Early Transcriptional Activation of the Hsp70.1 Gene by Osmotic ...

BIOLOGY OF REPRODUCTION 70, 1606–1613 (2004) Published online before print 6 February 2004. DOI 10.1095/biolreprod.103.024877

Early Transcriptional Activation of the Hsp70.1 Gene by Osmotic Stress in One-Cell Embryos of the Mouse1 Maria Teresa Fiorenza, Arturo Bevilacqua, Sonia Canterini, Simona Torcia, Marco Pontecorvi, and Franco Mangia2 Istituto Pasteur—Fondazione Cenci Bolognetti and Department of Psychology, Section of Neuroscience, University La Sapienza of Rome, 00185 Rome, Italy ABSTRACT In fertilized mouse eggs, de novo transcription of embryonic genes is first observed during the S phase of the one-cell stage. This transcription, however, is mostly limited to the male pronucleus and possibly uncoupled from translation, making the functional meaning obscure. We found that one-cell mouse embryos respond to the osmotic shock of in vitro isolation with migration of HSF1, the canonical stress activator of mammalian heat shock genes, to pronuclei and by transient transcription of the hsp70.1, but not hsp70.3 and hsp90, heat shock genes. Isolated growing dictyate oocytes also display a nuclear HSF1 localization, but, in contrast with embryos, they transcribe both hsp70.1 and hsp70.3 genes only after heat shock. Intranuclear injection of double-stranded oligodeoxyribonucleotides containing HSE, GAGA box or GC box consensus sequences, and antibodies raised to transcription factors HSF1, HSF2, Drosophila melanogaster GAGA factor, or Sp1 demonstrated that hsp70.1 transcription depends on HSF1 in both oocytes and embryos and that Sp1 is dispensable in oocytes and inhibitory in the embryos. Hsp70.1 thus represents the first endogenous gene so far identified to be physiologically activated and tightly regulated after fertilization in mammals.

developmental biology, early development, embryo, gene regulation, oocyte development

INTRODUCTION

Fertilization in mammals activates a developmental clock that leads the newly formed embryo to progressively shift from the maternal to the zygotic genome expression (zygotic genome activation, ZGA). In the mouse, this process is initiated very soon after fertilization and de novo synthesis of RNA is consistently observed for the first time during the S phase of the one-cell stage [1, 2]. Mouse zygotes, in fact, are transcriptionally competent and their transcriptional machinery, including RNA polymerases I, II, and III [3], is functional [2, 4–7]. Transcription at this stage, however, has several peculiar features that make it very different from that which will occur after the first embryo cleavage. In fact, it takes place while the zygotic genome has not yet been assembled into chromatin, as originally Supported by grants from Istituto Pasteur—Fondazione Cenci Bolognetti and COFIN 2002 to F.M. and Telethon GP0139Y01 to A.B. Correspondence: Franco Mangia, Department of Psychology, Section of Neuroscience, Via dei Sardi 70, 00185 Roma, Italy. FAX: 39 06 4991 7872; e-mail: [email protected]

1

2

Received: 30 October 2003. First decision: 1 December 2003. Accepted: 26 January 2004. Q 2004 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org

indicated by the finding that plasmids injected in the male pronucleus of one-cell embryos are transcribed irrespective of the presence of an enhancer [8], unless the DNA is assembled into chromatin by injection of purified H1 histones [9]. In addition, transcription is more prominent in the male pronucleus than in the female one [2, 10], in agreement with the higher levels of hyperacetylated H4 histones [11] and DNA demethylase activity [12] present in this pronucleus. It is also temporally and causally linked to DNA replication, appearing at the onset of the first embryonic S phase and being strongly decreased by inhibition of DNA synthesis with aphidicolin [2]. An intriguing feature of onecell embryo transcription is that it is possibly uncoupled from translation until the first embryonic G2 phase [13, 14], while it does not appear that nascent RNA chains are rapidly degraded and/or display splicing deficiencies [15, 16]. Even though such transcription-translation uncoupling does not have a known molecular basis, it raises the question of the functional significance of the one-cell embryo’s transcriptional activity. In this regard, it has been proposed that the expression of one-cell embryo genes is opportunistic and promiscuous [17, 18] and, thus, that it is merely dependent on unregulated transcription factor accessibility to open male gene promoters during protamine-histone exchange. During the G2 phase of the one-cell stage and/or the transition to the two-cell stage, the male and female genomes equally undergo a process of generalized assembly into chromatin, concomitant with the de novo synthesis of histones H2A, H2B, and H1 [19] and extensive histone H4 deacetylation [11], which eventually results in the appearance of novel locally acting mechanism(s) of chromatin remodeling at the early two-cell stage [20]. At this stage, in fact, expression of plasmids driven by a weak promoter requires the presence of an enhancer, unless the embryo chromatin is opened by histone deacetylase inhibitors [2, 21, 22]. The process of zygotic genome packaging into chromatin then continues through the two-cell stage and is accompanied by the progressive appearance of more and more stringent transcriptional regulation mechanism(s) [18], that is virtually completed by the four-cell stage. The early two-cell stage, however, marks the onset of diploid embryo genome activation with the transcription of a limited set of zygotic genes. Some of these genes, including the transcription-requiring complex [23], the heat shock genes hsp70.1 [14, 24] and hsp25[25], the translation initiation factor eIF-2A [26], and a few others, are expressed transiently, whereas other genes, including the RING finger protein genes Rnf33 and Rnf35 [18] and the Mhc gene [27], continue to be expressed during subsequent early embryo development. Among the early and transiently activated genes, previous work performed in our laboratory was fo-

1606

REGULATION OF hsp70.1 IN ONE-CELL MOUSE EMBRYOS

cused on the transcriptional regulation of hsp70.1 in the early two-cell embryos. We demonstrated that the spontaneous expression of this gene at this developmental stage is driven by transcription factor Sp1 and a murine GAGA box-binding factor (mGAF), whereas heat shock factor 1 (HSF1) is likely to be responsible for a stress-elicited hyperactivation of the endogenous hsp70.1 genes in the in vitro-cultured embryos [20]. Heat shock-dependent activation of hsp70 was also reported in two-cell bovine embryos [28]. In contrast with the information presently available on embryo transcriptional regulation taking place after the zygotic genome has been assembled into chromatin, molecular mechanisms that regulate the transient transcriptional wave occurring in the one-cell embryos are still unknown. In this article, we have addressed this issue by taking advantage of the unexpected finding that, in addition to the spontaneous expression at the early two-cell stage, the hsp70.1 gene is also transiently transcribed in the in vitrocultured, but not the in vivo-maintained, one-cell mouse embryo. We have compared the molecular regulation of hsp70.1 in the one-cell embryos with that occurring in heatshocked dictyate oocytes derived from preantral ovarian follicles [29] and report here that the hsp70.1 activation in one-cell embryos is mediated by HSF1 and is triggered by the osmotic stress to which embryos are subject in consequence of isolation from the tubal fluid environment. As far as we know, hsp70.1 thus represents the earliest endogenous gene physiologically expressed after fertilization that has been identified in mammals. The tight regulation of hsp70.1 at the one-cell stage indicates that stringent transcriptional regulation mechanisms are already present very soon after fertilization. MATERIALS AND METHODS

Animals Growing and full-grown ovarian oocytes at the dictyate stage of meiosis were obtained from ovaries of 12- to 14-day old and of eCG-treated, 40- to 60-day old Swiss-CD1 female mice, respectively. Preimplantation embryos were routinely obtained by crossing eCG-hCG-primed B6D2F1 females with B6D2F1 males. All mice were from Charles River Italia (Calco, Italy) or bred from their stocks in our colony. Experimental protocols and procedures were approved by the Ministry of Public Health, Rome, Italy. Animals were treated in accordance with La Sapienza University Guidelines for the Care and Use of Laboratory Animals.

Oocyte and Embryo Isolation and Heat Shock Growing/full grown ovarian oocytes at the dictyate stage of meiosis were isolated from surrounding granulosa/cumulus cells by pricking preantral/antral ovarian follicles with a needle in the presence of medium M2 and were then transferred to drops of medium M16 [30] under paraffin oil with an atmosphere of 5% CO2 in air. M2 and M16 media were supplemented with 2 mg/ml bovine serum albumin (BSA) and, in the case of maturation competent oocytes, with 200 mM dibutyril cAMP (M2/M16cAMP). One-cell embryos were routinely collected from the tubes of mated females 11–12 h postfertilization (p.f.), taking midnight after mating as fertilization time, and deprived of surrounding cumulus cells with hyaluronidase. Cell cycle phases were determined by relative pronuclei distance [11]. For the heat shock, isolated oocytes/embryos were transferred to 0.5 ml of M2 in a plastic tube, incubated at 438C for 30 min in a precision water bath. and then further transferred at 378C for 30 min in M16 to allow cell recovery and transcription of heat shock genes [29].

Determination and Modification of Tubal Fluid Osmolarity and Embryo Culture in the Presence of Normal/Experimentally Modified Tubal Fluid Tubal fluid osmolarity was determined on fluids obtained from 25 superovulated and mated female mice. To this purpose, dissected oviducts

1607

were submerged into paraffin oil and then torn with a sharp needle to release their tubal fluid-embryo content. Tubal fluids from different tubes were aspirated by a mouth-controlled micropipette, pooled, and centrifuged to remove cells and debris, obtaining a total volume of 35 ml. For osmolarity determination, the 35-ml volume was diluted to 70 ml with embryo-quality water (Sigma, St. Louis, MO) and then assayed using a 13/13DR-Autocal freezing-point microsmometer (Roebling, Berlin, Germany). In the experiments of in vitro culture, embryos were constantly maintained within their intact cumulus and in the presence of their own tubal fluid, with the only addition of fluid from other tubes, as needed to bring the total drop volume to 5 ml (determined by visual comparison with a 5ml drop of medium). Under this condition, one-cell embryos developed normally to the two-cell stage, but did not cleave to four cells, suggesting that the tubal fluid of one-cell embryos is not able to support the in vitro development beyond the two-cell stage. It is still to be determined whether this is due to an actual change in tubal fluid composition during embryo development and progression through the tube and/or a tubal-fluid alteration caused by the fluid maintenance in vitro under oil for more than 24 h.

Intranuclear Microinjection of Double-Stranded Oligodeoxyribonucleotides and Antibodies and Reverse Transcription-Polymerase Chain Reaction Assays Intranuclear microinjections of double-stranded oligodeoxyribonucleotides and antibodies were performed as previously described [20]. Briefly, double-stranded oligodeoxyribonucleotides carrying appropriate consensus sequences were dissolved in 10 mM Tris, 0.1 mM EDTA, pH 7.4 (TE), and injected into the germinal vesicle of growing ovarian oocytes or the male pronucleus of one-cell embryos, as previously described [20]. Antibodies were first microdialyzed against TE through MF filters (Millipore, Rome, Italy) and diluted to a final concentration of approximately 250 ng/ ml. Rabbit polyclonal antisera raised to recombinant mouse HSF1 and HSF2 polypeptides and preimmune sera [31] and to Drosophila melanogaster GAGA factor (dGAF) were generous gift of Drs. V. Zimarino (DIBIT-San Raffaele Scientific Institute, Milan, Italy) and C. Wu (National Cancer Institute, National Institutes of Health, Bethesda, MD), respectively. FITC-conjugated anti-rabbit IgG were from Sigma (Sigma-Aldrich, Milano, Italy). For determinations of qualitative hsp70.1, hsp70.3, hsp90, and S16 mRNAs in oocytes and one-cell embryos, reverse transcription-polymerase chain reaction (RT-PCR) amplifications were routinely performed on total RNA extracted from pools of 80 oocytes or embryos by using the TRIzol reagent (Invitrogen, Milano, Italy) and adding 1 mg/tube glycogen (Roche Diagnostic, Monza, Italy) as carrier. DNase I-treated RNA was reverse transcribed by using SuperScript RNase H2 reverse transcriptase and random hexamer primers (Invitrogen) according to manufacturer’s instructions. PCR amplifications were performed using the Advantage 2 Polymerase Mix (BD Clontech, Milano, Italy) and 5 mCi [a-32P]dCTP (NEN DuPont Italiana, Cologno Monzese, Italy) as tracer, with a first denaturation step at 958C for 2 min and 30 cycles of the following steps: denaturation at 958C for 30 sec, annealing at 588C for 30 sec, and elongation at 688C for 55 sec. Primer pairs were hsp70.1 (amplification fragment, 291 base pairs [bp]), 59-TGCTTGGGCACCGATTACTGTCAAGG-39 (2793– 2818 nucleotides [nt], sense), 59-GGCAGCTAGACTATATGTCTTCCC AGGCTACTG-39 (3053–3084 nt, antisense); hsp70.3 (amplification fragment, 230 bp), 59-AGATATGTGGCCTTGAGGACTGTCATTATTTC-39 (531–562 nt, sense), 59-CAAATCACATCAGCGGGGCAGTGCTGAAT TG-39 (731–761 nt, antisense), according to Dix et al. [32] with minor modifications; hsp90 (amplification fragment, 304 bp), 59-CATCAAGTTGTATGTTCGC-39 (1343–1361 nt, sense), 59-GCTCTGAAAGCT TCTTCCG-39 (1629–1647 nt, antisense); S16 (amplification fragment, 103 bp), 59-AGGAGCGATTTGCTGGTGTGG-39 (1451–1471 nt, sense), 59-GCTACCAGGGCCTTTGAGATGGA-39 (1621–1641 nt, antisense). In the case of semiquantitative hsp70.1 determinations during one-cell embryo development and in injected oocytes and embryos, RT-PCR reactions were performed on groups of five oocytes or embryos, taking the S16 ribosomal protein mRNA as internal standard, as previously described [33]. The rationale for using the S16 mRNA as internal standard comes from the finding that the cellular amount of this message progressively increases during early oocyte growth according to the developmental pattern of ribosomal protein synthesis during mouse oogenesis [34] but does not significantly vary in late growing, full grown, and maturing oocytes and in the early cleaving embryos, as previously described [33]. Under our experimental conditions, S16 mRNA was amplified linearly for at least 40 PCR cycles [33].

1608

FIORENZA ET AL.

FIG. 1. HSF1 and HSF2 immunolocalization in representative mid-one-cell embryos isolated and cultured in vitro under different conditions, as described in the text. Similar features were displayed by the 20– 30 embryos examined for each experimental condition. Immunolocalization was performed by laser scanning confocal microscopy. A) HSF1 immunolocalization: embryos were fixed within the Fallopian tubes. Bar 5 25 mm. B) HSF1 immunolocalization: isolated embryos were cultured in vitro in the presence of M16 medium. C) HSF1 immunolocalization: isolated embryos were cultured in vitro in the presence of undiluted tubal fluid. D) HSF1 immunolocalization: isolated embryos were cultured in vitro in the presence of M16 medium supplemented with 20% fetal calf serum. E) HSF1 immunolocalization: isolated embryos were cultured in vitro in the presence of tubal fluid made hypertonic by addition of NaCl. F) HSF1 immunolocalization: isolated embryos were cultured in vitro in the presence of tubal fluid made hypotonic by addition of H2O. G) HSF2 immunolocalization: isolated embryos were fixed within the Fallopian tubes. H) HSF2 immunolocalization: isolated embryos were cultured in vitro in the presence of M16 medium.

Immunolocalization All incubations were routinely performed at room temperature. Oocytes and embryos were rapidly deprived of the zona pellucida by a brief treatment with acidic Tyrode solution [35] and then fixed with 2.4% paraformaldehyde in M2 without BSA for 1 h. Fixed cells were incubated for 3 h in PBS containing 0.1 M glycine and 0.3 mg/ml BSA and then permeabilized with PBS containing 0.1% Triton X-100 for 15 min, washed in PBS supplemented with 1 mg/ml BSA (PBS-BSA), and eventually processed for immunostaining. Antibodies were dissolved in PBS-BSA. Incubation with the first antibody was carried out overnight. Cells were then thoroughly washed with PBS-BSA and incubated for 1 h in the presence of the secondary antibody. Rabbit polyclonal antisera to mouse HSF1 and HSF2 were diluted 1:500 and FITC-conjugated anti-rabbit IgGs were diluted 1:200.

RESULTS

Mid-One-Cell Embryos Respond to Isolation from the In Vivo Environment by Nuclear Migration of HSF1

Studying the intracellular localization of heat shock factors HSF1 and HSF2 during mouse preimplantation development, we found that, when mid-one-cell embryos were fixed within the Fallopian tubes, HSF1 had a cytoplasmic localization and was fully absent from pronuclei (Fig. 1A), as expected for unstressed mammalian cells [36, 37]. Unexpectedly, however, we also noticed that, if the embryos were isolated from their tubal environment and then maintained in vitro for 20–30 min under standard culture conditions (i.e., drops of BSA-supplemented medium under paraffin oil with an atmosphere of 5% CO2 in air) before fixation, HSF1 was sharply localized to both pronuclei, ir-

respectively of the cell cycle phase at which the isolation had been done (Fig. 1B). A similar phenomenon was also observed in two-cell embryos (data not shown). Because this finding suggested that the in vitro isolated embryos had somehow undergone a stress, we investigated the nature of this stress by different experimental conditions. We first evaluated the effect of a mild oxidative stress, by tearing the ampullae in paraffin oil previously equilibrated with 5% CO2 in air with no addition of culture medium (as described under Materials and Methods), thus allowing the embryos to escape from the tubes while maintained in the presence of their cumulus and undiluted tubal fluid. The dish was then incubated at 378C for 1 h under an atmosphere of 5% CO2 in air (containing approximately 21% O2). At the end of this period, no nuclear HSF1 localization was observed (Fig. 1C), ruling out the possibility that HSF1 mobilization to pronuclei was triggered by embryo exposure to the air O2 concentration. To determine whether the stress was due to the lack of a factor(s) present in the tubal fluid, we isolated the embryos in a drop of M16 supplemented with 20% fetal calf serum and then incubated them in vitro at 378C for 1 h. This condition, however, did not prevent the pronucleus shuttling of HSF1 (Fig. 1D), making it unlikely that the stress was due to the lack of a tubal factor(s) also present in the serum. We finally evaluated the effect of embryo exposure to a slight variation in tubal fluid osmolarity. To this aim, we first determined that the normal osmolarity of mouse tubal fluid (a pool of tubal fluids obtained from different animals,


1609

FIG. 2. Expression profile of hsp70.1 mRNA in one-cell embryos during a developmental interval encompassing the S and the early G2 phases. A) In vivo: one-cell embryos were isolated from the tubes and immediately processed for RT-PCR assays at the times p.f. indicated above lanes; in vitro: onecell embryos were isolated, cultured in vitro, and then processed for RT-PCR assays at the times p.f. indicated above lanes; in vitro a-amanitin: onecell embryos were isolated, cultured in vitro in the presence of a-amanitin, and then processed for RT-PCR assays at the times p.f. indicated above lanes. Numbers below lanes indicate hsp70.1/S16 incorporation ratios [20, 31]. Panels indicate cDNA amplification bands obtained in a typical experiment. B) Histograms represent the mean 6 SEM of hsp70.1/S16 incorporation ratios, calculated with at least four samples (each consisting in a pool of five developing embryos) for each developmental time and obtained in three independent experiments. Asterisks above bars indicate a significant difference (P , 0.001) from time 1200 h, determined by ANOVA. C) Expression of hsp70.1, hsp70.3, and hsp90 in isolated growing oocytes and onecell embryos. One-cell embryos were allowed to develop in vivo or in vitro at the times p.f. indicated above lanes and then processed for RT-PCR assays. HS: cells received a heat shock.

as described in Materials and Methods) is 302 6 4 mOsm (mean 6 SEM of three repeated measures of the same sample). This value approximated the 285–292 mOsm values of standard mouse embryo culture media [35]. In the in vitro culture experiments, the embryos were isolated and constantly maintained under oil in the presence of their own tubal fluid (brought to 5 ml by addition of fluid from other tubes) and within their intact cumulus, as described in Materials and Methods. Tubal fluid osmolarity was then modified by addition of 0.5 ml of embryo-quality water or 300 mM NaCl in water, causing a final hypo/hypertonic variation of approximately 10% (i.e., approximately 270 and 330 mOsm, respectively). Embryos were then cultured in vitro for 1 h and eventually assayed for HSF1 immunolocalization as described above. Both H2O and NaCl additions consistently elicited a sharp HSF1 mobilization to pronuclei (Fig. 1, E and F), ruling out the possibility that the HSF1 mobilization was triggered by embryo isolation from its cumulus and indicating that one-cell mouse embryos are extremely sensitive to an even small change in the osmotic environment per se. In contrast with HSF1, HSF2 was constantly localized to pronuclei irrespectively of the embryo culture in the presence of tubal fluid or M16 (Fig. 1, G and H) and under all experimental conditions described above (data not shown). In Vitro-Cultured One-Cell Embryos Undergo a Transient, HSF1-Dependent Transcription of the hsp70.1 Gene, But Not of hsp70.3 and hsp90

We then investigated whether the HSF1 nuclear mobilization described above was accompanied by transcriptional activation of stress-inducible heat shock genes, including hsp70.1, hsp70.3, and hsp90. In a first series of experiments we compared the levels of hsp70.1 mRNA in one-cell embryos cultured in vitro and maintained in vivo through a

developmental interval encompassing the S and early G2 phases [11] (Fig. 2A). In vitro-cultured embryos displayed a transient, 3.5-fold increase in hsp70.1 mRNA corresponding to the S phase and maximizing at 1530 h p.f., which was fully suppressed by a-amanitin and thus depended on de novo transcription (Fig. 2B). Similar results were also observed in one-cell embryos deprived of the female pronucleus but not in those deprived of the male pronucleus (data not shown), showing that hsp70.1 transcription occurred, for the most part, if not totally, in the male pronucleus. In contrast with hsp70.1, levels of hsp70.3 and hsp90 mRNAs were not affected by embryo isolation in vitro nor by the administration of a heat shock (Fig. 2C), showing full refractoriness of these genes to stress activation. The finding that hsp70.1 was transiently transcribed in in vitro-cultured, one-cell embryos provided an opportunity to determine how this gene is regulated before its spontaneous activation at the early two-cell stage. To this end, we intranuclearly microinjected double-stranded oligodeoxyribonucleotides containing HSE, GAGA box, or GC box consensus sequences (Fig. 3, A and B) or antibodies raised to transcription factors HSF1, HSF2, dGAF, or Sp1 (Fig. 3, C and D), as previously described [20]. Embryos were then incubated for 2 h at 378C and eventually processed for RTPCR determination of their hsp70.1 mRNA content. HSE oligodeoxyribonucleotides reduced the amount of hsp70.1 mRNA to approximately 45% that of noninjected embryos, whereas mutated HSE oligodeoxyribonucleotides had no effect, ruling out the possibility of a nonspecific effect of the injection. No effect was observed with GAGA box oligodeoxyribonucleotides, whereas those carrying the GC box elicited a significant 25–30% increase in hsp70.1 mRNA. As for antibodies, anti-HSF1 antibodies significantly reduced the hsp70.1 mRNA content to approximately 55% of control embryos, whereas anti-HSF2 antibodies

1610

FIORENZA ET AL.

FIG. 3. RT-PCR amplification assay of the hsp70.1 mRNA of mid-one-cell embryos, following the intranuclear injection of competing amounts of double-stranded oligodeoxyribonucleotides or antibodies. A) One-cell embryos received the injection of plain buffer (buffer) or saturating amounts of double-stranded GAGA box, HSE, GC box, mutated HSE (mutHSE), or mutated GC box (mutGC) oligodeoxyribonucleotides. Embryos were then cultured for 2 h and eventually processed for RT-PCR assay. Noninjected embryos were either immediately processed for RT-PCR assay (t0) or cultured for 2 h and then processed for the RT-PCR assay (noninj.). The panel shows a typical experiment. Numbers below lanes indicate hsp70.1/S16 incorporation ratios. B) Histograms represent the mean 6 SEM of hsp70.1/S16 incorporation ratios, calculated with at least six samples (each consisting in a pool of five injected/noninjected embryos) for each treatment and obtained in three independent experiments. Asterisks above bars indicate significant difference (** P , 0.001; * P , 0.05) from noninjected/buffer injected embryos, determined by ANOVA. C) Embryos received the injection of a 1/5 diluted solution of antibodies to HSF1 (aHSF1), HSF2 (aHSF2), Sp1 (aSp1), dGAF (adGAGA), or Sp1 preadsorbed with the antigenic peptide (preads Sp1) and were then treated as in A. The panel shows a typical experiment. D) Histograms represent the mean 6 SEM of hsp70.1/S16 incorporation ratios, calculated with at least six samples (each consisting in a pool of five injected/noninjected embryos) for each treatment and obtained in three independent experiments. Asterisks above bars indicate significant difference (** P , 0.001; * P , 0.05) from noninjected/buffer-injected embryos, determined by ANOVA.

had no effect and anti-Sp1 antibodies significantly increased the amount of hsp70.1 mRNA, in agreement with results obtained with oligonucleotides. We therefore conclude that transient hsp70.1 transcription of in vitro-cultured, one-cell embryos depends on HSF1, but not on HSF2 or GAGA factor, while Sp1 is apparently inhibitory. Isolated Growing Oocytes Display Nuclear HSF1 Localization and Canonical Activation of Both hsp70.1 and hsp70.3 Following Heat Shock

Because the dependence of hsp70.1 transcription on HSF1 indicated that one-cell embryos have an osmotic stress response, but not the ability to activate hsp70.3 following a heat shock, it was of interest to compare features of one-cell embryos with the canonical heat shock response of growing dictyate oocytes [29]. Interestingly, growing oo-

cytes isolated from their follicles displayed a nuclear HSF1 shuttling similar to that of isolated one-cell embryos (Fig. 4, A and B). The same feature was also observed in in vitro-isolated full grown oocytes (data not shown), which lack the heat shock response [38]. In contrast with embryos, no hsp70.1 transcription was ever observed in isolated growing oocytes under normal in vitro culture conditions. However, when the oocytes were subjected to a heat shock, a 3-fold, a-amanitin-sensitive increase in hsp70.1 mRNA content was observed (Fig. 4, C and D). Interestingly, a similar heat shock-dependent transcription was also observed for hsp70.3, but not hsp90 (Fig. 2C). Intranuclear injection of double-stranded HSE oligodeoxyribonucleotides (data not shown) or anti-HSF1 antibodies before the heat shock fully abolished the stress-dependent increase in both hsp70.1 and hsp70.3 mRNAs, whereas GAGA box


1611

and GC box oligodeoxyribonucleotides and anti-Sp1 or anti-HSF2 antibodies had no effect (data not shown). DISCUSSION Since the original observations of BrdU incorporation in the male pronucleus of one-cell mouse embryos [1, 2], the functional meaning of such an early transcriptional activity of a mammalian embryo has been unclear. In fact, RNA molecules synthesized at this stage and their encoded proteins, if any, still have to be identified, and their functional significance is still obscure. In this study, we have addressed this issue by identifying a gene expressed during the first embryonic S phase, the heat shock gene hsp70.1, which also represents one of the earliest zygotic genes spontaneously activated at the early two-cell stage. This has made it possible to directly compare the transcriptional regulation mechanisms acting on the same gene before and after the zygotic genome has been assembled into chromatin. This study is based on the finding that one-cell embryo isolation from the tubes elicits a rapid HSF1 mobilization to both pronuclei, a feature that we also observed at the mid-two-cell stage, whereas HSF2 constantly maintains a nuclear localization. The HSF1 migration to nucleus as a consequence of embryo isolation in vitro is a strong indication that early preimplantation embryos have a stress response. This conclusion is in agreement with the previous finding that oocytes and one-cell/two-cell, but not four-cell, embryos of the mouse contain a strong heat-inducible activity of protein binding to a DNA probe carrying heat shock element consensus sequences [39]. Among various cell signaling pathways that mediate the stress-dependent HSF1 entrance to the nucleus and transcriptional activation in mammalian cells [40–43], it is unlikely that the rho protein Rac1 GTPase, a mediator of the redox-dependent stress response in mammals [44], plays a relevant role in the embryos due to the inability of an increased O2 partial pressure to elicit HSF1 mobilization to pronuclei. However, the possibility of Rac1 GTPase function cannot be ruled out with O2 concentrations higher than were tested here. In contrast with oxidative stress, embryos were extremely sensitive to an osmotic shock as small as a 610% variation in medium osmolarity and this likely reflects a need of early embryos for protection from direct exposure to the tubal environment when the surrounding cumulus envelope is progressively lost. In this context, it is not surprising that hsp70.1 was transcriptionally activated in in vitro-cultured embryos, whereas hsp70.3 was fully silent and refractory even to a heat shock. Among the HSP70 heat shock gene family [45], the hsp70.1 [46] and hsp70.3 [47] genes are typically inducible by stress in adult somatic tissues. These genes are located in a tandem array on the major histocompatibility complex region of the mouse chromosome 17 [48, 49] and code for two proteins differing at only two residues, suggesting a very similar function in stressed cells. In spite of their similar structure and responsiveness to heat shock, however, hsp70.1 and hsp70.3 respond differently to a number of stresses. In fact, hsp70.1 is activated by oxidative stress and hyperosmotic shock in somatic cell lines, whereas hsp70.3 is not [50]. Accordingly, hsp70.1-knockout mice are sensitized to osmotic shock and exhibit increased apoptosis in the renal medulla following administration of an osmotic stress in vivo [51]. Hsp70.1 thus has a pivotal role in mouse stress response and is the major cell protector against osmotic stress in the mouse. We therefore conclude that mouse one-cell and two-cell embryos are extremely

FIG. 4. Nuclear HSF1 mobilization and hsp70.1 expression in growing dictyate oocytes. Similar features were displayed by the 20–30 oocytes examined for each experimental condition. A) Growing dictyate oocytes were fixed during the isolation procedure and were then processed for HSF1 immunolocalization. B) Growing dictyate oocytes were isolated from their follicle, then cultured in vitro for 1 h, and eventually processed for HSF1 immunolocalization. Bar 5 25 mm. C) Growing dictyate oocytes received the injection of saturating amounts of antibodies to HSF1 (aHSF1) or HSF2 (aHSF2). Oocytes were then subject to heat shock, allowed to recover at 378C in the absence or presence of a-amanitin (aaman), and eventually processed for RT-PCR assays. Control oocytes were constantly maintained at 378C. The panel shows a typical experiment. Numbers below lanes indicate hsp70.1/S16 incorporation ratios. D) Histograms represent the mean 6 SEM of hsp70.1/S16 incorporation ratios, calculated with at least six samples (each consisting in a pool of five developing embryos) for each treatment and obtained in three independent experiments. Asterisk above bars indicates significant difference (P , 0.001) from 438C, determined by ANOVA.

1612

FIORENZA ET AL.

sensitive to their osmotic environment and respond to an even small osmotic variation by HSF1 migration to the nucleus and specific activation of hsp70.1, but not of other stress-inducible heat shock genes. Even though the transcriptional activation of hsp70.1 by osmotic stress in somatic tissues is not yet understood [50], present experiments with oligodeoxyribonucleotide and antibody microinjection have conclusively shown that hsp70.1 activation in mouse one-cell embryos depends essentially on HSF1. In this context, it is interesting to note that HSF1 is activated by either a hypo- or hyperosmotic stress [51] and that this transcription factor is very relevant to early preimplantation development of the mouse. In fact, mouse embryos lacking maternally derived HSF1 die at the one-cell/twocell stage in spite of their ability to spontaneously activate the hsp70.1 gene at the early two-cell stage [52]. Actual HSF1 function(s) in mammalian preimplantation embryos other than mouse has (have) not been characterized yet [53]. A deeper insight on hsp70.1 transcriptional regulation was also provided by comparing in vitro-cultured one-cell embryos with heat-shocked growing oocytes, a cell type having a canonical heat shock response [29]. Results obtained in our experiments showed that these oocytes respond to the isolation from their follicle by a nuclear HSF1 migration similar to that of embryos, even though it was not accompanied by transcriptional activation of inducible heat shock genes unless the cells were preliminarily subject to heat shock. In this case, however, both hsp70.1 and hsp70.3 were activated, pinpointing the specificity of the osmotic stress response of one-cell embryos. As for transcription factors that mediate stress responses of one-cell embryos and growing oocytes, HSF1 was the essential mediator of hsp70.1 activation and mGAF was dispensable in both cell types. Sp1, however, was fully dispensable in growing oocytes and played an apparent inhibitory role in one-cell embryos. Sp1 plays a pivotal role as a positive transcriptional regulator during ZGA in the mouse [20]. The inhibitory role of Sp1 and the dispensability of mGAF in one-cell embryos represent the first indication that transcription factors relevant to zygotic gene activation undergo an activation process with zygotic genome assembly into chromatin. Sp1 is a highly phosphorylated and glycosylated protein [54] and may act as either positive or negative regulator of gene expression [55]. Even though the mechanism(s) of opposite Sp1 roles is (are) still unclear, it was shown that O-GlcN acylation of the Sp1 activation domain blocks the factor’s ability to homo-/heterodimerize with other Sp1 molecules and/or the TATA-binding protein-associated factor II 110 [56]. Western-blot analysis of Sp1 electrophoretic mobility in growing oocytes and one-cell versus two-cell embryos (data not shown) showed similar mobility, making unlikely the possibility of factor regulation by phosphorylation. We therefore favor the possibility that Sp1 activation during early embryo development is mediated by heterodimerization with another transcription factor(s) such as mGAF [20]. ACKNOWLEDGMENTS We thank Dr. Robert P. Erickson for critical reading of the manuscript and Drs. Vincenzo Zimarino and Carl Wu for the generous gift of antibodies.

REFERENCES 1. Bouniol C, Nguyen EPD. Endogenous transcription occurs at the 1cell stage in the mouse embryo. Exp Cell Res 1995; 218:57–62.

2. Aoki F, Worrad DM, Schultz RM. Regulation of transcriptional activity during the first and second cell cycles in the preimplantation mouse embryos. Dev Biol 1997; 181:296–307. 3. Nothias J-Y, Miranda M, DePamphilis ML. Uncoupling of transcription and translation during zygotic gene activation in the mouse. EMBO J 1996; 15:5715–5725. 4. Telford NA, Watson AJ, Schultz GA. Transition from the maternal to embryonic control in early mammalian development. Mol Reprod Dev 1990; 26:90–100. 5. Seshagiri PB, McKenzie DI, Bavister BD, Williamson JL, Aiken JM. Golden hamster embryonic genome activation occurs at the two-cell stage: correlation with major developmental changes. Mol Reprod Dev 1992; 32:229–235. 6. Ram PT, Schultz RM. Reporter gene expression in G2 of the 1-cell mouse embryo. Dev Biol 1993; 156:552–556. 7. Schultz RM. Regulation of zygotic gene activation in the mouse. Bioessays 1993; 15:531–538. 8. Martinez-Salas E, Linney E, Hassell J, DePamphilis ML. The need for enhancers in gene expression first appears during mouse development with formation of the zygotic nucleus. Genes Dev 1989; 3: 1493–1506. 9. Rastelli L, Robinson K, Xu Y, Majumder S. Reconstitution of enhancer function in paternal pronuclei of one-cell mouse embryos. Mol Cell Biol 2001; 21:5531–5540. 10. Beaujean N, Bouniol-Baly C, Monod C, Kissa K, Jullien D, Aulner N, Amirand C, Debey P, Kas E. Induction of early transcription in one-cell mouse embryos by microinjection of the nonhistone chromosomal protein HMG-I. Dev Biol 2000; 221:337–354. 11. Adenot PG, Mercier Y, Renard JP, Thompson EM. Differential H4 acetylation of paternal and maternal chromatin precedes DNA replication and differential transcriptional activity in pronuclei of 1-cell mouse embryos. Development 1997; 124:4615–4626. 12. Santos F, Hendrich B, Reik W, Dean W. Dynamic reprogramming of DNA methylation in the early mouse embryo. Dev Biol 2002; 241: 172–182. 13. Matsumoto K, Anzai M, Nagakata N, Takahashi A, Takahashi Y, Miyata K. Onset of paternal gene activation in early mouse embryos fertilized with transgenic mouse sperm. Mol Reprod Dev 1994; 39: 136–140. 14. Christians E, Campion E, Thompson EM, Renard J-P. Expression of the HSP70.1 gene, a landmark of early zygotic activity in the mouse embryo, is restricted to the first burst of transcription. Development 1995; 121:113–122. 15. Latham KE, Solter D, Schultz RM. Acquisition of a transcriptionally permissive state during the 1-cell stage of mouse embryogenesis. Dev Biol 1992; 149:457–462. 16. Schultz RM. The molecular foundations of the maternal to zygotic transition in the preimplantation embryo. Hum Reprod Update 2002; 8:323–331. 17. Ma J, Svoboda P, Schultz RM, Stein P. Regulation of zygotic gene activation in the preimplantation mouse embryo: global activation and repression of gene expression. Biol Reprod 2001; 64:1713–1721. 18. Choo KB, Chen HH, Liu TY, CPC. Different modes of regulation of transcription and premRNA processing of the structurally juxtaposed homologs, Rnf33 and Rnf35, in eggs and in preimplantation embryos. Nucleic Acids Res 2002; 30:4836–4844. 19. Wiekowski M, Miranda M, Nothias JY, DePamphilis ML. Changes in histone synthesis and modification at the beginning of mouse development correlate with the establishment of chromatin mediated repression of transcription. J Cell Sci 1997; 110:1147–1158. 20. Bevilacqua A, Fiorenza MT, Mangia F. A developmentally regulated GAGA box-binding factor and Sp1 are required for transcription of the hsp70.1 gene at the onset of mouse zygotic genome activation. Development 2000; 127:1541–1551. 21. Majumder S, Miranda M, DePamphilis ML. Analysis of gene expression in mouse preimplantation embryos demonstrates that the primary role of enhancers is to relieve repression of promoters. EMBO J 1993; 12:1131–1140. 22. Wiekowski M, Miranda M, DePamphilis ML. Requirements for promoter activity in mouse oocytes and embryos distinguish paternal pronuclei from maternal and zygotic nuclei. Dev Biol 1993; 159:366– 376. 23. Davis WJ, De Sousa PA, Schultz RM. Transient expression of translation initiation factor eIF-4C during the 2-cell stage of the preimplantation mouse embryo: identification by mRNA differential display and the role of DNA replication in zygotic gene activation. Dev Biol 1996; 174:190–201.

REGULATION OF hsp70.1 IN ONE-CELL MOUSE EMBRYOS 24. Bensaude O, Babinet C, Morange M, Jacob F. Heat shock proteins, first major products of zygotic gene activity in mouse embryo. Nature 1983; 305:331–333. 25. Kim M, Geum D, Khang I, Park YM, Kang BM, Lee KA, Kim K. Expression pattern of HSP25 in mouse preimplantation embryo: heat shock responses during oocyte maturation. Mol Reprod Dev 2002; 61: 3–13. 26. Davis W Jr, Schultz RM. Molecular cloning and expression of the mouse translation initiation factor eIF-1A. Nucleic Acids Res 1998; 26:4739–4747. 27. Arcellana-Panlilio MY, Schultz GA. Temporal and spatial expression of major histocompatibility complex class I H-2K in the early mouse embryo. Biol Reprod 1994; 51:169–183. 28. Chandolia RK, Peltier MR, Tian W, Hansen PJ. Transcriptional control of development, protein synthesis, and heat-induced heat shock protein 70 synthesis in 2-cell bovine embryos. Biol Reprod 1999; 61: 1644–1648. 29. Curci A, Bevilacqua A, Fiorenza MT, Mangia F. Developmental regulation of heat-shock response in mouse oogenesis: identification of differentially responsive oocyte classes during Graafian follicle development. Dev Biol 1991; 144:362–368. 30. Whittingham DG. Culture of mouse ova. J Reprod Fertil Suppl 1971; 14:7–21. 31. Fiorenza MT, Farkas T, Dissing M, Kolding D, Zimarino V. Complex expression of murine heat shock transcription factors. Nucleic Acids Res 1995; 23:467–474. 32. Dix DJ, Garges JB, Hong RL. Inhibition of hsp70-1 and hsp70-3 expression disrupts preimplantation embryogenesis and heightens embryo sensitivity to arsenic. Mol Reprod Dev 1998; 51:373–380. 33. Fiorenza MT, Mangia F. Quantitative RT-PCR amplification of RNA in single mouse oocytes and preimplantation embryos. BioTechniques 1998; 24:618–623. 34. LaMarca MJ, Wassarman PM. Program of early development in the mammal: changes in absolute rates of synthesis of ribosomal proteins during oogenesis and early embryogenesis in the mouse. Dev Biol 1979; 73:103–119. 35. Hogan B, Costantini F, Lacy E. Manipulating The mouse Embryo: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory; 1986. 36. Sistonen L, Sarge KD, Phillips B, Abravaya K, Morimoto RI. Activation of heat shock factor 2 during hemin-induced differentiation of human erythroleukemia cells. Mol Cell Biol 1992; 12:2087–2099. 37. Sarge KD, Murphy SP, Morimoto RI. Activation of heat shock gene transcription by heat shock factor 1 involves oligomerization, acquisition of DNA-binding activity, and nuclear localization and can occur in the absence of stress. Mol Cell Biol 1993; 13:1392–1407. 38. Curci A, Bevilacqua A, Mangia F. Lack of heat-shock response in preovulatory mouse oocytes. Dev Biol 1987; 123:154–160. 39. Mezger V, Renard JP, Christians E, Morange M. Detection of heat shock element-binding activities by gel shift assay during mouse preimplantation development. Dev Biol 1994; 165:627–638.

1613

40. Hong Y, Rogers R, Matunis MJ, Mayhew CN, Goodson ML, ParkSarge OK, Sarge KD, Goodson M. Regulation of heat shock transcription factor 1 by stress-induced SUMO-1 modification. J Biol Chem 2001; 276:40263–40267. 41. Mathew A, Mathur SK, Jolly C, Fox SG, Kim S, Morimoto RI. Stressspecific activation and repression of heat shock factors 1 and 2. Mol Cell Biol 2001; 21:7163–7171. 42. Hilgarth RS, Hong Y, Park-Sarge OK, Sarge KD. Insights into the regulation of heat shock transcription factor 1 SUMO-1 modification. Biochem Biophys Res Commun 2003; 303:196–200. 43. Hu Y, Mivechi NF. HSF-1 interacts with Ral-binding protein 1 in a stress-responsive, multiprotein complex with HSP90 in vivo. J Biol Chem 2003; 278:17299–17306. 44. Ozaki M, Deshpande SS, Angkeow P, Suzuki S, Irani K. Rac1 regulates stress-induced, redox-dependent heat shock factor activation. J Biol Chem 2000; 275:35377–35383. 45. Lindquist S, Craig EA. The heat-shock proteins. Annu Rev Genet 1988; 22:631–677. 46. Hunt C, Calderwood S. Characterization and sequence of a mouse hsp70 gene and its expression in mouse cell lines. Gene 1990; 87: 199–204. 47. Perry MD, Aujame L, Shtang S, Moran LA. Structure and expression of an inducible HSP70-encoding gene from Mus musculus. Gene 1994; 146:273–278. 48. Hunt CR, Gasser DL, Chaplin DD, Pierce JC, Kozak CA. Chromosomal localization of five murine HSP70 gene family members: Hsp70-1, Hsp70-2, Hsp70-3, Hsc70t, and Grp78. Genomics 1993; 16: 193–198. 49. Snoek M, Jansen M, Olavesen MG, Campbell RD, Teuscher C, van Vugt H. Three Hsp70 genes are located in the C4-H-2D region: possible candidates for the Orch-1 locus. Genomics 1993; 15:350–356. 50. Lee JS, Seo JS. Differential expression of two stress-inducible hsp70 genes by various stressors. Exp Mol Med 2002; 34:131–136. 51. Shim EH, Kim JI, Bang ES, Heo JS, Lee JS, Kim EY, Lee JE, Park WY, Kim SH, Kim HS, Smithies O, Jang JJ, Jin DI, Seo JS. Targeted disruption of hsp70.1 sensitizes to osmotic stress. EMBO Rep 2002; 3:857–861. 52. Christians E, Davis AA, Thomas SD, Benjamin IJ. Maternal effect of Hsf1 on reproductive success. Nature 2000; 407:693–694. 53. Pirkkala L, Nyka¨nen P, Sistonen L. Roles of the heat shock transcription factors in regulation of the heat shock response and beyond. FASEB J 2001; 15:1118–1131. 54. Jackson SP, Tjian R. O-glycosylation of eukaryotic transcription factors: implications for mechanisms of transcriptional regulation. Cell 1988; 55:125–133. 55. Song J, Ugai H, Kanazawa I, Sun K, Yokoyama KK. Independent repression of a GC-rich housekeeping gene by Sp1 and MAZ involves the same cis-elements. J Biol Chem 2001; 276:19897–19904. 56. Yang X, Su K, Roos MD, Chang Q, Paterson AJ, Kudlow JE. Olinkage of N-acetylglucosamine to Sp1 activation domain inhibits its transcriptional capability. Proc Natl Acad Sci U S A 2001; 98:6611– 6616.