Conservation of a Pumilio-Nanos complex from Drosophila germ plasm to human germ cells

Dev Genes Evol (2003) 213:120–126 DOI 10.1007/s00427-003-0303-2

ORIGINAL ARTICLE

Jadwiga Jaruzelska · Maciej Kotecki · Kamila Kusz · Anna Spik · Meri Firpo · Renee A. Reijo Pera

Conservation of a Pumilio-Nanos complex from Drosophila germ plasm to human germ cells Received: 4 October 2002 / Accepted: 19 December 2002 / Published online: 11 February 2003
Springer-Verlag 2003

Abstract Germ cells are the cells which ultimately give rise to mature sperm and eggs. In model organisms such as flies and worms, several genes that are required for formation and maintenance of germ cells have been identified and their interactions are rapidly being delineated. By contrast, little is known of the genes required for development of human germ cells and it is not clear whether findings from model organisms will translate into knowledge of human germ cell development, especially given observations that reproductive pathways may evolve more rapidly than somatic pathways. The Pumilio and Nanos genes have been especially well-characterized in model organisms and encode proteins that interact and are required for development of germ stem cells in one or both sexes. Here we report the first characterization of a mammalian Nanos homolog, human NANOS1 (NOS1). We show that human NOS1 protein interacts with the human PUMILIO-2 (PUM2) protein via highly conserved domains to form a stable complex. We also show that in men, the NOS1 and PUM2 proteins are particularly abundant in germline stem cells. These observations mirror those in distant species and document for the first time a conserved protein-protein interaction in germ cells from flies to humans. These results suggest the possibility that the interaction of PUM2 and NOS1 may play a conserved role in germ cell development and maintenance in humans as in model organisms. Edited by D.A. Weisblat J. Jaruzelska · M. Firpo · R. A. Reijo Pera ()) Center for Reproductive Sciences, Department of Obstetrics, Gynecology and Reproductive Sciences, Departments of Physiology and Urology and Programs in Human Genetics and Cancer Genetics, University of California at San Francisco, San Francisco, CA 94143-0546, USA e-mail: [email protected] Tel.: +1-415-4763178 Fax: +1-415-4763121 J. Jaruzelska · M. Kotecki · K. Kusz · A. Spik Institute of Human Genetics, Polish Academy of Sciences, Poznan, Poland

Keywords Nanos · Pumilio · Germ cell formation · Germ plasm · Germline expression

Introduction Pluripotent germ cells give rise to mature sperm and eggs and transmit genetic information from one generation to the next. Two broad strategies exist in metazoans to specify the germ cell lineage. In some organisms, such as flies, worms and frogs, germ plasm is formed in the oocyte and segregates with the cells destined to give rise to the germ cell lineage (Rongo et al. 1997). In others, germ plasm has not been observed and instead extrinsic signaling is thought to specify the germ cell lineage during embryonic development. Most notably, signaling by molecules of the Bone Morphogenetic Protein (BMP) family are implicated in formation and maintenance of primordial germ cells in mice (Lawson et al. 1999; Ying et al. 2001) and possibly, by extension, in humans. A number of genes required for germ cell formation and maintenance have been identified in model organisms such as the fruit fly, Drosophila, and have been shown to have counterparts in distant organisms (Wylie 2000). The Pumilio and Nanos genes were first described as genes required for differentiation of the anterior-posterior body axis in the Drosophila embryo (Lehmann and NusslenVolhard 1987; Wang and Lehmann 1991; Curtis et al. 1995; Murata and Wharton 1995). In flies, the proteins interact indirectly on hunchback RNA on the NRE (Nanos Responsive Element) sequences to regulate translation of the hunchback message (Murata and Wharton 1995; Wreden et al. 1997; Sonoda and Wharton 1999). More recently, the genes were shown to encode components of germ plasm and to be required for migration of primordial germ cells and later for oogenesis and spermatogenesis (Kobayashi et al. 1996; Lin and Spradling 1997; Bhat 1999; Deshpande et al. 1999; Parisi and Lin 1999). Both proteins contain RNA-binding domains and, where known, act to repress specific mRNAs in patterning, and in germ cell formation, migration, and differentiation

121

(Curtis et al. 1995; Murata and Wharton 1995; Lin and Spradling 1997; Wreden et al. 1997; Asaoka et al. 1998; Forbes and Lehmann 1998; Asaoka-Taguchi et al. 1999; Bhat 1999; Parisi and Lin 1999; Sonoda and Wharton 1999; Nakahata et al. 2001; Crittenden et al. 2002). Nanos homologs have also been identified in distant organisms, such as Hydra where the protein is expressed only in multipotent stem cells and germline stem cells (Mochizuki et al. 2000), in frogs where it is a component of germ plasm and interacts directly with Pumilio (Mosquera et al. 1993; Zhou and King 1996; Nakahata et al. 2001), and in zebrafish where it is essential for primordial germ cell development and migration (Koprunner et al. 2001). In this work, we sought to address whether the PumilioNanos complex is conserved from flies to humans. Our recent work reported that humans possess two PUMILIO genes and that PUMILIO-2 (PUM2) is predominantly expressed in embryonic and germline stem cells of the testis and ovary (Moore et al. 2003).

Materials and methods Cloning and expression analysis To identify the complete human NOS1 gene, a 1.2-kb insert of human EST (GenBank accession number AI 138906) was used to screen a human adult testis cDNA library (Clontech, Palo Alto, Calif.). Clones that hybridized to the EST were sequenced and used for northern blot analysis as described (Moore et al. 2003). For RTPCR, tissues were homogenized in Trizol reagent according to the manufacturer’s instructions (Gibco-BRL, Bethesda, Md.). Then 1 g total RNA was reverse-transcribed using oligo-dT primers (Roche, Indianapolis, Ind.). Following reverse transcription, 100 ng cDNA was subjected to PCR with NOS1 primers: GCTCCTGGAACGACTACCTG and GTCGTCGTCCTCGTCGTAGT. Northern blots were obtained from Clontech (Palo Alto, Calif.). Fetal tissues from week 14–20 of development were obtained from ABR Resources (Oakland, Calif.). Human embryonic stem cells were obtained from the Stem Cell Laboratory, UCSF.

Deletion analysis of PUM2 and NOS1 Deletions of the PUM2 open reading frame have been reported (Moore et al. 2003). NOS1 deletions were constructed using primers designed to clone two fragments of the NOS1 cDNA into the pACT2 vector adjacent to DNA that encoded the GAL4 activation domain; constructs were designed to encode fusion proteins. For deletion analysis, PUM2 deletion constructs were transformed into Y190 yeast along with a construct that encoded full length NOS1. Likewise, for NOS1 deletions, deletion constructs were transformed into yeast along with a construct that encoded the full-length PUM2:GAL4 activation domain fusion construct. Transformation of constructs into yeast strains, selection of transformants and testing for interaction with the X-Gal assay was done according to instructions (Clontech, Palo Alto, Calif.). To identify proteins that interact with PUM2, PUM2 cDNA was cloned in fusion with the GAL4 DNA-binding domain (the construct encoded amino acids from 261 to 1,064). This construct was used to screen a human testis cDNA library as described in Moore et al. (2003). Immunohistochemistry NOS1 antisera were prepared using the peptide EATPREERAPAWAAEPR as an antigen in rabbits (Research Genetics, Huntsville, Ala.). Immunohistochemistry on human testis sections was as described in Reijo et al. (2000). Essentially, human testis sections were obtained from men with physical obstruction of spermatogenic tubules and normal spermatogenic patterns; testes sections were obtained following institutional review board approval. Formalin-fixed, paraffin-embedded human tissues were sectioned at 5 mm. After baking at 60C for 1 h, sections were deparaffinized and rehydrated (through a xylene, ethanol, water series). Slides were blocked by treating with 3% hydrogen peroxide in methanol for 5 min at room temperature, then washed in water and transferred to phosphate-buffered saline (PBS) with 0.1% Tween 20, and blocked in 10% goat serum. Slides were incubated for 1 h at room temperature with antisera against NOS1 (at 1:200 dilution), antisera to PUM2 (1:200) or preimmune sera (1:200). Slides were washed three times in PBS, 0.1% Tween 20 and then incubated with secondary anti-rabbit antibody conjugated with HRP (horseradish peroxidase) for 1 h. After three washes in PBS, 0.1% Tween 20, slides were incubated for 10 min at room temperature with DAB (3,3'-diaminobenzidine tetrachloride) substrate for the HRP reaction. Sections were counterstained with 2% Gill’s hematoxylin.

Co-immunoprecipitation Co-immunoprecipitation was done using total protein from primate COS7 cell extracts that expressed PUM2 and/or NOS1 proteins. For co-immunoprecipitation, the PUM2 and NOS1 cDNAs were cloned into the pBud vector under the control of CMV and EF-1a promoters to express fusion proteins with myc and V5 epitopes, respectively (Invitrogen, Carlsbad, Calif.). Cells were transfected using Fugene 6 reagent (Roche, Indianapolis, Ind.), grown for 48 h, and lysed in buffer containing 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1% NP-40, 0.5% deoxycholic acid, 0.1% SDS, 1 mM phenylmethylsulfonyl fluoride, 2 mM sodium vanadate, 20 mM sodium fluoride, and protease inhibitors as described (Moore et al. 2003). Cell protein lysates were incubated overnight at 4C with rabbit polyclonal anti-PUM2 antibody (Moore et al. 2003) or preimmune rabbit serum as a negative control followed by incubation with protein A/G beads (Santa Cruz Biochemicals, Santa Cruz, Calif.) for 4 h. Beads were centrifuged and proteins were eluted as described (Moore et al. 2003). Proteins were electrophoresed and western blotting was done using antibodies that detect NANOS1 protein with an anti-V5 antibody.

Results Identification of the NOS1 open reading frame and expression of mRNA A search of public sequence databases indicated that a short human EST encodes a peptide similar to the carboxyl-terminal amino acid sequences of Xcat-2, a Nanos homolog in frogs that localizes to germ plasm in that organism (Mosquera et al. 1993). We characterized the complete sequence of a human NANOS gene by screening a human testis cDNA library for clones that hybridized to this EST. The gene, NANOS-1 (NOS1), is an unusual human gene in that it contains just a single exon; it lacks introns. NOS1 contains a long 3' untranslated (UTR) region, a feature typical of many developmental genes (Fig. 1A). The protein encoded by NOS1 mRNA is approximately 32 kDa in size and contains a conserved

122

proteins found in widely divergent organisms from the parazoans, the porifera (Ephydatia) and hydra (Hydra magnipapillata) to other eumetazoans such as worms, frogs, zebrafish, and mice. The positions of all the cysteines and histidines of the zinc-finger structure are identical in Nanos homologs from humans and other organisms with the exception of the spacing of these residues in the worm (Fig. 1C). However, in contrast to the conserved zinc finger, the N-terminus of Nanos homologs in metazoans is divergent. To determine whether we had obtained the entire coding region of NOS-1 and to assess the tissues in which the NOS-1 gene might function, we analyzed expression of the gene. Northern blot analysis of mRNA from human adult tissues indicates that human NOS1 message is expressed exclusively in the testis and is approximately 2.0 kb in size (Fig. 1D). More sensitive approaches such as RT-PCR analysis indicated that NOS1 is expressed predominantly in testis in adults but is also expressed in embryonic stem cells, and in fetal testis and fetal ovary (Fig. 1E). Interaction of NOS1 with PUM2

Fig. 1A–E Cloning of NANOS-1. A The open reading frame of the human NOS1 cDNA is represented by light gray shading; the two CCHC zinc fingers motifs (1 and 2) are in dark gray shading. B The amino acid sequence of the human NOS1 protein; the zinc finger motifs are underlined. C The zinc finger domains of the Nanos proteins are conserved in divergent organisms. Accession numbers for Nanos homologs are as follows: human NOS1 (AAL36982) and NOS2 (XP_089800), Ephydatia (Ephydatia fluviatilis, sponge) Nos-1 (BAB19253), frog (Xenopus) Xcat2 (I51603), fruitfly (Drosophila melanogaster) Nos-1 (P25724), hydra (Hydra magnipapillata) Nos-1(BAB01491) and Nos-2 (BAB01492), leech (Helobdella robusta) Nos-1 (AAB63111), mouse (Mus musculus) Nos-1 (XP_146605) and Nos-2 (XP_140766), worm (Caenorhabditis elegans) Nos-1 (NP_496358) and Nos-2 (AAB93424), zebrafish (Danio rerio) Nos-1 (NP_571953). D Northern blot analysis indicates that the NOS1 cDNA is predominantly expressed in testis (NOS1, upper blot; B-actin cDNA, lower blot). E RT-PCR analysis of expression of NOS1 in somatic tissues from the human embryo, and fetal gonads and embryonic stem cells

RNA-binding domain in the carboxyl terminus (Fig. 1B). The RNA-binding domain consists of two (CysCysHisCys)2 zinc fingers of approximately 52 amino acids that share 62% amino acid identity with the corresponding domain of Drosophila Nanos (Wang and Lehmann 1991). Remarkably, NOS1 protein is also homologous to Nanos

We next sought to determine whether the NOS1 protein might interact with another conserved germ cell protein, PUM2, and form a stable protein complex. To address whether PUM2 and NOS1 can interact, primate COS7 cells were cotransfected with constructs that direct expression of the PUM2 and NOS1 proteins. Then, either PUM2 antisera or negative-control preimmune antisera were used to immunoprecipitate PUM2 protein, along with interacting proteins, from cell supernatants. Proteins were electrophoresed and probed by western blotting with anti-NOS1 antibodies (against a V-5 fusion tag, upper panel of Fig. 2) or with anti-PUM2 antibodies (Fig. 2 lower panel). Results indicate that NOS1 protein is coimmunoprecipitated with PUM2-specific antibodies (Fig. 2 upper panel). No co-immunoprecipitation of NOS1 protein is observed with negative-control preimmune sera (Fig. 2 upper panel). These results indicated that human PUM2 and NOS1 can form a stable complex in vivo. Interacting domains If the interaction of PUM2 and NOS1 is specific, than we expect specific domains of each protein to be required for interaction. Thus, we also used the yeast two-hybrid system to verify the interaction between NOS1 and PUM2 and define domains required for this interaction (Fig. 3). We mapped regions of NOS1 and PUM2 that are required for interaction by assaying a series of truncated PUM2 fusion proteins for their ability to interact with NOS1 and vice versa. We found that the PUM2 RNA-binding domain, along with 155 amino acids upstream, is necessary for interaction with NOS1; the N-terminus of

123

Fig. 2 PUM2 and NOS1 proteins form a stable complex. Primate COS cells were cotransfected with constructs that encode PUM2 cDNA and NOS1 cDNA. Proteins were separated by gel electrophoresis and transferred by western blotting. Anti-V5 antibodies detect NOS1 protein (upper blot) and anti-PUM2 antibodies detect PUM2 (bottom blot). When total protein from cellular extracts was analyzed, both NOS1 (left hand lane, upper panel) and PUM2 (left hand lane, lower panel) were detected. Neither NOS1 (middle lane, upper panel) or PUM2 (middle lane, lower panel) were detected in co-immunoprecipitates with negative control preimmune sera. In contrast, NOS1 protein is detected in co-immunoprecipitates with anti-PUM2 (upper blot, right hand lane), as is PUM2 protein (lower blot, right hand lane)

PUM2 is not required (Fig. 3A). Next, we tested whether the N-terminal or the conserved zinc finger domains of NOS1 were required to bind PUM2. We found that the zinc finger domain was capable of interacting with PUM2, whereas the N-terminus was not (Fig. 3B, C). These results indicate that NOS1 and PUM2 can interact, apparently in the absence of the target RNA they regulate. In other organisms, ability to interact may or may not be RNA-dependent. For example, in Drosophila, interaction of pumilio and nanos requires association with RNA (Sonoda and Wharton 1999) whereas, in Xenopus, Nanos and Pumilio proteins are able to interact in the absence of their RNA target (Nakahata et al. 2001). PUM2 dimerization In Drosophila, pumilio binds to two NRE motifs in the 3' UTR of the hunchback mRNA (Zamore et al. 1997, 1999; Sonoda and Wharton 1999). Experiments designed to examine the structure of the pumilio complex suggest that a single pumilio RNA-interacting domain binds a single Drosophila NRE as a monomer; no cooperative interactions between adjacent RNA-interacting domains bound to NRE motifs were detected (Zamore et al. 1997, 1999; Sonoda and Wharton 1999). To gain further insight into the structure of the PUM2-NOS1 complex in human testis, we used the yeast two-hybrid system. In a screen for proteins that interact with PUM2, we identified PUM2 itself; this suggested that PUM2 may form homodimers (Fig. 3D). Moreover, when we assayed interaction, via the yeast two-hybrid system, of two PUM2 constructs that had been previously generated and encoded either the Nor the C-terminal halves of the protein as a fusion protein

Fig. 3A–D Mapping of domains required for interaction of the PUM2 and NOS1 proteins in the yeast two-hybrid system. A Deletion analysis of the PUM2 protein defines the minimal domain of PUM2 required for interaction with NOS1. Nine deletions of the PUM2:GAL4 fusion protein were assayed for their ability to interact with intact NOS1 protein. Positions of amino acids on PUM2 constructs are in reference to GenBank accession no. AF272350. The minimal region required for interaction is indicated by the black bar below the construct diagrams. B Deletion analysis of the NOS1 construct indicates that the zinc finger domains are sufficient for binding to PUM2. The black bar indicates the minimal region identified for interaction. C The X-Gal test used to detect interaction of PUM2 with the NOS1 zinc-finger domain. The dark reaction product (left hand figure) indicates a strong (+) interaction in vivo in yeast. The negative control used was an unrelated protein (laminin) which does not interact with PUM2 (). D PUM2 may act as a homodimer. The PUM2-PUM2 homodimer is formed through interaction of N-terminal regions of PUM2 indicated by the black bar

with the GAL4 activation domain (Moore et al. 2003), results indicated that the formation of potential homodimers required the N-terminal amino acids 261–550 of PUM2 (Fig. 3D). This is a region distinct from that required for interaction of PUM2 and NOS1. Taken together, these results suggested that PUM2 may interact

124

Fig. 4A–E Comparison of immunohistochemistry of PUM2 and NOS1 proteins. A Staining of testis tissue with antisera to PUM2 protein. B Western blot demonstrates detection of NOS1 protein in human testis tissue with negative control (lane 1) and NOS-1 (lane 2) antisera. C Staining of testis tissue sections with NOS1 antisera. D Negative control preimmune antisera to PUM2 and E NOS1 suggest the PUM2 and NOS1 antisera are specific. Note that two other genes, PUM1 and NOS2, have also been identified. The

antibodies used here are specific to PUM2 and NOS1. As demonstrated recently (Moore et al. 2003), PUM1 is expressed ubiquitously; expression of NOS2 has not been reported. Single spermatogonia and spermatocytes positive for NOS1 are shown in insets for better visualization of subcellular localization (Sg spermatogonia, Spc spermatocyte, Spd spermatid, SC sertoli cell). PUM2 antisera were as described in Moore et al. (2003)

specifically with other PUM2 monomers and exist in a NOS1-PUM2/PUM2-NOS1 mRNA complex.

abundant in spermatogonia, the stem cells of the germline (Fig. 4C). Like PUM2, NOS1 was also expressed during meiosis in spermatocytes (Fig. 4C). Both PUM2 and NOS1 proteins were not present in late, post-meiotic stage germ cells. Use of preimmune PUM2 control sera demonstrated specificity of PUM2 antisera (Fig. 4D); in addition, competition with the immunizing PUM2 protein abolishes the signal (Moore et al. 2003). Likewise, use of NOS1 preimmune sera also demonstrated the specificity of the NOS1 antisera; no staining of human testis cells was observed with preimmune sera (Fig. 4E). Note that closer examination of histological sections stained for the presence of NOS1 and PUM2 proteins indicated that both proteins were not evenly distributed across subcellular compartments but instead were most abundant in a perinuclear subregion of the cytoplasm (Fig. 4A, C). Also note that we did not observe the NOS-1 protein or mRNA in mature postmeiotic oocytes as shown by northern blotting and RT-PCR and immunohistochemistry (data not shown). Since the NOS-1 mRNA is present in fetal ovaries, however, the protein may be expressed in premeiotic female germ cells. This remains to be demonstrated.

Localization of NOS1 If the interaction of the PUM2 and NOS1 proteins occurs in vivo, the proteins must be present in the same cell types and subcellular compartments. Thus, we addressed whether PUM2 and NOS1 colocalize by comparing expression of both proteins in adult testis tissue. Histochemical localization of PUM2 protein has recently been reportedly (Moore et al. 2003). Essentially, expression of PUM2 is abundant in germ cells in the testis especially in the spermatogonia and meiotic spermatocytes (Fig. 4A). To compare expression of NOS1 with that of PUM2, we generated antisera to a NOS1-specific peptide. On western blots, these antisera detected a single protein band of approximately 32 kDa (Fig. 4B). We then used the NOS1 antisera to localize NOS1 protein in human testis sections. We found that like PUM2 (Moore et al. 2003), NOS1 proteins were specifically expressed during germline development. In adult tissues, NOS1 was most

125

Discussion In this work, we characterized the NOS-1 protein, a human protein that is highly conserved in sequence to Nanos proteins of widely divergent organisms from parazoans such as the sponge or hydra to eumetazoans such as flies, worm, frogs and zebrafish. We found that the protein encoded by NOS-1 can form a stable complex with the human Pumilio homolog, PUM2, through the functional domain that is required for RNA binding, protein-protein interactions and rescue of Pumilio mutations in flies. In addition, the distribution of the NOS1 and PUM2 proteins suggested subcellular localization in a distinct region of germ cells. Concentration of these RNA-binding proteins in this subcellular region is reminiscent of germ plasm which consists of clusters of RNA-binding proteins and RNAs that are required for allocation, maintenance and differentiation of the germline in model organisms such as flies, worms, frogs and fish. We suggest that, although allocation, development and maintenance of the germ cells seems to be independent of germ plasm in mammals, perhaps mammals like other organisms possess germ particles composed of proteins homologous to Vasa, Dazl, Pumilio and Nanos (a series of RNA-binding proteins) that all localize to germ plasm in lower organisms (Lehmann and Nusslen-Volhard 1987; Wang and Lehmann 1991; Mosquera et al. 1993; Houston et al. 1998; Maegawa et al. 1999; Subramaniam and Seydoux 1999; Castrillon et al. 2000; Houston and King 2000). In humans, all of these proteins are expressed predominantly in germ cells or embryonic stem cells (Dorfman et al. 1999; Reijo et al. 2000; Tanaka et al. 2000; Moore et al. 2003; this work). Considering that these proteins contain RNA-binding domains, we expect that they may function in a ribonucleoprotein complex as in germ plasm of lower organisms to direct the maintenance or development of the early germ cells through translational regulation of specific mRNAs. Demonstration of such complexes in humans, however, now requires their biochemical isolation and molecular dissection, as well as identification of mutations of the components in a human population. In addition to a potential role in early germ cell development, the expression of both PUM2 and NOS1 in later, meiotic stages points toward a potential role in differentiation of the germ cells of the adult (Fig. 4). In Drosophila, the Pumilio-Nanos complex plays multiple roles in germ cells by binding to different mRNAs: the complex enables development of the anterior-posterior axis by binding to hunchback mRNA and may also be required for development of PGCs during their migration to the gonad (Curtis et al. 1995; Murata and Wharton 1995; Wreden et al. 1997; Asaoka et al. 1998; AsaokaTaguchi et al. 1999). It is possible that among interactions of the human PUM2-NOS1 complex with specific mRNAs some are critical for maintenance, while others enable differentiation of the germ cells. Reproductive genes are thought to evolve rapidly (Hendry et al. 2000; Wyckoff et al. 2000; Swanson and

Vacquier 2002). Perhaps because of this, it is difficult to assess whether many or just a few molecular mechanisms for germ cell development are conserved even within closely-related species. For example, in the worm, C. elegans, transcriptional silencing of the entire genome occurs in germ cell precursor cells via the action of the PIE-1 gene (Seydoux et al. 1996). Yet, to date, homologs of this gene have not been identified in organisms such as frogs, zebrafish, mice or humans. Even within mammals, considerable divergence of the germ cell pathways has occurred as demonstrated by differences in requirements for the sex chromosomes. In humans, deletions of several regions of the X chromosome cause loss of oocytes and premature ovarian failure; complete monozygosity of the X chromosome in women results in few or no germ cells at birth (Davison et al. 1999). In contrast, in mice, a mammal with X-chromosome genes in common with humans, deletions or monosomy of the X chromosome do not alter germ cell development and/or maintenance. Similar observations have been made regarding the role of the genes on the human and mouse Y chromosomes and male germ cell development (Chandley 1979; Burgoyne 1987; Reijo et al. 1995; Lahn and Page 1997). Thus, in light of this divergence of reproductive genes between closely related species, it is particularly remarkable that a core germ cell machinery may be conserved and contain at least a few of the cytoplasmic RNA-binding proteins, such as Pumilio and Nanos, that localize to germline cells of organisms as diverse as parazoans and eumetazoans. Acknowledgements We thank Maria M. Konarska, Daniel F. Ortiz, Krzysztof Kula, Frederick L. Moore, Eugene Y. Xu, and Mark S. Fox for helpful discussions. This work was supported by grants from the Howard Hughes Medical Institute (M.K.), the Polish State Committee for Scientific Research (P05E 018 33 to J.J. and 6 P05E 001 21 to K.K.) and the National Institutes of Health, the Searle Foundation and the Sandler Family Foundation (R.A.R.P.).

References Asaoka M, Sano H, Obara Y, Kobayashi S (1998) Maternal Nanos regulates zygotic gene expression in germline progenitors of Drosophila melanogaster. Mech Dev 78:153–158 Asaoka-Taguchi M, Yamada M, Nakamura A, Hanyu K, Kobayashi S (1999) Maternal Pumilio acts together with Nanos in germline development in Drosophila embryos. Nature Cell Biol 1:431–437 Bhat KM (1999) The posterior determinant gene nanos is required for the maintenance of the adult germline stem cells during Drosophila oogenesis. Genetics 151:1479–1492 Burgoyne PS (1987) The role of the mammalian Y chromosome in spermatogenesis. Development Suppl 101:133–141 Castrillon DH, Quade BJ, Wang TY, Quigley C, Crum CP (2000) The human VASA gene is specifically expressed in the germ cell lineage. Proc Natl Acad Sci USA 97:9585–9590 Chandley AC (1979) The chromosomal basis of human infertility. Br Med Bull 35:181–186 Crittenden SL, Bernstein D, Bachorik J, Thompson B, Moulder G, Barstead R, Wickens M, Kimble J (2002) FBF and control of germline stem cells inCaenorhabditiselegans. Nature 417:600– 603

126 Curtis D, Apfeld J, Lehmann R (1995) Nanos is an evolutionarily conserved organizer of anterior-posterior polarity. Development 121:1899–1910 Davison RM, Davis CJ, Conway GS (1999) The X chromosome and ovarian failure. Clin Endocrinol 51:673–679 Deshpande G, Calhoun G, Yanowitz JL, Schedl PD (1999) Novel functions of nanos in downregulating mitosis and transcription during the development of the Drosophila germline. Cell 99:271–281 Dorfman DM, Genest DR, Reijo Pera RA (1999) Human DAZL1 encodes a candidate fertility factor in women that localizes to the prenatal and postnatal germ cells. Hum Reprod 14:2531– 2536 Forbes A, Lehmann R (1998) Nanos and Pumilio have critical roles in the development and function of Drosophila germline stem cells. Development 125:679–690 Hendry AP, Wenburg JK, Bentzen P, Volk EC, Quinn TP (2000) Rapid evolution of reproductive isolation in the wild: evidence from introduced salmon. Science 290:516–518 Houston DW, King ML (2000) A critical role for Xdazl, a germ plasm-localized RNA, in the differentiation of primordial germ cells in Xenopus. Development 127:447–456 Houston DW, Zhang J, Maines JZ, Wasserman SA, King ML (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 Kobayashi S, Yamada M, Asaoka M, Kitamura T (1996) Essential role of the posterior morphogen nanos for germline development in Drosophila. Nature 380:708–711 Koprunner M, Thisse C, Thisse B, Raz E (2001) A zebrafish nanosrelated gene is essential for the development of primordial germ cells. Genes Dev 15:2877–2885 Lahn BT, Page DC (1997) Functional coherence of the human Y chromosome. Science 278:675–680 Lawson KA, Dunn NR, Roelen BAJ, Zeinstra LM, Davis AM, Wright CVE, Korving JPWFM, Hogan BLM (1999) Bmp4 is required for the generation of primordial germ cells in the mouse embryo. Genes Dev 13:424–436 Lehmann R, Nusslen-Volhard C (1987) Involvement of the Pumilio gene in the transport of an abdominal signal in the Drosophila embryo. Nature 329:167–170 Lin H, Spradling AC (1997) A novel group of pumilio mutations affects the asymmetric division of germline stem cells in the Drosophila ovary. Development 124:2463–2476 Maegawa S, Yasuda K, Inoue K (1999) Maternal mRNA localization of zebrafish DAZ-like gene. Mech Dev 81:223–226 Mochizuki K, Sano H, Kobayashi S, Nishimiya-Fujisawa C, Fujisawa T (2000) Expression and evolutionary conservation of nanos-related genes in Hydra. Dev Genes Evol 210:591–602 Moore FL, Jaruzelska J, Fox MS, Urano J, Firpo MT, Turek PJ, Dorfman DM, Reijo Pera RA (2003) A human homologue of Drosophila Pumilio is expressed in embryonic stem cells and germ cells and interacts with DAZ (Deleted in AZoospermia) protein. Proc Natl Acad Sci USA 100:538–543 Mosquera L, Forristall C, Zhou Y, King ML (1993) A mRNA localized to the vegetal cortex of Xenopus oocytes encodes a protein with a nanos-like zinc finger domain. Development 117:377–386 Murata Y, Wharton RP (1995) Binding of Pumilio to maternal hunchback mRNA is required for posterior patterning in Drosophila embryos. Cell 80:747–756

Nakahata S, Katsu Y, Mita K, Inoue K, Nagahama Y, Yamashita M (2001) Biochemical identification of Xenopus Pumilio as a sequence-specific cyclin B1 mRNA-binding protein that physically interacts with a Nanos homolog, Xcat-2, and a cytoplasmic polyadenylation element-binding protein. J Biol Chem 276:20945–20953 Parisi M, Lin H (1999) The Drosophila Pumilio gene encodes two functional protein isoforms that play multiple roles in germline development, gonadogenesis, oogenesis, and embryogenesis. Genetics 153:235–250 Reijo R, Lee TY, Salo P, Alagappan R, Brown LG, Rosenberg M, Rozen S, Jaffe T, Straus D, Hovatta O, et al (1995) Diverse spermatogenic defects in humans caused by Y chromosome deletions encompassing a novel RNA-binding protein gene. Nature Genet 10:383–393 Reijo RA, Dorfman DM, Slee R, Renshaw AA, Loughlin KR, Cooke H, Page DC (2000) DAZ family proteins exist throughout male germ cell development and transit from nucleus to cytoplasm at meiosis in humans and mice. Biol Reprod 63:1490–1496 Rongo C, Broihier HT, Moore L, Van Doren M, Forbes A, Lehmann R (1997) Germ plasm assembly and germ cell migration in Drosophila. Cold Spring Harbor Symp Quant Biol 62:1–11 Seydoux G, Mello CC, Pettitt J, Wood WB, Preiss JR, Fire A (1996) Repression of gene expression in the embryonic germ lineage of C. elegans. Nature 382:713–716 Sonoda J, Wharton RP (1999) Recruitment of Nanos to hunchback mRNA by Pumilio. Genes Dev 13:2704–2712 Subramaniam K, Seydoux G (1999) nos-1 and nos-2, two genes related to Drosophila nanos, regulate primordial germ cell development and survival in Caenorhabditis elegans. Development 126:4861–4871 Swanson WJ, Vacquier VD (2002) The rapid evolution of reproductive proteins. Nature Rev Genet 3:137–144 Tanaka SS, Toyooka Y, Akasu R, Katoh-Fukui Y, Nakahara Y, Suzuki R, Yokoyama M, Noce T (2000) The mouse homolog of Drosophila Vasa is required for the development of male germ cells. Genes Dev 14:841–853 Wang C, Lehmann R (1991) Nanos is the localized posterior determinant in Drosophila. Cell 68:1177 Wreden C, Verrotti AC, Schisa JA, Lieberfarb ME, Strickland S (1997) Nanos and pumilio establish embryonic polarity in Drosophila by promoting posterior deadenylation of hunchback mRNA. Development 124:3015–3023 Wyckoff GJ, Wang W, Wu C-I (2000) Rapid evolution of male reproductive genes in the descent of man. Nature 403:304–309 Wylie C (2000) Germ cells. Curr Opin Genet Dev 10:410–413 Ying Y, Qi X, Zhao G-Q (2001) Induction of primordial germ cells from murine epiblasts by synergistic action of BMP4 and BMP8B signaling pathways. Proc Natl Acad Sci USA 98:7858– 7862 Zamore PD, Williamson JR, Lehmann R (1997) The Pumilio protein binds RNA through a conserved domain that defines a new class of RNA-binding proteins. RNA 3:1421–1433 Zamore PD, Bartel DP, Lehmann R, Williamson JR (1999) The PUMILIO-RNA interaction: a single RNA-binding domain monomer recognizes a bipartite target sequence. Biochem 38:596–604 Zhou Y, King ML (1996) RNA transport to the vegetal cortex of Xenopus oocytes. Development 179:173–183