Transcriptional Analysis of the Candidate Spermatogenesis GeneUbe1yand of the Closely RelatedUbe1xShows That They Are Coexpressed in Spermatogonia and Spermatids but Are Repressed in Pachytene Spermatocytes

DEVELOPMENTAL BIOLOGY 180, 336–343 (1996) ARTICLE NO. 0305

Transcriptional Analysis of the Candidate Spermatogenesis Gene Ube1y and of the Closely Related Ube1x Shows That They Are Coexpressed in Spermatogonia and Spermatids but Are Repressed in Pachytene Spermatocytes Teresa Odorisio,* Shantha K. Mahadevaiah,* John R. McCarrey,† and Paul S. Burgoyne* *Laboratory of Developmental Genetics, National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7, 1AA, United Kingdom; and †Department of Genetics, Southwest Foundation for Biomedical Research, P.O. Box 28147, San Antonio, Texas 78228

Ube1y is a Y-linked gene transcribed in the testis, which maps to a region of the mouse Y required for normal spermatogonial proliferation. Ube1y, together with a ubiquitously expressed homologue on the X chromosome (Ube1x), encodes ubiquitinactivating enzyme E1, an enzyme essential for eukaryotic cell proliferation. Ube1y is thus a strong candidate for the Y function in spermatogonial proliferation. Using probes specific for the two genes, we have used Northern analysis and RNase protection to assess transcript levels throughout testis development and, by using germ cell-deficient XXSxra testes and purified cell fractions, we have defined the testicular cell types in which transcription occurs. Ube1y transcripts are already detectable in the fetal testis at 12.5 dpc, with higher levels at 14.5 dpc and then falling to low levels by the time of birth. Postnatally levels rise sharply, peaking at 10 dpp. Analysis of XXSxra testes indicates that the bulk of the Ube1y transcription is in germ cells. The analysis of purified cell fractions shows that X- and Y-encoded transcripts are present in A spermatogonia, both are at very low levels (or perhaps absent) in pachytene spermatocytes and then return to high levels in round spermatids. The reactivation of transcription in round spermatids implies a requirement for the ubiquitination pathway at this time. The presence of Ube1x transcripts in A spermatogonia raises the question as to why Ube1y transcripts are required. This question is discussed in relation to the spermatogenic failure in XSxrbO mice which are deleted for Ube1y and it is argued that Ube1y serves to increase UBE1 production at a time of high demand. Ube1y transcripts were also detected in XXY and XY ovaries. q 1996 Academic Press, Inc.

INTRODUCTION Deletion mapping has established that the mouse Y chromosome, in addition to the testis determining function encoded by Sry, carries factors required for spermatogenesis (Burgoyne, 1987, 1992, 1993). One of these factors, Spy, which defines a Y requirement in spermatogonial proliferation, has been mapped to the Y short arm, within the ú900kb deletion that occurred when Sxrb arose from Sxra (Burgoyne et al., 1986; Sutcliffe and Burgoyne, 1989; Simpson and Page, 1991; Mitchell and Bishop, 1992). Molecular genetic approaches have led to the identification of a gene Ube1y within this deletion interval (Kay et al., 1991; Mitchell et al., 1991) that encodes ubiquitin activating enzyme

E1 (UBE1). Since ubiquitination is necessary for the progression of the mitotic cell cycle (Zacksenhaus and Sheinin, 1990a; Handley-Gearhart et al., 1994b), Ube1y is a very attractive candidate for the spermatogonial proliferation gene Spy. However, a closely related gene Ube1x is present on the X chromosome which also encodes a UBE1 protein (Kay et al., 1991; Mitchell et al., 1991; Imai et al., 1992). Furthermore, in man, no Y chromosomal copy of Ube1y has been detected [although it has been shown to be present in a wide variety of mammalian species including kangaroos (Mitchell et al., 1992)] which raises the question as to why a Y chromosomal Ube1 copy is needed in the mouse. In the present paper we have analysed Ube1y and Ube1x tran0012-1606/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

336

AID

DB 8397

/

6x16$$$$81

10-29-96 00:40:04

dba

AP: Dev Bio

337

Ube1y and Ube1x Transcription in the Mouse Testis

script levels throughout testis development and have asked whether Ube1x and Ube1y are coexpressed in spermatogenic cells.

MATERIALS AND METHODS Mice With the exception of the random bred CD1 males (Charles River) used for the purification of ‘‘primitive’’ type A spermatogonia, all mice were bred in our own colony on an MF1 random bred background. These comprised: (1) XYRIII males which have an RIII strain Y chromosome, (2) XXSxra and XXSxrb males which carry the RIII Y-derived sex-reversing factors Sxra (Cattanach et al., 1971) or Sxrb (McLaren et al., 1984), (3) XYSry0 and XXYSry0 females which carry a 129 Y chromosome which lacks Sry due to an 11-kb deletion (Gubbay et al., 1992; Mahadevaiah et al., 1992), and (4) XYd01 females which have a deletion which has brought Sry closer to the Y centromere, where it is transcriptionally repressed (Capel et al., 1993; Laval et al., 1995).

Mitchell et al. (1991); Ube1x F1, GTGCATTCCCCTAAGCCCCA; Ube1x R1, GGGTAATTATCCTTTTATTGGGAT—see Imai et al. (1992)]. The mouse actin probe (used to provide a loading control) was an a-actin cDNA 1.15-kb Pst fragment (Minty et al., 1981) which recognises a- and b-actin transcripts. The probes were labeled by random priming (Prime It kit, Stratagene). The Sap62 probe used as a loading control for RNase protection was probe A of Dresser et al. (1995). The Ube1y and Ube1x probes (both 325 nt) for RNase protection were located at the 3* end of the open reading frame and were generated as follows: PCR products amplified from spermatid cDNA (primers 5*-3*: Ube1y F2, GGCTCCAGAGTGTCATCAGTT; R2, TATGTCGTCTCCACTGTCACT; Ube1x F2, TGCTGCACCTCGTCACCAGTA; R2, GACGTCCTCGCCGCTTTCAT) were cloned into PCR II (Invitrogen). The labeled riboprobes were produced by linearising the plasmids with BamHI and transcribing with T7 RNA polymerase in the presence of 800 Ci/mmole [32P]UTP (Amersham) as described by Krieg and Melton (1987). The riboprobes were electrophoresed through a 6% polyacrylamide, 8 M urea gel, and full-length probes were excised from the gel, eluted with 0.5 M ammonium acetate, 1 mM EDTA, 0.1% SDS, and were then precipitated in ethanol.

RNA Extraction

Testis Cell Purification Pachytene spermatocytes and round spermatids were separated from testicular cell suspensions of adult XYRIII males by elutriation (Meistrich, 1977). Briefly, decapsulated testes were incubated for 15 min at 327C in a shaking incubator in Hepes-buffered Dulbecco’s modified minimal essential medium (DMEM, HyClone Europe Ltd.) containing 0.25 mg/ml collagenase (Type X1, Sigma) to separate the seminiferous tubules from the interstitial tissue. The seminiferous tubules were washed twice in DMEM and were then incubated in DMEM containing 2.5 mg/ml trypsin (GIBCO-BRL) and 50 mg/ml DNase (DN-25, Sigma) for 15 min at 327C in a shaking incubator. Trypsin digestion was stopped with 8% fetal calf serum and the digested tubules were gently pipetted to help produce a single cell suspension. Any remaining tubule fragments were allowed to settle and the cell suspension was then loaded into an elutriator rotor (5.2) in a Beckman J-6M/E centrifuge. Fractions (150 ml) were collected—the round spermatid fraction at a rotor speed of 3000 rpm and a flow rate of 40 ml/min and the pachytene fraction at a rotor speed of 2000 rpm and a flow rate of 28.5 ml/min. The purity of the fractions was assessed from air-dried Giemsa-stained cell preparations. Pachytene spermatocytes were 75–80% pure (contaminated predominantly with round and elongating spermatids); round spermatids were 80–85% pure (contaminated predominantly with elongating spermatids). Populations of ‘‘primitive’’ type A spermatogonia (purity § 85%) [equivalent to undifferentiated A / differentiating A1 and A2 spermatogonia of Sutcliffe and Burgoyne (1989)] and Sertoli cells (purity § 90%) were isolated from testes of 6-day-postpartum (dpp) CD1 males using a small StaPut chamber (Johns Scientific) as described by Bellve´ et al. (1977); purities were assessed by examination of cell morphology using phase microscopy.

Probes The 3* untranslated region Ube1y (156 bp) and Ube1x (189 bp) probes for Northern analysis were generated by PCR from genomic DNA [primers 5*-3*: Ube1y F1, TTACTGACTCCGTTCTTTCAGAT; Ube1y R1, CACAGGCTGAAAATTTATTATTAT—see

Poly(A)/ RNA for Northern analysis was isolated with a MicroFast Track or Fast-Track kit (InVitrogen). Total RNA for RNase protection and RT–PCR was isolated using the AGPC method (Chomczynski and Sacchi, 1987), except for the 6-dpp Sertoli and spermatogonia RNA samples which were isolated using the guanidinium isothiocyanate/cesium chloride method (Chirgwin et al., 1979).

Northern Analysis Poly(A)/ RNA (4–5 mg) from testis or testis cell preparations together with RNA size markers was electrophoresed through a 1.4% agarose gel containing 1.9 M formaldehyde. RNA was transferred to Hybond N membrane (Amersham) using 201 SSC and the membrane was hybridised overnight first to Ube1y, then to Ube1x, and finally to actin using 1–2 1 106 cpm of each probe. For Ube1y and Ube1x the filters were washed at low stringency (21 SSC, 1% SDS at 507C) and for actin at high stringency (0.11 SSC, 0.1% SDS at 657C). The filters were then exposed to phosphor screens to allow subsequent quantitation and then to autoradiographic film (Kodak X-OMAT AR). The phosphor screens were scanned on a PhosphorImager (Molecular Dynamics, Sunnydale, CA) and individual bands were quantitated using the ‘‘integrate volume’’ feature of the ImageQuant software, background being estimated from adjacent areas within each track. After each hybridisation filters were washed in 50% formamide in 10 mM NaH2PO4 (pH 6.8) at 657C for 1 hr to remove labeled probes.

RNase Protection Total RNA (1–10 mg) from testis or purified testis cell populations was hybridised overnight with 5 1 105 –106 cpm of each riboprobe (Ube1x / Sap or Ube1y / Sap) in 19 ml of 80% formamide with 40 mM Pipes (pH 6.7), 0.4 M NaCl, and 1 mM EDTA and then was digested with 10 U RNase T1 and 40 mg RNase A per milliliter, in 0.35 ml of 300 mM NaCl, 10 mM Tris, pH 7.5, 5 mM EDTA, at 377C for 45 min. After proteinase K treatment and phenol–chloro-

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID

DB 8397

/

6x16$$$$82

10-29-96 00:40:04

dba

AP: Dev Bio

338

Odorisio et al.

FIG. 1. Ube1y and Ube1x transcription throughout postnatal testis development. (A) Northern of poly(A)/ testis RNA probed with 3*-untranslated region Ube1y and Ube1x probes and with an actin probe to provide a loading control. XXSxrb testis RNA (Sxrb is deleted for Ube1y) serves as a negative control for Ube1y. Overlaying the original autorads showed that the Ube1y and Ube1x transcripts were indistinguishable in size (estimated to be 3.8 kb). (B) Relative changes in level of transcription of Ube1y and Ube1x (normalised against actin) throughout postnatal testis development.

form extraction, samples were ethanol precipitated and electrophoresed on 6% polyacrylamide, 8 M urea gels. Gels were exposed to phophor screens to allow subsequent quantitation and then to autoradiographic film. The phosphor screens were scanned on a PhosphorImager and relative band intensity was quantitated as described above under Northern analysis.

are also deleted for Ube1y) serve as a negative control for Ube1y and confirm the specificity of the probe. The length of the Ube1x transcript has previously been estimated to be 3.5 kb (Imai et al., 1992); the current Northern analysis shows that the Ube1x and Ube1y transcripts are identical in length but the size was estimated to be Ç3.8 kb based on RNA size markers (not shown). From the time course it can be seen that Ube1y transcripts are already detectable on the day of birth, increasing fourfold to a peak at 10 dpp, and falling thereafter except perhaps for a small secondary peak at 28 days. Ube1x transcripts on the other hand are present at high levels on the day of birth and do not fall below this level until after 10 days. There is a clear secondary peak at 28 days. The presence of large quantities of Ube1x transcripts in XXSxrb testes even though they lack germ cells (two X chromosomes are incompatible with spermatogenesis) is to be expected because Ube1x is ubiquitously expressed. Since Ube1y transcripts are already detectable on the day of birth we decided to look earlier. Because of the small amounts of material we used RNase protection. The ubiquitously expressed Sap62 (encoding a spliceosome protein) was used as a loading control since this has been used previously in this laboratory for studies of transcription in fetal gonads (Dresser et al., 1995; Hacker et al., 1995). Initially we looked at 14.5-, 16.5-, 18.5-dpc testis RNA and also included RNA samples from testes of newborns, 10 dpp, and adult to allow the results to be keyed in to the Northern results. Ube1y transcripts were detectable on all the days assayed (Fig. 2A), but relative to the levels of Sap transcripts, Ube1y

RT–PCR Total RNA (2 mg) from each tissue was reverse transcribed in a 60-ml reaction using standard procedures. A 2.5-ml aliquot was then added to a 25-ml PCR reaction containing 250 ng of primers Ube1y F1 and R1 (156-bp product) and 200 ng of primers Ube1x R2 and F2 (325-bp product). cDNA was amplified for 35 cycles of 10 sec at 967C, 30 sec at 627C, and 30 sec at 727C. The products were separated on a 3% agarose gel and stained with ethidium bromide. Because the Ube1y primers are from within an exon it is necessary to be certain that there is no DNA contamination. In principle, contaminating DNA should be revealed by the presence of a Ç1kb band amplified by the Ube1x primers. However, following the initial PCR, some critical RNA samples were DNase treated and the reactions were repeated with and without reverse transcriptase, prior to PCR amplification with Ube1y primers.

RESULTS Because the data so far available indicate very close sequence homology between Ube1y and Ube1x, there are difficulties in designing Y- and X-specific probes that can be used with confidence in Northern analysis. We have therefore used probes from the 3* untranslated region where there is more sequence divergence than in coding regions (M. Mitchell, personal communication). A time course of transcript levels for the two genes in the testis is given in Fig. 1. XXSxrb testes (which are not only germ cell deficient, but

FIG. 2. RNase protection analysis of Ube1y and Ube1x transcripts in the fetal testis. (A) Ube1y (B) Ube1y at 12.5 dpc and repeat of 14.5 dpc and 0 dpp. (C) Ube1x. [Note: Difficulty was experienced in getting conditions which allowed full protection of both Ube1y and Sap. In A most of the Ube1y is fully protected but Sap is a doublet. In B Sap is fully protected by Ube1y now has much more of the incompletely protected fragment. However, it is clear that the fully and incompletely protected fragments vary together with respect to levels at the different ages. Both fragments (when present) were included during quantitation].

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID

DB 8397

/

6x16$$$$82

10-29-96 00:40:04

dba

AP: Dev Bio

339

Ube1y and Ube1x Transcription in the Mouse Testis

FIG. 3. Summary of Ube1y expression throughout testis development. The RNase protection data (open circles) were keyed in to the Northern data (closed circles) by arbitrarily assigning a value of 1 for both at 10 dpp. The 0-dpp and adult levels estimated by RNase protection using Sap as a loading control are reassuringly close to the levels estimated by Northern analysis using actin as a loading control. The relative transcript levels at 14.5 dpc and 0 dpp from the second RNase protection (Fig. 2B) were almost identical to those plotted here.

transcript levels appeared to fall between 14.5 dpc and the day of birth. A second Ube1y protection assay was performed using RNA from 12.5 dpc, 14.5 dpc, and the day of birth. This confirmed the drop between 14.5 dpc and the day of birth and also showed that transcripts were present at 12.5 dpc (Fig. 2B). High levels of Ube1x transcripts were present at all ages assayed (Fig. 2C). A summary of the changes in level of Ube1y transcripts throughout testis development is given in Fig. 3. In order to see if Ube1y transcription is germ cell dependent, we have looked at Ube1y transcript levels in XXSxra testes (RNA from five testes from five males, testis weights 17–26 mg) by Northern analysis. XXSxra testes, like XXSxrb testes, lack germ cells, but differ in having the potential for Ube1y transcription since the gene is present. As can be seen from Fig. 4A, Ube1y transcripts are detectable in XXSxra testes, but relative to actin the levels are 15-fold lower than those in XY testes by PhosphorImager analysis. This supports earlier RT–PCR data from two studies (Kay et al., 1991; Mitchell et al., 1991) suggesting that Ube1y transcription in the adult testis is largely germ cell dependent, although the studies conflicted in that Kay et al. detected low levels, while Mitchell et al., detected no transcripts in XXSxra testes. Occasionally XXSxra testes may have a patch of XSxraO germ cells (Lyon et al., 1981) due to the loss of the second X in a progenitor cell. Although the testis weights (õ30 mg) of the XXSxra testes used for the Northern make this unlikely, we decided to use RNase protection on individual XXSxra testes to see if Ube1y transcripts are consistently present. Total RNA from the contra-

lateral testis of four of the five XXSxra males used for the Northern was used for the protection analysis and in all four Ube1y transcripts were detected (Fig. 5B). The similar modulation of transcript levels of Ube1y and Ube1x after 10 dpc apparent in the Northern analysis (Fig. 1) suggested that both genes may be subject to stage-dependent regulation of transcription in spermatogenic cells. Attempts to utilise the 3* probes for in situ hybridisation were unsuccessful so we looked instead at purified spermatogenic cell populations. Initially we looked at pachytene spermatocytes and round spermatids purified by elutriation. Northern analysis revealed very low levels of both Ube1y and Ube1x transcripts in the sample from purified pachytene spermatocytes but there were at least sixfold higher levels in round spermatids (Fig. 5A). The low levels at pachytene are to be expected (even this low level could be due to the contamination of the sample with round spermatids) since it has long been known that the sex chromosomes are transcriptionally repressed during pachytene (Monesi, 1965). However, the return to high transcript levels in round spermatids is significant in that it implies reactivation of Ube1 transcription at this time. It also accounts for the rise in Ube1y and Ube1x transcript levels between 21 and 28 dpp seen in Fig. 1, although the rise for Ube1y is surprisingly small. This may indicate that the Ube1y transcription in round spermatids is short-lived, being limited to a subset of round spermatid stages. In view of the coexpression of X- and Y-encoded Ube1 transcripts in spermatids it seemed likely that they would also be coexpressed in spermatogonial stages. In order to address this directly, we purified samples of A spermatogonia from 6-dpp males (using the StaPut method) and measured Ube1y and Ube1x transcript levels by RNase protection. Transcript levels in purified 6-dpp Sertoli cell samples and whole testes were assessed to provide a comparison. It is clear from Fig. 5B that both transcripts are present in A

FIG. 4. Ube1y transcripts in adult testes lacking germ cells. (A) Northern of poly(A)/ RNA from five testes from five XXSxra males and from a normal male. The XXSxra track has been overloaded (see hybridisation with Ube1x and actin) in order to detect the low level of Ube1y transcripts. (B) Ube1y RNase protection of total RNA from single testes from four of the five XXSxra males used for the Northern analysis.

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID

DB 8397

/

6x16$$$$83

10-29-96 00:40:04

dba

AP: Dev Bio

340

Odorisio et al.

ern blotting and probing with the Ube1y-specific 3* probe (data not shown).

DISCUSSION

FIG. 5. Analysis of Ube1y and Ube1x transcription in purified testis cell fractions. (A) Northern analysis of poly(A)/ RNA from purified pachytene spermatocytes and from round spermatids compared with total adult testis. High levels are seen in the spermatid fraction, but very low levels of both transcripts are detectable at pachytene (these transcripts at least in part are from contaminating round spermatids). (B) RNase protection analysis of Ube1y and Ube1x in purified 6-dpp A spermatogonia compared with whole 6dpp testis and purified 6-dpp Sertoli cells. Both transcripts are clearly expressed in A spermatogonia, Ube1x levels being some 2.5-fold higher than Ube1y (quantitating both Ube1y bands).

spermatogonia. In so far as the protected fragments for Ube1y and Ube1x are the same size and the labeling for both was done at the same time, this analysis should also give some idea of the relative amounts of the two transcripts. Quantitation by PhosphorImager analysis estimates the level of X-encoded transcripts in A spermatogonia to be Ç2.5-fold higher than the level of Y-encoded transcripts. In reality, this overestimates the difference, because the Ç15% somatic cell contamination will have diluted the Ube1y transcripts, while leaving Ube1x levels unaffected (Ube1x levels are equivalent in the purified Sertoli and spermatogonia fractions—see 5B). From Fig. 5B it can also be seen that Ube1y transcripts are present in the Sertoli cell fraction (purity §90%) of 6-dpp testes at an estimated one-quarter of the level in gonia. It is impossible to judge whether this is entirely due to contamination of this fraction with spermatogonia or whether Sertoli cells are the somatic source of Ube1y transcripts implied by the analysis of XXSxra testes. We have also assayed Y-bearing ovaries, for Ube1y transcripts using RT–PCR. The Y-bearing ovaries were from XY, Sry0 and XXY, Sry0 females which lack the testis-determining gene Sry (Mahadevaiah et al., 1992) and from XYd01 females which fail to express Sry (Capel et al., 1993; Laval et al., 1995). Ube1y transcripts are clearly present in all the Y-bearing ovaries assayed (Fig. 6). This RT–PCR also indicates a low level of Ube1y transcripts in the adrenal and this was confirmed on a DNase-treated RNA sample with the amplified product positively identified by South-

Although 5 years have elapsed since the cloning of Ube1y, no further information has been reported which either adds to or detracts from its candidature for the spermatogonial proliferation factor Spy. There has not even been a detailed transcriptional analysis, perhaps because of the technical difficulties arising from the close sequence similarity between Ube1y and the ubiquitously expressed Ube1x. The limited RT–PCR-based transcriptional analysis that has been published gave conflicting results as to the germ cell specificity of Ube1y transcription (Kay et al., 1991; Mitchell et al., 1991). By using a combination of Northern analysis with Ube1yand Ube1x-specific 3*-untranslated probes, RNase protection, and analysis of purified spermatogenic cell populations, we have provided a more detailed description of Ube1y and Ube1x transcription throughout testis development. The salient features of Ube1y transcription are: (1) Ube1y is transcribed predominantly in the germ line, although a low level of transcription was detected in XXSxra testes lacking germ cells, as previously reported by Kay et al. (1991). (2) There is a peak of Ube1y expression in the fetal testis at 14.5 dpc followed by a fall to low levels at birth. (3) After birth, levels rise steeply to a peak at 10 dpp, after which there is again a steep fall. Significantly, the analysis of purified spermatogonia at 6 dpp shows high levels of Ube1y transcripts. The abrupt fall in transcript levels after 10 dpp is undoubtedly due to the low level (or absence) of transcription in pachytene spermatocytes which first appear at this time. (4) Despite the absence or near absence

FIG. 6. A. RTP–PCR analysis of Ube1y and Ube1x transcription in ovaries of Y-bearing females and various control tissues. All samples are positive for the ubiquitously expressed Ube1x. Ube1y transcripts are detected in all three ovary samples (XYSry0, XXYSry0, XYd01) and in XY and XXSxra testes (but not in XXSxrb which is deleted for Ube1y); all other tissues are negative except for a weak signal in adrenal. (B) Repeat Ube1y RT–PCR for XYSry0 and XYd01 ovaries after treating the RNA with DNase I, including samples with (/) and without (0) reverse transcriptase.

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID

DB 8397

/

6x16$$$$83

10-29-96 00:40:04

dba

AP: Dev Bio

341

Ube1y and Ube1x Transcription in the Mouse Testis

of transcripts in pachytene spermatocytes, high levels are detected in purified round spermatids. In the present context, the important features of Ube1x expression are that it is also transcribed in A spermatogonia and that subsequently there is suppression of transcription at pachytene followed by a resumption in round spermatids, mirroring the pattern seen for Ube1y. The almost exclusive expression of Ube1y in spermatogenic cells (present results), and the retention of a Y-linked Ube1 in all mammals analysed except man, chimpanzee, and old world monkeys (Mitchell et al., 1996), argues strongly for a role in spermatogenesis. However, the coexpression in spermatogonia and round spermatids of the Xlinked UBE1 raises the question as to the need for a Yencoded product. One possibility is that the Y-encoded protein may have some properties that distinguish it from the X-encoded protein and provided some male germ line-specific function. Examination of the complete Ube1y amino acid sequence (M. Mitchell, personal communication) indicates that the Y-encoded protein retains the features thought to be required for ubiquitin binding (Burch and Haas, 1994), for nuclear localisation (Grenfell et al., 1994; Handley-Gearhart et al., 1994a), and for phosphorylation by Cdc2 kinase (Nagai et al., 1995), so there is no obvious basis for a difference in function between the two proteins. An alternative possibility is that Ube1y expression simply serves to increase the amount of product during periods of high demand. A number of features of the spermatogenic block in XSxrbO mice (which are deleted for Ube1y) support this possibility. The XSxrbO spermatogenic block has been analysed in detail (Sutcliffe and Burgoyne, 1989) and can be summarised as follows: (1) There are normal numbers of T1 prospermatogonia at birth but only about half of these resume mitosis, (2) those that do resume mitosis then successfully complete two or three further rounds of division without any evidence of a reduced mitotic rate, (3) there is then total cell cycle arrest during the differentiating A spermatogonial stages, and (4) after an ill-defined interval following the initial arrest, groups of differentiating spermatogonia sporadically resume mitosis, some of which complete the mitotic divisions and progress as far as the pachytene stage of meiosis. The important features of the spermatogenic failure in these mice lacking Ube1y are that the early spermatogonial divisions are unaffected and that after a period of ‘‘rest’’ some groups of spermatogonia go on to complete the later divisions. This clearly shows that there is no mandatory requirement for Ube1y during spermatogonial mitosis; rather, the function of Ube1y seems to be to enable all the mitoses to be completed without a break. It is instructive to consider the possible relationship between UBE1 deficiency and the XSxrbO spermatogenic failure in more detail. UBE1 is known to be essential at two stages of the mitotic cell cycle: for exit from metaphase because this requires ubiqutination of cyclins (Glotzer et al., 1991) and for the S/G2 transition (Mita et al., 1980;

Zacksenhaus and Sheinin, 1990b). As a working hypothesis we propose the following explanation of the features of XSxrbO spermatogenesis: (1) There is a reduction in the number of T1 prospermatogonia that resume mitosis after birth because of the reduced amount of UBE1. (2) The fact that the following two to three divisions proceed normally suggests that lower rates of ubiquitination are required for progression of mitotic cycles than for the initial triggering of mitosis in the G1-arrested T1 prospermatogonia. (3) During these mitotic cycles UBE1 is depleted more quickly than it is produced and this eventually reduces UBE1 levels sufficiently to induce cell cycle arrest, either at metaphase or at the S/G2 transition. Groups of spermatogonia conjoined by intercellular bridges are likely to arrest at the same stage of the cycle. (4) Many of the arrested spermatogonia are eliminated (there is increased cell death at this stage—see Sutcliffe and Burgoyne, 1989), probably by apoptosis. (5) Occasionally, groups of spermatogonia survive long enough for Ube1x transcripts to replenish UBE1 levels to the point where the level of ubiquitination allows mitosis to resume. An interesting feature of the present transcriptional analysis is the sharp increase in Ube1y and Ube1x transcripts in round spermatids after their near absence at the pachytene stage. This is in agreement with growing evidence that the period of maximal repression of sex-linked gene transcription during male meioisis is limited to the later stages of the first meiotic prophase. Indeed, recent evidence on the chromatin structure of X-linked structural genes suggests that X inactivation may be restricted to the pachytene stage (Kumari et al., 1996) and may therefore be related to requirements for a specific chromatin structure during recombination events in the male (McKee and Handel, 1993). Whether X-linked genes become reactivated in round spermatids therefore presumably reflects the extent of the requirement for the products of those genes at this stage. Since Mhr6a, another X-linked gene involved in ubiquitination, is also transcribed in round spermatids (Hendricksen et al., 1995), this points to a need to replenish transcripts for these essential components of the ubiquitination pathway in order to meet the requirements for ubiquitination in spermatids (e.g., for the replacement of histones with protamines; Mita et al., 1980). In our view, the transcriptional analysis presented here provides support for the view that deletion of Ube1y is responsible for the spermatogenic failure in XSxrbO mice. Of particular significance is our finding that Ube1y and Ube1x are coexpressed in A spermatogonia and are both repressed in pachtene spermatocytes. This rules out the possibility that Ube1y is ‘‘standing in’’ for Ube1x when the X copy is inactivated, and it also provides a plausible basis for the ‘‘leak’’ in the XSxrbO spermatogenic block. If we are correct in thinking that the function of Ube1y is primarily to increase the rate of production of UBE1 at a time of high demand, then it should be possible to reproduce the XSxrbO spermatogenic block, not only by the targeted knock out of Ube1y, but also by the downregulation of Ube1x in spermatogenic cells.

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID

DB 8397

/

6x16$$$$83

10-29-96 00:40:04

dba

AP: Dev Bio

342

Odorisio et al.

ACKNOWLEDGMENTS ´ ine Rattigan for excellent We thank Paul Molland and A technical assistance, Meena Kumari for helping to isolate the spermatogonial RNA, and Michael Mitchell for sharing unpublished sequence information. Sohaila Rastan and Graeme Penny helped to get this project started, and Robin Lovell-Badge and many of his colleagues, together with Salvatore Ulisse, provided advice and support throughout. T.O. was the recipient of a European Communities HCM fellowship.

REFERENCES Bellve´, A. R., Cavicchia, C. C., Millette, C. F., O’Brien, D. A., Bhatnagar, Y. M., and Dym, M. (1977). Spermatogenic cells of the prepuberal mouse: isolation and morphological characterization. J. Cell Biol. 74, 68–85. Burch, T. J., and Haas, A. L. (1994). Site-directed mutagenesis of ubiquitin. Differential roles of arginine in the interaction with ubiquitin-activating enzyme. Biochemistry 33, 7300–7308. Burgoyne, P. S. (1987). The role of the mammalian Y chromosome in spermatogenesis. Development 101(Suppl.), 133–141. Burgoyne, P. S. (1992). Y-chromosome function in mammalian development. Adv. Dev. Biol. 1, 1–29. Burgoyne, P. S. (1993). Deletion mapping the functions of the mouse Y chromosome. In ‘‘Sex Chromosomes and Sex-Determining Genes’’ (K. C. Reed and J. A. M. Graves, Eds.), pp. 357– 372. Harwood Academic, Chur, Switzerland. Burgoyne, P. S., Levy, E. R., and McLaren, A. (1986). Spermatogenic failure in male mice lacking H-Y antigen. Nature 320, 170–172. Capel, B., Rasberry, C., Dyson, J., Bishop, C. E., Simpson, E., Vivian, N., Lovell-Badge, R., Rastan, S., and Cattanach, B. M. (1993). Deletion of Y chromosome sequences located outside the testis determining region can cause XY female sex reversal. Nature Genet. 5, 301–307. Cattanach, B. M., Pollard, C. E., and Hawkes, S. G. (1971). Sex reversed mice: XX and XO males. Cytogenetics 10, 318–337. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J., and Rutter, W. J. (1979). Isolation of biologically active ribonucleic acid from sources enriched with ribonuclease. Biochemistry 18, 5294– 5299. Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156–159. Dresser, D. W., Hacker, A., Lovell-Badge, R., and Guerrier, D. (1995). The genes for a spliceosome protein (SAP62) and the antiMu¨llerian hormone (AMH) are contiguous. Hum. Mol. Genet. 4, 1613–1618. Glotzer, M., Murray, A. W., and Kirschner, M. W. (1991). Cyclin is degraded by the ubiquitin pathway. Nature 349, 132–138. Grenfell, S. J., Trausch-Azar, J. S., Handley-Gearhart, Ciechanover, A., and Schwartz, A. L. (1994). Nuclear localization of the ubiquitin-activating enzyme, E1, is cell-cycle-dependent. Biochem. J. 300, 701–708. Gubbay, J., Vivian, N., Economou, A., Jackson, D., Goodfellow, P., and Lovell-Badge, R. (1992). Inverted repeat structure of the Sry locus in mice. Proc. Natl. Acad. Sci. USA 89, 7953–7957. Hacker, A., Capel, B., Goodfellow, P., and Lovell-Badge, R. (1995).

Expression of Sry, the mouse sex determining gene. Development 121, 1603–1614. Handley-Gearhart, P. M., Stephen, A. G., Trausch-Azar, J. S., Ciechanover, A., and Schwartz, A. L. (1994a). Human ubiquitin-activating enzyme E1: Indication of potential nuclear and cytoplasmic subpopulations using epitope-tagged cDNA constructs. J. Biol. Chem. 269, 33171–33178. Handley-Gearhart, P. M., Trausch-Azar, J. S., Ciechanover, A., and Schwartz, A. L. (1994b). Rescue of the complex temperature-sensitive phenotype of Chinese hamster ovary E36ts20 cells by expression of the human ubiquitin-activating enzyme cDNA. Biochem. J. 304, 1015–1020. Hendriksen, P. J. M., Hoogerbrugge, J. W., Themmen, A. P. N., Koken, M. H. M., Hoeijmakers, J. H. J., Ooostra, B. A., van der Lende, T., and Grootegoed, J. A. (1995). Postmeiotic transcription of X and Y chromosomal genes during spermatogenesis in the mouse. Dev. Biol. 170, 730–733. Imai, N., Kaneda, S., Nagai, Y., Seno, T., Ayusawa, D., Hanaoka, F., and Yamao, F. (1992). Cloning and sequencing of a functionally active cDNA encoding the mouse ubiquitin-activating enzyme E1. Gene 118, 279–282. Kay, G. F., Ashworth, A., Penny, G. D., Dunlop, M., Swift, S., Brockdorff, N., and Rastan, S. (1991). A candidate spermatogenesis gene on the mouse Y chromosome is homologous to ubiquitin-activating enzyme E1. Nature 354, 486–489. Krieg, P. A., and Melton, D. A. (1987). In vitro RNA synthesis with SP6 RNA polymerase. Methods Enzymol. 155, 397–416. Kumari, M., Stroud, J. C., Anji, A., and McCarrey, J. R. (1996). Differential appearance of DNase I-hypersensitive sites correlates with differential transcription of Pgk genes during spermatogenesis in the mouse. J. Biol. Chem. 271, 14390–14397. Laval, S. H., Glenister, P. H., Rasberry, C., Thornton, C. E., Mahadevaiah, S. K., Cooke, H. J., Burgoyne, P. S., and Cattanach, B. M. (1995). Y chromosome short arm-Sxr recombination in XSxr/Y males causes deletion of Rbm and XY female sex reversal. Proc. Natl. Acad. Sci. USA 92, 10403–10407. Lyon, M. F., Cattanach, B. M., and Charlton, H. M. (1981). Genes affecting sex differentiation in mammals. In ‘‘Mechanisms of Sex Differentiation in Animals and Man’’ (C. R. Austin and R. G. Edwards, Eds.), pp. 329–386. Academic Press, New York. Mahadevaiah, S. K., Lovell-Badge, R., and Burgoyne, P. S. (1992). Tdy-negative XY, XXY and XYY female mice: breeding data and synaptonemal complex analysis. J. Reprod. Fertil. 97, 151–160. McKee, B. D., and Handel, M. A. (1993). Sex chromosomes, recombination, and chromatin conformation. Chromosoma 102, 71– 80. McLaren, A., Simpson, E., Tomonari, K., Chandler, P., and Hogg, H. (1984). Male sexual differentiation in mice lacking H-Y antigen. Nature 312, 552–555. Meistrich, M. L. (1977). Separation of spermatogenic cells and nuclei from rodent testes. In ‘‘Methods in Cell Biology’’ (D. M. Prescott, Ed.), pp. 15–54. Academic Press, New York. Minty, A. J., Caravatti, M., Robert, B., Cohen, A., Daubas, P., Weydert, A., Gros, F., and Buckingham, M. E. (1981). Mouse actin mRNAs. Construction and characterization of a recombinant plasmid molecule containing a complementary DNA transcript of mouse a-actin mRNA. J. Biol. Chem. 256, 1008–1014. Mita, S., Yasuda, H., Marunouchi, T., Ishiko, S., and Yamada, M. (1980). A temperature-sensitive mutant of cultured mouse cells defective in chromosome condensation. Exp. Cell Res. 126, 407– 416. Mitchell, M. J., Woods, D. R., Tucker, P. K., Opp, J. S., and Bishop,

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID

DB 8397

/

6x16$$$$84

10-29-96 00:40:04

dba

AP: Dev Bio

343

Ube1y and Ube1x Transcription in the Mouse Testis

C. E. (1991). Homology of a candidate spermatogenic gene from the mouse Y chromosome to the ubiquitin-activating enzyme E1. Nature 354, 483–486. Mitchell, M. J., Woods, D. R., Wilcox, S. A., Graves, J. A. M., and Bishop, C. E. (1992). Marsupial Y chromosome encodes a homologue of the mouse Y-linked candidate spermatogenesis gene Ube1y. Nature 359, 528–531. Mitchell, M. J., and Bishop, C. E. (1992). A structural analysis of the Sxr region of the mouse Y chromosome. Genomics 12, 26– 34. Mitchell, M. J., Scheffler, J., Hearn, J., and Bishop, C. (1996). Ube1y homologues on the primate Y chromosome. Cytogenet. Cell Genet. 73, 74. Monesi, V. (1965). Synthetic activities during spermatogenesis in the mouse. Exp. Cell Res. 39, 197–224. Nagai, Y., Kaneda, S., Nomura, K., Yasuda, H., Seno, T., and Yamao, F. (1995). Ubiquitin-activating enzyme, E1, is phosphorylated in mammalian cells by the protein kinase Cdc2. J. Cell Sci. 108, 2145–2152.

Simpson, E. M., and Page, D. C. (1991). An interstitial deletion in mouse Y chromosomal DNA created a transcribed Zfy fusion gene. Genomics 11, 601–608. Sutcliffe, M. J., and Burgoyne, P. S. (1989). Analysis of the testes of H-Y negative XOSxrb mice suggests that the spermatogenesis gene (Spy) acts during the differentiation of the A spermatogonia. Development 107, 373–380. Zacksenhaus, E., and Sheinin, R. (1990a). Molecular cloning of the human A1S9 locus: An X-linked gene essential for progression through S phase of the cell cycle. Somat. Cell Mol. Genet. 15, 545–553. Zacksenhaus, E., and Sheinin, R. (1990b). Molecular cloning, primary structure and expression of the human X linked A1S9 gene cDNA which complements the ts A1S9 mouse L cell defect in DNA replication. EMBO J. 9(9), 2923–2929. Received for publication May 6, 1996 Accepted September 6, 1996

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID

DB 8397

/

6x16$$$$84

10-29-96 00:40:04

dba

AP: Dev Bio