Previously, several studies have investigated genes expressed at specific stages or in particular cell type during spermatogenesis [4
]. Although these studies provided inclusive information about the expression profile of a large number of germ-cell genes, comprehensive understanding of spermatogenesis requires further systematic identification and analysis of uncharacterized genes with germ cell-specific expression. UniGene is a large and widely used transcript sequence database containing a large amount of unexplored information about genes. The sequences are divided according to tissue type or developmental stage from the UniGene database, which provides a resource for identifying novel tissue-, cell-, or stage-specific gene transcripts. In the present study, analyzing the mouse spermatocyte UniGene library (Lib.6787), we disclosed that significant proportion (11%) of the spermatocyte genes are testis-specific and about half of the testis-specific genes are unknown or uncharacterized. Previously, a similar approach was applied by our group using the round spermatid UniGene library [11
], revealing that 22% (467 of 2124 genes) of genes expressed in round spermatids are testis-specific and functions of 74% of the testis-specific genes are unexplored. In the present investigation, the initial number of uncharacterized, testis-specific genes selected from the earlier version of the spermatocyte library is 134. These 134 genes were narrowed down to 24 authentic genes considered to be abundantly and specifically expressed in the testis by various expression analyses. The other 110 genes were eliminated from consideration because they displayed unreliable coding sequences (53 genes), were undetected or detected with unexpected sizes in the PCR assay (25 genes), not specifically or predominantly expressed in the testis (19 genes) or undetected in the Northern blot analysis (13 genes).
Our data provide extensive information on the 24 genes at the genomic and transcript levels. Genomic analysis disclosed orthologues for 13 mouse genes in the human genome and 11 other were identified as mouse-specific genes. The proportion of mouse genes with a single identifiable ortholog in the human genome is about 80%. The other 20% of mouse genes lack a strict 1:1 relationship, due to differential expansion in at least one of the two genomes [16
]. Mostly, those mouse-specific genes were involved in reproduction, olfaction and immunity. Similarly, a global view of human and mouse proteases revealed that the mouse degradome is more complex, and several genes in the mouse genome encode proteases involved in reproductive functions [17
]. One such example is the testis-specific or predominant ADAM genes in postmeiotic germ cells [18
]. Thus, the 11 mouse-specific genes identified in our study are related to aspects of reproductive physiology. At the transcript level, the Northern blot analysis revealed that four genes are transcribed into products of more than one size. In addition, the analysis demonstrated that transcript sizes from the database are consistent with those determined experimentally for most of the genes. A special feature of genes whose expression is strongly favored in male germ cells is developmentally regulated during meiotic and postmeiotic phases [1
]. Consistently, the expression patterns of the 24 genes during postnatal testicular development, found in the present study, are indicative of developmental regulation. The pachytene spermatocyte stage is significant during spermatogenesis. It involves genetic recombination, which occurs only in germ cells through cross-over between paired chromosomes and increases RNA and protein synthesis in preparation for the next phase [19
]. Transcription of more than half (16 genes) the total genes was found to start from the pachytene spermatocyte stage.
Germ cell-specific and developmentally regulated genes could be directly responsible for the spermatogenesis or fertilization. We also investigated the genes at the protein and cellular levels, providing functional perspectives of the genes. The proteins encoded by 14 out of the 24 genes were analyzed in living GC-2 cells. No expression of the other 10 genes might be due to their peculiar protein natures, such as instability, translational delay and toxicity to the cells. Cellular localization of the 14 genes was divided into nucleus, endoplasmic reticulum, Golgi apparatus, and cytoplasm. To further gain an insight into the characteristics of the proteins, we generated antibodies to five proteins. The Western blot analysis disclosed that two proteins were restricted to testicular spermatogenic cells and testicular sperm, while the others were present at all stages, including testicular spermatogenic cells, testicular sperm, and mature sperm. Results from the immunofluorescence analysis of testis sections and mature sperm corroborate and extend the Western blot data. Taken together, our transfection and immuno-analyses provided new information about 16 genes at the protein and cellular levels (Figs. , and ).
Among the 16 genes with the in vitro
data, eight genes have in silico
information congruous with the in vitro
results (Table and Figure ). We attempted to categorize these eight genes based on all the in silico
and in vitro
data, and relate them to potential functions in reproduction (Table ). Three (Mm.290718, Mm86671 and Mm.373242) of the gene products are likely to be involved in transcriptional regulation. All of these proteins were found to be localized in the nucleus of GC-2 cells transfected with the corresponding cDNAs. Mm.437189, predicted to be present in a perinuclear region, was targeted to the nucleus of GC-2 cells. This protein might be related to nuclear activity or integrity of spermatogenic cells. According to a recent report, Mm.437189 belongs to the cysteine-rich perinuclear theca family with potential functions in the remodelling of the spermatid nucleus [20
]. Three of the genes seem to encode proteins implicated in sperm structure and motility. It should be noted that these three genes have been named and reported previously [21
]. Nonetheless, we did not eliminate them because we have obtained new information about these proteins in this investigation. Mm.23377, named Tep22, has been suggested to be involved in the biogenesis of the acrosome and the midpiece of the sperm tail [22
]. Our Western blot analysis newly revealed that the Mm.23377 protein made as a 22 kDa-protein in testicular cells is changed to a higher molecular weight form between the stages of testicular sperm and mature sperm, suggesting post-translational modification. Consistent with this, the protein contains several putative glycosylation and phosphorylation sites [22
]. Mm.159795, identified as CatSper3 [21
], was found localized to the endoplasmic reticulum in GC-2 cells in this study. Other CatSper family members, CatSper1 and CatSper2, are known to be specifically expressed in sperm and linked to sperm motility [24
]. In fact, the expression pattern of the CatSper3 gene and its essential role in sperm motility and male fertility were reported during the preparation of the present paper [27
]. Mm.23534 has been named Tektin3 which belongs to the TEKTIN family [23
]. Tektin2 and Tektin4 are microtubule- or outer dense fiber-associated proteins in sperm flagella [32
]. Here, we provide the first information about the Mm23534 protein, Tektin3. This protein was found to be present at the sperm flagella. It should be noted that the molecular size of the Mm23534 protein was increased during spermiogenesis, suggestive of post-translational modification.
Finally, we also obtained original findings on Mm.333010. The protein encoded by Mm.333010 was targeted to the Golgi apparatus in GC-2 cells. The immuno-analysis uncovered that the Mm333010 protein, 28 kDa, is present in both spermatogenic cells and mature sperm. In particular, the protein was located in the acrosomal region of mature sperm. It is important to mention that Mm.333010 is predicted to contain a trypsin-like serine protease domain. The acrosome is a Golgi-derived secretory granule which is formed during spermiogenesis and positioned at the apex of mature sperm [34
]. When sperm reach the egg extracellular coat, the zona pellucida (ZP), during fertilization, they bind to it and undergo acrosome reaction, releasing the acrosomal contents at the site of sperm-egg binding. The hydrolytic and proteolytic enzymes comprising the acrosomal contents digest the ZP and, thus, enable sperm to penetrate the ZP. The sperm acrosome contains both unique enzymes and common enzymes present in somatic cells [35
]. To date, only a handful of unique enzymes have been identified and enzymes directly responsible for the fertilization process are unknown [36
]. Thus, the Mm.333010 protein is a candidate for a type of protease involved in the penetration of the ZP during fertilization.