One goal of our research is to delineate the genetic network that regulates spermatogenesis. Using cDNA microarrays, we are studying changes in gene expression in type A spermatogonia (Sg), pachytene spermatocytes (Sc), and round spermatids (Sd) of the mouse. We employed two types of microarrays, namely, Nylon-membrane-based Mouse GeneFilters® containing 5,184 mouse genes (ResGen®) and glass-slide microarrays printed with the NIA 15K mouse cDNA clone set. In the GeneFilters® experiments, we identified 79 differentially expressed genes and ESTs in the three types of germ cells. Quantitative real-time PCR confirmed the differential expression of a number of these genes. In the glass-slide microarray experiments, we studied the changes in gene expression in the transition from Sc to Sd. We identified 161 differentially expressed genes. More than one-quarter (43/161) were uncharacterized genes. A larger number of genes (110/161) were preferentially expressed in Sd. Functional categorization indicated that genes of signal transduction, energy metabolism, biosynthesis, and cellular transport were preferentially expressed in Sd while genes for chromatin remodeling were expressed only in Sc. Several testis-specific genes (feminization 1b homolog, phosphatidylcholine transfer protein-like, sperm-specific antigen 1, and cDNA moderately similar to casein kinase) were preferentially expressed in Sd. We achieved confirmation of the results by both quantitative real-time PCR and serial analysis of gene expression (SAGE).

We performed SAGE on Sc and Sd by using the I-SAGE kit (Invitrogen Corp). We sequenced 101,068 and 106,212 tags of the Sc and Sd library, respectively. Excluding singletons, they represented 10,717 and 10,135 genes, respectively. In the Sc library, four tags were present at > 0.5 percent; they matched a mitochondrial sequence (0.65 percent), t-complex-associated testis 3 (0.57 percent), and Y box protein 2 (0.50 percent). The third most abundant tag (0.56 percent) had multiple hits in the SAGE map database. In the Sd library, four tags were present at > 0.5 percent. The most abundant tag was protamine 2 (1.31 percent) followed by that matching FK506 binding protein (1.18 percent) and a tag that matched a mitochondrial sequence (0.64 percent). The fourth most abundant tag (0.56 percent) had multiple hits. Virtual subtraction of the two libraries yielded 4,344 Sc-specific tags and 4,155 Sd-specific tags. The majority of the cell stage–specific tags were present at less than five copies. We found that 353 Sc-specific tags were present at > 5 copies and that only 38 of these were present at > 10 copies. The corresponding figures for Sd-specific tags were 266 and 27. The most abundant Sc-specific tag matched Janus kinase 3 (43 tags, 0.04 percent of library) followed by that matching WW domain binding protein 4 (37 tags, 0.034 percent) and dynein (28 tags, 0.026 percent). Two of the three most abundant Sd-specific tags, CAGAAGGCGG and TATTAAAGCT, both at 18 copies (0.017 percent), were novel with no hit in the SAGE map database. The other tag present at 18 copies matched a RIKEN cDNA.

Comparison of the results obtained by cDNA microarray hybridization and SAGE indicated a high degree of concordance (>80 percent) between the two methods. Discordance was limited to genes of low expression level. Our work identifies a large number of genes previously not known to be expressed in germ cells (characterized genes + ESTs > 12,270) as well as novel genes. It is the first investigation to present a detailed comparison of the gene expression pattern in mouse germ cells. It provides the basis for investigation of genetic regulation of spermatogenesis and for the definition of these processes in pathological conditions.

Table 1

Functional categorization of genes identified in the 15K mouse cDNA microarray experiment. Genes preferentially expressed in pachytene-specific cell type are highlighted.

Physiological and Genetic Effects of Disease-Causing Mutations of the Luteinizing Hormone Receptor
Leung, Wu, Aziz, Pang, Rennert, Chan; in collaboration with Leschek, Martin, Al-Muslim
Constitutively activating mutations of the human luteinizing hormone/chorionic gonadotropin receptor (LHR) cause familial male-limited precocious puberty (FMPP), a non-central form of gonadotropin-independent precocious puberty. In collaboration with Dr. Malcolm Martin and Dr. Ellen Leschek, we have identified two FMPP patients who developed testicular neoplasia. To study the impact of constitutive activation of the LH/hCG signaling pathway on spermatogenesis and sexual development as well as the potential tumorigenic consequences of a constitutively activated LHR, we have generated an in vitro cell model and are in the process of generating a transgenic animal model. We transfected MA-10 cells with LHR carrying activating mutations and compared the cDNA microarray profile of expressed genes in cells expressing the mutated LHR with that of control cells. Preliminary studies of one mutated LHR indicated up-regulation of genes associated with cell proliferation and down-regulation of genes associated with differentiation. Interestingly, several genes known to be involved in spermatogenesis were down-regulated in cells expressing the mutated LHR. We are expanding this experimental approach with the transfection of different mutated LHR genes and an analysis employing more extensive mouse cDNA microarrays.
The antithesis of FMPP is Leydig cell hypoplasia (LCH). In LCH patients, mutation inactivates the LHR, resulting in reduced production of testosterone, which causes hypergonadotrophic hypogonadism or male pseudohermaphroditism. We recently identified a novel inactivating homozygous mutation in the LHR of a 19-year-old patient with male pseudohermaphroditism. The single base substitution T1505C caused replacement of Leu-502 by Pro in the transmembrane helix (TM) IV of LHR. This change disrupted the a helical structure of TM IV and resulted in the inactivation of the LHR. This is the first disease-causing mutation identified in TM IV of the LHR. The transfected mutant receptor is unable to trigger cAMP production following hCG stimulation. The role of TM IV in signal transduction of the LHR is not known. This mutation provides a tool with which to investigate the role of the TM IV in the active-inactive conformation transition of the receptor. It is known that cells process the mutated LHRs abnormally in either activated or inactivated form. To investigate the trafficking of mutated LHR in vitro, we have fused the coding sequence of green fluorescent protein (GFP) to that of wild-type and mutated LHR. We will study the trafficking of the fused protein by fluorescent microscopy. The information generated from this approach should enhance our knowledge on the cellular processing of the LHR.

The clinical impact of the activating mutation of LHR has always been considered to be limited to sexual development. The abnormal social behavior of patients was thought to be secondary to precocious sexual maturation. Expression of the LHR in the brain has been demonstrated. We speculate that the abnormal behavior of FMPP patients is caused by the expression of the mutated LHR in the brain. To examine this hypothesis, we plan to identify the cellular localization of LHR in the brain. To generate the transgenic animals, we are making constructs containing a 2 Kb 5'UTR putative promoter sequence of murine LHR and a GFP-coupled wild-type and mutated LHR for introduction into mouse ES cells. We will use the animal model to study the impact of constitutively activated LHR on spermatogenesis and sexual and neurological development.

Identification of Genetic Risk Factors for Pseudotumor Cerebri in Turkish Population
Dogulu, Leung, Baxendale, Rennert, Chan; in collaboration with Kansu, Ozguc
Pseudotumor cerebri (PTC) is a syndrome characterized by symptoms and signs of isolated intracranial hypertension leading to catastrophic effects on the visual system in the absence of clinical, laboratory, or radiological evidence of a space-occupying lesion or hydrocephalus. The clinical picture is caused by increased cerebro-spinal fluid (CSF) pressure; however, its pathogenesis is not well understood. The most prevalent hypothesis is that the increased CSF pressure is attributable to reduced CSF absorption through the arachnoid villi. Although the majority of cases are non-familial, reports of familial PTC raise the possibility of some genetic variations/defects that lead to clinical manifestations following exposure to a precipitating agent. We hypothesize that PTC is multifactorial and that an underlying genetic thrombotic risk factor predisposes the patients to develop local thrombi lining arachnoid villi, which in turn leads to increased intracranial pressure without demonstrable cerebral venous thrombosis by conventional imaging techniques. Such a genetic variation might occur in coagulation factor V.

Coagulation Factor V is an enzyme cofactor with pivotal functions in hemostasis. Several polymorphisms/mutations have been identified among the 25 exons of the Factor V gene. Altered activity of mutated Factor V is the most common hereditary blood coagulation disorder predisposing to thrombosis. A number of mutations/polymorphisms of the Factor V gene, including VHong Kong (Arg306Gly), VCambridge (Arg306Thr), Arg485Lys, VLeiden (Arg506Gln), and the R2 allele (Arg-1299), all known to be associated with thrombotic risk, are located in exons 7, 10, and13. We scanned exons 7 and 10 of Factor V gene for mutations in 51 PTC patients and 69 controls from Turkey. The G1628A single nucleotide substitution (Arg485Lys replacement) in exon 10 was previously associated with increased risk of coronary artery heart disease in the Chinese population (Le et al., Clin Genet. 2000;57:296). It was found to be significantly associated with the development of PTC in the Turkish population. This is the first study to document a genetic risk factor for PTC. We are in the process of analyzing exon 13 of Factor V and the Factor II gene, which also is an important predisposing component of thrombosis development.