Association of the Mouse Infertility Factor DAZL1 with Actively Translating Polyribosomes 1

Transcription

1 BIOLOGY OF REPRODUCTION 62, (2000) Association of the Mouse Infertility Factor DAZL1 with Actively Translating Polyribosomes 1 Shanli Tsui, 3 Tiane Dai, 3 Stephen T. Warren, 4 Eduardo C. Salido, 5 and Pauline H. Yen 2,3 Department of Pediatrics, 3 Harbor-UCLA Medical Center, Torrance, California Howard Hughes Medical Institute, 4 Departments of Biochemistry, Pediatrics and Genetics, Emory University School of Medicine, Atlanta, Georgia Unidad de Investigacion, 5 Hospital Universitario de Canarias and Department of Pathology, Universidad de La Laguna, E Tenerife, Spain ABSTRACT The DAZ (Deleted in AZoospermia) gene family was isolated from a region of the human Y chromosome long arm that is deleted in about 10% of infertile men with idiopathic azoospermia. DAZ and an autosomal DAZ-like gene, DAZL1, are expressed in germ cells only. They encode proteins with an RNA recognition motif and with either a single copy (in DAZL1) or multiple copies (in DAZ) of a DAZ repeat. A role for DAZL1 and DAZ in spermatogenesis is supported by their homology to a Drosophila male infertility protein Boule and by sterility of Dazl1 knock-out mice. The biological function of these proteins remains unknown. We found that DAZL1 and DAZ bound similarly to various RNA homopolymers in vitro. We also used an antibody against the human DAZL1 to determine the subcellular localization of DAZL1 in mouse testis. The sedimentation profiles of DAZL1 in sucrose gradients indicate that DAZL1 is associated with polyribosomes, and further capture of DAZL1 on oligo(dt) beads demonstrates that the association is mediated through the binding of DAZL1 to poly(a) RNA. Our results suggest that DAZL1 is involved in germ-cell specific regulation of mrna translation. INTRODUCTION A requirement of the Y chromosome long arm (Yq) for normal spermatogenesis was recognized more than two decades ago when six infertile men were found to have Yq deletions [1]. More recent molecular screening has detected Yq microdeletions in 10 15% of males with idiopathic azoospermia [2 4]. The DAZ (Deleted in AZoospermia) gene family was isolated from one of the deleted regions [5]. There are seven DAZ genes clustered within a 1-Mbp region in interval 6 of Yq [6]. DAZ homologues are present on the Y chromosomes of only great apes and old-world monkeys [7], yet all mammals contain a single-copy DAZlike gene, DAZL1, on their autosomes [7 12]. DAZ and DAZL1 encode proteins with an RNA recognition motif (RRM) and varying numbers of copies of a DAZ repeat. DAZ has 8 24 copies of the repeat, and DAZL1 has a single copy [5, 13]. DAZ and DAZL1 have 85% sequence similarity over most of their lengths, including the RRM 1 This work was supported by NIH grants HD28009 and HD36347 (P.H.Y.). S.T.W. is an Investigator of the Howard Hughes Medical Institute. 2 Correspondence: Pauline Yen, Division of Medical Genetics, Harbor- UCLA Medical Center, 1124 W. Carson Street, Torrance, CA FAX: ; pyen@rei.edu Received: 13 October First decision: 5 November Accepted: 7 January by the Society for the Study of Reproduction, Inc. ISSN: domain and the DAZ repeats, but they have different C- terminal sequences due to a frame shift downstream of the DAZ repeat region. It was proposed that the DAZ genes originated from translocation of an ancestral DAZL1 gene to the Y chromosome, followed by amplification and pruning [10]. A role for DAZ and DAZL1 in spermatogenesis is supported by their exclusive expression in germ cells, their homology to a Drosophila male infertility gene, boule [14], and the sterility of Dazl1 knock-out mice [15]. Male flies with the boule mutation have morphologically normal primary spermatocytes, which fail to enter into meiotic division, suggesting that boule is required for the G2/M transition. Dazl1 knock-out mice are sterile in both sexes, and male mice exhibit a spermatogenic defect significantly different from that of boule flies [15]. The seminiferous tubules contain only a few premeiotic spermatogonia that rarely progress into meiosis, suggesting that Dazl1 is required for both the development and the maintenance of the germ cells. The spermatogenic defects of boule flies and Dazl1 knock-out mice were partially rescued by a Xenopus Xdazl gene [16] and a human DAZ gene [17], respectively, indicating conservation of the function of DAZ and Dazl1. The biological function of DAZ and DAZL1 is unknown. Immunostaining of mouse testicular sections detected DAZL1 abundantly in the cytoplasm of pachytene spermatocytes and to a lesser degree in the cytoplasm of type- B spermatogonia and preleptotene and zygotene spermatocytes [15]. The whereabouts of the human DAZ protein is less certain. It has been localized either to late spermatids and sperm tails [18] or to spermatogonia and primary spermatocytes [19]. In order to gain insights into their function, we studied the ability of DAZ and DAZL1 to bind RNA molecules in vitro and determined the subcellular localization of DAZL1 in mouse testis. MATERIALS AND METHODS Generation of Anti-DAZL1 Antibody A 229-base pair Sau3AI fragment encoding the last 76 amino acid residues of DAZL1 was isolated from a human DAZL1 cdna clone f2 [11] and cloned into the BamHI site of an expression vector pet32b (Novagen, Madison, WI). The construct was introduced into Escherichia coli to direct the synthesis of a fusion protein between thioredoxin and DAZL1, which was purified on His-Bind metal chelation resins according to the manufacturer s manual, and injected into rabbits to generate antibody. In Vitro RNA Binding Assay S 35 -Labeled proteins were synthesized in vitro using the TNT Coupled Reticulocyte Lysate System (Promega, Mad-

2 1656 TSUI ET AL. ison, WI). In vitro binding of the labeled proteins to RNA homopolymers was carried out according to Siomi et al. [20]. Briefly, equal amounts of S 35 -labeled proteins were added to 250 l of a 10% suspension of RNA homopolymers immobilized on agarose beads in a binding buffer containing 10 mm Tris ph 7.4, 2.5 mm MgCl 2, 0.5% Triton X-100, 2 mg/ml heparin, and 0.1 M NaCl unless otherwise stated. After rocking at 4 C for 10 min, the beads were washed 5 times with 1 ml of ice-cold binding buffer. Bound proteins were eluted by heating in 50 l of double-strength Laemmli sample buffer at 100 C for 5 min. After cooling on ice immediately afterward, the beads were spun down, and analysis of the supernatants on 10% SDS-PAGE gels was followed by autoradiography. Poly U, poly C, and poly G were purchased from Sigma (St. Louis, MO), and poly A was purchased from Pharmacia (Piscataway, NJ). Generation of Deletion Constructs of DAZ and DAZL1 Deletion constructs of DAZL1 and DAZ were generated from cdna clones DAZL1-f2 [11] and DAZ-e11 [13], respectively. Deletions were created either by using existing restriction sites within the coding regions or by polymerase chain reaction amplification of two regions followed by ligation. Correct joining at the junctions was verified by DNA sequencing. DAZL1-f2 encodes a protein of 295 residues, with the RRM domain spanning residues 32 to 115 and the DAZ repeat spanning 167 to 190. Deletion constructs DAZL-R1, DAZL-Z1, DAZL-C1, and DAZL-RZ1 encode proteins lacking residues , , , and plus , respectively. DAZ-e11 encodes a protein of 414 amino acid residues with 9 DAZ repeats. The RRM domain spans residues 32 to 115, and the DAZ repeat region spans 167 to 382. Deletion constructs DAZ-R1, DAZ-R2, DAZ-Z1, DAZ-Z2, and DAZ- RZ1 encode proteins lacking residues , , , , and plus , respectively. Trx-RRM and Trx-DAZ were constructed by cloning restriction fragments of DAZ cdna that contained the RRM domain and the DAZ repeats, respectively, in frame into the pet-32 vectors (Novagen). They encode fusion proteins between thioredoxin and DAZ fragments. Detection of DAZL1 in Mouse Tissue Extracts Tissues were removed from 2- to 3-mo-old mice immediately after they were killed. About 0.2 g of tissues were homogenized in 1 ml of an extraction buffer (EB) containing 20 mm Tris ph 7.5, 100 mm KCl, 5 mm MgCl 2, 0.3% Igepal CA-630 (Sigma), 40 U/ml ribonuclease inhibitor, 1 mm PMSF, and 1 g/ml each of aprotinin and leupeptin, in a Wheaton Potter-Elvehjem tissue grinder [21]. Homogenates were centrifuged at 1000 g for 10 min to remove cell debris and nuclei. The supernatants were centrifuged again at g for 10 min to generate a pellet containing largely mitochondria, and a postmitochondrial supernatant (PMS). The pellet was resuspended in EB to a final volume equal to that of PMS. Analysis of aliquots on 10% SDS-PAGE gels was followed by Western blotting with anti-dazl1 antibody [22]. Inhibition of Western Detection of DAZL1 Unlabeled DAZL1 was synthesized in vitro using the TNT Coupled Reticulocyte Lysate System (Promega). T3 and T7 RNA polymerase transcribed the Dazl1 cdna clone in the sense and antisense directions, respectively. Varying amounts of the in vitro synthesis mixtures were incubated with 2 l of anti-dazl1 antiserum in a total volume of 200 l of TBS-T buffer (50 mm Tris ph 7.4, 150 mm NaCl, 0.1% Tween-20) on ice for 1 h. The solutions were diluted to 4 ml with 10% milk in TBS-T before incubation with Western blot membranes containing equal amounts of liver and testis extracts. Sucrose Gradient Analysis of Mouse Testicular Extracts Mouse testicles were homogenized in either standard EB or modified EB in which 5 mm MgCl 2 was replaced with 30 mm EDTA ph 7.5, or KCl was added to a final concentration of 0.5 M. After centrifugation at g for 10 min, 0.5 ml of PMS was layered on top of 15 45% or 5 30% sucrose gradients in 20 mm Tris ph 7.5, 0.1 M KCl, and the same concentration of MgCl 2, EDTA, or KCl as in the extracts. After centrifugation in a Beckman SW41 rotor (Beckman Instruments, Palo Alto, CA) at rpm for 2hat4 C, 0.5-ml fractions were collected from the bottom, and the OD 254 was determined. Aliquots of every other fraction were analyzed on 10% SDS-PAGE gels and then by Western blotting with antibodies against DAZL1 and the fragile X mental retardation protein FMRP, and a human autoantibody against ribosomal P antigens (ImmunoVision, Springdale, AR). For ribonuclease (RNase) treatment, RNase A was added to the PMS in standard EB to a final concentration of 300 g/ml. After incubation at room temperature for 5 min, the solution was analyzed by sucrose gradient centrifugation as above. Capture of DAZL1-Associated Messenger Ribonucleoprotein (mrnp) Particles on Oligo(dT) Beads Mouse testicular PMS was subjected to mrnp capture using the Oligotex mrna purification kit (Qiagen, Santa Clarita, CA) according to the manufacturer s protocol. Briefly, 0.25 ml of PMS was mixed with 0.25 ml of doublestrength buffer OBB (20 mm Tris ph 7.5, 1 M NaCl, 2 mm EDTA, and 0.2% SDS) and 20 l of Oligotex beads. After rocking at room temperature for 10 min, the beads were spun down and washed twice with 0.5 ml of wash buffer OW2 (10 mm Tris ph 7.5, 150 mm NaCl, and 1 mm EDTA). Bound proteins were eluted by heating in 30 l of double-strength Laemmli sample buffer at 100 C for 5 min. Aliquots of the load, the wash, and the bound fractions were analyzed by Western blotting with antibodies against DAZL1 and FMRP. Pretreatment of PMS with EDTA (30 mm, 37 C, 15 min), RNase (1.2 mg/ml, 37 C, 10 min), or NaOH (0.2 N, 37 C, 10 min, then neutralized to ph 8.0 with 1 N HCl) before capture was carried out according to Feng et al. [21]. RESULTS Binding of DAZ and DAZL1 to RNA Molecules The abilities of DAZ and DAZL1 to bind RNA molecules were studied using an assay in which in vitro-synthesized 35 S-labeled proteins were incubated with RNA homopolymers immobilized on agarose beads, and the bound proteins were analyzed by SDS-PAGE. As shown in Figure 1a, human DAZ and DAZL1, as well as mouse DAZL1, bound preferentially to poly U and poly G, similar to the Xenopus DAZL1 homologue, Xdazl, and FMRP [16, 21]. On the other hand, the protein product of an anonymous human cdna clone KIAA0058 [23] and the firefly luciferase, which contain no RNA binding motifs, failed to bind

3 ASSOCIATION OF DAZL1 WITH POLYRIBOSOMES 1657 FIG. 1. In vitro binding of DAZ and DAZL1 to RNA homopolymers. a) Equal amounts of S 35 -labeled proteins were incubated with equal bed volumes of RNA homopolymers immobilized on agarose beads in the presence of 0.1 M NaCl. Bound proteins were analyzed by SDS-PAGE followed by autoradiography. The lanes are A, poly A; U, poly U; C, poly C; and G, poly G. b) The binding assay was carried out at the indicated salt concentration (Molar). The T lane contains an amount equivalent to 50% of the material used in the binding assay. DAZL, DAZL1. to any of the RNA homopolymers (data not shown). In the presence of increasing amounts of salt, the binding of DAZ and DAZL1 to poly U persisted, whereas the binding to poly G diminished, suggesting that DAZ and DAZL1 bind more strongly to poly U (Fig. 1b). These results verified the RNA-binding properties of DAZ and DAZL1. However, the biological significance of the preferential binding observed in the in vitro assays has yet to be demonstrated. To delineate the regions required for RNA binding, a series of constructs with deletions in the RRM domain, the DAZ repeat region, or the C-terminal portion were generated (Fig. 2). These constructs were transcribed and translated in vitro, and the abilities of the resultant truncated proteins to bind poly U and poly G in the presence of 0.25 M NaCl were studied (Fig. 2). For DAZL1, deletion of RRM abolished its RNA-binding ability, whereas deletion of the single DAZ repeat affected the binding to poly G but not poly U, and that of the C-terminal portion showed little effect. It appears, therefore, that DAZL1 binds to RNA mainly through the RRM domain. The results for DAZ are less clear-cut. Deletion of RRM in either DAZ-R1 or DAZ- R2 diminished, but did not abolish, the RNA-binding ability. Similarly, deletion of seven of the nine DAZ repeats (in DAZ-Z1) or the C-terminal portion including the entire DAZ repeat region (in Z2) diminished but did not abolish the binding. The results suggest that the binding of DAZ to RNA does not depend solely on the RRM domain. To test whether the RRM domain and the C-terminal portion of DAZ can bind RNA homopolymers independently, fusion proteins were generated in which either the RRM domain or the C-terminal portion of DAZ was fused to the thioredoxin protein [24]. Both fusion proteins bound to the RNA polymers, though at different strengths. Generation of Anti-DAZL1 Antibodies A polyclonal antiserum was generated against the last 76 amino acid residues of the human DAZL1 protein, which is 86% identical to the corresponding region of mouse DAZL1 [11]. On Western blots, the anti-dazl1 antiserum detected a major band in mouse testicular extracts that comigrated with in vitro-synthesized DAZL1 and that was absent in the liver and the brain (Fig. 3a). The apparent molecular size of the putative DAZL1 was larger than the 33 kda predicted from the cdna sequence. A discrepancy between predicted molecular size and mobility in the SDS- PAGE gel was also reported for the Drosophila Boule protein [25]. Preincubation of the antiserum with increasing FIG. 2. Mapping regions within DAZL1 and DAZ that are required for RNA binding. The structures of DAZL1 and DAZ and the regions retained in the truncated proteins are shown, with the RNA recognition motif (RRM) and the DAZ repeat region indicated. See Materials and Methods for the amino acid residues deleted in the proteins. Binding of proteins to poly U (U) and poly G (G) in the presence of 0.25 M NaCl are shown to the right. The T lane contains an amount equivalent to 50% of the material used in the binding reaction. amounts of in vitro-synthesized DAZL1 reduced the signal of the putative DAZL1 but not of other minor bands, further supporting the authenticity of the protein (Fig. 3b). Detection of DAZL1 in Mouse Testicular Fractions The anti-dazl1 antiserum was used to trace DAZL1 during fractionation of mouse testicular extracts. When the FIG. 3. Identification of DAZL1 in mouse tissue extracts. a) Detection of DAZL1 by Western blotting with anti-dazl1 antibody. Extracts of mouse liver, brain, and testis, and in vitro-synthesized DAZL1 were fractionated by SDS-PAGE and Western blotted with anti-dazl1 antiserum. The arrowhead points to the putative DAZL1 protein. b) Inhibition of immunodetection of DAZL1 by in vitro-synthesized DAZL1. Anti-DAZL1 antiserum was preincubated with the indicated volume of TNT protein synthesis reaction before incubation with membranes containing equal amounts of mouse liver (L) and testis (T) extracts. T3 and T7 RNA polymerases transcribed the Dazl1 cdna clone in the sense and antisense directions, respectively.

4 1658 TSUI ET AL. FIG. 5. Association of DAZL1 with poly(a) mrna. a) Poly(A) containing RNP particles in mouse PMS were captured on oligo(dt) beads, and the presence of DAZL1 and FMRP in the captured RNP particles was analyzed by 10% SDS-PAGE followed by Western blotting. L, An amount equivalent to 8% of the initial load to the beads; W, the wash; and B, the bound fraction. b) Mouse testis PMS was pretreated with EDTA, RNase, or NaOH before oligo(dt) capture, and the presence of DAZL1 in the captured fraction was analyzed. polysomes into monosomes and caused a significant reduction in the sedimentation of DAZL1 (Fig. 4d). The sedimentation profiles of DAZL1 under the various conditions indicate that most DAZL1 in mouse testes is associated with polysomes. FIG. 4. Sucrose gradient analyses of DAZL1 in mouse testes. Mouse testis PMS in the presence of 5 mm MgCl 2 (a), 30 mm EDTA (b, e), or 0.5 M KCl (c), or after RNase A treatment (d), were fractionated by centrifugation through 15 45% sucrose gradients (except 5 30% in e). Aliquots of the fractions were analyzed on 10% SDS-PAGE gels and then Western blotted with antibodies against DAZL1, FMRP, and a human autoantibody against ribosomal P antigens (P-Ag). extracts were separated into nuclear, mitochondrial, and postmitochondrial fractions by differential centrifugation, over 75% of DAZL1 was detected in the PMS fraction (data not shown). Further fractionation of PMS on 15 45% sucrose gradients showed that most DAZL1 comigrated with the polysomes (Fig. 4a), similar to FMRP, which was shown previously to be associated with polysomes [21]. Other proteins that cross-reacted with the antiserum stayed at the top of the gradients (data not shown). Inclusion of 30 mm EDTA in the testicular extracts dissociated polysomes into 60S and 40S subunits, and caused a shift of DAZL1 to the top fractions of the gradients (Fig. 4b). On a 5 30% sucrose gradient, the DAZL1 protein in the EDTA-treated extracts showed a more heterogeneous distribution than that of FMRP (Fig. 4e). Addition of 0.5 M KCl, which is known to remove translation factors and aminoacyl-trna synthetases, but not most intrinsic ribosomal proteins [26], dissociated most of DAZL1 from the polysomes (Fig. 4c). RNase A treatment of PMS separated the Capture of DAZL1-Associated mrnp To investigate whether DAZL1 is associated with polysomes through its binding to mrna molecules, poly(a) containing mrnp particles in PMS were captured on oligo(dt) beads, and the presence of DAZL1 in the captured mrnp fractions was analyzed by SDS-PAGE followed by Western blotting. DAZL1, but not other cytoplasmic proteins that cross-reacted with anti-dazl1, was captured on oligo(dt) beads (Fig. 5a). FMRP was also captured on oligo(dt) beads, as previously reported [21, 27]. Capture of DAZL1 on oligo(dt) was not affected by 30 mm EDTA but was reduced by preincubation with RNase and abolished by NaOH treatment, further supporting the binding of DAZL1 to poly(a) RNA (Fig. 5b). DISCUSSION We present evidence for the association of DAZL1 with actively translating polyribosomes in testes. Our results suggest a role for DAZL1 in translational regulation of a subset of mrnas in germ cells. Germ cell-specific translational regulation is well documented for transition proteins and protamines [28]. Their mrnas are synthesized in round spermatids and stored as cytoplasmic mrnp particles for up to a week before being translated in elongated spermatids. Premature translation of Prm-1 mrna in transgenic mice leads to dominant male sterility [29]. Translational repression of protamine mrnas requires sequences within the 3 untranslated region, and several RNA-binding proteins that bind to the region have been isolated [30 32]. Among them PRBP (Prm-1 RNA-binding protein, encoded by the Tarbp2 gene) and TB-RBP (testis brain RNA-binding protein) were shown to repress protamine mrna translation in vitro [30, 31], and TB-RBP was found to be associated with translationally dormant free mrnp particles [33]. However, a recent study on the expression of protamines in the testes of Tarbp2 knock-out mice indicated

5 ASSOCIATION OF DAZL1 WITH POLYRIBOSOMES 1659 unexpectedly that PRBP is required for proper translational activation, but not repression, of protamine mrna [34]. How PRBP activates the translation of protamine mrna and whether it is associated with polyribosomes remain to be determined. While this report was in preparation, Maines and Wasserman [35] reported that the Drosophila homologue of DAZL1 Boule controlled the translation of Twine, a meiotic Cdc25-type phosphatase. They found that in boule mutants the translation, but not the transcription, of twine was significantly reduced, and that heterologous expression of Twine partially rescued the spermatogenic defect of boule mutants. Their results not only support our proposition that DAZL1 plays a role in translational regulation but also suggest Cdc25 transcripts as possible targets of DAZL1. Cdc25 phosphatases activate cyclin-dependent kinases and play an important role in cell cycle regulation. Three mammalian Cdc25 genes have been identified. Both Cdc25A and Cdc25C are expressed predominantly in the testis and could be the targets of DAZL1 [36, 37]. However, DAZL1 is likely to have additional targets besides Cdc25, as suggested by the incomplete rescue of the boule phenotype by Twine. DAZL1 may facilitate the translation of its target genes by sequestering their transcripts to polysomes, by enhancing their translational rate, or by protecting the transcripts from degradation. Whether DAZ is also involved in mrna translation remains to be determined. It has been proposed that DAZ plays little, or a limited, role in spermatogenesis, on the basis of evolutionary considerations [38]. The ability of a human DAZ transgene to rescue the spermatogenic defect of Dazl1 knock-out mice [17] argues against such a proposal and suggests that DAZ serves a function similar to that of DAZL1. Structurally, DAZ and DAZL1 share high sequence identity except for the C-terminal segments. The RNA-binding characteristics of DAZ and DAZL1 are very similar. Whether DAZ is also associated with polysomes remains to be determined. Because DAZ is present in great apes and Old World monkeys only, its subcellular localization awaits the availability of testicular tissues from these higher mammals. If DAZ and DAZL1 are shown to serve a similar function, future studies could address the therapeutic potential of DAZL1 over-expression in infertile males with DAZ deletion. ACKNOWLEDGMENTS We thank H. Cooke for the Dazl1 cdna clone and R. Swerdloff s group for helpful discussion. REFERENCES 1. Tiepolo L, Zuffardi O. Localization of factors controlling spermatogenesis in the nonfluorescent portion of the human Y chromosome long arm. Hum Genet 1976; 34: Roberts KP. Y-Chromosome deletions and male infertility: state of the art and clinical implications. J Androl 1998; 19: Cooke HJ. Y chromosome and male infertility. Rev Reprod 1999; 4: McElreavey K, Krausz C. Male infertility and the Y chromosome. Am J Hum Genet 1999; 64: Reijo R, Lee T-Y, Salo P, Allagappan R, Brown LG, Rosenberg M, Rozen S, Jaffe T, Straus D, Hovatta O, de la Chapelle A, Silber S, Page DC. Diverse spermatogenic defects in humans caused by Y chromosome deletions encompassing a novel RNA-binding protein gene. Nat Genet 1995; 10: Gläser B, Yen PH, Schempp W. Fiber-FISH unravels apparently seven DAZ genes or pseudogenes clustered within a Y chromosome region frequently deleted in azoospermic males. Chromosome Res 1998; 6: Shan Z, Hirschmann P, Seebacher T, Edelmann A, Jauch A, Morell J, Urbitsch P, Vogt PH. A SPGY copy homologous to the mouse gene Dazla and the Drosophila gene boule is autosomal and expressed only in the human male gonad. Hum Mol Genet 1996; 5: Cooke HJ, Lee M, Kerr S, Ruggiu M. A murine homologue of the human DAZ gene is autosomal and expressed only in male and female gonads. Hum Mol Genet 1996; 5: Reijo R, Seligman J, Dinulos MB, Jaffe T, Brown LG, Disteche CM, Page DC. Mouse autosomal homolog of DAZ, a candidate male sterility gene in humans, is expressed in male germ cells before and after puberty. Genomics 1996; 35: Saxena R, Brown LG, Hawkins T, Alagappan RK, Skaletsky H, Reeve MP, Reijo R, Rozen S, Dinulos MB, Disteche CM, Page DC. The DAZ gene cluster on the human Y chromosome arose from an autosomal gene that was transposed, repeatedly amplified and pruned. Nat Genet 1996; 14: Yen PH, Chai NN, Salido EC. The human autosomal gene DAZLA: testis specificity and a candidate for male infertility. Hum Mol Genet 1996; 5: Seboun E, Barbaux S, Bourgeron T, Nishi S, Algonik A, Egashira M, Kikkawa N, Bishop C, Fellous M, McElreavey K, Kasahara M. Gene sequence, localization and evolutionary conservation of DAZLA, a candidate male sterility gene. Genomics 1997; 41: Yen PH, Chai NN, Salido EC. The human DAZ genes, a putative male infertility factor on the Y chromosome, are highly polymorphic in the DAZ repeat region. Mamm Genome 1997; 8: Eberhart CG, Maines JZ, Wasserman SA. Meiotic cell cycle requirement for a fly homologue of human Deleted in Azoospermia. Nature 1996; 381: Ruggiu M, Speed R, Taggart M, McKay SJ, Kilanowski F, Saunders P, Dorin J, Cooke HJ. The mouse DAZLA gene encodes a cytoplasmic protein essential for gametogenesis. Nature 1997; 389: Houston DW, Zhang J, Maines JZ, Wasserman SA, King ML. A Xenopus DAZ-like gene encodes an RNA component of germ plasm and is a functional homologue of Drosophila boule. Development 1998; 125: Slee R, Grimes B, Speed RM, Taggart M, Maguire SM, Ross A, McGill MI, Saunders PTK, Cooke HJ. A human DAZ transgene confers partial rescue of the mouse Dazl null phenotype. Proc Natl Acad Sci USA 1999; 96: Habermann B, Mi HF, Edelmann A, Bohring C, Bäckert IT, Kiesewetter F, Aumüller A, Vogt PH. DAZ (Deleted in Azoospermia) genes encode proteins located in human late spermatids and in sperm tails. Hum Reprod 1998; 13: Dorfman DM, Genest DR, Pera RAR. Human DAZL1 encodes a candidate fertility factor in women that localizes to the prenatal and postnatal germ cells. Hum Reprod 1999; 14: Siomi H, Siomi MC, Nussbaum RL, Dreyfuss G. The protein product of the fragile X gene, FMR1, has characteristics of an RNA-binding protein. Cell 1993; 74: Feng Y, Absher D, Eberhart DE, Brown V, Malter HE, Warren ST. FMRP associates with polyribosomes as an mrnp, and the I304N mutation of severe fragile X syndrome abolishes this association. Mol Cell 1997; 1: Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; Nomura N, Nagase T, Miyajima N, Sazuka T, Tanaka A, Sato S, Seki N, Kawarabayashi Y, Ishikawa K, Tabata S. Prediction of the coding sequences of unidentified human genes. II. The coding sequences of 40 new genes (KIAA0041-KIAA0080) deduced by analysis of cdna clones from human cell line KG-1. DNA Res 1994; 1: LaVallie ER, DiBlasio EA, Kovacic S, Grant KL, Schendel PF, Mc- Coy JM. A thioredoxin gene fusion expression system that circumvents inclusion body formation in the E. coli cytoplasm. Biotechnology 1993; 11: Cheng MH, Maines JZ, Wasserman SA. Biphasic subcellular localization of the DAZL-related protein Boule in Drosophila spermatogenesis. Dev Biol 1998; 204: Spirin AS. Ribosome preparation and cell-free protein synthesis. In: Hill WE, Dahlberg A, Garrett RA, Moore PB, Schlessinger D, Warner JR (eds.), The Ribosomes: Structure, Function, & Evolution. Washington, DC: American Society for Microbiology; 1990: Corbin F, Bouillon M, Fortin A, Morin S, Rousseau F, Khandjian W. The fragile X mental retardation protein is associated with poly(a) mrna in actively translating polyribosomes. Hum Mol Genet 1997; 6:

6 1660 TSUI ET AL. 28. Braun RE. Post-transcriptional control of gene expression during spermatogenesis. Semin Cell Dev Biol 1998; 9: Lee K, Haugen H, Clegg CH, Braun RE. Premature translation of protamine 1 mrna causes precocious nuclear condensation and arrests spermatid differentiation in mice. Proc Natl Acad Sci USA 1995; 92: Kwon YK, Hecht NB. Binding of a phosphoprotein to the 3 untranslated region of the mouse protamine 2 mrna temporally repressed its translation. Mol Cell Biol 1993; 13: Lee K, Fajardo MA, Braun RE. A testis cytoplasmic RNA-binding protein that has the properties of a translational repressor. Mol Cell Biol 1996; 16: Schumacher JM, Lee K, Edelhoff S, Braun RE. Spnr, a murine RNAbinding protein that is localized to cytoplasmic microtubules. J Cell Biol 1995; 129: Han JR, Gu W, Hecht NB. Testis-brain RNA-binding protein, a testicular translational regulatory RNA-binding protein, is present in the brain and binds to the 3 untranslated regions of transported brain mrnas. Biol Reprod 1995; 53: Zhong J, Peters AHFM, Lee K, Braun RE. A double-stranded RNAbinding protein required for activation of repressed messages in mammalian germ cells. Nat Genet 1999; 22: Maines JZ, Wasserman SA. Post-transcriptional regulation of the meiotic Cdc25 protein Twine by the Dazl orthologue Boule. Nat Cell Biol 1999; 1: Wickramasinghe D, Becker S, Ernst MK, Resnick JL, Centanni JM, Tessarollo L, Grabel LB, Donovan PJ. Two CDC25 homologues are differentially expressed during mouse development. Development 1995; 121: Wu S, Wolgemuth DJ. The distinct and developmentally regulated patterns of expression of members of the mouse Cdc25 gene family suggest differential functions during gametogenesis. Dev Biol 1995; 170: Agulnik AI, Zharkikh A, Boettger-Tong H, Bourgeron T, McElreavey K, Bishop CE. Evolution of the DAZ gene family suggest that Y- linked DAZ plays little, or a limited, role in spermatogenesis but underlines a recent African origin for human populations. Hum Mol Genet 1998; 7: