Insertional Mutation of the Murine Kisimo Locus Caused a Defect in Spermatogenesis
2000; Elsevier BV; Volume: 275; Issue: 20 Linguagem: Inglês
10.1074/jbc.c901047199
ISSN1083-351X
AutoresNoriyuki Yanaka, Kinji Kobayashi, Koji Wakimoto, Eriko Yamada, Hiroshi Imahie, Yuji Imai, Chisato Mori,
Tópico(s)Animal Genetics and Reproduction
ResumoSpermatogenesis is a developmental process that occurs in several phases and is regulated by a large number of gene products. An insertional transgenic mouse mutant (termed kisimo mouse) has been isolated that results in abnormal germ-cell development, showing abnormal elongated spermatids in the lumina of seminiferous tubules. We cloned the disrupted locus of kisimo and identified a novel testis-specific gene, THEG, which is specifically expressed in spermatids and was disrupted in the transgenic mouse. The yeast two-hybrid screening method revealed that THEG protein strongly interacts with chaperonin containing t-complex polypeptide-1ε, suggesting that THEG protein functions as a regulatory factor in protein assembly. Our findings indicate that the kisimo locus is essential for the maintenance of spermiogenesis and that a gene expression disorder may be involved in male infertility. Spermatogenesis is a developmental process that occurs in several phases and is regulated by a large number of gene products. An insertional transgenic mouse mutant (termed kisimo mouse) has been isolated that results in abnormal germ-cell development, showing abnormal elongated spermatids in the lumina of seminiferous tubules. We cloned the disrupted locus of kisimo and identified a novel testis-specific gene, THEG, which is specifically expressed in spermatids and was disrupted in the transgenic mouse. The yeast two-hybrid screening method revealed that THEG protein strongly interacts with chaperonin containing t-complex polypeptide-1ε, suggesting that THEG protein functions as a regulatory factor in protein assembly. Our findings indicate that the kisimo locus is essential for the maintenance of spermiogenesis and that a gene expression disorder may be involved in male infertility. phosphodiesterase testicular haploid expressed gene t-complex polypeptide chaperonin containing TCP-1 thermal asymmetric interlaced polymerase chain reaction Dulbecco's modified Eagle's medium fetal calf serum kisimo fluorescence in situ hybridization kilobase cAMP-responsive element modulator The integration of foreign DNA into the mouse germ line by retroviral infection or microinjection can result in insertional disruption of endogenous genes with important roles in development (1.Pellas T.C. Ramachandran B. Duncan M. Pan S.S. Marone M. Chada K. Proc. Natl. Acad. Sci., U. S. A. 1991; 88: 8787-8791Crossref PubMed Scopus (82) Google Scholar, 2.MacGregor G.R. Russell L.D. Van Beek M.E. Hanten G.R. Kovac M.J. Kozak C.A. Meistrich M.L. Overbeek P.A. Proc. Natl. Acad. Sci., U. S. A. 1991; 87: 5016-5020Crossref Scopus (53) Google Scholar, 3.Keller S.A. Liptay S. Hajra A. Meisler M.H. Proc. Natl. Acad. Sci., U. S. A. 1990; 87: 10019-10022Crossref PubMed Scopus (18) Google Scholar). Foreign DNA insertion provides an approach to the cloning of disrupted host loci. By using the introduced DNA as a probe to screen genomic libraries from mutant animals, it has been possible in a few instances to isolate clones that contain DNA flanking the exogenous integrated material and, thus, include portions of the interrupted gene (4.Krulewski T.F. Neumann P.E. Gordon J.W. Proc. Natl. Acad. Sci., U. S. A. 1989; 86: 3709-3712Crossref PubMed Scopus (46) Google Scholar, 5.Magram J. Bishop J.M. Proc. Natl. Acad. Sci., U. S. A. 1991; 88: 10327-10331Crossref PubMed Scopus (19) Google Scholar). We have produced a series of 12 transgenic mouse lines with the human phosphodiesterase 5A (PDE5A)1 gene (6.Yanaka N. Kotera J. Ohtsuka A. Akatsuka H. Imai Y. Michibata H. Fujishige K. Kawai E. Takebayashi S. Okumura K. Omori K. Eur. J. Biochem. 1998; 255: 391-399Crossref PubMed Scopus (110) Google Scholar). As the mice were bred, it became evident that many males of one line were sterile and that the sterility arose from a defect in spermatogenesis (we named this mutant kisimo for a Japanese goddess of easy delivery). Because the sterility segregated with the hemizygous transgene and occurred in the absence of the detectable expression of the transgene, we concluded that the abnormal phenotype was due to mutagenesis by insertion of the transgene. In this study, to clone the junctions between the inserted transgene and adjoining cellular DNA, we used thermal asymmetric interlaced polymerase chain reaction (TAIL-PCR) (7.Liu Y.G. Whittier R.F. Genomics. 1995; 25: 674-681Crossref PubMed Scopus (1001) Google Scholar) and defined a genomic locus important for spermatogenesis.The process of spermatogenesis in the mouse has been well characterized at the morphological level. The spermatogenic process can be subdivided into three main phases. Spermatogonia, the germinal stem cells, undergo mitosis to produce additional spermatogonia, a portion of which develop into primary spermatocytes. The spermatocytes enter meiosis and proceed through two cell divisions to give rise to haploid round spermatids. These, in turn, undergo a complex morphological transformation involving nuclear condensation and elongation resulting in the production of mature spermatozoa. However, at the molecular level, relatively little is known about the control of cellular differentiation and the architectural changes during spermatogenesis.In this study, we found that an insertion of foreign DNA results in abnormal male germ-cell development, showing abnormal elongated spermatids in the lumina of seminiferous tubules with severely abnormal or absent flagella, and that a novel testis-specific gene was disrupted in the transgenic mouse. This novel mouse autosomal recessive mutant exhibited a phenotype similar to asthenospermia and provides us an approach to understand the mechanisms underlying the formation of flagella during spermiogenesis.RESULTS AND DISCUSSIONAll twelve independent transgenic lines carrying the human PDE5A gene (6.Yanaka N. Kotera J. Ohtsuka A. Akatsuka H. Imai Y. Michibata H. Fujishige K. Kawai E. Takebayashi S. Okumura K. Omori K. Eur. J. Biochem. 1998; 255: 391-399Crossref PubMed Scopus (110) Google Scholar) were bred to homozygosity to screen for recessive insertional mutations. One of these mice transmitted the transgene to its progeny, but when they were intercrossed the resulting homozygotes (ki/ki mice) were infertile. The average testis weight ofki/ki mice (46.9 ± 8.2 mg; n = 5) was 60% that of wild-type littermates (78.0 ± 6.3 mg;n = 5) at 12 weeks of age. In wild and heterozygous mice, none of the spermatogenic cell types at stages II-III, V, VII, and XI showed either pathological or quantitative changes. On the contrary, ki/ki mice had virtually no spermatozoa in the lumina of seminiferous and epididymal tubules. As shown in Fig.1 B, elongated spermatids in the vicinity of the lumina of seminiferous tubules showed vacuolation and were occasionally phagocytosed by Sertoli cells. Several studies on male mice that were sterile because of blocked spermatogenesis have demonstrated the appearance of multinucleated giant cells and numerous spermatocytes undergoing apoptotic cell death (13.Roest H.P. van Klaveren J. de Wit J. van Gurp C.G. Koken M.H. Vermey M. van Roijen J.H. Hoogerbrugge J.W. Vreeburg J.T. Baarends W.M. Bootsma D. Grootegoed J.A. Hoeijmakers J.H. Cell. 1996; 86: 799-810Abstract Full Text Full Text PDF PubMed Scopus (343) Google Scholar, 14.Dix D.J. Allen J.W. Collins B.W. Mori C. Nakamura N. Poorman-Allen P. Goulding E.H. Eddy E.M. Proc. Natl. Acad. Sci., U. S. A. 1996; 93: 3264-3268Crossref PubMed Scopus (455) Google Scholar). However, analysis of cross-sections of these testes by Tdt-mediated dUTP-biotin nick-end labeling assay showed the absence of apoptotic cells in the lumina of seminiferous tubules from ki/ki mice (data not shown). Further characterization of spermatids by electron microscopy revealed the elongated spermatids to have abnormal or completely nonexistent flagella. In the elongated spermatids of wild-type mice, the axoneme was composed of microtubules emanating straight from the centriole at the base of the spermatid nucleus (Fig. 1 C). On the contrary, the microtubules and coarse fibers were arranged in a whirl, the nuclei were misshapen, and the cytoplasmic electron density was increased in elongated spermatids of ki/ki mice (Fig.1 D). Intracytoplasmic vacuoles and autophagolysosomes were increased in elongated and some round spermatids. In addition, in the quantitative evaluation of spermatogenic cells in seminiferous tubules, elongated spermatids were significantly decreased at all stages (stage VII is shown in Fig. 1 E). Neither the number of spermatogonia nor that of spermatocytes exhibited significant differences. Furthermore, we analyzed testis RNAs derived from wild-type and ki/ki mice for markers of testis development, post-meiotic-specific cAMP-responsive element modulator (CREM), and haploid-specific protamine 2 transcripts. Neither CREM nor protamine 2 RNA levels were significantly affected by the insertional mutation (Fig. 2 C), also indicating that spermatogenesis was not suppressed in the proliferative and meiotic phases.Figure 2Identification of the mouse kilocus. A, physical map of the kilocus. The direction of THEG transcription is indicated by an arrow. The genomic DNA flanking the transgene insertion site obtained by TAIL-PCR is shown by a bar. Thehatched box represents a DNA fragment used as a probe for Northern blot analysis. Restriction enzymes are indicated as follows:K, KpnI; X, XhoI.B, fluorescence in situ hybridization. The transgene insertion locus was determined to be localized chromosome 10.C, Northern blot analysis probed with theKpnI/XhoI 5-kb genomic fragment fromki locus (top panel) or with testis-specific probes corresponding to protamine 2 (prm-2) or CREM cDNA.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Several copies of the transgene were integrated into theki/ki genome (data not shown). The chromosomal localization of the transgene insertion site was determined by FISH using the whole transgene as a probe, which resulted in a single pair of symmetrical signals mapped to the C1 region of chromosome 10, showing multiple copies of the transgene to be integrated in tandem at a single locus (Fig. 2 B). To clone the junctions between the inserted transgene and adjoining cellular DNA, we employed TAIL-PCR by using degenerated primers and specific primers corresponding to the nucleotide sequence of the transgene. One junction fragment containing DNA flanking the transgene insertion site was identified. A wild-type mouse genomic phage library was screened using the flanking DNA fragment as a probe. These isolated phage clones covered about 25 kb of the mouse genome, showing that the integration of the transgene produced a deletion of about 20 kb in the genomic locus (Fig.2 A).To identify exons in the genomic locus, Northern blot analysis was performed using several DNA fragments (3–6 kb) obtained by endonuclease digestion, covering the deleted region, as a probe. When using 5 kb of the KpnI/XhoI fragment as a probe, a 1.6-kb transcript was detected in the testes from wild-type mice. However, Northern blot analysis showing no signal in testis RNA fromki/ki mice revealed ki/ki mice to be a null strain for this transcript (Fig. 2 C). Next, a mouse testis cDNA library was screened with the 5-kbKpnI/XhoI fragment to isolate the corresponding cDNA. We cloned three cDNAs generated by alternative splicing. The proteins predicted from these nucleotide sequences are 313, 351, and 375 amino acids long (Fig.3 A). In vitrotranscription/translation analysis demonstrated that the mouse cDNAs encode corresponding proteins (data not shown). To isolate a human counterpart, a human testis cDNA library was screened at a reduced stringency using the mouse full-length cDNA as a probe. The isolated human cDNA encoded a 380-amino acid protein and showed 59.6% identity with the mouse protein (Fig.3 B). A search of the database using the nucleotide and amino acid sequences revealed that the isolated cDNA is identical to a novel THEG (15.Nayernia K. von Mering M.H. Kraszucka K. Burfeind P. Wehrend A. Kohler M. Schmid M. Engel W. Biol. Reprod. 1999; 60: 1488-1495Crossref PubMed Scopus (19) Google Scholar). Expression of the THEG is highly specific to the testes of mice (data not shown) and humans (Fig.4 A). To identify the cell types that express the THEG, we performed in situhybridization with a cRNA probe (Fig. 4 B). In testis, the transcripts were detectable only in round and elongated spermatids, consistent with the abnormal spermiogenic process of ki/kimice.Figure 3A, three isoforms of the mouse THEG protein (mTHEG1a, mTHEG1b, andmTHEG1c) are produced by alternative splicing. The hatched line represents the 24-amino acid insertion. mTHEG1c has an independent typical polyadenylation signal in the 3′ untranslated region. B, alignment of amino acid sequences of mTHEG1a (m1a) and its human homologue (h). Identical amino acids are indicated by an asterisk.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 4Expression of THEG . A, Northern blot analysis probed with full-length human THEG cDNA. Each lane contained 2 μg of poly(A) + RNA (human MTN blot,CLONTECH). B, In situhybridization analysis of mouse THEG mRNA expression in testis. Antisense probe gave no signal in testis from ki/kimice (data not shown).View Large Image Figure ViewerDownload Hi-res image Download (PPT)To investigate the potential functions of THEG protein and to determine the mechanism by which these functions are carried out, we employed the yeast two-hybrid system using THEG as bait and a mouse testis MATCHMAKER cDNA library. We found a single positive clone, CCTε, that could interact with the THEG protein (Fig.5 A). To confirm this interaction between THEG protein and CCTε in intact cells, we co-expressed a FLAG epitope-tagged THEG (FLAG-mTHEG1b) with an Xpress epitope-tagged CCTε (Xpress-CCTε) in 293 cells. The ability of antibody against the Xpress epitope to precipitate a complex of THEG and CCTε suggests that the two proteins interact in the cytoplasm (Fig. 5 B). The TCP-1 gene is located in the mouset-complex on chromosome 17 (16.Phillips D.M. Pilder S.H. Olds-Clarke P.J. Silver L.M. Biol. Reprod. 1993; 49: 1347-1352Crossref PubMed Scopus (23) Google Scholar, 17.Silver L.M. Annu. Rev. Genet. 1985; 19: 179-208Crossref PubMed Scopus (261) Google Scholar). A mouset-complex mutation was discovered to produce a phenotype with a tail-less sperm and, to date, TCP-1 identical to the α subunit of CCT has been shown to be highly expressed during haploid stages of spermatogenesis (18.Silver L.M. Kleene K.C. Distel R.J. Hecht N.B. Dev. Biol. 1987; 119: 605-608Crossref PubMed Scopus (42) Google Scholar). The CCT complex is reported to be required for the proper folding or assembly of cytoskeletal proteins under bothin vitro and in vivo conditions (19.Kubota H. Hynes G. Willison K. Eur. J. Biochem. 1995; 230: 3-16Crossref PubMed Scopus (256) Google Scholar, 20.Sternlicht H. Farr G.W. Sternlicht M.L. Driscoll J.K. Willison K. Yaffe M.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 9422-9426Crossref PubMed Scopus (280) Google Scholar, 21.Yaffe M.B. Farr G.W. Miklos D. Horwich A.L. Sternlicht M.L. Sternlicht H. Nature. 1992; 358: 245-248Crossref PubMed Scopus (376) Google Scholar). In the testis, one possible candidate for such a substrate may be α tubulin; α-tubulin has constitutively expressed subtypes and testis-specific subtypes (22.Villasante A. Wang D. Dobner P. Dolph P. Lewis S.A. Cowan N.J. Mol. Cell. Biol. 1986; 6: 2409-2419Crossref PubMed Scopus (223) Google Scholar, 23.Pratt L.F. Okamura S. Cleveland D.N. Mol. Cell. Biol. 1987; 7: 552-555Crossref PubMed Scopus (36) Google Scholar). Taken together with the previous observations that mutations in individual CCT subunits affect microtubule assembly and produce morphologically abnormal structures detected by anti-α-tubulin antibodies in yeast (24.Ursic D. Culbertson M.R. Mol. Cell. Biol. 1991; 11: 2629-2640Crossref PubMed Scopus (141) Google Scholar), abnormal or absent flagella in ki/ki mice appear to be associated with impairment of the assembly of cytoskeletal proteins such as the tubulins that are major structural proteins of flagella.Figure 5A, two-hybrid analysis of THEG and CCTε stained by in situ β-galactosidase assay.B, interaction of THEG with CCTε in vivo. 293 cells were transiently transfected with FLAG-mTHEG1b an/or Xpress-CCTε constructs, and cell extracts were immunoprecipitated (IP) with antibody specific to Xpress (IP: Xpress) and immunoblotted with anti-FLAG (Blot: FLAG). Total cell extracts were blotted with anti-Xpress or anti-FLAG antibody.View Large Image Figure ViewerDownload Hi-res image Download (PPT)In the sperm flagellum, as in the epithelial cilia, the cytoskeleton is highly differentiated to permit motility. The basic element of the cytoskeleton is the axoneme, which has an identical structure in both cilia and flagella. Structural anomalies of the axoneme, which are associated with impaired motility, have been described inChlamydomonas (25.Mitchell D.R. Sale W.S. J. Cell Biol. 1999; 144: 293-304Crossref PubMed Scopus (102) Google Scholar). On the other hand, asthenozoospermia is a very frequent cause of male infertility. Previous reports demonstrated that ultrastructural abnormality of spermatozoa is observed in asthenozoospermic or teratozoospermic sterile men (26.Gentleman S. Kaiser-Kupfer M.I. Sherins R.J. Caruso R. Robison Jr., W.G. Lloyd-Muhammad R.A. Crawford M.A. Pikus A. Chader G.J. Hum. Pathol. 1996; 27: 80-84Crossref PubMed Scopus (7) Google Scholar, 27.Chemes H.E. Olmedo S.B. Carrere C. Oses R. Carizza C. Leisner M. Blaquier J. Hum. Reprod. 1998; 13: 2521-2526Crossref PubMed Scopus (136) Google Scholar). A high incidence of flagellar pathology was found to be the underlying cause of motility disorders in severely asthenozoospermic patients (28.Escalier D. David G. Biol. Cell. 1984; 50: 37-52Crossref PubMed Scopus (80) Google Scholar). In mammalian spermatozoa the flagellum is distinguished by adjoining structures, particularly the dense fibers and fibrous sheath. In particular, poor development of outer dense fibers is considered to be a major cause of tail abnormality; however, research in this area has been limited by the lack of appropriate animal models. The present study indicated that availability of this novel mouse autosomal recessive mutant showing a phenotype similar to asthenospermia enables us to investigate the developmental mechanisms underlying the formation of flagella. The integration of foreign DNA into the mouse germ line by retroviral infection or microinjection can result in insertional disruption of endogenous genes with important roles in development (1.Pellas T.C. Ramachandran B. Duncan M. Pan S.S. Marone M. Chada K. Proc. Natl. Acad. Sci., U. S. A. 1991; 88: 8787-8791Crossref PubMed Scopus (82) Google Scholar, 2.MacGregor G.R. Russell L.D. Van Beek M.E. Hanten G.R. Kovac M.J. Kozak C.A. Meistrich M.L. Overbeek P.A. Proc. Natl. Acad. Sci., U. S. A. 1991; 87: 5016-5020Crossref Scopus (53) Google Scholar, 3.Keller S.A. Liptay S. Hajra A. Meisler M.H. Proc. Natl. Acad. Sci., U. S. A. 1990; 87: 10019-10022Crossref PubMed Scopus (18) Google Scholar). Foreign DNA insertion provides an approach to the cloning of disrupted host loci. By using the introduced DNA as a probe to screen genomic libraries from mutant animals, it has been possible in a few instances to isolate clones that contain DNA flanking the exogenous integrated material and, thus, include portions of the interrupted gene (4.Krulewski T.F. Neumann P.E. Gordon J.W. Proc. Natl. Acad. Sci., U. S. A. 1989; 86: 3709-3712Crossref PubMed Scopus (46) Google Scholar, 5.Magram J. Bishop J.M. Proc. Natl. Acad. Sci., U. S. A. 1991; 88: 10327-10331Crossref PubMed Scopus (19) Google Scholar). We have produced a series of 12 transgenic mouse lines with the human phosphodiesterase 5A (PDE5A)1 gene (6.Yanaka N. Kotera J. Ohtsuka A. Akatsuka H. Imai Y. Michibata H. Fujishige K. Kawai E. Takebayashi S. Okumura K. Omori K. Eur. J. Biochem. 1998; 255: 391-399Crossref PubMed Scopus (110) Google Scholar). As the mice were bred, it became evident that many males of one line were sterile and that the sterility arose from a defect in spermatogenesis (we named this mutant kisimo for a Japanese goddess of easy delivery). Because the sterility segregated with the hemizygous transgene and occurred in the absence of the detectable expression of the transgene, we concluded that the abnormal phenotype was due to mutagenesis by insertion of the transgene. In this study, to clone the junctions between the inserted transgene and adjoining cellular DNA, we used thermal asymmetric interlaced polymerase chain reaction (TAIL-PCR) (7.Liu Y.G. Whittier R.F. Genomics. 1995; 25: 674-681Crossref PubMed Scopus (1001) Google Scholar) and defined a genomic locus important for spermatogenesis. The process of spermatogenesis in the mouse has been well characterized at the morphological level. The spermatogenic process can be subdivided into three main phases. Spermatogonia, the germinal stem cells, undergo mitosis to produce additional spermatogonia, a portion of which develop into primary spermatocytes. The spermatocytes enter meiosis and proceed through two cell divisions to give rise to haploid round spermatids. These, in turn, undergo a complex morphological transformation involving nuclear condensation and elongation resulting in the production of mature spermatozoa. However, at the molecular level, relatively little is known about the control of cellular differentiation and the architectural changes during spermatogenesis. In this study, we found that an insertion of foreign DNA results in abnormal male germ-cell development, showing abnormal elongated spermatids in the lumina of seminiferous tubules with severely abnormal or absent flagella, and that a novel testis-specific gene was disrupted in the transgenic mouse. This novel mouse autosomal recessive mutant exhibited a phenotype similar to asthenospermia and provides us an approach to understand the mechanisms underlying the formation of flagella during spermiogenesis. RESULTS AND DISCUSSIONAll twelve independent transgenic lines carrying the human PDE5A gene (6.Yanaka N. Kotera J. Ohtsuka A. Akatsuka H. Imai Y. Michibata H. Fujishige K. Kawai E. Takebayashi S. Okumura K. Omori K. Eur. J. Biochem. 1998; 255: 391-399Crossref PubMed Scopus (110) Google Scholar) were bred to homozygosity to screen for recessive insertional mutations. One of these mice transmitted the transgene to its progeny, but when they were intercrossed the resulting homozygotes (ki/ki mice) were infertile. The average testis weight ofki/ki mice (46.9 ± 8.2 mg; n = 5) was 60% that of wild-type littermates (78.0 ± 6.3 mg;n = 5) at 12 weeks of age. In wild and heterozygous mice, none of the spermatogenic cell types at stages II-III, V, VII, and XI showed either pathological or quantitative changes. On the contrary, ki/ki mice had virtually no spermatozoa in the lumina of seminiferous and epididymal tubules. As shown in Fig.1 B, elongated spermatids in the vicinity of the lumina of seminiferous tubules showed vacuolation and were occasionally phagocytosed by Sertoli cells. Several studies on male mice that were sterile because of blocked spermatogenesis have demonstrated the appearance of multinucleated giant cells and numerous spermatocytes undergoing apoptotic cell death (13.Roest H.P. van Klaveren J. de Wit J. van Gurp C.G. Koken M.H. Vermey M. van Roijen J.H. Hoogerbrugge J.W. Vreeburg J.T. Baarends W.M. Bootsma D. Grootegoed J.A. Hoeijmakers J.H. Cell. 1996; 86: 799-810Abstract Full Text Full Text PDF PubMed Scopus (343) Google Scholar, 14.Dix D.J. Allen J.W. Collins B.W. Mori C. Nakamura N. Poorman-Allen P. Goulding E.H. Eddy E.M. Proc. Natl. Acad. Sci., U. S. A. 1996; 93: 3264-3268Crossref PubMed Scopus (455) Google Scholar). However, analysis of cross-sections of these testes by Tdt-mediated dUTP-biotin nick-end labeling assay showed the absence of apoptotic cells in the lumina of seminiferous tubules from ki/ki mice (data not shown). Further characterization of spermatids by electron microscopy revealed the elongated spermatids to have abnormal or completely nonexistent flagella. In the elongated spermatids of wild-type mice, the axoneme was composed of microtubules emanating straight from the centriole at the base of the spermatid nucleus (Fig. 1 C). On the contrary, the microtubules and coarse fibers were arranged in a whirl, the nuclei were misshapen, and the cytoplasmic electron density was increased in elongated spermatids of ki/ki mice (Fig.1 D). Intracytoplasmic vacuoles and autophagolysosomes were increased in elongated and some round spermatids. In addition, in the quantitative evaluation of spermatogenic cells in seminiferous tubules, elongated spermatids were significantly decreased at all stages (stage VII is shown in Fig. 1 E). Neither the number of spermatogonia nor that of spermatocytes exhibited significant differences. Furthermore, we analyzed testis RNAs derived from wild-type and ki/ki mice for markers of testis development, post-meiotic-specific cAMP-responsive element modulator (CREM), and haploid-specific protamine 2 transcripts. Neither CREM nor protamine 2 RNA levels were significantly affected by the insertional mutation (Fig. 2 C), also indicating that spermatogenesis was not suppressed in the proliferative and meiotic phases.Several copies of the transgene were integrated into theki/ki genome (data not shown). The chromosomal localization of the transgene insertion site was determined by FISH using the whole transgene as a probe, which resulted in a single pair of symmetrical signals mapped to the C1 region of chromosome 10, showing multiple copies of the transgene to be integrated in tandem at a single locus (Fig. 2 B). To clone the junctions between the inserted transgene and adjoining cellular DNA, we employed TAIL-PCR by using degenerated primers and specific primers corresponding to the nucleotide sequence of the transgene. One junction fragment containing DNA flanking the transgene insertion site was identified. A wild-type mouse genomic phage library was screened using the flanking DNA fragment as a probe. These isolated phage clones covered about 25 kb of the mouse genome, showing that the integration of the transgene produced a deletion of about 20 kb in the genomic locus (Fig.2 A).To identify exons in the genomic locus, Northern blot analysis was performed using several DNA fragments (3–6 kb) obtained by endonuclease digestion, covering the deleted region, as a probe. When using 5 kb of the KpnI/XhoI fragment as a probe, a 1.6-kb transcript was detected in the testes from wild-type mice. However, Northern blot analysis showing no signal in testis RNA fromki/ki mice revealed ki/ki mice to be a null strain for this transcript (Fig. 2 C). Next, a mouse testis cDNA library was screened with the 5-kbKpnI/XhoI fragment to isolate the corresponding cDNA. We cloned three cDNAs generated by alternative splicing. The proteins predicted from these nucleotide sequences are 313, 351, and 375 amino acids long (Fig.3 A). In vitrotranscription/translation analysis demonstrated that the mouse cDNAs encode corresponding proteins (data not shown). To isolate a human counterpart, a human testis cDNA library was screened at a reduced stringency using the mouse full-length cDNA as a probe. The isolated human cDNA encoded a 380-amino acid protein and showed 59.6% identity with the mouse protein (Fig.3 B). A search of the database using the nucleotide and amino acid sequences revealed that the isolated cDNA is identical to a novel THEG (15.Nayernia K. von Mering M.H. Kraszucka K. Burfeind P. Wehrend A. Kohler M. Schmid M. Engel W. Biol. Reprod. 1999; 60: 1488-1495Crossref PubMed Scopus (19) Google Scholar). Expression of the THEG is highly specific to the testes of mice (data not shown) and humans (Fig.4 A). To identify the cell types that express the THEG, we performed in situhybridization with a cRNA probe (Fig. 4 B). In testis, the transcripts were detectable only in round and elongated spermatids, consistent with the abnormal spermiogenic process of ki/kimice.Figure 3A, three isoforms of the mouse THEG protein (mTHEG1a, mTHEG1b, andmTHEG1c) are produced by alternative splicing. The hatched line represents the 24-amino acid insertion. mTHEG1c has an independent typical polyadenylation signal in the 3′ untranslated region. B, alignment of amino acid sequences of mTHEG1a (m1a) and its human homologue (h). Identical amino acids are indicated by an asterisk.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 4Expression of THEG . A, Northern blot analysis probed with full-length human THEG cDNA. Each lane contained 2 μg of poly(A) + RNA (human MTN blot,CLONTECH). B, In situhybridization analysis of mouse THEG mRNA expression in testis. Antisense probe gave no signal in testis from ki/kimice (data not shown).View Large Image Figure ViewerDownload Hi-res image Download (PPT)To investigate the potential functions of THEG protein and to determine the mechanism by which these functions are carried out, we employed the yeast two-hybrid system using THEG as bait and a mouse testis MATCHMAKER cDNA library. We found a single positive clone, CCTε, that could interact with the THEG protein (Fig.5 A). To confirm this interaction between THEG protein and CCTε in intact cells, we co-expressed a FLAG epitope-tagged THEG (FLAG-mTHEG1b) with an Xpress epitope-tagged CCTε (Xpress-CCTε) in 293 cells. The ability of antibody against the Xpress epitope to precipitate a complex of THEG and CCTε suggests that the two proteins interact in the cytoplasm (Fig. 5 B). The TCP-1 gene is located in the mouset-complex on chromosome 17 (16.Phillips D.M. Pilder S.H. Olds-Clarke P.J. Silver L.M. Biol. Reprod. 1993; 49: 1347-1352Crossref PubMed Scopus (23) Google Scholar, 17.Silver L.M. Annu. Rev. Genet. 1985; 19: 179-208Crossref PubMed Scopus (261) Google Scholar). A mouset-complex mutation was discovered to produce a phenotype with a tail-less sperm and, to date, TCP-1 identical to the α subunit of CCT has been shown to be highly expressed during haploid stages of spermatogenesis (18.Silver L.M. Kleene K.C. Distel R.J. Hecht N.B. Dev. Biol. 1987; 119: 605-608Crossref PubMed Scopus (42) Google Scholar). The CCT complex is reported to be required for the proper folding or assembly of cytoskeletal proteins under bothin vitro and in vivo conditions (19.Kubota H. Hynes G. Willison K. Eur. J. Biochem. 1995; 230: 3-16Crossref PubMed Scopus (256) Google Scholar, 20.Sternlicht H. Farr G.W. Sternlicht M.L. Driscoll J.K. Willison K. Yaffe M.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 9422-9426Crossref PubMed Scopus (280) Google Scholar, 21.Yaffe M.B. Farr G.W. Miklos D. Horwich A.L. Sternlicht M.L. Sternlicht H. Nature. 1992; 358: 245-248Crossref PubMed Scopus (376) Google Scholar). In the testis, one possible candidate for such a substrate may be α tubulin; α-tubulin has constitutively expressed subtypes and testis-specific subtypes (22.Villasante A. Wang D. Dobner P. Dolph P. Lewis S.A. Cowan N.J. Mol. Cell. Biol. 1986; 6: 2409-2419Crossref PubMed Scopus (223) Google Scholar, 23.Pratt L.F. Okamura S. Cleveland D.N. Mol. Cell. Biol. 1987; 7: 552-555Crossref PubMed Scopus (36) Google Scholar). Taken together with the previous observations that mutations in individual CCT subunits affect microtubule assembly and produce morphologically abnormal structures detected by anti-α-tubulin antibodies in yeast (24.Ursic D. Culbertson M.R. Mol. Cell. Biol. 1991; 11: 2629-2640Crossref PubMed Scopus (141) Google Scholar), abnormal or absent flagella in ki/ki mice appear to be associated with impairment of the assembly of cytoskeletal proteins such as the tubulins that are major structural proteins of flagella.Figure 5A, two-hybrid analysis of THEG and CCTε stained by in situ β-galactosidase assay.B, interaction of THEG with CCTε in vivo. 293 cells were transiently transfected with FLAG-mTHEG1b an/or Xpress-CCTε constructs, and cell extracts were immunoprecipitated (IP) with antibody specific to Xpress (IP: Xpress) and immunoblotted with anti-FLAG (Blot: FLAG). Total cell extracts were blotted with anti-Xpress or anti-FLAG antibody.View Large Image Figure ViewerDownload Hi-res image Download (PPT)In the sperm flagellum, as in the epithelial cilia, the cytoskeleton is highly differentiated to permit motility. The basic element of the cytoskeleton is the axoneme, which has an identical structure in both cilia and flagella. Structural anomalies of the axoneme, which are associated with impaired motility, have been described inChlamydomonas (25.Mitchell D.R. Sale W.S. J. Cell Biol. 1999; 144: 293-304Crossref PubMed Scopus (102) Google Scholar). On the other hand, asthenozoospermia is a very frequent cause of male infertility. Previous reports demonstrated that ultrastructural abnormality of spermatozoa is observed in asthenozoospermic or teratozoospermic sterile men (26.Gentleman S. Kaiser-Kupfer M.I. Sherins R.J. Caruso R. Robison Jr., W.G. Lloyd-Muhammad R.A. Crawford M.A. Pikus A. Chader G.J. Hum. Pathol. 1996; 27: 80-84Crossref PubMed Scopus (7) Google Scholar, 27.Chemes H.E. Olmedo S.B. Carrere C. Oses R. Carizza C. Leisner M. Blaquier J. Hum. Reprod. 1998; 13: 2521-2526Crossref PubMed Scopus (136) Google Scholar). A high incidence of flagellar pathology was found to be the underlying cause of motility disorders in severely asthenozoospermic patients (28.Escalier D. David G. Biol. Cell. 1984; 50: 37-52Crossref PubMed Scopus (80) Google Scholar). In mammalian spermatozoa the flagellum is distinguished by adjoining structures, particularly the dense fibers and fibrous sheath. In particular, poor development of outer dense fibers is considered to be a major cause of tail abnormality; however, research in this area has been limited by the lack of appropriate animal models. The present study indicated that availability of this novel mouse autosomal recessive mutant showing a phenotype similar to asthenospermia enables us to investigate the developmental mechanisms underlying the formation of flagella. All twelve independent transgenic lines carrying the human PDE5A gene (6.Yanaka N. Kotera J. Ohtsuka A. Akatsuka H. Imai Y. Michibata H. Fujishige K. Kawai E. Takebayashi S. Okumura K. Omori K. Eur. J. Biochem. 1998; 255: 391-399Crossref PubMed Scopus (110) Google Scholar) were bred to homozygosity to screen for recessive insertional mutations. One of these mice transmitted the transgene to its progeny, but when they were intercrossed the resulting homozygotes (ki/ki mice) were infertile. The average testis weight ofki/ki mice (46.9 ± 8.2 mg; n = 5) was 60% that of wild-type littermates (78.0 ± 6.3 mg;n = 5) at 12 weeks of age. In wild and heterozygous mice, none of the spermatogenic cell types at stages II-III, V, VII, and XI showed either pathological or quantitative changes. On the contrary, ki/ki mice had virtually no spermatozoa in the lumina of seminiferous and epididymal tubules. As shown in Fig.1 B, elongated spermatids in the vicinity of the lumina of seminiferous tubules showed vacuolation and were occasionally phagocytosed by Sertoli cells. Several studies on male mice that were sterile because of blocked spermatogenesis have demonstrated the appearance of multinucleated giant cells and numerous spermatocytes undergoing apoptotic cell death (13.Roest H.P. van Klaveren J. de Wit J. van Gurp C.G. Koken M.H. Vermey M. van Roijen J.H. Hoogerbrugge J.W. Vreeburg J.T. Baarends W.M. Bootsma D. Grootegoed J.A. Hoeijmakers J.H. Cell. 1996; 86: 799-810Abstract Full Text Full Text PDF PubMed Scopus (343) Google Scholar, 14.Dix D.J. Allen J.W. Collins B.W. Mori C. Nakamura N. Poorman-Allen P. Goulding E.H. Eddy E.M. Proc. Natl. Acad. Sci., U. S. A. 1996; 93: 3264-3268Crossref PubMed Scopus (455) Google Scholar). However, analysis of cross-sections of these testes by Tdt-mediated dUTP-biotin nick-end labeling assay showed the absence of apoptotic cells in the lumina of seminiferous tubules from ki/ki mice (data not shown). Further characterization of spermatids by electron microscopy revealed the elongated spermatids to have abnormal or completely nonexistent flagella. In the elongated spermatids of wild-type mice, the axoneme was composed of microtubules emanating straight from the centriole at the base of the spermatid nucleus (Fig. 1 C). On the contrary, the microtubules and coarse fibers were arranged in a whirl, the nuclei were misshapen, and the cytoplasmic electron density was increased in elongated spermatids of ki/ki mice (Fig.1 D). Intracytoplasmic vacuoles and autophagolysosomes were increased in elongated and some round spermatids. In addition, in the quantitative evaluation of spermatogenic cells in seminiferous tubules, elongated spermatids were significantly decreased at all stages (stage VII is shown in Fig. 1 E). Neither the number of spermatogonia nor that of spermatocytes exhibited significant differences. Furthermore, we analyzed testis RNAs derived from wild-type and ki/ki mice for markers of testis development, post-meiotic-specific cAMP-responsive element modulator (CREM), and haploid-specific protamine 2 transcripts. Neither CREM nor protamine 2 RNA levels were significantly affected by the insertional mutation (Fig. 2 C), also indicating that spermatogenesis was not suppressed in the proliferative and meiotic phases. Several copies of the transgene were integrated into theki/ki genome (data not shown). The chromosomal localization of the transgene insertion site was determined by FISH using the whole transgene as a probe, which resulted in a single pair of symmetrical signals mapped to the C1 region of chromosome 10, showing multiple copies of the transgene to be integrated in tandem at a single locus (Fig. 2 B). To clone the junctions between the inserted transgene and adjoining cellular DNA, we employed TAIL-PCR by using degenerated primers and specific primers corresponding to the nucleotide sequence of the transgene. One junction fragment containing DNA flanking the transgene insertion site was identified. A wild-type mouse genomic phage library was screened using the flanking DNA fragment as a probe. These isolated phage clones covered about 25 kb of the mouse genome, showing that the integration of the transgene produced a deletion of about 20 kb in the genomic locus (Fig.2 A). To identify exons in the genomic locus, Northern blot analysis was performed using several DNA fragments (3–6 kb) obtained by endonuclease digestion, covering the deleted region, as a probe. When using 5 kb of the KpnI/XhoI fragment as a probe, a 1.6-kb transcript was detected in the testes from wild-type mice. However, Northern blot analysis showing no signal in testis RNA fromki/ki mice revealed ki/ki mice to be a null strain for this transcript (Fig. 2 C). Next, a mouse testis cDNA library was screened with the 5-kbKpnI/XhoI fragment to isolate the corresponding cDNA. We cloned three cDNAs generated by alternative splicing. The proteins predicted from these nucleotide sequences are 313, 351, and 375 amino acids long (Fig.3 A). In vitrotranscription/translation analysis demonstrated that the mouse cDNAs encode corresponding proteins (data not shown). To isolate a human counterpart, a human testis cDNA library was screened at a reduced stringency using the mouse full-length cDNA as a probe. The isolated human cDNA encoded a 380-amino acid protein and showed 59.6% identity with the mouse protein (Fig.3 B). A search of the database using the nucleotide and amino acid sequences revealed that the isolated cDNA is identical to a novel THEG (15.Nayernia K. von Mering M.H. Kraszucka K. Burfeind P. Wehrend A. Kohler M. Schmid M. Engel W. Biol. Reprod. 1999; 60: 1488-1495Crossref PubMed Scopus (19) Google Scholar). Expression of the THEG is highly specific to the testes of mice (data not shown) and humans (Fig.4 A). To identify the cell types that express the THEG, we performed in situhybridization with a cRNA probe (Fig. 4 B). In testis, the transcripts were detectable only in round and elongated spermatids, consistent with the abnormal spermiogenic process of ki/kimice. To investigate the potential functions of THEG protein and to determine the mechanism by which these functions are carried out, we employed the yeast two-hybrid system using THEG as bait and a mouse testis MATCHMAKER cDNA library. We found a single positive clone, CCTε, that could interact with the THEG protein (Fig.5 A). To confirm this interaction between THEG protein and CCTε in intact cells, we co-expressed a FLAG epitope-tagged THEG (FLAG-mTHEG1b) with an Xpress epitope-tagged CCTε (Xpress-CCTε) in 293 cells. The ability of antibody against the Xpress epitope to precipitate a complex of THEG and CCTε suggests that the two proteins interact in the cytoplasm (Fig. 5 B). The TCP-1 gene is located in the mouset-complex on chromosome 17 (16.Phillips D.M. Pilder S.H. Olds-Clarke P.J. Silver L.M. Biol. Reprod. 1993; 49: 1347-1352Crossref PubMed Scopus (23) Google Scholar, 17.Silver L.M. Annu. Rev. Genet. 1985; 19: 179-208Crossref PubMed Scopus (261) Google Scholar). A mouset-complex mutation was discovered to produce a phenotype with a tail-less sperm and, to date, TCP-1 identical to the α subunit of CCT has been shown to be highly expressed during haploid stages of spermatogenesis (18.Silver L.M. Kleene K.C. Distel R.J. Hecht N.B. Dev. Biol. 1987; 119: 605-608Crossref PubMed Scopus (42) Google Scholar). The CCT complex is reported to be required for the proper folding or assembly of cytoskeletal proteins under bothin vitro and in vivo conditions (19.Kubota H. Hynes G. Willison K. Eur. J. Biochem. 1995; 230: 3-16Crossref PubMed Scopus (256) Google Scholar, 20.Sternlicht H. Farr G.W. Sternlicht M.L. Driscoll J.K. Willison K. Yaffe M.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 9422-9426Crossref PubMed Scopus (280) Google Scholar, 21.Yaffe M.B. Farr G.W. Miklos D. Horwich A.L. Sternlicht M.L. Sternlicht H. Nature. 1992; 358: 245-248Crossref PubMed Scopus (376) Google Scholar). In the testis, one possible candidate for such a substrate may be α tubulin; α-tubulin has constitutively expressed subtypes and testis-specific subtypes (22.Villasante A. Wang D. Dobner P. Dolph P. Lewis S.A. Cowan N.J. Mol. Cell. Biol. 1986; 6: 2409-2419Crossref PubMed Scopus (223) Google Scholar, 23.Pratt L.F. Okamura S. Cleveland D.N. Mol. Cell. Biol. 1987; 7: 552-555Crossref PubMed Scopus (36) Google Scholar). Taken together with the previous observations that mutations in individual CCT subunits affect microtubule assembly and produce morphologically abnormal structures detected by anti-α-tubulin antibodies in yeast (24.Ursic D. Culbertson M.R. Mol. Cell. Biol. 1991; 11: 2629-2640Crossref PubMed Scopus (141) Google Scholar), abnormal or absent flagella in ki/ki mice appear to be associated with impairment of the assembly of cytoskeletal proteins such as the tubulins that are major structural proteins of flagella. In the sperm flagellum, as in the epithelial cilia, the cytoskeleton is highly differentiated to permit motility. The basic element of the cytoskeleton is the axoneme, which has an identical structure in both cilia and flagella. Structural anomalies of the axoneme, which are associated with impaired motility, have been described inChlamydomonas (25.Mitchell D.R. Sale W.S. J. Cell Biol. 1999; 144: 293-304Crossref PubMed Scopus (102) Google Scholar). On the other hand, asthenozoospermia is a very frequent cause of male infertility. Previous reports demonstrated that ultrastructural abnormality of spermatozoa is observed in asthenozoospermic or teratozoospermic sterile men (26.Gentleman S. Kaiser-Kupfer M.I. Sherins R.J. Caruso R. Robison Jr., W.G. Lloyd-Muhammad R.A. Crawford M.A. Pikus A. Chader G.J. Hum. Pathol. 1996; 27: 80-84Crossref PubMed Scopus (7) Google Scholar, 27.Chemes H.E. Olmedo S.B. Carrere C. Oses R. Carizza C. Leisner M. Blaquier J. Hum. Reprod. 1998; 13: 2521-2526Crossref PubMed Scopus (136) Google Scholar). A high incidence of flagellar pathology was found to be the underlying cause of motility disorders in severely asthenozoospermic patients (28.Escalier D. David G. Biol. Cell. 1984; 50: 37-52Crossref PubMed Scopus (80) Google Scholar). In mammalian spermatozoa the flagellum is distinguished by adjoining structures, particularly the dense fibers and fibrous sheath. In particular, poor development of outer dense fibers is considered to be a major cause of tail abnormality; however, research in this area has been limited by the lack of appropriate animal models. The present study indicated that availability of this novel mouse autosomal recessive mutant showing a phenotype similar to asthenospermia enables us to investigate the developmental mechanisms underlying the formation of flagella. We thank Y. Hamasaki, H. Chiba, and K. Muguruma for their technical support. We are grateful to Y. Kondo, H. Sakurai, K. Omori, T. Nishimura, and C. Aruga for their technical advice. We also thank S. Nito and M. Sugiura for their continuous kindness.
Referência(s)