Mouse staufen genes are expressed in germ cells during oogenesis ...

Molecular Human Reproduction vol.6 no.11 pp. 983–991, 2000

Mouse staufen genes are expressed in germ cells during oogenesis and spermatogenesis

P.T.K.Saunders1, S.Pathirana2, S.M.Maguire1, M.Doyle2, T.Wood2 and M.Bownes2,3 1Medical

Research Council, Human Reproductive Sciences Unit, Centre for Reproductive Biology, 37 Chalmers Street, Edinburgh, EH3 9ET, and 2Institute of Cell and Molecular Biology, The University of Edinburgh, Darwin Building King’s Buildings, Mayfield Road, Edinburgh, EH9 3JR, Scotland, UK

3To

whom correspondence should be addressed at: Institute of Cell and Molecular Biology, The University of Edinburgh, Darwin Building King’s Buildings, Mayfield Road, Edinburgh, EH9 3JR, Scotland, UK. E-mail: [email protected]

The Drosophila melanogaster staufen gene encodes an RNA binding protein (Dm Stau) required for the localization and translational repression of mRNAs within the Drosophila oocyte. In mammals translational repression is important for normal spermatogenesis in males and storage of mRNAs in the oocytes of females. In the present study we identified two mouse cDNA expressed sequence tags (ESTs), encoding proteins with significant homology to Dm Stau and used these firstly to screen a mouse kidney cDNA library and secondly to determine whether staufen mRNAs are expressed in the ovaries and testes of mice and rats. Sequence analysis of the cDNAs revealed that they originated from two different genes. Using Northern blots of RNAs from kidneys, ovaries and testes, both cDNAs hybridized to mRNA species of ~3 kb in all three tissues. On sections of mouse ovaries, staufen mRNA was localized specifically to oocytes. On sections of mouse testes, staufen mRNA was expressed in spermatocytes found in seminiferous tubules at stages VI–XII of the spermatogenic cycle. In conclusion, we have shown that the mammalian homologues of Dm stau are expressed in germ cells in both male and female mice, consistent with a role for these RNA binding proteins in mammalian gametogenesis. Key words: gametogenesis/ovary/RNA binding/staufen/testis

Introduction In Drosophila, during oogenesis some mRNAs are translationally repressed and remain stored until fertilization, when they are translated and their protein products become available to direct the initial development of the embryo. The Drosophila zygotic genome is not activated until the syncytial blastoderm stage. In addition, specific localization of mRNAs during oogenesis is essential for correct anterior–posterior axis formation of the embryo. The Drosophila staufen gene was identified in a genetic screen (Schu¨pbach and Wieschaus, 1986) and the Staufen protein (Stau) has been shown to contain five copies of a double stranded RNA binding domain (St Johnston et al., 1992). Binding studies undertaken with Stau protein have demonstrated that it recognizes secondary structures within the 3⬘ untranslated regions (UTR) of three mRNAs, i.e. bicoid and oskar in the oocyte (St Johnston et al., 1992) and prospero in the central nervous system (Broadus et al., 1998). Stau is essential for the localization of osk mRNA to the posterior of the oocyte (Kim-Ha et al., 1991) and bic mRNA to the anterior of the oocyte (St Johnston et al., 1991; Ferrandon et al., 1994; Hurst et al., 1999). In mutants, e.g. gurken, which alter the polarity of the oocyte, Stau protein and osk mRNAs mislocalize to the same ectopic sites (Gonzalez-Reyes et al., 1995). Recent analysis of the RNA binding domains within Dm Stau has revealed that when mutations encompassing domain 3 are incorporated into a full-length stau transgene, this construct is unable to localize osk or bic mRNAs correctly (Ramos et al., © European Society of Human Reproduction and Embryology

2000). Microtubules play an important role in localization of mRNAs during oocyte development. Data from studies using bacterially expressed fusion proteins prepared using cDNAs encoding human or mouse Stau confirm that Stau cross-links cytoskeletal and RNA components (Wickham et al., 1999). Other authors (Micklem et al., 2000) have shown recently that in Drosophila melanogaster Stau RNA binding domains 1, 3 and 4 bind double-stranded RNA in vitro whereas domains 2 and 5 do not. Furthermore, their results demonstrate that domain 2 is required for microtubule-dependent localization of osk mRNA, but domain 5 is involved in derepression of osk mRNA once it is correctly localized. The characteristics of Dm Stau including its specific mRNA binding ability, its role in translational repression, and its role in polarity determination via transcript localization, led us to consider that the mammalian equivalent of Dm Stau could play an important role in mammalian gametogenesis. In the mouse, transcription of the embryonic genome is not initiated until the 2-cell stage (Flach et al., 1982). In humans and pigs, transcription begins at the 4-cell stage and in sheep and cattle it does not initiate until the 8-cell stage (Braude et al., 1988). Both development of the embryo immediately after fertilization, and embryonic genome activation, are dependent upon maternally-derived mRNA (Harper and Monk, 1983; Latham et al., 1991; Wang and Latham, 1997). We were also interested to see whether mammalian stau was expressed in the testicular germ cells, as RNA binding 983

P.Saunders et al.

proteins have been shown to play an important role in male fertility (Elliott and Cooke, 1997). In the testis, during their development within the seminiferous epithelium, male germ cells undergo sequential mitoses, meiosis and structural remodelling. One important aspect of mammalian male germ cell development is the conversion of a round stem cell into a compact, motile cell; a process that involves nuclear condensation (Ward et al., 1983). The net result of these changes is to convert a transcriptionally active nucleus into the compact, quiescent nucleus of the spermatozoon. This process is tightly controlled and requires the synthesis of DNA binding proteins from mRNAs, the translation of which is controlled by specific RNA binding proteins (Braun et al., 1989; Kwon and Hecht, 1991, 1993). One of these RNA binding proteins has been shown to localize translationally repressed mRNA to microtubules (Han et al., 1995), a function analagous to that of that performed by Stau in the Drosophila oocyte. At the start of the study, database screening identified two mouse expressed sequence tags (ESTs) which had significant homology to Dm stau. The cDNA insert of one of these was used to isolate cDNAs from a mouse 11 day embryo cDNA library. To investigate the pattern of expression of mouse staufen mRNA in mammalian gonads, Northern blots and in-situ hybridization were undertaken. The results obtained demonstrate that stau is expressed in somatic cells at low levels and show for the first time that stau mRNAs are expressed at high levels in both male and female germ cells in the mouse.

Materials and methods Preparation of M1 and M2 cDNAs Individual cDNAs were amplified from the M1 and M2 plasmids (Mouse Staufen ESTs, mouse stau and mouse stau 2 respectively, Figure 1A) by polymerase chian reaction (PCR) with primers corresponding to the SK and T7 sites on the Bluescript vector (94°C 0.5 min, 48°C 0.5 min, 72°C 1.5 min, 30 cycles; Saiki et al., 1988). Amplified products were separated on agarose gels (1% w/v) containing ethidium bromide, run in TAE (Tris-acetate) buffer according to standard protocols (Davis et al., 1986) and the size of the cDNA inserts determined by comparison to makers run in a parallel lane. Selected PCR reactions (50 µl) were purified by spin column chromatography on TE (Tris-EDTA) 100 columns (Clontech, Cambridge, UK) and radiolabelled with [32P]-labelled dCTP, by random priming (Rediprime II kit; Pharmacia Amersham Biotech Ltd, St Albans, UK). Screening of cDNA libraries An 11 day mouse 5⬘ stretched λgt11 library was screened. This library was constructed by Clontech for the MRC Human Genetics Unit, Edinburgh, UK. The RNA source was normal, whole embryos pooled from Swiss Webster/NIH mice. The library was screened with M2 EST as a probe following Clontech’s protocol for screening λgt11 libraries. Northern blot hybridization Total RNA was extracted from testes, ovaries and kidneys obtained fresh from adult mice and rats using Tri reagent (Sigma-Aldrich Co. Ltd, Poole, Dorset, UK), according to the manufacturer’s instructions. Samples were dissolved in RNase-free water containing 0.01% sodium

984

dodecyl sulphate (SDS; Sigma) and the concentration of RNA determined by absorbance at 260 nm using a Genequant spectrophotometer (Pharmacia Amersham Biotech Ltd). RNA was stored at –70°C. Total RNA (20 µg/lane) was separated on denaturing agarose gels according to standard methods (Saunders et al., 1992), transferred by capillary blotting to Bright Star membranes (AMS Biotechnology Europe Ltd, Abingdon, Oxon, UK) using 20⫻ sodium chloride/ sodium itrate (SSC) and fixed by UV light. Membranes were prehybridized for a minimum of 2 h in 0.2 mol/l phosphate buffer pH 7.2, containing 1% (w/v) bovine serum albumin (BSA), 7% (w/v) SDS, and 15% (v/v) deionized formamide at 60°C. Radio-labelled M1 or M2 cDNAs (0.5–1⫻106 cpm/ml), prepared as above, were added and hybridization continued for 16–24 h. Membranes were washed with 40 mmol/l phosphate buffer containing 1% SDS at 65°C, exposed to a phosphorimager screen for 2 days and then visualized using the imager software (Molecular Dynamics, Chesham, Bucks, UK). Membranes were stripped (0.1% SDS) and reprobed with a labelled oligonucleotide specific for 18S ribosomal RNA (Saunders et al., 1992). In-situ hybridization Testes and ovaries from immature (day 16) and adult mice were immersion fixed in Bouin’s fluid for 6 h and then processed into paraffin wax according to standard methods (Millar et al., 1993). Sense and antisense cRNA transcipts were prepared from linearized Bluescript M1 and M2 plasmids by incubation with T3 or T7 RNA polymerases in the presence of [35S]-labelled UTP (PharmaciaAmersham) according to methods described in detail elsewhere (Maguire et al., 1992, 1997). The hybridization was carried out overnight at 50°C and the hybridization buffer contained the following constituents – 50% formamide, 4 ⫻ sodium chloride – Tris-EDTA, 1 ⫻ Denhardts, 125 µg/ml salmon testes DNA, 125 µg/ml yeast tRNA, 10 mM DTT, 10% dextran sulphate. After hybridization, slides were dipped in undiluted NBT-2 photographic emulsion (Kodak, Rochester, NY, USA) and exposed in a light tight box at 4°C for 16– 33 days. After processing, sections were viewed under light and dark field illumination on an Olympus Provis microcope (Olympus Optical Co, London, UK) equipped with a Kodak DC420 camera. Captured images were stored on a Mackintosh G3 computer and montages were assembled using Photoshop 5.0 (Adobe, Mountain View, CA, USA).

Results Identification of mouse homologues of Dm Stau At the time of the initial database search there were 14 mammalian EST sequences encoding proteins related to Dm Stau; eight in human and six in mice deposited in the database. Potential mouse staufen (m stau) homologues were identified by carrying out a BLAST search with the complete Dm stau sequence. Confirmation that these sequences were Dm stau-like was obtained by independently analysing sequence homology using GCG 10 http://www.hgmp.mic.ac.uk. Two mouse ESTs were selected for further study, M1 (Genbank accession no. AA104967) and M2 (Genbank accession no. AA106776). The sequence lodged for the M1 EST is 490 bp long, encodes a putative protein with 21% identity to Dm Stau, and corresponds to a region spanning from amino acids 356–574 on Dm Stau. The sequence for EST M2 was 401 bp long, and the encoded protein displayed 16% identity to Dm Stau. PCR amplification of the cDNA inserts of M1 and M2 containing plasmids revealed that they were ~1400 and 800 bp respectively, and

staufen expression in mouse gonads

therefore additional sequencing was undertaken. M2 was established to be 765 nucleotides long, the sequence reads through in frame 2 indicating that it represents part of an open reading frame (ORF), with neither the 5⬘ nor 3⬘ of the sequence being complete. This sequence is identical with a longer mouse EST more recently lodged in the database as mouse staufenlike (Genbank accession no. AJ244015). M2 also has two regions of high sequence similarity to the Dm stau, these coincide with RNA binding domain 4 (Figure 1A). M1 is 1376 nucleotides long, and appears to be a splice variant of m stau (Wickham et al., 1999) containing a very short 3⬘ UTR. This variant also lacks RNA binding domain 5. However, neither the m stau lodged in the database nor M1 appear to represent a complete cDNA as both read through from the start and appear to lack a 5⬘ sequence end. This may explain why alignment of the sequences with that of Dm Stau revealed only four conserved RNA binding domains, whereas Dm Stau has five such domains.

are likely to be two staufen genes in mammals was obtained by searching the human EST databases. A human homologue of Dm stau has been described (Wickham et al., 1999; Genbank accession no. AF061941) and other studies (Marion et al., 1999) and sequence comparisons have shown this to be homologous with mouse M1. A human homologue of m stau 2 was also submitted to the database (Genbank accession no. AK002152) as being similar to staufen. Interestingly, the two different human staufens described map to different chromosomal positions, H stau maps to chromosome 20 while H stau 2 is reported to map to chromosome 15. These findings therefore confirm that they are distinct genes, not differentially spliced forms of the same gene. The fact that two genes equivalent to the M1 and M2 cDNAs exist on different chromosomes in humans adds weight to our suggestion that these transcripts represent two different genes in mice. The chromosomal locations are unavailable for the mouse staufens.

Isolation of m stau cDNAs A good mouse ovary cDNA library was not available at the time of the experiments therefore, screening was undertaken using: (i) a mouse kidney cDNA library (obtained from Jonathan Bard, Department of Anatomy, University of Edinburgh, Edinburgh, UK) and (ii) an 11 day mouse embryo cDNA library (obtained from the Western General Hospital, Edinburgh, UK). cDNA amplified from plasmid M1 was used for screening both because it was the longer cDNA but also because at the time of screening it was shown to have the highest sequence homology to Dm stau. Several clones with insert sizes ranging from 1 to 3 kb (not shown) were obtained from the 11 day embryo library. Sequencing of three cDNAs amplified from purified lambda confirmed that they contained sequences identical to the M1 EST at their 5⬘ ends (Figure 1A). Subsequent comparison to sequences in Genbank revealed that the longest cDNA obtained had an identical sequence to the m stau (Genbank accession no. NM011490), submitted to the database in 03/99 (Wickham et al., 1999). Therefore, the data are not presented again in this paper. Sequence alignment of the individual RNA binding domains show that domains 2, 3, 4 and 5 are extremely well conserved between Drosophila and mouse (Figure 1B). Domains 1, 3 and 4 in Drosophila are very similar (St Johnston et al., 1992) and this is maintained in the mouse. Sequence alignments suggest that the M1 and M2 cDNAs represent different Dm stau-like genes. This finding is based on sequence comparisons between the two mouse staufens and Dm stau. For example, at the amino acid level, good alignments can be made between M1, M2 and Dm stau in the region of the RNA binding domain 4 (Figure 1B) with ⬎80% identity at the protein level. However, at the DNA level M1 is only 50% identical to M2 over this domain, which makes the differences between M1 and M2 as great as the differences between mammals and flies. The DNA sequence differences detected are much greater than could be explained by polymorphisms in the DNA, and since this carries an homologous RNA binding domain 4 (Figure 1B), it is unlikely to be a splicing variant at the DNA level. Further evidence that there

Northern analysis Northern blot analysis was undertaken with full length cDNAs amplified from plasmids M1 and M2, on replica samples of RNA isolated from testis, ovaries and kidney from adult mouse and rat (Figure 2). An mRNA of ~3 kb was detected in the gonads from both species, and the transcript was less abundant in adult kidney (Figure 2, lanes 1 and 7). The hybridization signal using the M1 cDNA was always more intense than that observed with the M2 plasmid. The transcript detected in rat gonads using M1 cDNA appeared consistently larger than that detected in mouse gonads (Figure 2, lanes 5, 6). Cellular sites of expression of m stau in mouse gonads Analysis of the pattern of expression of m stau using cRNA prepared from plasmid M1 revealed that silver grains were concentrated over germ cells (Figures 3 and 4). In the mouse ovary mRNA was detected in oocytes within follicles containing a single layer of granulosa cells, e.g. in ovaries on day 8 (Figure 3a,b), and was maintained as the follicle developed to contain several layers of granulosa cells (e.g. day 15, Figure 3c,d) and to form an antrum (adult, Figure 3e arrowed A). In the mouse testis stau mRNA was also localized to germ cells and this expression was dependent upon their progression through the process of meiosis in adults. No significant expression of mRNA was noted on day 16 (Figure 4a), although some tubules at this age contained germ cells up to and including (asterisks) early pachytene spermatocytes, equivalent to those seen in stages in the first half of the spermatogenic cycle in the adult (Oakberk, 1956). In the adult mouse testis, the pattern of expression was stage dependent and silver grains were localized to those pachytene spermatocytes present in the stages VI–XII of the cycle (Oakberk, 1956) with highest levels seen at stages VII and VIII. A similar pattern of expression was noted in testes using the cRNA prepared from the M2 plasmid (m stau 2); however the numbers of silver grains were much reduced when compared with identical samples hybridized to the m stau cRNA (not shown). 985

P.Saunders et al.

Figure 1. (A) Schematic sequence alignment of M1 and M2 plasmid cDNAs with staufen cDNA from Drosophila and two mouse staufenlike genes lodged in Genbank. Expressed sequence tags (ESTs), M1 (AA104976) encoded 1023 bp, corresponding to the open reading frame (ORF) of mouse Stau (Genbank accession no. NM011490; Wickham et al., 1999). Stau (*) has a 600 bp 3⬘ untranslated region (UTR), while EST M1 has a short 3⬘ UTR of 100 bp identical to the 3⬘ end of the m stau 3⬘ UTR. EST M2 encoded 767 bp corresponding to the translated region of mouse Stau 2 (Genbank accession no. AJ244015). The figure illustrates the regions of strong homology represented by the Dm stau binding sites. The UTRs are also marked on the figure. Although there is a potential start methionine near the beginning of m stau transcript, there is no indication that this is a real start methionine as there is no consensus Kozak sequence or stop codon upstream (Kozak, 1996). m stau 2 is also incomplete; missing its 5⬘ end. (B) Shows the alignment of Dm Stau RNA binding domains with corresponding regions of two mouse proteins described in this paper. This alignment was carried out using GCG10 (pileup) and Box Shade software. Staufen domain 2 alignment shows the alignment of Dm Stau binding domain 2 (DstauD2 amino acids 490–559) with mouse Stau 1 binding domain 2 (MStauLD2 aa111–183) showing a high degree of conservation between them. Staufen domains 1, 3 and 4 alignment shows the alignment of three Dm stau RNA binding domains (DStauD1 amino acids 308–380, DStauD3 amino acids 575–647 and DStauD4 amino acids 708–782) with the corresponding RNA binding domains observed in mouse Stau (MStauLD3 amino acids 201– 273 and MStauLD4 amino acids 301–374) and the region of mouse Stau 2-like currently available, containing RNA binding domain 4(M2StauD4 aa 50–124). St Johnston et al. (1992) illustrated that Drosophila RNA binding domains 1, 3 and 4 were similar to each other. A high degree of conservation is observed between Dm Stau and the two mouse Staufens. The highest similarity is observed between MStauLD4 and M2StauD4. Staufen domain 5 alignment illustrates the alignment between Dm Stau domain 5 (DStauD5 amino acids 949– 1020) and mouse Stau, domain 5 (MLStauD5 amino acids 503–576). Earlier studies had showed that Dm Stau domains 2 and 5 are more closely related to each other than to domains 1, 3 and 4 (St Johnston et al., 1992). The very close similarity between Dm Stau and mouse Stau for each of these domains makes it clear that they are quite distinct and that the mouse/Drosophila similarities within a domain are much greater than the similarity between domains 2 and 5 within a species.

986

staufen expression in mouse gonads

Figure 2. Northern analysis of expression of stau mRNA in mouse and rat tissues. Total RNA (20 µg) per lane from adult animals was loaded as follows: mouse kidney, 1,7; mouse testis, 2, 8; mouse ovary, 3, 4, 9, 10; rat testis, 5, 11; rat ovary 6, 12. (A) Lanes 1–6 were hybridized to M1 cDNA and lanes 7–12 with M2 cDNA, and after washing membranes were exposed to phosphorimager screens for 5 days. (B) Membranes were stripped and reprobed with an oligonucleotide specific for 18S ribosomal RNA to confirm successful loading and transfer of RNA, with exposure to imager screens for 16 h.

Discussion The preliminary expression data presented here suggest a role for the RNA binding protein, Staufen, in both oogenesis and spermatogenesis in rodents. Whilst these studies were underway papers describing the cloning of the human and mouse homologues of Dm stau were published (Marion et al., 1999; Wickham et al., 1999). Staufen homologues have also been identified in several other species including rat (Genbank accession no. AJ010200) and Caenorhabditis elegans (Genbank accession no. U67949, see Wickham et al., 1999). Sequence comparisons show that similarities between the Staufen proteins identified are almost exclusively confined to the RNA binding domains (Micklem et al., 2000). It is notable that human Staufen has been shown to have several splice variants and have been identified in multiple human tissues using Northern blots, however one group (Wickham et al., 1999) did not include any gonadal RNA in their tissue screen. Antibodies against human Staufen have been used to immunolocalize the protein to rat hippocampal neurones (Kiebler et al., 1999; Marion et al., 1999) and have shown that it is enriched in the vicinity of smooth endoplasmic reticulum and microtubules near synaptic contacts. In living hippocampal neurones Staufen-green fluorescent fusion protein was found to associate with granules containing RNA and to move through the cell in a microtubule-dependent manner (Kohrmann et al., 1999). In addition to an association with microtubules, cell transfection experiments revealed that Stau also localized to the rough endoplasmic reticulum and may, therefore, target mRNAs to their site of translation (Wickham et al., 1999). It will be exciting in the future to use antibodies to determine the cellular location of mouse Staufens in the oocyte and spermatocytes. Of particular interest is the possibility that there is a splice

variant of m stau, called M1, which lacks RNA binding domain 5. This domain has been shown to be involved in RNA localization. This means that there could be Staufen proteins which repress translation without localizing the RNA. Several RNA binding proteins, conserved in many species, from Drosophila to human, have already been shown to be important for mammalian fertility. A role for RNA binding proteins in the process of gametogenesis is to be expected, as translational control represents a key mechanism for gene regulation in germ cell differentiation and early embryogenesis. Perturbations in this process can therefore have a major impact on normal development. For example, deletion of the RNA binding protein dazl, in mice results in infertility in both males and females (Ruggiu et al., 1997), consistent with expression of dazl in germ cells in both the ovary and testis in mice (Ruggiu et al., 1997, 2000) and human (Seligman and Page, 1998). Homologues of dazl have been identified in a wide range of species including Xenopus (xDazl; Houston et al., 1998), zebra fish (zDazl; Maegawa et al., 1999) and Drosophila (boule; Eberhart et al., 1996). It is notable that whereas deletion of dazl in mice results in infertility of both males and females (Ruggiu et al., 1997), deletion of boule in Drosophila is associated with infertility in males alone (Eberhart et al., 1996) and in Caenorhabditis elegans loss of dazl function has no affect on sperm production, but is associated with a meiotic block during oogenesis (Karashima et al., 2000). These findings are an interesting parallel to those of the present study in which staufen transcripts have been identified in the germ cells of both male and female mice, but appear to play a role in only oogenesis in Drosophila. In Drosophila spermatogenesis, translational control is mediated via translational control elements (TCE) located in the 5⬘ UTRs of testis RNA (Kempe et al., 1993; Schafer et al., 1993, 1995). Y-box proteins are also believed to function as translational repressors in germ cells. Mammalian homologues of the Y-box proteins FRGY-1 and FRGY-2, first identified in Xenopus oocytes (Murray et al., 1991), are both expressed in the testis (Kwon et al., 1993). MSY2, the FRGY-2 homologue is expressed in both male and female germ cells and is inherited in the early embryo (Gu et al., 1998). In the testis, expression occurs in pachytene spermatocytes but was maximal in post-meiotic round spermatids, the cell type which contains abundant stored messenger ribonuclear proteins (Gu et al., 1998). In the ovaries, MSY-2 was expressed in oocytes which are arrested at diplotene within follicles, an identical pattern of expression to that observed using the staufen mRNA. Many transcripts have been found to have testis-specific 5⬘ and 3⬘ UTRs, which are likely to be crucial in their translational regulation and are probably recognized by specific sets of RNA binding proteins. There is also a tendency for the translationally-controlled testis transcripts to have shortened poly A tails (see review by Hecht, 1998). One of the proteins that operates on 3⬘ UTRs is testis brain–RNA binding protein (TB-RBP). Interestingly this localizes to the nucleus in pachytene spermatocytes during meiosis but to the cytoplasm of the round spermatids after meiosis. Additional RNA binding proteins, not known to be represented in species other than mammals, have been identified as 987

P.Saunders et al.

Figure 3. (A) In-situ analysis of site of expression of m stau in mouse ovaries indicates that expression is specific to the developing oocyte. Sense and antisense cRNAs prepared from the m stau M1 cDNA were hybridized to sections obtained from mouse ovaries on post natal days 8 (a, b), 9 (not shown), 15 (c, d) and adults (e, f). Dark field (a, c, e) and light field (b, d, f) of the same sections are shown. Silver grains were localized over oocytes (arrows) at all ages examined consistent with expression of m stau in oocytes within preantral follicles containing only one or two layers of granulosa cells (day 8, a, b; stage 2) and persisting into oocytes from antral follicles (e.g. e, arrowed A; stage 5/6). There was no specific signal using the sense probe on any section examined (inset c). All original magnifications ⫻20. (B) Summary of the levels of stau mRNA in relation to morphogenesis of the follicle. The stages of mouse oogenesis are shown at the top and diagramatically beneath. There is a period of oocyte growth during stages 2–5. Thecal layer formation begins in stage 4 and from stage 6 the follicle size increases dramatically and the antral cavity is formed. staufen mRNA accumulates in the oocyte during the period of oocyte growth and remains at this level until the oocyte matures.

important in the storage and translational repression of the mRNAs encoding the protamines. These are highly basic proteins that replace the transition proteins used for DNA 988

packaging with spermatozoa (Braun, 1998; Hecht, 1998). Of particular interest is the protein found in the spermatids called perinuclear RNA-binding protein (SPNR). SPNR is a

staufen expression in mouse gonads

Figure 4. Expression of m stau in testis is germ cell specific and stage dependent. In-situ analysis using sense and antisense cRNA prepared from M2 revealed that expression of m stau in the mouse testis is germ cell specific; dark field (a, c, e) and light field (b, d, f) views are shown from the same section. No specific hybridization signal was detected on sections of testes from day 15 mice, although these contained somatic Sertoli cells and germ cells up to and including early pachytene spermatocytes (asterisks in a and b). In adult animals the localization of the silver grains was consistent with expression of staufen mRNA in spermatocytes present in tubules during the second half of the cycle (stages VI–XII; arrows c, e). (B) Summary of the levels of staufen mRNA in relation to morphogenesis of the spermatozoon. The stages of spermatogenesis are shown, staufen mRNA increases and peaks in primary spermatocytes during the first meiotic division when there are high levels of transcription. Transcripts then degrade in secondary spermatocytes and have disappeared when the sperm cell differentiates. This suggests a function for staufen in primary spermatocytes when there is active transcription and translational represssion, but some of the transcripts will not be translated until the sperm begins to differentiate and the staufen mRNA has degraded. It is possible that Staufen protein is still present after the mRNA degrades.

microtubule associated protein which binds the 3⬘ UTR of Protamine 1 mRNAs, suggesting a role in subcellular localization as well as translational repression and storage prior to activation of translation (Schumacher et al., 1998). It is notable that in common with Drosophila Staufen, one of these has been

identified as reponsible for the localization of translationallyrepressed mRNA to microtubules (Han et al., 1995). In mouse testes, expression of staufen mRNA within pachytene spermatocytes is consistent with a role in translational repression of some germ cell encoded mRNAs. Exactly when Staufen 989

P.Saunders et al.

protein is present remains to be established, as repression by a stable Staufen protein could easily continue well beyond the stage when the mRNA is degraded. Further work will be needed to identify specific mRNA targets. In conclusion, expression of staufen mRNA in mammalian germ cells extends the number of RNA binding proteins which may act as key regulators of fertility in a wide range of animal species.

Acknowledgements The authors thank Mike Millar and Joseph Gaughan (HRSU) for assistance with the in-situ studies and sequence analysis respectively; the MRC Human Genetics Unit for λgt11 embryo library; and Sheila Milne for preparing the manuscript.

References Braude, P., Bolton, V. and Moore, S. (1988) Human gene expression first occurs between the four- and eight-cell stages of preimplantation development. Nature, 332, 459–461. Braun, R.E., Peschon, J.J., Behringer, R.R., et al. (1989) Protamine 3⬘untranslated sequences regulate temporal translational control and subcellular localizaton of growth hormone in spermatids of transgenic mice. Genes Dev., 3, 793–802. Braun, R.E. (1998) Post-transcriptional control of gene expression during spermatogenesis. Semin. Cell Dev. Biol., 9, 483–489. Broadus, J., Fuerstenberg, S. and Doe, C.Q. (1998) Staufen-dependent localization of prospero mRNA contributes to neuroblast daughter-cell fate. Nature, 391, 792–795. Davis, L.G., Dibner, M.D. and Battey, J.F. (1986) Basic Methods in Molecular Biology. Elsevier Inc, New York, USA. Eberhart, C.G., Maines, J.Z. and Wasserman, S.A. (1996) Meiotic cell cycle requirement for a fly homologue of human Deleted in Azoospermia. Nature, 381, 783–785. Elliott, D.J. and Cooke, H.J. (1997) The molecular genetics of male infertility. Bioessays, 19, 801–809. Ferrandon, D., Elphick, L., Nusslein Volhard, C. et al. (1994) Staufen protein associates with the 3⬘UTR of bicoid mRNA to form particles that move in a microtubule-dependent manner. Cell, 79, 1221–1232. Flach, G., Johnson, M.H., Braude, P.R. et al. (1982) The transition from maternal to embryonic control in the 2-cell mouse embryo. EMBO J., 1, 681–686. Gonzalez-Reyes, A., Elliott, H. and St Johnston, D. (1995) Polarization of both major body axes in Drosophila by gurken-torpedo signalling. Nature, 375, 654–658. Gu, W., Tekur, S., Reinbold, R. et al. (1998) Mammalian male and female germ cells express a germ cell-specific Y-protein, MSY2. Biol. Reprod., 59, 1266–1274. Han, J.R., Yui, G.K. and Hecht, N.B. (1995) Testis/brain RNA-binding protein attaches translationally repressed and transported mRNAs to microtubules. Proc. Natl Acad. Sci. USA, 92, 9550–9554. Harper, M.I. and Monk, M. (1983) Evidence for translation of HPRT enzyme on maternal messenger-RNA in early mouse embryos. J. Embryology and Experimental Morphology, 74, 15–28. Hecht, N.B. (1998) Molecular mechanisms of male germ cell differentiation. Bioessays, 20, 555–561. Houston, D.W., Zhang, J., Maines, J.Z. et al. (1998) A Xenopus DAZ-like gene encodes an RNA component of germ plasm and is a functional homologue of Drosophila boule. Development, 125, 171–180. Hurst, S., Talbot, N.J. and Stebbings, H. (1999) A staufen-like RNA-binding protein in translocation channels linking nurse cells to oocytes in Notonecta shows nucleotide-dependent attachment to microtubules. J. Cell, Sci., 112, 2947–2955. Karashima, T., Sugimoto, A. and Yamamoto, M. (2000) Caenorhabditis elegans homologue of the human azoospermia factor DAZ is required for oogenesis but not for spermatogenesis. Development, 127, 1069–1079. Kempe, E., Muhs, B. and Schafer, M. (1993) Gene regulation in Drosophila spermatogenesis: analysis of protein binding at the translational control element TCE. Dev. Genet., 14, 449–459. Kiebler, M.A., Hemraj, I.A., Verkade, P. et al. (1999) The mammalian staufen

990

protein localizes to the somatodendritic domain of cultured hippocampal neurons: Implications for its involvement in mRNA transport. J. Neurosci., 19, 288–297. Kim-Ha, J., Smith, J.L. and Macdonald, P.M. (1991) oskar mRNA is localized to the posterior pole of the Drosophila oocyte. Cell, 66, 23–35. Kohrmann, M., Luo, M., Kaether, C. et al. (1999) Microtubule-dependent recruitment of Staufen-green fluorescent protein into large RNA-containing granules and subsequent dendritic transport in living hippocampal neurons. Mol. Biol. Cell,, 10, 2945–2953. Kozak, M. (1996) Interpreting cDNA sequences: some insights from studies on translation. Mamm. Genome, 7, 563–574. Kwon, Y. and Hecht, N. (1991) Cytoplasmic protein binding to highly conserved sequences in the 3⬘ untranslated region of mouse protamine 2 mRNA, a translationally regulated gene of male germ cells. Proc. Natl Acad. Sci. USA, 88, 3584–3588. Kwon, Y. and Hecht, N. (1993) Binding of a phosphoprotein to the 3⬘ untranslated region of the mouse protamine 2 mRNA temporally represses its translation. Mol. Cell,. Biol., 13, 6547–6557. Kwon, Y.K., Murray, M.T. and Hecht, N.B. (1993) Proteins homologous to the Xenopus germ cell-specific RNA-binding proteins p54/p56 are temporally expressed in mouse germ cells. Dev. Biol., 158, 90–100. Latham, K.E., Garrels, J.I., Chang, C. et al. (1991) Quantitative-analysis of protein-synthesis in mouse embryos. 1. Extensive reprogramming at the one-cell and 2-cell stages. Development, 112, 921–932. Maegawa, S., Yasuda, K. and Inoue, K. (1999) Maternal mRNA localization of zebrafish DAZ-like gene. Mech. Dev., 81, 223–226. Maguire, S.M., Millar, M., Sharpe, R.M. et al. (1992) Stage-dependent expression of mRNA for cyclic protein-2 during spermatogenesis is modulated by elongate spermatids. Mol. Cell. Endocrinol., 94, 79–88. Maguire, S.M., Millar, M.R., Sharpe, R.M. et al. (1997). Investigation of the potential role of the germ cell complement in control of the expression of transferrin mRNA in the prepubertal and adult rat testis. J. Mol. Endocrinol., 19, 67–77. Marion, R.M., Fortes, P., Beloso, A. et al. (1999) A human sequence homologue of Staufen is an RNA-binding protein that is associated with polysomes and localizes to the rough endoplasmic reticulum. Mol. Cell. Biol., 19, 2212–2219. Micklem, D.R., Adams, J., Grunert, S. et al. (2000) Distinct roles of two conserved Staufen domains in oskar mRNA localization and translation. EMBO J., 1366–1377. Millar, M.R., Sharpe, R.M., Maguire, S.M. et al. (1993) Cellular localisation of messenger RNAs in rat testis: application of digoxigenin labelled probes to embedded tissue. Cell Tissue Res., 273, 269–277. Murray, M.T., Krohne, G. and Franke, W.W. (1991) Different forms of soluble cytoplasmic mRNA binding proteins and particles in Xenopus laevis oocytes and embryos. J. Cell Biol., 112, 1–11. Oakberk, E.F. (1956) Duration of spermatogenesis in the mouse and timing of stages of the cycle of the seminiferous epithelium. American Journal of Anatomy, 99, 507–516. Ramos, A., Grunert, S., Adams, J. et al. (2000) RNA recognition by a Staufen double-stranded RNA-binding domain. EMBO J., 19, 997–1009. Ruggiu, M., Speed, R., Taggart, M. et al. (1997) The mouse Dazla gene encodes a cytoplasmic protein essential for gametogenesis. Nature, 389, 73–77. Ruggiu, M., Saunders, P.T. and Cooke, H.J. (2000) Dynamic subcellular distribution of the DAZL protein is confined to primate male germ cells. J. Androl., 21, 470–477. Saiki, R.K., Gelfand, D.H., Stoffel, S. et al. (1988). Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science, 239, 224–230. Saunders, P.T.K., Millar, M.R., Maguire, S.M. et al. (1992) Stage-specific expression of rat transition protein-2 mRNA and possible localization to the chromatoid body of step 7 spermatids by in situ hybridization using a nonradioactive riboprobe. Mol. Reprod. Dev., 33, 385–391. Schafer, M., Borsch, D., Hulster, A. et al. (1993) Expression of a gene duplication encoding conserved sperm tail proteins is translationally regulated in Drosophila melanogaster. Mol. Cell, Biol., 13, 1708–1718. Schafer, M., Nayernia, K., Engel, W. et al. (1995) Translational control in spermatogenesis. Dev. Biol., 172, 344–352. Schu¨ pbach, T. and Wieschaus, E. (1986) Germline autonomy of maternaleffect mutations altering the embryonic body pattern of Drosophila. Dev. Biol., 113, 443–448. Schumacher, J.M., Artzt, K. and Braun, R.E. (1998) Spermatid perinuclear ribonucleic acid-binding protein binds microtubules in vitro and associates with abnormal machettes in vivo in mice. Biol. Reprod., 59, 69–76.

staufen expression in mouse gonads Seligman, J. and Page, D.C. (1998) The Dazh gene is expressed in male and female embryonic gonads before germ cell sex differentiation. Biochem. Biophys Res. Commun., 245, 878–882. St Johnston, D., Beuchle, D. and Nusslein Volhard, C. (1991) staufen, a gene required to localize maternal RNAs in the Drosophila egg. Cell, 66, 51–63. St Johnston, D., Brown, N.H., Gall, J.G. et al. (1992) A conserved doublestranded RNA-binding domain. Proc. Natl Acad. Sci. USA, 89, 10979– 10983. Ward, S., Hogan, E. and Nelson, G.A. (1983) The initiation of spermiogenesis in the nematode Caenorhabditis elegans. Dev. Biol., 98, 70–79. Wang, Q.X. and Latham, K.E. (1997) Requirement for protein synthesis during embryonic genome activation in mice. Mol. Reprod. Dev., 47, 265–270. Wickham, L., Duchaine, T., Luo, M. et al. (1999) Mammalian staufen is a double-stranded-RNA- and tubulin-binding protein which localizes to the rough endoplasmic reticulum. Mol. Cell, Biol., 19, 2220–2230. Received on June 5, 2000; accepted on August 24, 2000

991