Infertility-associated DAZL mutations and their effects on DAZL functions

DAZL (DAZ-Like) is a member of the Deleted in Azoopermia (DAZ) family of RNA-binding proteins, consisting of DAZ, DAZL and BOULE. DAZ is the founding member of the family, and deletions of the Y-chromosome encoded DAZ genes are commonly observed in infertile men with non-obstructive azoospermia (absence of viable sperm within seminiferous tubules). However, DAZ is only present in Old world primates and humans, making its study difficult. On the other hand, DAZL is a vertebrate-specific autosomal homolog of DAZ, and studies have shown that it is essential for germ cell development in males and/or females of many model organisms including X. laevis and mice. Expression of DAZL is virtually germ cell specific, and its critical role in fertility appears to be the regulation of post-transcriptional gene expression of mRNAs encoding proteins required for germ cell development. Post-transcriptional regulation of mRNAs is well studied in germ cells, in which it is particularly important due to periods of transcriptional quiescence, where changes in the pattern of protein synthesis have to be achieved via the activation, repression or destruction of pre-existing mRNAs. The best characterised mechanism of post-transcriptional regulation is the dynamic changes in poly(A) tail length. mRNAs with short poly(A) tails are stored as translationally silent in the cytoplasm, and become translationally activated by addition of a new poly(A) tail (cytoplasmic polyadenylation). Conversely mRNAs which undergo deadenylation (removal of the poly(A) tail) become translationally silenced. The best characterised DAZL function is activation of mRNA-specific translation when bound to the 3’ untranslated region (UTR) of target mRNAs. This role is independent of the presence of poly(A) tail, providing a mechanism for the activation of mRNAs that do not undergo cytoplasmic polyadenylation. Furthermore, unpublished work from Gray lab has also identified a second DAZL function in protecting target mRNAs from default deadenylation during oocyte maturation, thereby maintaining them in a translationally active state. Deadenylation is also the first and rate limiting step in mRNA turnover. Despite its critical role in gametogenesis in model organisms, and knowledge of its molecular functions and target mRNAs, the role of DAZL in human fertility has not yet been established. However, several mutations have been identified by candidate gene sequencing in a small number of patients with reduced fertility, and in a small population of infertile men. Nonetheless, it remains unclear whether these mutations are causative, due to an absence of functional studies and familial genetic studies. Here I address the putative role for DAZL in human fertility by examining whether the subfertility-associated DAZL mutations and mutations from human populations, which I identified by in silico screening of human genome database, affect its known molecular functions in post-transcriptional control. To address the effect of mutations on the known functions of DAZL I use established tether-function assays in Xenopus laevis oocytes, where I artificially bind DAZL to a reporter mRNA, circumventing potential effects of mutations on RNA binding, allowing me to study the effects of mutations directly on the post-transcriptional regulation of bound mRNAs. The effects of mutations were studied in both tethered human DAZL, as well as mouse DAZL, to validate potential future mouse models. A minimal functional region of DAZL, which is a region essential for its function, has previously been described, and I hypothesised that non-conservative mutations within this region may disrupt function. Consistent with this hypothesis, mutations from infertile patients found outside of the minimal functional region of DAZL did not affect its function in stimulating translation of the bound mRNA. However, two mutations found in infertile patients within the minimal active region reduced DAZL function, with this effect being most profound for a nonconservative mutation. Putative mutations identified by in silico studies of general populations did not affect ability of DAZL to stimulate translation, regardless of their position. The observed effects of mutations in human construct and mouse construct were largely consistent. I also used tether-function assay, followed by poly(A) tail (PAT) assay, to study effects of mutations on impeding of deadenylation. Interestingly, the role in impeding of deadenylation was not affected by any of these mutations, suggesting that the two functions of DAZL are mechanistically separate. As the effects of mutations were tested in the context of tethered DAZL, the observed effect on translation cannot be explained by an effect on mRNA binding. DAZL has been reported to interact with a number of partner proteins, however interaction with poly(A)-binding protein (PABP) has been shown to be essential for both its ability to stimulate translation and to impede deadenylation. Moreover, the most disruptive mutation is located in the mapped PABP-binding site. Therefore, to understand the mechanism by which DAZL function in translation was disrupted, I carried out coimmunoprecipitation to test whether DAZL-PABP interactions are disrupted. My results show that there is a decreased, but not completely abolished, interaction between mutant DAZL and PABP, providing a mechanistic understanding of how this mutation impacts DAZL function. Taken together my results establish that mutations within the minimal functional region of DAZL can disrupt its role in regulating gene expression. This finding opens the avenue for future mouse models containing this mutation to establish its effects on mammalian gametogenesis and supports the idea of future larger scale genetic assessment of the role of DAZL as a candidate human infertility factor