Human Molecular Genetics, 2005, Vol

Human Molecular Genetics, 2005, Vol. 14, No. 24 3899–3909 doi:10.1093/hmg/ddi414 Advance Access published on November 8, 2005 Dazl binds in vivo to speciﬁc transcripts and can regulate the pre-meiotic translation of Mvh in germ cells

Nicola Reynolds1, Brian Collier1,2, Klio Maratou1,{, Victoria Bingham1, Robert M. Speed1, Mary Taggart1, Colin A. Semple1, Nicola K. Gray1,2 and Howard J. Cooke1,* Downloaded from https://academic.oup.com/hmg/article/14/24/3899/2355858 by guest on 27 September 2021 1MRC Human Genetics Unit, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, UK and 2School of Molecular and Clinical Medicine, 3rd Floor Outpatients Department, Western General Hospital, University of Edinburgh, Edinburgh EH4 2XU, UK

Received September 23, 2005; Revised and Accepted November 2, 2005

Gametogenesis is a complex process subject to strict controls at both levels of transcription and translation. Members of a family of conserved RNA-binding proteins encoded by the DAZ genes are required for the translational regulation of gene expression essential for this process. Although loss of DAZ family genes is associated with infertility in several organisms including humans, the identity of the transcripts regulated in vivo is unknown. Using a combination of immunoprecipitation and microarray analysis, we have identiﬁed a number of mRNAs that are bound by the murine Dazl protein both in vivo and in vitro. Sequence analysis shows that these transcripts contain binding sites for Dazl, which have been conserved during evolution between human, rat and mouse. We have focussed on mouse vasa homologue (Mvh), a gene that is essential for male gametogenesis, and show that Dazl stimulates translation via the Mvh 30-UTR. Finally, we show that germ cells of Dazl null mice contain reduced levels of Mvh protein, indicating that Dazl-mediated regulation of Mvh translation is crucial for mammalian spermatogenesis.

INTRODUCTION that cells have been seen to advance along the meiotic pathway is to leptotene of meiotic prophase I (10). This indi- The DAZ family of genes encodes RNA-binding proteins that cates multiple roles for Dazl: in mitotic proliferation before are essential for gametogenesis in metazoans. The family com- meiosis, during spermatogonial development and in the lepto- prises homologues of BOULE which are found in all metazo- tene to zygotene transition. It is possible that Dazl is also ans; DAZL, found in all vertebrates; and DAZ, which is present required for other speciﬁc events in meiosis. on the Y chromosome of Old World monkeys and humans (1). Female Dazl knockout mice are also infertile and lack germ In humans, loss of the Y chromosomal DAZ genes is associ- cells and follicles in the adult ovary (6), whereas mutations in ated with oligozoospermia or azoospermia. The DAZ genes the Caenorhabditis elegans daz-1 result in female infertility are strong candidates for the AZFc azoospermia factor, with multiple abnormalities visible at the onset of meiotic pro- one of the most common genetic causes of male infertility phase (8). In contrast, knockdown of Xenopus Xdazl disrupts (2–4). Loss of function of the other autosomal DAZ family germ cell development at a much earlier stage, resulting in a members results in failure to produce mature gametes in failure of germ cell migration and proliferation (7). The Dro- organisms as diverse as mouse, frog, fruit ﬂy and nematode sophila boule mutant has a spermatogenic arrest at meiotic (5–8). entry and there is strong evidence that the protein is required In male mice, loss of Dazl results in multiple defects includ- for the translation of the meiotic CDC25, twine (5,11,12). ing a modest reduction in germ cell numbers before birth and a Recent evidence also suggests a possible role for its regulation failure to progress from Aaligned to A1 spermatogonia (6,9). of twine in the Drosophila CNS (13). In humans, the pheno- Although some cells progress beyond this point, the furthest type resulting from the AZFc deletion is highly variable

*To whom correspondence should be addressed. Tel: þ44 1314678427; Fax: þ44 1314678456; Email: [email protected] {Present address: Neural Plasticity Unit, Institute of Child Health, 30 Guilford Street, London WC1N 1EH, UK.

# The Author 2005. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] 3900 Human Molecular Genetics, 2005, Vol. 14, No. 24 ranging from impaired fertility to Sertoli cell only syndrome (reviewed in 14,15). This variability reflects variation in the structure of the Y chromosome in different haplogroups, differences in the age at which patients are examined, as well as genetic background and environmental factors. Because of these complexities, model organisms are essential to study the role of the DAZ family proteins and the mouse remains the closest model to humans. Gametogenesis is subject to complex regulation at the levels of transcription and translation. It has been proposed that the DAZ-related proteins bind to RNA in the cytoplasm of germ cells and control gametogenesis at the level of translation. Downloaded from https://academic.oup.com/hmg/article/14/24/3899/2355858 by guest on 27 September 2021 To date, all work points towards a role for Dazl in stimulating rather than repressing the translation of specific mRNAs. However, it remains to be seen if this is true in all cases, as few in vivo targets of the DAZ-related proteins have been studied in detail. A role for the DAZ-related proteins in translational regulation was first suggested by the identification of a genetic interaction between boule and twine in Drosophila. A correlation between the levels of Boule and Twine proteins, along with the fact that heterologous expression of twine can rescue the meiotic entry defect of boule mutant flies, showed that twine is a target for post-transcriptional regulation by Boule (12). In addition, Dazl associates with polysomes in mouse testis (16). In vitro studies using zebra fish Dazl Figure 1. Isolation and identification of RNAs associated with Dazl protein showed a modest but direct stimulatory effect on the trans- in vivo.(A) Outline of experimental approach. Immunoprecipitations were carried out with three antibody combinations, RNA isolated and identified lation of target sequences (17). More recently, experiments by hybridization to adult mouse, testis-specific microarrays. Comparison of have shown that DAZ family proteins, including murine transcripts present under these different conditions enables specific inter- Dazl, act directly to stimulate translation of RNAs to which actions to be determined. (B) Western blot to show proteins present in they are bound through the recruitment of 80S ribosomes immunoprecipitates. Immunoprecipitations from rat testis homogenate were via an interaction with poly(A)-binding protein (PABP; 18). carried out using antibodies directed against Dazl, SF2 or negative control (anti-Dazl þ peptide against which it was raised). Although a number of studies have identified specific mRNAs that are bound by the DAZ-related proteins in vitro (17,19–22), no direct interactions have been demonstrated arrays (data not shown, see Materials and Methods). Array in vivo. Here we describe the identification of a number of analysis was also carried out to compare total rat testis RNA transcripts that are bound by murine Dazl in vivo and show before and after amplification and showed that amplification that at least one of them is subject to translational regulation did not preferentially change the levels of any particular tran- by the Dazl protein. scripts (data not shown). Each array consists of 1345 unique, testis-expressed genes represented multiple times in duplicate. Of the genes rep- RESULTS resented, 11 showed 5-fold or greater enrichment in the Dazl compared with Sf2 co-immunoprecipitate and at least 2-fold Immunoprecipitation and identification of enrichment when compared with the anti-Dazl þ peptide Dazl/RNA complexes negative control at a statistically significant level (P , 0.05). The approach taken to identify mRNAs associated with Dazl These genes are listed in Table 1. protein in vivo is illustrated in Figure 1A. Endogenous Dazl In male mice, spermatogenesis begins shortly after birth and protein was immunoprecipitated from rat testis homogenate initially occurs with a high degree of synchrony. Transcrip- and co-purifying RNA isolated. Control immunoprecipitations tional profiling of gene expression across this first wave of were carried out with either an antibody directed against spermatogenesis has proven to be a powerful method of ana- the splicing factor, SF2 (another RRM-type (RNA Recog- lysing the complex process of meiosis (24,26–29). We postu- nition Motif) RNA-binding protein) (23), or the anti-Dazl anti- lated that genuine in vivo Dazl targets would not be subject to body in the presence of the peptide to which it was raised normal translational regulation in the absence of Dazl protein. (negative control) (Fig. 1B). A decrease of translation could lead indirectly to a decrease Co-purifying RNA in each case was amplified, labelled and in stability levels of the mRNA (30), which would result in hybridized to adult mouse testis cDNA microarrays using a differences in transcript levels between Dazl null and wild- triple-dye hybridization system and adult rat testis total type animals. We, therefore, compared the transcripts ident- RNA as a reference (24,25). Control experiments were ified in the co-immunoprecipitation experiments to expression carried out to confirm that there was a high level of cross- profile data at 5 days after birth for Dazl knockout and wild- hybridization between rat transcripts and mouse cDNA type animals (24). This is before any major changes in the Human Molecular Genetics, 2005, Vol. 14, No. 24 3901

Table 1. Transcripts enriched in anti-Dazl immunoprecipitates compared with controls

Gene name Accession Function number

Enriched in Dazl IP (5x.Sf2, 2x.Dazl þ Peptide IP) Odc1 NM_013614 Ornithine decarboxylase, polyamine biosynthesis Sycp3 NM_011517 Synaptonemal complex formation and male fertility (46) Col9a3 NM_009936 Testis-speciﬁc procollagen Hspa2 NM_008301 Heat shock protein, regulation of CDC2 activity in prophase I of meiosis in male mice (47) Slc2a3 NM_011401 Solute carrier family 2 (facilitated glucose transporter), member 3 Calm2 NM_007589 Calmodulin, mitotic cell cycle progression (48) Smac/DIABLO NM_023232 Regulation of apoptosis (49)

Usp2 NM_016808 Ubiquitin-speciﬁc protease Downloaded from https://academic.oup.com/hmg/article/14/24/3899/2355858 by guest on 27 September 2021 Actg1 NM_009609 Cytoskeletal gamma-actin 4930453N24Rik BC020029 Not determined Hnrpul1 BC027844 Heterogeneous nuclear ribonucleoprotein U-like 1 Different in Dazlþ/þ and Dazl2/2 at 5 days post-partum and enriched in Dazl IP (3x.Sf2 and Dazl þ Peptide IP) Sycp3 NM_011517 Synaptonemal complex formation and male fertility (46) Tex19 NM_028602 Testis expressed gene 19 (50) Mvh NM_010029 Ddx4, DEAD family protein required for progression through meiotic prophase I in male mice (32) Stk31 NM_029916 Serine threonine kinase (50) Dazl NM_010021 Deleted in azoospermia like, essential for gametogenesis in mouse (6) Tex14 NM_031386 Testis-speciﬁc protein kinase (51) Tubulin alpha 3 NM_009446 Tubulin alpha 3 B130016L12Rik NM_144835 Not determined Fthl17 NM_031261 Ferritin, heavy polypeptide-like 17, iron storage

germ cell complement of the Dazl null testis has been Matches to the motif U(2-10)[G/C]U(2-10), based on the reported. In this way, a further eight candidate genes were general consensus sequence of Venables et al. (19), were identified (Table 1). By taking this dual approach to the also detected. This general pattern generated 251 matches analysis of the microarray data, we expected to identify in across all 15 genes (P , 0.002) and a total of 14 genes were vivo targets with a high degree of confidence rather than matched by this pattern when the longest 30-UTR data set create an exhaustive list of target mRNAs. A combination of was searched (Usp2/NM_016808 lacked a match in its this conservative analysis of the microarray data and the rela- 30-UTR), generating 158 hits (P , 0.002). The Jiao et al. tively large number of transcripts with which the Dazl protein (20) data set was also found to contain multiple matches to associates in vivo resulted in very few genes being identified in this consensus (91 hits in total) with all eight genes showing both analyses. more than one match (P , 0.002). In contrast, when the longest 50-UTR data set was searched only 10 hits (P ¼ 0.274) to the consensus were generated, the randomiz- Sequence analysis ations suggest that these sequences contain somewhat fewer As for many DNA- and RNA-binding proteins, attempts to matches per sequence than would be expected by chance identify a consensus binding sequence for murine Dazl have (Table 2). The number of matches to transcripts preferentially produced differing results. This probably reflects both the enriched in Sf2 immunoprecipitates was also no greater than differences in experimental approaches taken and the multi- expected by chance. tude and complexity of in vivo interactions. Jiao et al. (20) In summary, the motif U(2-10)[G/C]U(2-10) is statistically defined a 26 bp motif present in eight mRNAs that were over-represented within 30-UTR regions (but not 50-UTR bound by recombinant GST-tagged Dazl in mouse testis regions) of the 15 transcripts specifically co-immunoprecipitated extract. Venables et al. (19) took a combined three-hybrid with Dazl protein. The same is true of transcripts from and SELEX approach and suggested the general consensus the Jiao et al. data set. This constitutes strong evidence for binding sequence (G/CUn)n. the presence of this motif in transcripts bound by the Dazl Sequences of 15 of the transcripts that were specifically protein. co-immunoprecipitated with Dazl were examined for the To confirm the biological relevance of this motif, compara- presence of each of these motifs, as were data sets consisting tive genomics were used to detect conservation in and around of the longest 30-UTR and 50-UTR sequences for each the putative Dazl-binding sites. Genes orthologous to the 15 (Supplementary Material, Table S1). Only six of these 15 genes under study were identified in the human, rat and genes (Calm2, Smac/DIABLO, Usp2, Slc2a3, Dazl and chicken genomes (see Materials and Methods) and the best Tex14) contained 30-UTR matches to the motif defined by conserved putative Dazl-binding sites were identified for Jiao et al. (20), even when matches were sought to a hidden each gene (Supplementary Material, Tables S2 and S3). All Markov model (HMM) rather than a strict consensus. In of the genes containing the motif in their 30-UTR except one addition, the number of matches found was not significantly (Slc2a3) exhibited some level of evolutionary conservation greater than that expected by chance alone (P ¼ 0.062). (Supplementary Material, Table S3). Conservation at this 3902 Human Molecular Genetics, 2005, Vol. 14, No. 24

Table 2. Number of matches per transcript (n) to the general consensus pattern U(2,10)[GC]U(2,10) in various sets of sequences

Sequence set Number of Proportion of Signiﬁcance matches, n random sets n of difference

15 transcripts enriched in 11.29 0 NA Dazl IP Eight transcripts from Jiao 11.37 0.002 0.978 et al. (20) 50-UTRs of 15 transcripts 2 0.548 0.028 enriched in Dazl IP Seven transcripts highly 4.40 0.258 0.029

enriched in Sf2 IP Downloaded from https://academic.oup.com/hmg/article/14/24/3899/2355858 by guest on 27 September 2021 Figure 2. Semi-quantitative RT–PCR to show relative amounts of specific 23 transcripts enriched in 3.44 0.276 0.001 transcripts in anti-Dazl and negative control samples from UV cross-linked Sf2 IP mouse testis for two representative immunoprecipitation experiments. Results are shown for six messages that were specifically enriched in Matches are within the 30-UTRs of transcripts unless stated otherwise. the Dazl immunoprecipitates (Mvh, Sycp3, Odc1, Tex14, Fthl17, Calm2) The rightmost column shows the proportion of randomized sets with and three that did not show specific enrichment (Prm2, Fdft1, greater than or equal to the original frequency of matches per gene. 2810408A11Rik). In each case, lanes 1 and 3 show anti-Dazl immunoprecipi- Also indicated is the significance (unpaired t-test) of the difference tation, lanes 2 and 4 anti-Dazl þ peptide negative control, lane 5 negative between n for each set compared with that calculated for the first set in control (water), and lane 6 adult mouse testis cDNA as a positive control the table. for PCR reaction. level, particularly in untranslated sequences, is likely to reflect In vitro binding of Dazl protein to RNA a significant functional relevance in vivo. Of the genes identified as putative targets of Dazl, we were particularly interested in mouse vasa homologue (Mvh). This In vivo association between Dazl and mRNAs gene has a clearly defined meiotic function and thus is a poten- Cell lysis enables interactions between protein and RNA mol- tially important target for translational regulation by the Dazl ecules that are not present in the same cell types in vivo and protein in vivo. In particular, the Mvh knockout phenotype in can therefore lead to the appearance of false-positives in male mice is a block at leptotene to zygotene of meiotic pro- immunoprecipitation experiments (31). To confirm in vivo phase I (32), which corresponds to one of the distinct blocks in interactions for our putative Dazl targets, UV cross-linking meiosis that has been described in Dazl null animals (10). A of intact tubules from adult mouse testes was carried out lack or reduction in translation of Mvh could therefore con- prior to cell lysis, immunoprecipitation and quantitative tribute to the phenotype seen in Dazl null animals. The Mvh RT–PCR. For six putative Dazl target RNAs that were transcript was enriched 4-fold in the anti-Dazl compared analysed in this way (Odc1, Calm2, Tex14, Fthl17, Mvh, with the anti-Dazl þ peptide immunoprecipitate and 6.5-fold Sycp3), enrichment was consistently seen in the anti-Dazl compared with the anti-SF2 immunoprecipitate in our micro- compared with the anti-Dazl þ peptide control in three or array analysis. Association between the Mvh transcript and more independent experiments. Average levels of enrichment the Dazl protein in vivo was confirmed by UV cross-linking were between 5- and 100-fold. In contrast, three mRNAs and quantitative RT–PCR (Fig. 2). In order to further define which were identified as not specifically bound by Dazl in the interaction between Dazl protein and Mvh mRNA, biotiny- vivo (Prm2, Fdft1 and 2810408A11Rik) showed very little or lated RNA corresponding to the full-length 30-UTR of the gene no enrichment in the Dazl immunoprecipitate (0.9–1.4-fold was transcribed in vitro and immobilized on streptavidin- enrichment in Dazl immunoprecipitates compared with con- conjugated Dynabeads. An antisense transcript was used as a trols) (Fig. 2 and data not shown). This validates our approach control for non-specific binding of the protein. The ability of for identifying genuine in vivo targets for Dazl. the RNAs to pull down either wild-type or an RNA-binding Human PRM2 RNA was identified in a screen for tran- domain mutant of GST-Dazl fusion protein was determined scripts bound by both PUM2 and DAZL proteins using by western blotting. Truncated GST-Dazl proteins encompass- recombinant proteins and a human testis RNA library (22). ing the entire RNA-binding domain (amino acids 1–137) were Filter binding experiments using the full-length murine used due to problems with solubility of the full-length con- Prm2 30-UTR and recombinant GST-Dazl protein confirm structs. This truncation of Dazl retains specificity in RNA- that this interaction occurs in vitro (Fig. 5A). However, analy- binding in vitro (19). Assays were carried out in mouse sis of transcripts that were co-immunoprecipitated with Dazl testis extract and therefore in the presence of endogenous con- from rat testis showed that Prm2 was not specifically enriched centrations of proteins and RNA to increase interaction speci- (4.6-fold enrichment in the negative control, anti-Dazl þ ficity. Binding was detected only with the sense transcript of peptide immunoprecipitation, compared with that using the the 30-UTR of Mvh mRNA in combination with GST-Dazl anti-Dazl antibody alone). This lack of interaction for mouse protein with an intact RNA-binding domain (Fig. 3), confirm- Dazl and Prm2 in vivo was confirmed using UV cross-linked ing that this protein is able to interact specifically with the mouse testis (Fig. 2). This inconsistency between in vivo Mvh transcript. Neither the 50-UTR nor any part of the and in vitro binding emphasizes the importance of confirming coding sequence was bound by GST-Dazl in these assays the physiological relevance for each interaction individually. (data not shown). Human Molecular Genetics, 2005, Vol. 14, No. 24 3903 Downloaded from https://academic.oup.com/hmg/article/14/24/3899/2355858 by guest on 27 September 2021

Figure 3. In vitro binding of Dazl protein to the Mvh 30-UTR. Biotinylated RNA was immobilized on streptavidin-conjugated Dynabeads, mixed with recombinant GST-Dazl in the presence of mouse testis homogenate and bound protein detected by western blotting. The full-length, sense transcript of Mvh 30-UTR is bound by GST-Dazl with a functional RNA-binding domain but not by an RNA-binding mutant. No binding was detected to the antisense transcript.

The minimal region within the Mvh 30-UTR bound by Dazl was initially identified by comparing binding of the protein to a series of truncated RNAs. This approach led to the identification of the 200-nucleotide region between nucleotides 100 and 300 of the 30-UTR as a Dazl-binding region (Fig. 4A and C). This region of the 30-UTR corresponds closely to a region containing all five of the evolutionarily conserved Dazl-binding sites identified by sequence analysis, between 30-UTR positions 93 and 265 (Fig. 4B). These five predicted Dazl-binding sites are all conserved to some degree in other mammals. More specifically, according to the current data of Kent et al. (33) these sites occupy a region of the mouse Mvh 30-UTR which shows 78 and 77% identity with human and dog genomes, respectively. This approaches the level of similarity seen in Mvh Figure 4. Characterization of minimal binding region for GST-Dazl within the Mvh 30-UTR. (A) Serial truncations of the Mvh 30-UTR were assayed for coding sequence (84 and 85% identity with human and binding to GST-Dazl. Nucleotides 1–200 (1) sense and (2) antisense tran- dog, respectively). When RNA encompassing those sequences scripts, nucleotides 100–300 (3) sense and (4) antisense transcripts, nucleo- was transcribed and assayed in an equivalent experiment, tides 200–400 (5) sense and (6) antisense transcripts, nucleotides 300–500 it was bound by Dazl protein (Fig. 4C). Mutation of all (7) sense and (8) antisense transcripts, nucleotides 400–609 (9) sense and five of the putative binding sites completely abolished (10) antisense transcripts, (11) input protein. Strong binding is evident for nucleotides 100–300. Reproducible, though low-level binding occurs Dazl binding. The same was true for mutation of the between nucleotides 200 and 400. Binding to antisense controls was absent 0 first three sites. However, the two most 3 sites could be or low in all cases. (B) Sequence of Mvh 30-UTR. Putative Dazl-binding altered with no noticeable effect on Dazl binding (Fig. 4C sites which have been evolutionarily conserved are marked in bold. Five 0 and D). sites were identified in the Mvh 3 -UTR (numbered 1–5). One additional, potential binding site is underlined although this is not conserved in other The results of binding assays of a number of mutant and 0 0 species. (C) Mutation of putative Dazl-binding sites within the Mvh 3 -UTR truncated Mvh 3 -UTR sequences are summarized in abolishes binding of the Dazl protein. (1) No RNA control, (2) Mvh 30-UTR Figure 4D. Since RNAs containing either putative binding nucleotides 85–270 with all five putative binding sites intact, (3) sites 1–3 sites 1–3 or sites 2–4 interact in vitro with recombinant mutated, (4) sites 4 and 5 mutated, (5) all five sites mutated and (6) protein Dazl protein, binding must either be dependent on the pre- input only. (D) Summary of results for binding of GST-Dazl protein and Mvh 30-UTR. Truncation analysis shows Dazl binding correlates to the presence of a minimum number of these sites within the RNA, sence of multiple, putative binding sites in the Mvh 30-UTR. Mutation of or the context of the sites is important. sites 1–3, but not of 4 and 5 abolishes the interaction. A fragment of the 30- Filter binding experiments showed that the interaction UTR with sites 2–5 is bound tightly, RNA containing sites 4, 5 and a sixth, between Dazl protein and the 30-UTR of Mvh is direct non-conserved potential binding site is also bound by the protein. and does not require the presence of any other proteins (Fig. 5A). Binding was competed by the addition of unlabelled UUCUUCUGUUCUU corresponding to putative binding sites transcript and no binding was detected between an antisense 2and3inthe30-UTR was bound directly by GST-Dazl with transcript and Dazl protein with an intact RNA-binding a functional RRM, but a control sequence in which uracils domain (data not shown). A multimer of the Mvh sequence and cytosines had been interchanged was not (Fig. 5B). 3904 Human Molecular Genetics, 2005, Vol. 14, No. 24 Downloaded from https://academic.oup.com/hmg/article/14/24/3899/2355858 by guest on 27 September 2021

Figure 5. Filter binding assays to show interaction between GST-Dazl and RNA. Equal amounts of radiolabelled RNA were incubated with an increasing concentration of GST fusion protein and binding detected in the form of retention of the RNA on a nitrocellulose filter. (A) GST-Dazl with a functional RNA- binding domain can bind to 30-UTRs of Mvh (filled circles) and Prm2 (filled diamonds) in the absence of other proteins. Transcripts are not bound by an RNA-binding domain mutant version of the protein (open circles and open diamonds, respectively). (B) A multimer of the putative Dazl-binding site is bound directly by GST-Dazl in vitro. (UUCUUCUGUUCUU)5 is bound by GST-Dazl with a functional RNA-binding domain (filled diamonds), but an equivalent transcript in which uracils and cytosines were interchanged, (CCUCCCUGCCUCC)5, is not (filled triangles). Neither (UUCUUCUGUUCUU)5 (filled squares) nor (CCUCCCUGCCUCC)5 (open circles) are bound by GST-Dazl with a mutated RNA-binding domain.

Stimulation of translation by Dazl via 30-UTR sequences with wild-type (Fig. 7A). Quantification of the immunofluor- Members of the DAZ family of proteins have been shown to escence signals for Mvh in a total of 40 cells for each genotype indicated that Mvh protein levels in Dazl null germ cells were, stimulate translation of specific transcripts (17,18). We, there- on average, decreased to 25% that of wild-type (Fig. 7B). fore, asked whether Dazl had an effect on translation via the This in vivo decrease in Mvh protein levels strongly 30-UTR of Mvh using a translation assay in Xenopus laevis supports the hypothesis that the Mvh transcript is a target for oocytes which has previously been used to show that murine translational regulation by Dazl and suggests that decreased Dazl can directly activate translation of reporter mRNAs levels of Mvh protein contribute to the leptotene block seen (18). In this assay, stage VI oocytes are injected with RNA in Dazl null animals. encoding either Dazl or a control RNA-binding protein (U1A) and incubated to allow production of the protein. The oocytes are then co-injected with RNAs encoding a luciferase reporter fused to either the 30-UTR of Mvh or of Dengue virus, DISCUSSION which serves as a negative control, as well as RNA for This study is the first attempt to isolate endogenous Dazl b-galactosidase as an internal control. Where Dazl protein is 0 protein–RNA complexes directly rather than reconstituting recruited to the target 3 -UTR RNA, stimulation of translation those interactions using recombinant proteins or RNA is reflected in an increase in luciferase activity (18). 0 libraries. We present evidence that Dazl protein binds to and When the luciferase reporter was fused to the Mvh 3 -UTR, regulates the translation of specific transcripts important for translation was increased 5.5-fold relative to the U1A control spermatogenesis. We have identified transcripts that are in (based on an average of four independent experiments) close association with Dazl protein in vivo and focussed on (Fig. 6). No stimulation of translation was detected when the 0 Mvh, a gene essential for mammalian spermatogenesis. Dazl control Dengue virus 3 -UTR was used, indicating that this binds Mvh mRNA directly in vitro via the 30-UTR, which con- effect is specific to the Mvh 30-UTR. This shows that the trans- 0 tains five putative Dazl-binding sequences that are conserved lation of mRNAs containing the 3 -UTR of Mvh can be in mammals. Mutation of these sequences abolishes binding specifically regulated by Dazl. by Dazl, and multimers of the motif alone can be bound directly by Dazl protein in vitro, suggesting that these sequences may contribute to the recognition of transcripts by In vivo protein levels of Mvh protein Dazl in vivo. There is a good overlap between the cell types To determine the effect of Dazl on Mvh translation in vivo, the and stage of spermatogenesis in which Mvh is transcribed levels of Mvh protein in testes from wild-type and knockout and the presence of the Dazl protein in vivo. In addition, the mice at 4 and 5 days post-partum was compared using an anti- interaction between Dazl and the Mvh 30-UTR leads to stimu- body to germ cell nuclear antigen (GCNA) to identify germ lation of translation in Xenopus oocytes and there is a strong cells. This corresponds to a time just before the gonocytes correlation between the absence of Dazl and a reduction in settle onto the basement membrane and resume mitosis. At Mvh protein levels in vivo. this stage, there were far fewer gonocytes in the Dazl null We, therefore, propose that Mvh is subject to translational animals compared with wild-type littermates (1078 GCNA- regulation by Dazl in vivo and that a loss of or reduction in positive cells per cross-section of testis for wild-type com- its translation contributes to the phenotypes observed in pared with an average of 9.4 per section for Dazl null Dazl null mice. The scope of this analysis has also enabled animals). In GCNA-positive cells in the Dazl knockout us to define elements within the 30-UTRs of genes that mark animals, Mvh protein levels tended to be low when compared them as potential in vivo targets of Dazl and will allow Human Molecular Genetics, 2005, Vol. 14, No. 24 3905 Downloaded from https://academic.oup.com/hmg/article/14/24/3899/2355858 by guest on 27 September 2021

Figure 6. Dazl stimulates translation via the Mvh 30-UTR. Xenopus oocytes producing either Dazl or U1A were co-injected with luciferase reporter mRNA with the 30-UTR of either Mvh or Dengue virus (DV) and a b-galactosidase reporter as an injection control. Relative translation is assayed as luciferase activity normalized with respect to b-galactosidase activity. Errors were calculated as standard error of the mean. Figure 7. Mvh protein levels in germ cells from wild-type and Dazl null animals. (A) Immunofluorescence of testis sections from 5 days post-partum animals using anti-GCNA (green) and anti-Mvh (red) antibodies. The signal from Mvh alone is shown beside each merged image. Germ cells in further study of the mechanism of action as well as the in vivo Dazlþ/þ animals have high levels of Mvh protein in all GCNA-positive 2/2 function of Dazl and other family members. cells whereas levels in Dazl GCNA-positive cells are lower. (B) Quantifi- cation of protein levels by immunofluorescence at the single-cell level. Signal The Dazl knockout phenotype is complex with a failure of intensities for GCNA and Mvh were measured for a total number of 40 cells in diploid cell differentiation in the adult and a block in leptotene at least three animals for Dazlþ/ þ (white bars) and Dazl2/2 animals (black in the first wave of spermatogenesis. This suggests a role for bars). Standard errors of the mean are indicated. Dazl in the translational regulation of a number of different mRNAs important for multiple, distinct functions in gametogenesis. To identify those RNA molecules associated specifi- perhaps also require other specific factors for their transla- cally with Dazl in rodent testis, we excluded genes that were tional regulation. The motif defined by Venables et al. is also co-purified with the Sf2 protein. This will almost certainly present and has been conserved evolutionarily in the 30-UTR mean that some genuine targets of Dazl were lost from sub- of 14 out of 15 of these transcripts. This level of conservation sequent analysis, as there is no evidence to suggest that in the non-coding region of a gene is indicative of an import- some transcripts might not be regulated by both proteins. ant function. Our results also support the role of the minimal Although this resulted in a decreased number of potential motif, GUUC, identified as a requirement for binding of targets for further study, those remaining have an increased zebra fish Dazl (17). Identification of a specific subset of tran- probability of relevance because abundant RNAs which are scripts bound in vitro by both PUM2 and DAZL proteins led to prone to non-specific interactions with RNA-binding proteins the definition of a U-rich element in the 30-UTR of SDAD1. were excluded. Both human SDAD1 and murine Sdad1 contain sequence Previous studies have identified a number of potential elements that fit our minimal binding motif although these targets and conserved sequences required for binding by the do not lie within regions that were bound tightly in vitro DAZ family of proteins (12,17,19,20,22). The multitude of by DAZL protein (22). This would suggest that the Dazl targets means that there has been little overlap in the U2-10[G/C]U2-10 element proposed here is not the sole identity of specific transcripts identified to date. In addition, element bound by Dazl proteins, or that there are differences cell lysis can lead to associations between molecules that are in binding specificity between the human and murine proteins. not representative of biologically relevant interactions (31). One of the blocks in gametogenesis that has been described Indeed we were able to detect an interaction between Dazl in detail for animals lacking Dazl is at leptotene of meiotic and the 30-UTR of Prm2 in vitro although no such interaction prophase in both male and female mice (10). This is the was detectable in vivo using our approach. In vivo UV cross- point in male Mvh null mice at which meiotic progression linking has confirmed interaction in vivo for a number of the ceases (32) and we propose that the observed decrease in potential targets identified here. Mvh protein levels could contribute to the Dazl null phenotype Both of the sequence elements that have been defined for in males. No defects in female gametogenesis have been binding of murine Dazl (19,20) have been found in these reported for Mvh knockout mice although the protein is transcripts. The conserved sequence defined by Jiao et al. is present throughout oogenesis (32,34). It is, therefore, likely present in six out of 15 of the mRNAs specifically enriched that there are a number of transcripts that cannot be efficiently in the Dazl immunoprecipitates and these messages could translated in the absence of Dazl, whose protein products are therefore reflect a subset of messages regulated by Dazl that required at leptotene/zygotene in both male and female mice. 3906 Human Molecular Genetics, 2005, Vol. 14, No. 24

The reduction in Mvh protein levels in Dazl null cells is in (Stratalinker 1800, Stratagene). Lysis was carried out in 1 agreement with the hypothesis that Mvh is an in vivo target for PXL (1 PBS containing 0.1% SDS, 0.5% deoxycholate, translational regulation by Dazl. Mvh protein levels were not 0.5% NP-40) and immunoprecipitations carried out as reduced completely to zero but to 25% that seen in the equiv- above. RNA was reverse-transcribed using an oligo(dT) alent wild-type cells. These protein levels might be explained primer and a first-strand cDNA kit (Roche Diagnostics), by a low, basal level of translation that occurs in the absence followed by either semi-quantitative or real-time PCR to of the Dazl protein rather than a complete cessation of Mvh determine relative levels of transcripts present. PCR primers protein production. It is possible that these levels of Mvh were designed against the 30-UTR of genes and a mock protein are sufficient for completion of early germ cell devel- reverse transcription reaction was carried out in all cases as opment but are insufficient for passage through later stages, a control for contamination with genomic DNA. Gene-specific culminating in a common phenotype or Mvh and Dazl null primers used are listed in Supplementary Material, Table 4. mice of a final block at leptotene of meiotic prophase I. Downloaded from https://academic.oup.com/hmg/article/14/24/3899/2355858 by guest on 27 September 2021 It is not possible to distinguish whether this in vivo reduction of Mvh protein is caused directly by changes in Microarray analysis the translation or stability of the mRNA, or by changes in Amplified aRNA was hybridized to adult mouse, testis- the rate of protein turnover. Translation and mRNA stability specific, subtracted and normalized cDNA microarrays as are often intimately linked, meaning that a decrease in the described previously (24) using a triple-dye labelling approach translation of an mRNA can lead to a dramatic effect on its (25). In brief, total sample RNA was aminoallyl-labelled with stability (30). More work is required to fully characterize the Cy3 and Cy5 dyes and Alexa 594. Labelled total adult rat mechanism by which Dazl regulates translation of target testis RNA was used as a reference sample in all cases. Scan- mRNAs in murine spermatogenesis although, at least in ning and analysis of the microarrays was carried out as Xenopus oocytes, Dazl has been shown to have a direct described previously (24). Levels of hybridization to the effect on translation initiation (18). microarrays for total RNA from adult rat testis and adult The DAZ genes are essential for fertility in a number of mouse testis were compared to confirm a high level of organisms and their loss has been linked to male infertility cross-hybridization between rat and mouse genes. Statistical in humans. Determining genuine in vivo targets of the DAZ- analysis and fold difference filtering were carried out related proteins is fundamental to our understanding of the using Genespring (version 5.1; Silicon Genetics). One-way biological function of this protein family, as well as to a ANOVA was used to obtain lists of genes that were signifi- wider understanding of mammalian gametogenesis and the cantly different between the immunoprecipitation conditions aetiology of human infertility. and data corrected for multiple testing using the Benjamini and Hochberg false discovery rate, using a P-value cutoff of 0.05. MATERIALS AND METHODS Immunoprecipitation and isolation of RNA Sequence analysis Adult rat testes were homogenized in ice-cold IP buffer Identification of orthologous genes was carried out as follows. (20 mM HEPES pH 7.6, 10 mM KCl, 1.5 mM MgCl2, Non-redundant proteome data sets for human, mouse and rat 100 mM NaCl, 0.5% Triton, 0.5 mM DTT) supplemented were retrieved from the EBI Proteome Analysis site (http:// with protease inhibitors (Complete, Roche Diagnostics Ltd). www.ebi.ac.uk/proteome/index.html). Putative orthologs Insoluble material was removed by centrifugation and the were detected as reciprocal best BLAST hits between the protein concentration of the resulting supernatant was two proteomes of interest (36,37). For chicken, where a non- measured by Bradford assay. Glycerol was added to a final redundant proteome data set is not yet available, a two-stage concentration of 40% and aliquots stored at 2708C. process was followed. First, mouse proteins were searched Immunoprecipitation of proteins was carried out using an against a large (though incomplete) chicken mRNA collection affinity purified polyclonal antibody to murine Dazl (6), mono- (available from ftp://hgdownload.cse.ucsc.edu/goldenPath/ clonal antibody to human SF2 (23) or with the affinity purified galGal2/bigZips/) to find the best reciprocal matches. Then polyclonal antibody to Dazl in the presence of the peptide to the draft chicken genome (available from http://genome. which it was raised. Wash conditions were adjusted to wustl.edu/projects/chicken/) was searched to verify that each minimize co-purifying proteins (IP buffer þ 500 mM NaCl). mRNA match was the best in the chicken genome. Compari- Aliquots were taken for western blot analysis and RNA sons between each mouse mRNA and its orthologous group isolated from the remainder by proteinase K digestion and of mRNA sequences were performed using Multi-VISTA (38). phenol/chloroform extraction. Three immunoprecipitations Multiple sequence alignments were performed using were carried out for each antibody or the antibody þ peptide. CLUSTALW (39). Hidden Markov models were constructed The RNA was pooled and amplified using a T7-Oligo(dT) and searched against mRNA sequences using HMMER primer and the MessageAmp aRNA Kit (Ambion). Version 1.8.4 (40). Simple motif searches were carried out UV cross-linking and immunoprecipitations from mouse using the patmatdb utility from the EMBOSS suite of pro- testes were carried out as above with modifications based grams (41) and sequences were randomized using the shuffle- on (35). Testes from adult C57BL/6 mice were dissociated seq utility from the same package. For control analyses, 500 to single tubules in Hanks Balanced Salt Solution (GIBCO) randomized sets of sequences with identical base composition on ice before cross-linking three times with 4000 mJofUV to those in the starting set were produced and each of these Human Molecular Genetics, 2005, Vol. 14, No. 24 3907 randomized sets was searched in the same way as the original sequence was excised from pGEM-50Leader with EcoRI/SacI, set of sequences. gel-purified and cloned into the polylinker of pUC19 generat- Approximate P-values were calculated for over-representation ing pUC19-Leader. of motifs in the starting set, according to the proportion of ran- The luciferase open-reading frame (ORF) was PCR domized sets yielding the same or a greater number of motif amplified from pLG-MS2 (43) and subcloned into pGEM-T matches. Easy, generating pGEM-LUC (see Supplementary Material, Table S4 for primer sequences). pGEM-LUC was digested BssHII/SalI excising the luciferase ORF, which was then In vitro assays for protein/RNA interactions cloned into pUC19-Leader digested BssHII/SalI generating Full-length Mvh 30-UTR was cloned by PCR from adult mouse pUC19-LUC. testis cDNA into pGEM-T Easy (Promega) to create pGEM- pUC19LUC-Mvh (30-UTR). The full-length 30-UTR of Mvh 0 0 TEasyMvh3 -UTR, which was sequenced fully. Truncations was PCR amplified from pGEM-T EasyMvh3 -UTR using a Downloaded from https://academic.oup.com/hmg/article/14/24/3899/2355858 by guest on 27 September 2021 and mutations were introduced by PCR products cloned into 50 primer with a SalI and a 30primer with a HindIII restriction pGEM-T Easy and all inserts fully sequenced. Transcription site (see Supplementary Material, Table S4) and cloned into from linearized plasmid was carried out using T7 or SP6 the SalI and HindIII sites of pUC19LUC. RNA polymerases (Roche Diagnostics) as appropriate. Oligo- pUC19LUC-DV(30-UTR). The Dengue virus 30-UTR was nucleotide sequences are listed in Supplementary Material, PCR amplified from the DEN4 full-length cDNA-2A Table S4. plasmid (44) using primers detailed (Supplementary Material, RNA was biotin labelled by in vitro transcription in the Table S4). This was subcloned into pGEM-T Easy prior to presence of biotin-16-UTP (Biotin RNA Labelling Mix, Roche excision and cloning into pUC19LUC (SalI/HindIII). Diagnostics). Equal quantities of sense and antisense RNA were immobilized on streptavidin-conjugated Dynabeads M-280 (Dynal Biotech ASA) according to the manufacturer’s RNA synthesis and microinjection of Xenopus oocytes instructions. RNA and beads were washed once in cold IP RNA synthesis was carried out according to Gray et al. (43). buffer before mixing with rat testis extract containing Briefly, stage VI Xenopus oocytes were collected and injected 10 ng/ml recombinant GST-Dazl proteins for 30 min at 48C. with 1 mg/ml of mDazl or U1A RNA. Following 5–6 h incu- Beads were then washed five times with IP buffer and co- bation at 188C, the oocytes were injected with a mix of a target purifying proteins detected by western blotting using an reporter RNA at a concentration of 30 ng/ml and the internal antibody directed against GST (Sigma). Recombinant control RNA b-Gal at a concentration 15 ng/ml. The oocytes GST-Dazl and GST-Dazl RRM- were bacterially 1 –137 1 –137 were incubated overnight at 188C before collection and produced using expression constructs described previously (19). assay as described in Gray et al. (43). Filter binding assays were carried out essentially as in (42) using GST-fusion Dazl proteins described earlier. Proteins were diluted in 2 Binding Buffer (10 mM HEPES pH 7.9, Immunofluorescence and quantification of protein levels 100 mM KCl, 1 mM DTT, 5% glycerol, 500 ng BSA, 5 mM Tm1hgu/Tm1hgu MgCl2) and mixed with an equal volume of radiolabelled Dazl mice (6) were housed under standard con- RNA in the presence of 150 ng/ml tRNA. Following incu- ditions and fed ad libitum. Generation and breeding of mice bation at 378C for 10 min, samples were applied to a nitro- was covered by Home Office project licence and was approved cellulose filter (NC45, Schleicher and Schuell), washed by institutional ethics committees. twice with 500 mlof1 Binding Buffer and the radioactivity Testes were recovered from mice on days 4 and 5 after birth retained on each filter measured by scintillation counting. (where day 0 is the day of birth) and tail tips taken at the time of death for genotyping. Tissues were fixed for 4–6 h in Bouin’s fixative and processed into paraffin wax using stan- Translation assays dard methods. Sections were cut (6 mm) and mounted on Plasmid construction. pGEM-mDazl was generated by PCR slides before being dewaxed and rehydrated in graded amplification of the mDazl ORF from MSPN-mDazl (18) alcohols. Samples were processed using a method based on prior to cloning into pGEM-T Easy vector (Promega) (see MacPherson et al. (45). In brief, slides were microwaved Supplementary Material, Table S4 for primer sequences). on full power (Panasonic, 800 W microwave) for 18 min in pMS2-U1A has been previously described in (43). pUC19- citrate buffer (0.01 M sodium citrate, pH 6.0), washed once LUC was generated by PCR amplification of the region in PBST (PBS þ 0.1% Tween) and then twice in PBS. Follow- (GGAGAAAATA CCGCATCAGG CGCCATTCGC CATT ing a 30 min incubation in blocking solution (5% horse serum, CAGGCT GCGCAACTGT TGGGAAGGGC GATCGGT 2% BSA in PBST), sections were incubated with primary anti- GCG GGCCTCTTCG CTATTACGCC AGCTGGCGAA A) body diluted in 5% horse serum in PBST (anti-Mvh at 1 in of pUC19 to act as a 50Leader sequence using the oligonucleo- 5000 and anti-GCNA at 1/20) overnight at 48C. Sections tide 50-GAATTCtgta atacgactca ctatagggcg CTCGAGGGAG were washed once in PBST and twice in PBS before incu- 0 AAAATACCGC-3 which contains an EcoRI site followed bation with secondary antibodies (Alexa594 anti-rabbit 0 by a T7 promoter and the oligonucleotide 3 -GAGCT for anti-Mvh, Alexa488 anti-rat for anti-GCNA, both diluted CGCGCGC TTTCGCCAGC TGGC-50 containing a SacI 1/500 in 5% horse serum in PBST) for 2 h at room tempera- restriction site. The 50Leader sequence was subcloned into ture. Sections were washed as earlier before mounting for pGEM-T Easy generating pGEM-50Leader. The 50Leader microscopy and image capture. 3908 Human Molecular Genetics, 2005, Vol. 14, No. 24

The imaging system used comprises a Coolsnap HQ CCD 7. Houston, D.W. and King, M.L. (2000) A critical role for Xdazl, a germ camera (Photometrics Ltd, Tucson, AZ, USA) Zeiss Axioplan plasm-localized RNA, in the differentiation of primordial germ cells in Xenopus. Development, 127, 447–456. II fluorescence microscope with Plan-neofluar objectives, a 8. Karashima, T., Sugimoto, A. and Yamamoto, M. (2000) Caenorhabditis 100 W Hg source (Carl Zeiss, Welwyn Garden City, UK) elegans homologue of the human azoospermia factor DAZ is required for and Chroma #83000 triple band-pass filter set (Chroma Tech- oogenesis but not for spermatogenesis. Development, 127, 1069–1079. nology Corp., Rockingham, VT, USA) with the excitation 9. Schrans-Stassen, B.H., Saunders, P.T., Cooke, H.J. and de Rooij, D.G. 2/ 2 filters installed in a motorized filter wheel (Ludl Electronic (2001) Nature of the spermatogenic arrest in Dazl mice. Biol. Reprod., 65, 771–776. Products, Hawthorne, NY, USA). Image capture and analyses 10. Saunders, P.T.K., Turner, J.M.A., Ruggiu, M., Taggart, M., were performed using in-house scripts written for IPLab Spec- Burgoyne, P.S., Elliott, D. and Cooke, H.J. (2003) Absence of mDazl trum (Scanalytics Corp., Fairfax, VA, USA). Quantification of produces a final block on germ cell development at meiosis. signals was carried out on samples that had been processed for Reproduction, 126, 589–597. microscopy in precisely the same way. Un-manipulated raw 11. Castrillon, D.H., Gonczy, P., Alexander, S., Eberhart, C.G., Viswanathan, Downloaded from https://academic.oup.com/hmg/article/14/24/3899/2355858 by guest on 27 September 2021 S., DiNardo, S. and Wasserman, S.A. (1993) Toward a molecular genetic images that had been captured with identical exposure times analysis of spermatogenesis in Drosophila melanogaster: characterization were used for all calculations, which were performed using of male-sterile mutants generated by single P element mutagenesis. IPLab Spectrum (Scanalytics Corp.). Samples from at least Genetics, 135, 489–505. three animals of each genotype were used in each analysis. 12. Maines, J.Z. and Wasserman, S.A. (1999) Post-transcriptional regulation of the meiotic Cdc25 protein Twine by the Dazl orthologue Boule. Nat. Cell Biol., 1, 171–174. 13. Joiner, M.L. and Wu, C.F. (2004) Nervous system function for the SUPPLEMENTARY MATERIAL testis RNA-binding protein boule in Drosophila. J. Neurogenet., 18, 341–363. Supplementary Material is available at HMG Online. 14. Vogt, P.H. and Fernandes, S. (2003) Polymorphic DAZ gene family in polymorphic structure of AZFc locus: artwork or functional for human spermatogenesis? APMIS, 111, 115–126. ACKNOWLEDGEMENTS 15. Vogt, P.H. (2005) Azoospermia factor (AZF) in Yq11: towards a molecular understanding of its function for human male fertility and We would like to thank Brian Hendrich, Ian Adams, Andrew spermatogenesis. Reprod. Biomed. Online, 10, 81–93. Childs and Mary O’Connell for useful discussions and com- 16. Tsui, S., Dai, T., Warren, S.T., Salido, E.C. and Yen, P.H. (2000) Association of the mouse infertility factor DAZL1 with actively ments on the manuscript as well as the following people, translating polyribosomes. Biol. Reprod., 62, 1655–1660. who generously provided reagents: Jeremy Sanford and 17. Maegawa, S., Yamashita, M., Yasuda, K. and Inoue, K. (2002) Zebra fish Javier Caceres (anti-SF2 antibody), Prof. Noce (anti-Mvh anti- DAZ-like protein controls translation via the sequence ‘GUUC’. Genes body), Prof. Enders (GCNA antibody) and Ching-Juh Lai Cells, 7, 971–984. (DEN4 cDNA). We would also like to thank Bill Richardson 18. Collier, B., Gorgoni, B., Loveridge, C., Cooke, H.J. and Gray, N.K. (2005) The DAZL family proteins are PABP-binding proteins that and Ross Anderson for technical assistance, Paul Perry for regulate translation in germ cells. EMBO J., 24, 2656–2666. help with microscopy and Sandy Bruce for figure production. 19. Venables, J.P., Ruggiu, M. and Cooke, H.J. (2001) The RNA-binding This work was funded by the UK Medical Research Council. specificity of the mouse Dazl protein. Nucleic Acids Res., 29, 2479–2483. B.C. and N.K.G. are funded by an MRC Career Development 20. Jiao, X., Trifillis, P. and Kiledjian, M. (2002) Identification of target Award and an MRC Senior Fellowship awarded to N.K.G. messenger RNA substrates for the murine deleted in azoospermia-like RNA-binding protein. Biol. Reprod., 66, 475–485. 21. Urano, J., Fox, M.S. and Reijo Pera, R.A. (2005) Interaction of the Conflict of Interest statement. The authors declare that there is conserved meiotic regulators, BOULE (BOL) and PUMILIO-2 (PUM2). no conflict of interest. Mol. Reprod. Dev., 71, 290–298. 22. Fox, M., Urano, J. and Reijo Pera, R.A. (2005) Identification and characterization of RNA sequences to which human PUMILIO-2 (PUM2) and deleted in Azoospermia-like (DAZL) bind. Genomics, 85, 92–105. REFERENCES 23. Hanamura, A., Caceres, J.F., Mayeda, A., Franza, B.R., Jr and 1. Haag, E.S. (2001) Rolling back to BOULE. Proc. Natl Acad. Sci. USA, 98, Krainer, A.R. (1998) Regulated tissue-specific expression of antagonistic 6983–6985. pre-mRNA splicing factors. RNA, 4, 430–444. 2. Reijo, R., Lee, T.Y., Salo, P., Alagappan, R., Brown, L.G., Rosenberg, M., 24. Maratou, K., Forster, T., Costa, Y., Taggart, M., Speed, R.M., Ireland, J., Rozen, S., Jaffe, T., Straus, D., Hovatta, O. et al. (1995) Diverse Teague, P., Roy, D. and Cooke, H.J. (2004) Expression profiling of the spermatogenic defects in humans caused by Y chromosome deletions developing testis in wild-type and Dazl knockout mice. Mol. Reprod. encompassing a novel RNA-binding protein gene. Nat. Genet., 10, Dev., 67, 26–54. 383–393. 25. Forster, T., Costa, Y., Roy, D., Cooke, H.J. and Maratou, K. (2004) 3. Vogt, P.H., Edelmann, A., Kirsch, S., Henegariu, O., Hirschmann, P., Triple-target microarray experiments: a novel experimental strategy. Kiesewetter, F., Kohn, F.M., Schill, W.B., Farah, S., Ramos, C. et al. BMC Genomics, 5, 13. (1996) Human Y chromosome azoospermia factors (AZF) mapped to 26. Shima, J.E., McLean, D.J., McCarrey, J.R. and Griswold, M.D. (2004) different subregions in Yq11. Hum. Mol. Genet., 5, 933–943. The murine testicular transcriptome: characterizing gene expression in the 4. Ferlin, A., Moro, E., Garolla, A. and Foresta, C. (1999) Human male testis during the progression of spermatogenesis. Biol. Reprod., 71, infertility and Y chromosome deletions: role of the AZF-candidate genes 319–330. DAZ, RBM and DFFRY. Hum. Reprod., 14, 1710–1716. 27. Schlecht, U., Demougin, P., Koch, R., Hermida, L., Wiederkehr, C., 5. Eberhart, C.G., Maines, J.Z. and Wasserman, S.A. (1996) Meiotic cell Descombes, P., Pineau, C., Jegou, B. and Primig, M. (2004) Expression cycle requirement for a fly homologue of human deleted in Azoospermia. profiling of mammalian male meiosis and gametogenesis identifies novel Nature, 381, 783–785. candidate genes for roles in the regulation of fertility. Mol. Biol. Cell, 15, 6. Ruggiu, M., Speed, R., Taggart, M., McKay, S.J., Kilanowski, F., 1031–1043. Saunders, P., Dorin, J. and Cooke, H.J. (1997) The mouse Dazla gene 28. Small, C.L., Shima, J.E., Uzumcu, M., Skinner, M.K. and Griswold, M.D. encodes a cytoplasmic protein essential for gametogenesis. Nature, 389, (2005) Profiling gene expression during the differentiation and 73–77. development of the murine embryonic gonad. Biol. Reprod., 72, 492–501. Human Molecular Genetics, 2005, Vol. 14, No. 24 3909

29. Costa, Y., Speed, R., Ollinger, R., Alsheimer, M., Semple, C.A., Gautier, P., 40. Eddy, S.R. (1998) Proﬁle hidden Markov models. Bioinformatics, 14, Maratou, K., Novak, I., Hoog, C., Benavente, R. and Cooke, H.J. (2005) 755–763. Two novel proteins recruited by synaptonemal complex protein 1 41. Rice, P., Longden, I. and Bleasby, A. (2000) EMBOSS: the European (SYCP1) are at the centre of meiosis. J. Cell Sci., 118, 2755–2762. Molecular Biology Open Software Suite. Trends Genet., 16, 276–277. 30. Jacobson, A. and Peltz, S.W. (1996) Interrelationships of the pathways of 42. Gallo, A., Keegan, L.P., Ring, G.M. and O’Connell, M.A. (2003) An mRNA decay and translation in eukaryotic cells. Annu. Rev. Biochem., 65, ADAR that edits transcripts encoding ion channel subunits functions as a 693–739. dimer. EMBO J., 22, 3421–3430. 31. Mili, S. and Steitz, J.A. (2004) Evidence for reassociation of 43. Gray, N.K., Coller, J.M., Dickson, K.S. and Wickens, M. (2000) Multiple RNA-binding proteins after cell lysis: implications for the interpretation portions of poly(A)-binding protein stimulate translation in vivo. of immunoprecipitation analyses. RNA, 10, 1–3. EMBO J., 19, 4723–4733. 32. Tanaka, S.S., Toyooka, Y., Akasu, R., Katoh-Fukui, Y., Nakahara, Y., 44. Lai, C.J., Zhao, B.T., Hori, H. and Bray, M. (1991) Infectious RNA Suzuki, R., Yokoyama, M. and Noce, T. (2000) The mouse homolog of transcribed from stably cloned full-length cDNA of dengue type 4 virus. Drosophila vasa is required for the development of male germ cells. Proc. Natl Acad. Sci. USA, 88, 5139–5143. Genes Dev., 14, 841–853.

45. MacPherson, H., Keir, P., Webb, S., Samuel, K., Boyle, S., Bickmore, W., Downloaded from https://academic.oup.com/hmg/article/14/24/3899/2355858 by guest on 27 September 2021 33. Kent, W.J., Sugnet, C.W., Furey, T.S., Roskin, K.M., Pringle, T.H., Forrester, L. and Dorin, J. (2005) Bone marrow-derived SP cells can Zahler, A.M. and Haussler, D. (2002) The human genome browser at contribute to the respiratory tract of mice in vivo. J. Cell Sci., 118, UCSC. Genome Res., 12, 996–1006. 2441–2450. 34. Toyooka, Y., Tsunekawa, N., Takahashi, Y., Matsui, Y., Satoh, M. 46. Yuan, L., Liu, J.G., Zhao, J., Brundell, E., Daneholt, B. and Hoog, C. and Noce, T. (2000) Expression and intracellular localization of mouse (2000) The murine SCP3 gene is required for synaptonemal complex Vasa-homologue protein during germ cell development. Mech. Dev., assembly, chromosome synapsis, and male fertility. Mol. Cell, 5, 73–83. 93, 139–149. 35. Ule, J., Jensen, K.B., Ruggiu, M., Mele, A., Ule, A. and Darnell, R.B. 47. Zhu, D., Dix, D.J. and Eddy, E.M. (1997) HSP70-2 is required for CDC2 (2003) CLIP identiﬁes Nova-regulated RNA networks in the brain. kinase activity in meiosis I of mouse spermatocytes. Development, 124, Science, 302, 1212–1215. 3007–3014. 36. Chervitz, S.A., Aravind, L., Sherlock, G., Ball, C.A., Koonin, E.V., 48. Rasmussen, C.D. and Means, A.R. (1989) Calmodulin is required for Dwight, S.S., Harris, M.A., Dolinski, K., Mohr, S., Smith, T. et al. (1998) cell-cycle progression during G1 and mitosis. EMBO J., 8, 73–82. Comparison of the complete protein sets of worm and yeast: orthology 49. Okada, H., Suh, W.K., Jin, J., Woo, M., Du, C., Elia, A., Duncan, G.S., and divergence. Science, 282, 2022–2028. Wakeham, A., Itie, A., Lowe, S.W., Wang, X. and Mak, T.W. (2002) 37. Tatusov, R.L., Galperin, M.Y., Natale, D.A. and Koonin, E.V. (2000) Generation and characterization of Smac/DIABLO-deﬁcient mice. The COG database: a tool for genome-scale analysis of protein functions Mol. Cell Biol., 22, 3509–3517. and evolution. Nucleic Acids Res., 28, 33–36. 50. Wang, P.J., McCarrey, J.R., Yang, F. and Page, D.C. (2001) An 38. Brudno, M., Poliakov, A., Salamov, A., Cooper, G.M., Sidow, A., abundance of X-linked genes expressed in spermatogonia. Nat. Genet., 27, Rubin, E.M., Solovyev, V., Batzoglou, S. and Dubchak, I. (2004) 422–426. Automated whole-genome multiple alignment of rat, mouse, and human. 51. Wu, M.H., Rajkovic, A., Burns, K.H., Yan, W., Lin, Y.N. and Genome Res., 14, 685–692. Matzuk, M.M. (2003) Sequence and expression of testis-expressed gene 39. Higgins, D.G., Thompson, J.D. and Gibson, T.J. (1996) Using CLUSTAL 14 (Tex14): a gene encoding a protein kinase preferentially expressed for multiple sequence alignments. Methods Enzymol., 266, 383–402. during spermatogenesis. Gene Expr. Patterns, 3, 231–236.