex determination provided that it is expressed within a
specific time window in pre-Sertoli cells to pre-empt the establishment of the female pathway. Data
from luciferase assays with the Amh promoter suggested that the molecular and biochemical
operation of SRY and SOX3 are different from SOX9 function. Therefore, any testis formation that
may be triggered by SRY and SOX3 is not likely to be caused by direct activation of targets of
SOX9. Several constructs were generated to test SOX3/SRY interchangeability in sexdetermination in transgenic mouse models. However, before the appropriate transgenic mice were
generated, data from another laboratory indicated that expression of SOX3 in genital ridges was
associated with phenotypic female-to-male sex reversal in transgenic mice. This independent
evidence supports the notion of SOX/SRY interchangeability in sex determination. The
implications of this finding are discussed in this thesis.
Thirdly, given that Sox9, Sox10 and Sox8 are co-expressed in the developing mouse testis, I
investigated whether these three genes may cooperate to bring about testis formation. I report
in vitro evidence showing that SOX10 can synergize with SF1 to trans-activate the Amh promoter
in a similar way to SOX9 and SOX8. This finding supports the notion that these three transcription
factors are functionally and biochemically equivalent in regulation of genes that mediate male
development. This prompted us to analyze whether the compound ablation of both Sox10 and Sox8
has an effect on testis determination in vivo. It was observed that SOX9 alone, is sufficient to drive
testis development. A likely explanation for this dominant role of SOX9 is discussed. I also
describe experiments aimed at the generation of Sox10 gain-of-function transgenic mice.
Finally, Sry is normally expressed in a small group of embryonic cells, pre-Sertoli cells, for a short
interval of time. This, together with the small size of mouse gonads at the time of Sry expression,
makes it difficult to disentangle experimentally the underlying relationships between Sry and other
gonadal genes. A novel gonad cell line transgenic for Sry and established in the Koopman
laboratory, recapitulates the up-regulation of endogenous Sox9, and may provide a means to
perform this kind of analysis in vitro. A preliminary analysis of this cell line is reported herein.