DEVELOPMENTAL

BIOLOGY

145, 13-27

(19%)

Metallothionein Gene Expression and Metal Regulation Preimplantation Mouse Embryo Development (MT mRNA during Early Development)

during

GLEN K. ANDREWS,*~~ YVETTE M. HUET-HUDSON,?*’ BIBHASH C. PAR&t MICHAEL T. MCMASTER,* SWAPAN K. DE,* AND SUDHANSU K. DEY-~

In order to provide information concerning gene expression and regulation in the preimplantation mammalian embryo, and to explore the roles of metallothionein (MT) during this period of development, the constitutive and metal-induced MT mRNA levels in mouse ova, preimplantation embryos, and oviducts were determined. These results were correlated with the effects of transient exposure to high levels of metals (zinc (Zn) or cadmium (Cd)) on the continued development of preimplantation embryos into blastocysts in culture. RN.4 from preimplantation mouse embryos at different stages of development (Days 1 through 4 of gestation; Dl = vaginal plug) was analyzed using the reverse transcriptase-polymerase chain reaction (RT-PCR) to specifically amplify MT-I and MT-II mRNA transcripts. MT-I mRNA in ova, preimplantation embryos, and oviducts was detected using i?c si(u hybridization. This mRNA in the oviduct was also analyzed by Northern blotting. The results establish that the mouse MT genes are coordinately and constitutively expressed at low basal levels in ova and preimplantation mouse embryos. In unfertilized (ova), fertilized (one-cell) eggs, and two-cell embryos, the MT-I gene was not detectably responsive to metal ions, whereas in later cleavage stage embryos (four- and eight-cell) the MT-I gene was detectably responsive to metals in some blastomeres of some of the embryos, In contrast, after the third cleavage this gene was highly metal-inducible in essentially all cells of the embryo (morula/blastocyst). Surprisingly, the appearance of metal responsiveness of the MT genes during development correlated with decreased Zn toxicity and increased Cd toxicity; two-cell embryos were Zn-sensitive and Cd-resistant, whereas eight-cell and older embryos were Zn-resistant and Cd-sensitive. In the oviduct, MT-I mRNA was not abundant in total RNA, but was detected specifically in the epithelial cells of the isthmus region and was elevated in theses cells on D3 and D4 of gestation. In the oviduct, only isthmus cpithelial cells responded to metals (Zn or Cd) by increased accumulation of this mRNA. These studies suggest that preimplantation mouse embryo develops the capacity to respond to metals in the environmental milieu by induction of MT gene expression at about the third cleavage. Whether the lack of responsiveness of these genes before this stage reflects transcriptional repression or attenuated metal ion influx and/or enhanced efflux remains to be determined. Sensitivity and resistance of preimplantation embryos to acute metal toxicity involve mechanisms other than MT gene expression in preimplantation mouse embryos. The cell-specific expression of MT genes in the isthmus epithclia of the oviduct suggests that these cells play a role in metal ion homeostasis on D3 and D4 of pregnancy, perhaps by creating a local environment rich in the essential metals Zn and copper and/or by protecting from the toxic effects of Cd. ac 1991 Academic Press, Inc.

from metal toxicity which occurs following exposure to higher than normal concentrations of essential metals, such as zinc (Zn) and copper, and following exposure to very low levels of the nonessential metal cadmium (Cd) (reviewed by Karin, 1985; Hamer, 1986; Webb, 1987). The mouse MT gene locus contains two linked MT genes that encode isoprotein forms (MT-I and MT-II) which differ slightly in amino acid sequence and total charge (Searle et al., 1984). Transcription of these genes is dramatically increased, in most cells, following exposure to heavy metals. This transcriptional effect is mediated via interactions between upstream DNA sequences, termed metal regulatory elements, and metal-dependent transacting factors (Stuart et al., 1985; Andersen et al., 1987; Mueller et crl., 1988; Imbert et al., 1989). MT gene tran-

INTRODUCTION

The metallothioneins (MTs)~ are small, cysteine rich, intracellular proteins that bind metal ions with high affinity. These proteins are considered to play an important role in metal metabolism, perhaps by protecting i To whom correspondence and reprint requests should be addressed at Department of Biochemistry and Molecular Biology, WHE 4018, linioersity of Kansas Medical Center, 39th and Rainbow Blvd., Kansas City, KS. Fax: (913) 588-5677. *Present address: Monsanto Co., AA4C, 700 Chesterfield Village Parkway, St. Louis, MO 63198. 3ilbbreviations used are: Cd, cadmium; D, day of gestation; MT, metallothionein; RT-PCR, reverse transcriptase-polymerase chain reaction; sc, subcutaneous; Zn, zinc. 13

0012.1606/91 Copyright All rights

$3.00

‘sj 1991 by Academic Press, Inc. of reproduction in any form reserved.

14

DEVELOPMENTALBIOLOGY

scription is also induced by glucocorticoids and a variety of cytokines (see Hamer, 1986; De et al., 1990b). That the MTs may serve critical functions during embryonic development (Webb, 1987) is suggested by the tissue-specific expression of these genes in the reproductive tract and conceptus during the postimplantation period. From the time of implantation to late in gestation, the mouse embryo is surrounded by cells, interposed between the maternal and embryonic environments, which actively express the MT genes (De et al., 1989). This expression is first noted in the deciduum, and subsequently in the spongiotrophoblasts of the placenta (De et al., 1989) and the visceral yolk sac endoderm (Andrews et al., 1984). During late embryonic and fetal development of the mouse, high level expression of the MT genes is apparently restricted to the developing hepatocytes (Webb, 1987; De et ah, 1990a). Possible roles for MT in the preimplantation period are suggested by studies of the effects of heavy metal ions on the development of the preimplantation embryo. Maternal dietary Zn deficiency leads to abnormal preimplantation development of rat embryos (Hurley and Shrader, 19’75), and Cd is embryotoxic to preimplantation mouse, rat, and rabbit embryos in vitro (Pedersen and Lin, 19’78; Schmid et al., 1983; Yu et al., 1985; Abraham et al., 1986; Yu and Chan, 1986; Andrews et al., 1987; Spielmann and Vogel, 1989). Despite the toxic effects of Cd in vitro, a systemic injection of Cd during the preimplantation period has been reported to have little effect on embryo development (Chiquoine, 1965; Giavini et ab, 1980). Thus, the oviduct or other maternal organs may play a role in protecting the preimplantation embryo from Cd toxicity. The rabbit blastocyst expresses low levels of MT that are dramatically increased at the mRNA and protein levels in response to in vitro exposure to heavy metals (Andrews et al., 1987). Except for these studies, little else is known about MT gene expression and regulation during preimplantation development of the mammalian embryo. In the sea urchin embryo, MT mRNA is a maternal messenger that is polyadenylated after fertilization and is first transcribed at the eight-cell cleavage stage (Nemer et al., 1984). During early morphogenesis, expression of MT in the sea urchin occurs in a cell type-specific manner (Angerer et al., 1986). The ectoderm expresses a 0.7-kb MT-a mRNA, and the endomesoderm a 0.85-kb MT-0 mRNA (Wilkinson and Nemer, 1987). Given the pivotal role that MT plays in metabolism of essential and toxic metals, and the utility of the MT gene system for analyzing inducible gene expression in the preimplantation mammalian embryo, we have studied the expression and metal regulation of the MT genes in ova, during development of the preimplantation mouse embryo, and in the oviduct. These results were correlated with changes in metal toxicity during preimplantation embryo development.

VOLUME 145,199l METHODS

Animals

Female CD-l mice (48 days old; Charles River Laboratories) were mated, and the gestational age of the embryo was calculated from the day on which a vaginal plug was detected (designated Dl of gestation). Mice were injected subcutaneously (SC) with 100 pmoles ZnCl,/kg body wt or with vehicle alone (normal saline). Oviducts, ovaries, and liver were recovered 5 hr after injection and analyzed by in situ hybridization and Northern blotting as described below. Recovery and Culture of Preimplantatimz Embryos

Preimplantation embryos at one-cell, two-cell, fourcell, six- to eight-cell, morula, and blastocyst stages were recovered by flushing the reproductive tract with Whitten’s medium (Whitten, 1971) on Dl-4 of pregnancy. Embryos from several animals were pooled, washed four times in medium, divided into batches of 10 to 20 embryos per 1 ml of medium, and cultured at 37°C under an atmosphere of 90% N, + 5% CO, + 5% O2 as described previously (Whitten, 1971; Dey and Johnson, 1980). The effects of transient exposure to Zn or Cd on subsequent in vitro development of preimplantation embryos into blastocysts were determined. Preimplantation embryos at the following stages of development were selected from among those flushed from the reproductive tract: two-cell (D2, 1500 hr), four-cell (D3, 0400 hr), eight-cell (D3, 0900 hr), and compacted eight-cell/ morula (D3,1600 hr). Stage-selected embryos were first cultured in the presence or absence of ZnCl, (50 palm) for 8 or 18 hr, or CdCl, (50 palm) for 8 hr. They were then washed four times in fresh medium and recultured to assess their potential to develop into blastocysts. The total culture periods, including metal exposure times, allowed for development to the blastocyst stage were as follows: 66 hr for two-cell, 53 hr for four-cell, 48 hr for eight-cell, and 42 hr for morula stage embryos. Embryonic development was examined under a dissecting microscope. Preimplantation embryos, analyzed by in situ hybridization for the effects of Zn or Cd on MT mRNA levels, or by reverse transcriptase-polymerase chain reaction (RT-PCR) for the presence of MT-I and MT-II mRNAs, were obtained as follows: Embryos were collected on the mornings of Dl through D4 of gestation. Due to asynchronous development of preimplantation embryos, a mixture of four-cell and eight-cell stage embryos was obtained on D3, and on D4, late morulae and blastocysts were collected. Embryos were cultured in the presence or absence of ZnCl, (50 PM) or CdCl, (10 &f) for 5 hr and then processed for in situ hybridization

ANDREWS

15

ET AL.

as described below. For RT-PCR, about 80 embryos per group (Dl to D4) were collected. Care was taken to avoid any contamination of embryos with maternal cells by repeated washing under a dissecting microscope (four times) of groups of 10 to 20 embryos in fresh medium. Washed embryos were quick frozen in a minimal volume (5 ~1) in the bottom of a 400-111 microcentrifuge tube.

based on the sequences for these cDNAs and on the organization of the MT genes (Glanville et al, 1981, Searle et ah, 1984) as follows: MT-I MT-II MT-I MT-II

Isolation

of Total RNA

RNA was extracted from ovaries, oviducts, and liver using sodium dodecyl sulfate (SDS)-phenol-chloroform buffers as described by Andrews et al. (1987). This procedure was modified for recovery of RNA from preimplantation embryos as follows: Preimplantation embryos, collected and frozen as described above, were suspended in 50 ~1 of SDS buffer (0.5% SDS, 25 mMEDTA, 75 mM NaCl, pH 8.0) plus 50 ~1 of phenol saturated with SDS buffer. Escherichia coli ribosomal RNA (20 pg; Boehringer-Mannheim, Indianapolis, IN) was added, and the mixture homogenized by repeated passage through a 27-ga needle using a l-cc syringe. This mixture was chilled to 4°C and centrifuged for 10 min at 4°C in a 400-~1 microcentrifuge tube in a microcentrifuge. The organic phase was removed, and the aqueous phase and interphase were reextracted with an equal volume of phenol/chloroform/isoamyl alcohol (24:24:1 v/v). Following centrifugation, as described above, the aqueous phase was carefully recovered, transferred to a 1.5-ml microcentrifuge tube, and 3 vol of 4 M ammonium acetate was added to precipitate the RNA. Following incubation in ice water for 45 min, the RNA precipitate was collected by centrifugation at 50,OOOg for 30 min at 4°C in a TL-100 tabletop ultracentrifuge (Beckman Instruments, Palo Alto, CA). The RNA pellet was carefully washed in cold 85% ethanol, dried briefly under vacuum, and dissolved in 10 ~1 of water. Reverse Transcriptase-Polymerase

Chain

Reaction

Methods for RT-PCR were modified from previously published protocols (Rappolee et al., 1988, 1989; Saiki et al., 1988). RNA (5 ~1 from about 40 embryos) was reverse transcribed in 10 ~1 of reaction buffer containing: 50 mM Tris-HCl, pH 8.3,60 mM KCl, 3 mM MgCl,, 10 mM dithiothreitol, 1 mMeach dATP, dGTP, dCTP, and dTTP, 25 &ml Ok0 dTc,-,,,, 10 pug/ml RNase free BSA (Promega Biotec, Madison, WI), 450 units/ml recombinant RNasin (Promega), and 9 units of recombinant M-MuLV reverse transcriptase (RT) (Boehringer-Mannheim). The reaction mixture was incubated at 37°C for 20 min, heated to 94°C for 3 min, and chilled on ice. Another 9 units of RT was added, and the reaction mixture incubated again at 37°C for 20 min. Oligodeoxyribonucleotide (oligos) primers (27 mers) for mouse MT-I or MT-II mRNAs were synthesized

sense fence

antisense antisense

ATG

GAC

CCC

AAC

TGC

TCC

TGC

TCC

ACC

ATG

GAC

CCC

AAC

TGC

TCC

TGT

GCC

TCC

GGA

AGA

GGG

TGG

AAC

GGT

CTA

TTT

TGT ACA

ATA CAG

ATG

TGG

CGC GGA

TGG CCC

The sense strand oligos begin at the MT-I or MT-II translation initiation codons and span the first exon of each of these MT genes. These oligos are 89% homologous to each other. However, the antisense strand oligos are complementary to the 3’ untranslated regions of their respective mRNAs and share little sequence similarity (15%). Amplification using the polymerase chain reaction (PCR) was based on the methods of Saiki et al. (1988). The PCR reaction buffer contained: 10 mM Tris-HCl, pH 8.3,2.5 mMMgCl,, 50 mMKCl,600 PMeach of dATP, dTTP, dGTP, dCTP, 100 pg/ml RNase free BSA, 0.5 PLM oligo primers (both sense and antisense), and 20 units/ ml of Taq polymerase (Perkin-Elmer Cetus, Norwalk, CT). PCR reaction mixtures (50 ~1) were assembled on ice, and an aliquot (3 ~1) of the RT reaction mixture (from about 10 embryo equivalents of RNA based on the original sample size) was added. Control PCR reactions in which the RT reaction products were omitted, or which contained total RNA that had not been reverse transcribed, were performed in parallel. Samples were overlayed with 30 /*l of silicone oil (200 fluid; Dow Corning Corp., Midland, MI) which had been passed through a 0.45-pm filter. In all instances, PCR was for 45 cycles with an annealing temperature of 50°C. PCR products (5 ~1; equivalent to that obtained from about one embryo based on the original sample size) were analyzed by 2% agarose gel electrophoresis with 1 pg/ml ethidium bromide in the running and gel buffer (40 mM Tris-20 mM acetic acid, 5 mM sodium acetate, 2 mM EDTA). Sizes of PCR products were estimated by comparison with a 123bp DNA ladder (Bethesda Research Laboratories, Gaithersburg, FL). In Situ Hybridization The methods for in situ hybridization have been described in detail by De et al. (1989) and were adapted from procedures published by Angerer and colleagues (Angerer and Angerer, 1981; Deleon et al., 1983; Cox et al., 1984). Mouse oviducts and ovaries were perfusionfixed with 4% paraformaldehyde in phosphate-buffered saline (PBS), dehydrated, cleared, paraffin embedded, and sectioned (7 vm). The following method was developed to obtain serial sections of preimplantation embryos for in situ hybridization: Embryos were recovered and cultured in the presence or absence of excess metal

16

DEVELOPMENTALBIOLOGY

VOLUME 145,1991

246 123

I

Cl

I

I

I

1

234 Day of Gestation C

MT-I

MT-II

369 246

6 Ml-l

MT-II

Pl 1 I

1 t

74 I I

AGT % 65 94 I 1 I , ATG BamHl

P2 236 I ’ TGA Sstll _p2 236 290 I I I 1 TGA Pvull

333 I

392 bp J

123

385 bp J

FIG. 1. Detection of MT-I and MT-II mRNAs in preimplantation mouse embryos using the reverse transcriptase-polymerase chain reaction (RT-PCR). Preimplantation mouse embryos (about 80 per group) on the indicated days of gestation (Dl = day of vaginal plug) were recovered, and total RNA was extracted in the presence of carrier E coli ribosomal RNA. RNA was reverse transcribed and subjected to 45 cycles of PCR using oligo primers specific for MT-I or MT-II mRNAs. PCR products, equivalent to that obtained from about one embryo equivalent of RNA, were separated by agarose-gel electrophoresis in the presence of ethidium bromide, and DNA was visualized under uv light. (A) RT-PCR of MT-I and MT-II mRNAs in normal Dl (one-cell), D2 (two-cell), D3 (four- and eight-cell), and D4 (late morulae/blastocysts) embryos (labeled l-4). The results indicate that both MT mRNAs are present in one-cell fertilized eggs, as well as in cleavage stage, morulae, and blastocyst stage embryos. In the control reactions (C), RNA samples were omitted from the PCR. The predominant reaction products were near the predicted sizes (259 and 225 bp) for proper amplification of the MT-I and MT-II cDNAs, respectively, and were much smaller that those expected from amplification of the MT genes in genomic DNA (>616 bp). Minor reactions products in this size range were noted in the MT-II PCR products, and these likely represent amplified primer sequences as they were detected in control reactions. However, control reactions were negative for the major reaction products after 45 cycles of PCR. (B) Restriction map of mouse MT-I (Durnam et aL, 1980) and MT-II (Searle et al., 1984)

ANDREWS ET AL.

MT mRNA during Early Development

ions as described above. Following termination of cultures, embryos were washed four times in fresh medium, and each batch of embryos was then rapidly transferred into an oviduct obtained from a nonpregnant female. The oviduct served as a convenient holder for the embryos during subsequent processing. Oviducts containing embryos were fixed in ice-cold 4% paraformaldehyde in PBS for 2 hr, dehydrated, cleared, paraffin embedded, and sectioned (7 pm). Serial sections were examined under a compound microscope, and those containing embryos were identified. Within a given experiment, sections from all experimental samples were mounted onto the same microscope slide, and several slides were prepared. Slides were then processed for in situ hybridization using ?S-labeled antisense or sense strand MT-I RNA probes as described previously (De et al, 1989). In order to provide quantitative estimates of relative MT mRNA levels, autoradiographic grains per embryo section were counted for multiple sections of multiple embryos, and background signal was subtracted to yield a final value. Northern

Blot Hybridization

RNA was separated by formaldehyde-agarose gel electrophoresis and analyzed by Northern blot hybridization using a %P-labeled MT-I cRNA probe (sp act 2 x 10’ dpm/pg) as described previously (Andrews et al., 1987; De et aZ., 1989). In all experiments, duplicate gels were stained with acridine orange to ensure integrity of the RNA sample and to confirm that equal amounts of RNA had been loaded onto each lane. RESULTS

The MT-I and MT-II Genes Are Constitutively Coordinately Expressed in Preimplantation Mouse Embryos

and

RNA from Dl to D4 preimplantation mouse embryos was analyzed, using RT-PCR, for the presence of MT-I and MT-II mRNAs (Fig. 1A). Following 45 cycles of PCR, both of these MT transcripts were detected in reverse transcribed RNA from one-cell fertilized eggs collected on Dl (Fig. 1A). This suggests that the MT mRNAs are maternal messengers stored in the mouse egg. MT-I and MT-II transcripts were also readily detected in cleavage stage embryos (D2 and D3) and in morulae and blastocysts (Fig. 1A). Thus, the mouse MTI and MT-II genes are coordinately expressed through-

17

out the preimplantation period of embryonic development. Relative efficiencies of recovery of total RNA, reverse transcription, and PCR amplification of each RNA sample were not monitored. The specificity of the RT-PCR was confirmed by restriction enzyme analysis of the MT-I and MT-II PCR products (Fig. 1C). PCR products in both cases were near the predicted sizes (259 and 225 bp) for proper amplification of the MT-I and MT-II mRNAs, respectively, and had predicted restriction enzyme cleavage sites (Fig. 1B) based on previously published nucleotide sequence information (Durnam et al, 1980; Glanville et al, 1981; Searle et aC, 1984). The major PCR products did not result from amplification of genomic DNA contaminating the RNA preparations, because this would have resulted in the amplification of much larger DNA fragments (953 and 616 bp, respectively, for MT-I and MT-II) due to the presence of two introns in each of the MT genes (Glanville et al., 1981; Searle et al., 1984). Only minor PCR products were detected following45 cycles of amplification of RNA samples which had not been reverse transcribed, thus confirming that cDNAs served as templates in the PCR and were not contaminating chromosomal DNA (not shown). Mock PCRs in which the RNA sample was omitted altogether were negative for the major PCR products (Fig. 1A). The MT Genes Are Metal Ion Responsive at w ajkr Third Cleavage (Eight-Cell/Mwula) Stage of Development

the

The temporal and spatial patterns of expression of MT-I mRNA in preimplantation embryos and the effects of an acute in vitro exposure to Zn or Cd on levels of this mRNA were analyzed by in situ hybridization (Fig. 2). Following treatment in culture, embryos were placed into the oviduct from a nonpregnant female. The oviduct was subsequently fixed and processed for in situ hybridization. Specificity of the hybridization procedure was ensured by (1) maintaining high stringency during the hybridization, (2) detecting RNase resistant hybrids, and (3) control experiments using sense strand probes, as described previously (De et aL, 1989). Furthermore, all control and experimental samples were analyzed in parallel on the same slides, and the oviduct served as the internal control for the background signal for each sample. Multiple sections of multiple embryos were examined.

cDNAs. Transcription start point is designated as 1 bp. The locations of translation start and stop codons and sites of polyadenylation are indicated, as are the locations of the sense strand (Pl) and antisense strand (P2) primers used in the PCR. All sites of cleavage of relevant restrictions enzymes are indicated, and these enzymes cleave MT-I or MT-II cDNA, but not both. (C) RT-PCR products were cleaved with the indicated restriction enzymes and analyzed by agarose-gel electrophoresis. The small DNA fragments generated by cleavage of MT-II RT-PCR products using PvuII (54 bp) or BamHI (29 bp) are not visible on these gels. The pattern of enzyme cleavage establishes that correct PCR amplification of MT-I and MT-II mRNAs had occurred.

18

DEVELOPMENTALBIOLOGY

A Day 1

Day 2

Day 3

Control

VOI

Metal-Treated

Control

blastocyst

morula

FIG. 2. Detection of MT-I mRNA in preimplantation mouse embryos using it/ .si(/r hybridization. Preimplantation mouse embryos on the indicated days of gestation were recovered and incubated for 5 hr in the presence or absence of Zn (50 &‘) or Cd (10 PM). Embryos were transferred hack into the ampulla region of the oviduct, fixed in paraformaldehyde. sectioned, and mounted onto polylysine-coated slides. Sections within a given experiment were placed on the same slide. Slides were hybridized with a ““S-labeled MT-I antisense strand RN.4 probe, and hybrids were detected by autoradiography for 5 days. A sense strand MT probe was used as a control for specificity of the hybridization (see B, panel p), and the oviduct served as an internal control to establish background signals. (A) Dark-field photomicrom-aphs (200X magnification) in which the autoradiographic grains appear as white dots. Dl embryos are one-cell fertilized eggs; D2 are two-cell; D3 are mostly eight-cell with a few four-cell embryos (f). Abbreviations are: EMB, embryos; 01) ampulla of the oviduct. Control embryos (a, c, e); Zn-treated embryos (b, d, f-h). The arrow in h points to heavily labeled blastomeres of a four- to eight-cell embryo. (B) Day il embryos are at the late morula or hlastocyst stages of development. Shown here are dark-field photomicrographs (200X) in which the autoradiographic grains appear as white dots (a-c and g), and bright-field photomicrographs (400X) in which the grains appear as black dots (d-f). Ahhreviations are: ICM, inner cell mass; Tr, trophectoderm. Control emhrgos (a, d); Zn-treated embryos (h, e-g), Cd-treated cmhryos (c). (g) Zn-treated hlastocyst hybridized with a sense strand MT probe as a control.

20

DEVELOPMENTALBIOLOGY

VOLUME 145.1997

ANDREWS

-

MT-

21

ET AL.

+

-

+

I

I

I

L

L

Zn

i

I

Dl

I

I

I

D2 D3 D4 OD

I

L

I

D3 D3 OD

FIG. 4. Northern blot detection of MT-I mRNA in the oviduct during the preimplantation period. Mouse oviducts (OD) were collected on the indicated days of pregnancy (DlLD4). Oviducts and liver from D3 pregnant females were collected 5 hr after a SC injection of 100 pmoles ZnCl,/kg body wt. Total RNA was extracted and analyzed by Northern blot hybridization using a mouse MT-I cRNA probe. Hybrids were detected by autoradiography for 12 hr. Duplicate agarose gels were analyzed by acridine staining to ensure that equal amounts of RNA had been loaded.

Low basal levels of MT-I mRNA were detected in preimplantation embryos on Dl to D3 (Fig. 2A). Autoradiographic grain counts per embryo section averaged about 10 grains/cell (twice background densities) on Dl and D2, regardless of prior metal exposure. The relatively large cell numbers per embryo on D3 and D4 make it difficult to accurately determine grains per cell. However, between D3 and D4, grain counts per embryo increased about eightfold (Fig. 2B). Although the number of cells per embryo also increases at least 2-fold between these days, these results suggest that the basal levels of MT mRNA increase significantly at the blastocyst stage. Incubation of preimplantation embryos for 5 hr in medium containing Zn or Cd had little effect on MT-I mRNA levels before the eight-cell stage (D3) (Fig. 2A). However, small changes in MT mRNA levels are unlikely to be detected by this method. Some blastomeres of some four- to eight-cell embryos had higher grain densities than others following exposure to metal ions (Fig. 2A; panel h), but essentially all blastomeres of aeight-cell embryos were heavily labeled following metal treatment (Fig. 2A, panel g). Autoradiographic grain counts indicate that metal exposure resulted in a six- to eight-fold increase in grain counts per embryo for

aeight-cell embryos. These results were reproducible in all embryos examined (n > 30). Given the high density of grains in the metal-treated morulae and blastocysts, the fold induction by metals is likely to be underestimated. Examination of serial sections of D4 embryos (morulae and blastocysts) indicated that in vitro exposure to metal ions up-regulated MT mRNA levels in essentially every cell (Fig. 2B). Both Zn and Cd were effective inducers (Fig. 2B, panels b and c, respectively). Therefore, it can be concluded that the MT genes in preimplantation mouse embryos respond to metal ion concentrations in the immediate environment at about the third cleavage (eight-cell stage of development), and in morulae and blastocysts the metal response is dramatic. The MT Genes Are not Metal Ion Responsive in Mouse Ova, In order to examine MT gene expression and metal responsiveness in developing ova, sections of ovaries from normal and metal-treated mice, collected 5 hr after an injection of Zn (100 pmoles ZnClJkg body wt), were analyzed using in situ hybridization (Fig. 3). Northern blot (not shown) and solution hybridization experiments established that low, but detectable levels of MT mRNA were present in the untreated mouse ovary, and metal treatment resulted in a rapid 4-fold increase in both MT-I and MT-II mRNAs (De et al., 1990b). In normal ova within follicles at various stages of development (Greenwald and Terranova, 1988), in situ hybridization detected low basal levels of MT mRNA (Fig. 3; panels a and b). High grain densities were detected over both interstitial and some granulosa cells in metal-treated ovaries (Fig. 3, panels c-f), but grain densities per ova remained unchanged following metal treatment (about 10 grains/large ovum or twice the background levels). Thus, the MT genes are not constitutively expressed at high levels nor are they detectably metal responsive in mouse ova at various stages of development. The MT Genes Are Expressed in a Cell-Speci& and Temporally Regulated Manner in the Oviduct during the Preimplantation Period Preimplantation embryo development takes place largely inside the oviduct. Fertilization of the egg (Dl)

FIG. 3. Detection of MT-I mRNA in mouse ova using in sift hybridization. Ovaries were obtained from normal cycling mice before or 5 hr after injection of Zn (100 rmoles ZnCl,/kg body wt) (control and metal-treated, respectively). Samples were perfusion-fixed, and analyzed by in situ hybridization as described in the legend to Fig. 2. Sections within a given experiment were placed on the same slide, and samples from at least two animals per group were analyzed. Sections were hybridized with a %-labeled MT-I cRNA probe, and hybrids were detected by autoradiography for 11 days. A sense strand MT probe was used as a control for specificity of the hybridization (data not shown). (a, c, e) Bright-field photomicrographs (200x magnification). (b, d, f) Dark-field photomicrographs of panels a, c, and e, respectively. Autoradiographic grains appear as white dots. Control, panels a and b; metal-treated, panels c-f. Abbreviations are: TC, thecal cell; GC, granulosa cell; IC, interstitial cell; arrowhead, ova.

22

DEVELOPMENTALBIOLOGY VOLUME145,1991

occurs in the ampulla. Cleavage stage embryos (D2 to D3) are transported through the oviduct lumen and by the morula stage (D3-early D4) are located in the isthmus region near the uterotubal junction. During D4, the late morulae and blastocysts traverse the isthmus-uterine junction and are located in the uterine lumen prior to implantation. The expression and regulation of the MT-I gene in the oviduct during the preimplantation period were examined. Northern blot analysis detected only low levels of MT-I mRNA in the oviduct (Dl-D4), but these levels were higher on D3 and D4 than on Dl and D2 (Fig. 4). An injection of Zn had a dramatic effect on MT-I mRNA levels in the liver, but only a slight induction of this mRNA was noted in the oviduct (Fig. 4). In situ hybridization established that MT-I mRNA is elevated specifically in the epithelial cells of the isthmus region of the oviduct and levels of this mRNA increased on D3 and D4 (Fig. 5A). No other oviductal cell types contained elevated MT-I mRNA (Fig. 5B, control). Furthermore, injection of Zn on D3 (other stages were not examined) led to a modest increase in MT-I mRNA levels exclusively in the isthmus epithelial cells (Fig. 5B). In Vitro Development of Preimplantation Mouse Embryos Is Altered by Exposure to Metals in an Age-Dependent Manner These experiments were designed to determine the relationship between the ability of preimplantation mouse embryos to respond to metal ions by increased expression of the MT genes and the effects of a transient exposure to high concentrations of Zn or Cd on subsequent in vitro development. Embryos were recovered on various days of gestation (D2 to D4) and cultured in vitro in medium alone or in medium containing Zn (50 PLMfor 18 hr) or Cd (50 PLMfor 8 hr). Embryos were then washed free of excess metals ions and returned to culture medium for continued development to the blastocyst stage. Eighty to 90% of untreated two-cell embryos developed into blastocysts within 72 hr of culture. Most of these blastocysts developed to term following transfer into the uterus of a synchronized pseudopregnant recipient female (data not shown). Regardless of the stage of development (two- or four-cell) when exposure to elevated Zn occurred, many preimplantation embryos continued to develop in vitro to the morula stage before degenerating (data not shown). However, the ability of the embryos to develop to the blastocyst stage was dependent on the stage of development when the exposure to Zn occurred (Table 1). Analysis of the data using the x2 test (Hoel, 1984) indicates that incubation with Zn leads to a greatly reduced frequency of blastocyst formation from two-cell embryos (P < 0.0001). This effect was diminished in four-cell and eight-cell em-

bryos (P = 0.0091 and 0.1495, respectively) and absent in morulae (P = 0.8165). In subsequent experiments, similar results were obtained following incubation of preimplantation embryos in Zn-containing medium for 8 hr rather than 18 hr (data not shown). Exposure of preimplantation embryos to Cd was also embryotoxic, and embryotoxicity was dependent on the stage of development when Cd exposure occurred (Fig. 6). Surprisingly, growth and differentiation of two-cell embryos were essentially unaffected following culture in medium containing Cd (P = 0.15’78), and about 50% of four-cell embryos also continued to develop into blastocysts following this treatment. In contrast, the vast majority of older embryos (&eight-cell) rapidly degenerated following incubation with Cd (P < 0.0001). These data suggest that Zn and Cd toxicities involve different mechanisms in the preimplantation mouse embryo and that metal toxicity cannot be correlated directly with the ability of preimplantation embryos to respond to metals in the environment by induction of the MT genes. DISCUSSION MT genes are expressed at basal levels in most higher eukaryotic cell types, and induction of MT gene expression by metal ions is a transcriptional response that has been highly conserved during evolution (Hamer, 1986). Our studies of basal expression of the MT genes indicate that these genes are expressed at low levels in mouse ova, fertilized eggs, and early embryos. Both the MT-I and the MT-II genes are constitutively and coordinately expressed throughout the preimplantation period of embryonic development. Most maternal poly(A)+ RNA is degraded by the two-cell stage, and the onset of transcription of the embryonic genome is thought to occur at the two-cell stage (reviewed by Magnuson and Epstein, 1987; First and Barnes, 1989). These observations, and the fact that MT mRNAs can be detected by RT-PCR in two-cell and older embryos, suggests that the MT genes may be among the first to be transcribed from the embryonic genome. However, the possibility that maternal MT mRNA is maintained in the cleavage stage mouse embryo cannot be excluded. Members of the heat shock stress protein gene family are transcribed soon after the first cleavage (see Magnuson and Epstein, 1987). The presence of maternal MT mRNA and the early activation of transcription of the MT genes during cleavage stages in the sea urchin embryo (Nemer et al., 1984), as well as the constitutive presence of MT-I and MT-II mRNAs in the preimplantation mouse embryo as shown in the present study, suggest that these proteins may serve important functions during early embryogenesis. Precisely what these functions are remains to be determined. However, as discussed below, it is unlikely that embryonic MT serves a significant protective role with

ANDREWSETAL.

MT VIRNA during Early Llrwlopmrnt

regard to acute metal toxicity in the preimplantation stages of development. Analysis of metal regulation of MT mRNA levels established that these genes do not respond appreciably to changing concentrations of metal ions in the environment until about the third cleavage (four- to eight-cell stage), after which the metal response is pronounced. The rabbit blastocyst also responds to metal ions by a rapid lo-fold induction of MT (Andrews et al., 1987). Another stress response, the heat shock response, is also operative in the morula-blastocyst stages of mouse and rabbit preimplantation embryos, but is not detected during the cleavage stages (Heikkila et al., 1985). Thus, both of these stress responses are apparently attenuated or lacking in early cleavage stage mammalian embryos. The reasons for the lack of metal ion induction of MT genes in early cleavage stage mouse embryos are unknown. Expressions of MT promoter-thymidine kinase (Brinster et al., 1982) and of MT promoter-fl-galactosidase fusion genes (Stevens et al., 1989), microinjected into one-cell fertilized eggs, are increased following incubation of the injected eggs and subsequent cleavage stage embryos in medium containing elevated levels of metal. Therefore, the trans-acting factors required for metal responsiveness of the MT genes are present and functional in early mouse embryos. This suggests the possibility that the structure of the MT genes in the embryonic genome precludes their induction by metal ions. Both the chromatin structure and the cytosine methylation pattern of the mouse MT-I gene have been implicated in regulating metal responsiveness (Andrews and Adamson, 1987; Compere and Palmiter, 1981). However, an alternative explanation for the lack of metal response could be that Cd influx is largely prevented and/or efflux is heightened in the early embryo relative to that in the morula. Increased efflux of Cd has been implicated in the attenuation of metal responsiveness of the hamster MT genes in cultured cells (Morris and Huang, 1989). Studies of metal responsiveness of MT promoter-thymidine kinase fusion genes microinjetted into the fertilized egg utilized Cd at an extremely high concentration (50 PM) (Brinster et al., 1982). At present, nothing is known about metal ion transport in the preimplantation mammalian embryo. The development of metal ion responsiveness of the MT genes during preimplantation embryogenesis correlated positively with resistance to Zn toxicity, but

23

inversely with resistance of the embryo to the embryotoxic effects of Cd. This paradoxical result was surprising, and to our knowledge is unique to the preimplantation embryo. In many cells lines, we and others have noted that Cd is far more toxic than Zn. The results suggest that the mechanism of embryotoxicity for Cd and Zn is different, as might be predicted since Zn is an essential metal, whereas Cd has no known biological role. A striking finding was that two-cell embryos are resistant to the embryotoxic effects of a brief exposure to a high concentration of Cd (50 PM). Pedersen and Lin (1978) reported that continuous culture of two-cell embryos in medium containing Cd (10 PM) leads to the arrest of almost all of the embryos at the eight-cell stage. Therefore, sensitivity to Cd toxicity appears after the third cleavage, and brief exposure to Cd during earlier cleavages does not lead to irreversible damage. The mechanisms of Cd resistance in two-cell embryos are unknown, but this finding is consistent with the above mentioned hypothesis that Cd influx may be largely prevented and/or efflux is heightened in two-cell mouse embryos. The finding reported here that MT mRNA is present only at low basal levels in ova and fertilized eggs, and that Zn is toxic to the two-cell embryo, suggests that these embryos do not contain substantial stores of maternal MT proteins that could perhaps confer protection from heavy metals. However, a recent study reported strong immunostaining of MT in the cytoplasm of the rat ovum (Nishimura et al., 1989). Whether Zn toxicity to the preimplantation embryo is exerted by an intracellular mechanism remains to be determined. Overall, the studies reported here suggest that expression and metal regulation of the MT genes during preimplantation embryo development cannot account for changes in the embryotoxicity of metals. Another novel finding in this study was the specific expression and metal regulation of MT genes in epithelial cells in the isthmus region of the oviduct on Days 3 and 4 of pregnancy. This suggests that the isthmus epithelia may play a role in homeostasis of essential metals in the preimplantation embryo at these stages of development, perhaps by contributing to maintenance of locally optimal concentrations of essential metals such as Zn and copper. Elevated MT in the oviduct epithelia may also serve a protective role against Cd, which is a potent toxin for Day 3 and Day 4 embryos in vitro. However, this seemsunlikely given the relatively low levels of MT mRNA found in the oviduct. Whether Cd is embryotoxic

FIG. 5. Detection of MT mRNA in the mouse oviduct using irk sifu hybridization. (A) Mouse oviducts from the indicated days of pregnancy (DlLD4) were perfusion-fixed and analyzed by i?~ situ hybridization as described in the legend to Fig. 2. Autoradiography was for 8 days and bright-field photomicrographs (200x) of the isthmus region are shown. E, epithelium; S, stroma; M, muscle. (B) D3 oviducts from control and 5 hr after a SC injection of 100 pmoles of ZnCl,/kg body wt (Zn) were analyzed by i?c situ hybridization. Autoradiography was for 4 days, and bright-field photomicrographs (200x) of the ampulla and isthmus regions of these oviducts are shown. Note that the length of autoradiography in A is twice that for B.

24

DEVELOPMENTALBIOLOGY

VOLUME 145, 1991

ANDREWSETAL.

MT wRNA during

Eurly

Lkwloprt~rr~t

25

26

DEVELOPMENTAL BIOLOGY

the preimplantation period requires further investigation. Two earlier studies have indicated that Cd injection during the preimplantation period has little effect on embryo development (Chiquoine, 1965; Giavini et al., 1980). However, in those studies a limited number of animals were examined and no dose-response data were generated. Several instances of cellspecific expression of the MT genes in the reproductive tract and embryo have been documented recently (Andrews et ab, 1984; De et al., 1989; Nishimura et al., 1989). What regulates oviduct MT gene expression remains to be determined, but the restricted expression of the MT genes to the oviduct isthmus epithelium cannot be ascribed to the selective accessibility of these cell to metal ions, since the blood borne components supplying the oviduct epithelium must first traverse the stromal compartment. This tightly regulated cell-specific expression is suggestive of an autocrine/paracrine type of regulation. In summary, these studies utilized the preimplantation mouse embryo as a model system in which to study the effects of metal ions on the regulation of MT gene expression during development. This model should also prove useful for examining the effects of a variety of other agents (cytokines, glucocorticoids, chemotherapeutic drugs, X-irradiation) that can up-regulate MT gene expression, and which may also influence embryonic development.

VOLUME 145,199l

in vivo during

TABLE 1 EFFECTS OF EXPOSURE TO ELEVATED ZINC ON THE IN VITRO DEVELOPMENT OF PREIMPLANTATION MOUSE EMBRYOS INTO BLASTOCYSTS Stage of development”

Number of embryos

Two-cell Four-cell Eight-cell Morulae

52 38 60 42

4

Cells

(38%) (55%) (72%) (88%)

“Embryos at specific stages of development were selected from among those flushed from the reproductive tract as follows: two-cell (D2, 1500 hr), four-cell (D3, 0400 hr), eight-cell (D3, 0900 hr), and compacted eight-cell/morula (D3, 1600 hr). These embryos were cultured in the presence or absence of 50 FM ZnC1, for 18 hr, washed four times in fresh medium, and recultured to assess their potential to develop into blastocysts. The total culture periods, including metal exposure time, allowed for development to the blastocyst stage were as follows: 66 hr for two-cell, 53 hr for four-cell, 48 hr for eight-cell, and 42 hr for morula stage embryos. Embryonic development was examined under a dissecting microscope. Embryos not progressing to the blastocyst stage during these culture periods degenerated after continued culture and did not form blastocysts. Results are based on four to five independent experiments each, and all experiments included control embryos cultured without exposure to excess zinc-containing medium, in which case an average of 85% (range 82 to 89%‘) developed to the blastocyst stage. Data were analyzed by the x2 test (Heel, 1984) as discussed under Results.

12 per

morulae

Embryo

FIG. 6. Age-dependent cadmium toxicity in preimplantation mouse embryos. Preimplantation mouse embryos were collected at the indicated stages of development and cultured in vitro for 8 hr in the presence of 50 PM Cd. Embryos were washed and allowed to develop in &ro to the blastocyst stage. The percentage of embryos that developed into blastocysts was recorded in three separate experiments involving a total of about 50 embryos per group. In the absence of Cd, an average of 85% of the embryos from each stage (range 82 to 89%) developed to the blastocyst stage. x2 analysis of the data (Hoel, 1984) establish that Cd has significant (P < 0.001) effects on blastocyst formation from the four-cell and older embryos. In experiments involving two-cell embryos, the viability of the blastocysts was confirmed by embryo transfer into a foster mother.

This work was supported by grants from the NIH (ES 04725) to G.K.A., and (HD 12304) to S.K.D. M.T.M. is a March of Dimes predoctoral fellow. Y.M.H. was supported by an NSF graduate fellowship, and Swapan De was supported by a Wesley Foundation Scholar Program in Cancer Research postdoctoral fellowship. We are indebted to Mrs. Cathy O’Rourke for excellent technical assistance.

Blastocysts 20 21 43 37

8

REFERENCES ABRAHAM, R., CHARLES, A. K., MANKES, R., LEFEVRE, R., RENAK, V., and ASHOK, L. (1986). In vitro effects of cadmium chloride on preimplantation rat embryos. Ecoto~icol. Environ. Safety 12,213-219. ANDERSEN, R. D., TAPLITZ, S. J., WONG, S., BRISTOL, G., LARKIN, B., and HERSCHMAN, H. R. (1987). Metal-dependent binding of a factor in viva to the metal-responsive elements of the metailothionein-I gene promoter. Mol. Cell. Biol. 7, 3574-3581. ANDREWS, G. K., and ADAMSON, E. D. (1987). Butyrate selectively activates the metallothionein genes in teratocarcinoma cells and induces hypersensitivity to metal induction. Nucleic Acids Res. 15, 5461-5475.

ANDREWS, G. K., ADAMSON, E. D., and GEDAMU, L. (1984). The ontogeny of expression of murine metallothionein: Comparison with the n-fetoprotein gene. Dev. Biol. 103, 294-303. ANDREWS, G. K., HUET, Y. M., LEHMAN, L. D., and DEY, S. K. (1987). Metallothionein gene regulation in the preimplantation rabbit blastocyst. Development 100, 463-465. ANGERER, L. M., and ANGERER, R. C. (1981). Detection of poly(A)+ RNA in sea urchin eggs and embryos by quantitative in situ hybridization. Nucleic Acids Res. 9,2819-2840. ANGERER, L. M., KAWCZYNSKI, G., WILKINSON, D. G., NEMER, M., and ANGERER, R. C. (1986). Spatial patterns of metallothionein mRNA expression in sea urchin embryos. Dev. Biol. 116,543-547.

BRINSTER, R. L., CHEN, H. Y., WARREN, R., SARTHY, A., and PALMITER, R. D. (1982). Regulation of metallothionein-thymidine kinase fusion plasmids injected into mouse eggs. Nature 296, 39-42. CHIQUOINE, A. D. (1965). Effects of cadmium chloride on the pregnant albino mouse. J. Reprod. Fe&l. 10, 263-265. COMPERE, S. J., and PALMITER, R. D. (1981). DNA methylation controls inducibility of the mouse metallothionein-I gene in lymphoid cells. Cdl 25, 233-240. Cox, K. H., DELEON, D. V., ANGERER, L. M., and ANGERER, R. C. (1984). Detection of mRNAs in sea urchin embryos by in situ hybridization using asymmetric RNA probes. Den Bid. 101, 485-502. DE, S. K., DEY, S. K., and ANDREWS, G. K. (1990a). Cadmium teratogcnicity and its relationship with metallothionein gene expression in midgestation mouse embryos. Tozicolo~y, 64, 899104. DE, S. K., MCMASTER, M., and ANDREWS, G. K. (1990b). Endotoxin induction of murine metallothionein gene expression. J. Biol. Che?~. 265, l&267-15,274. DE, S. K., MCMASTER, M. T., DEY, S. K., and ANDREW& G. K. (1989). Cell-specific metallothionein gene expression in mouse decidua and placentae. D~wlopmw~t 107, 611-621. DELEON, D. V., Cox, K. H., ANGERER, L. M., and ANGERER, R. C. (1983). Most early-variant histone mRNA is contained in the pronucleus of sea urchin eggs. Dev. Bid. 100, 197206. DEY, S. K., and JOHNSON,D. C. (1980). Histamine formation by mouse preimplantation embryos. J. Reprod. Fertil. 60, 457-460. DURNAM, D. M., PERRIN, F., GANNON, F., and PALMITER, R. D. (1980). Isolation and characterization of the mouse metallothionein-I gene. PIW. Nutl. Acclcl. Sci. CBA 77, 65116515. FIRST, N. L., and BARNES, F. L. (1989). Development of preimplantation mammalian embryos. Iyl “Development of Preimplantation Mammalian Embryos and Their Environment” (K. Yoshinaga, and T. Mori, Eds.), pp. 15-170. A. R. Liss, New York. GIAVINI, E., PRATI, M., and VISMARA, C. (1980). Effects of cadmium, lead and copper on rat preimplantation embryos. BrrlL Environ. Cm tu m. Tarid

25, 702-705.

GLANVILLE, N., DURNAM, D. M., and PALMITER, R. D. (1981). Structure of mouse metallothionein-I gene and its mRNA. Nature 292, 267269. GREENWALD, G. S., and TERRANOVA, P. F. (1988). Follicular secretion and its control. 1~ “The Physiology of Reproduction” (E. Knobil, and J. Neil, ct ul., Eds.), pp. 387-445. Raven Press, New York. HAMER, Ll. H. (1986). Metallothioneins. A~??IL. Rw. Bioch~n~ 55, 913951. HEIKKILA, J. J., MILLER, J. O., SCHULTZ, G. A., KLOC, M., and BROWDER,L. (1985). Heat shock expression during early animal development. In “Changes in Gene Expression in Response to Environmental Stress” (G. iltkinson, and D. B. Walden, Eds.), pp. 135-158. Academic Press, New York. HOEL, P. G. (1984). “Introduction to Mathematical Statistics,” fifth ed., pp. 252257. Wiley, New York. HURLEY, L. S., and SHRADER, R. E. (1975). Abnormal development of preimplantation rat eggs after three days of dietary Zn deficiency. h’utcrrr254, 427-429. IMBERT, J., ZAFARULLAH, M., CULOTTA, V. C., GEDAMU, J., and HAMER, D. (1989). Transcription factor MBF-I interacts with metal regulatory elements of higher eukaryotic metallothionein genes. Mol. Cell. Bid. 9, 5315-5323. KARIN, M. (1985). Metallothioneins: Proteins in search of function. Cd/ 41, 9-10. MAGNUSON, T., and EPSTEIN, C. J. (1987). Gene expression during early mouse development, 111“The Mammalian Preimplantation Embryo: Regulation of Growth and Differentiation” (B. D. Bavister, Ed.), pp. 133150. Plenum, New York.

S., and HUANG, P. C. (1989). Intracellular metallothionein concentration and the rate of zinc or cadmium influx and MT mRNA accumulation in a CHO Cdr variant. Eq. CeIZ Res. 1X5,166175. MUELLER, P. R., SALSER, S. J., and WOLD, B. (1988). Constitutive and metal-inducible protein: DNA interactions at the mouse metallothionein I promoter examined by irk llivo and irr vitro footprinting. Gews Del>. 2, 412-427. NEMER, M., TRAVAGLINI, E. C., RONDINELLI, E., and D’ALONZO, J. (1984). Developmental regulation, induction, and embryonic tissue specificity of sea urchin metallothionein gene expression. Dw. Bid. 102,471l482. NISHIMURA, N., NISHIMURA, H., and TOHYAMA, C. (1989). Localization of metallothionein in female reproductive organs of rat and guinea pig. J Hiotochcm. C~tochrm. 37, 16Oll1607. PEDERSEN, R. A., and LIN, P. (1978). Cadmium toxicity in preimplantation mouse embryos. 171 “Developmental Toxicology of EnergyRelated Pollutants” (D. D. Mahlum, M. R. Sikov, P. L. Hackett, and F. D. Andrew, Eds.), pp. 600-613. Technical Information Center IJ.S. Department of Energy. RAPPOLEE, D. A., BRENNER, C. A., SCHULTZ, R., MARK, D., and WERB, Z. (1988). Developmental expression of PDGF, TGF-(U and TGF-0 genes in preimplantation mouse embryos. Science 241,1823-1825. RAPPOLEE, D. A., WANG, A., MARK, D., and WERB, Z. (1989). Novel method for studying mRNA phenotypes in single or small numbers of cells. J. Cfdl. Biochem. 39, l-11. SAIKI, R. K., GELFAND, D. H., STOFFEL, S., SCHARF, S. J., HIGUCHI, R., HORN, G. T., MULLIS, K. B., and ERLICH, H. A. (1988). Primerdirected enzymatic amplification of DNA with a thermostable DNA polymerase. Scione~ 239, 487-494. SCHMID, B. P., HALL, J. L., GOULDING, E., FABRO, S., and DIXON, R. (1983). h vitro exposure of male and female mice gametes to cadmium chloride during the fertilization process, and its effects on pregnancy outcome. To.ricol. Appl. Pharmad 69, 326-332. SEARLE, P. F., DAVISON, B. L., STUART, G. W., WILKIE, T. M., NORSTEDT, G.. and PALMITER, R. D. (1984). Regulation, linkage and sequence of mouse metallothioncin-I and II genes. Mol. Cd. Bid. 4, 1221-1230. SPIELMANN, H., and VOGEL, R. (1989). Unique role of preimplantation embryos to understand mechanisms of embryotoxicity in early pregnancy. Crit. Rel: Tosicol. 20, 51-64. STEVENS, M. E., MENESES, J. J., and PEDERSEN, R. A. (1989). Expression of a mouse mrtallothionein-Escherichia coli @-galactosidase fusion gene (MT-figal) in early mouse embryos. Eq. Cd Res. 183, 319-325. STUART, G. W., SEARLE, P. F., and PALMITER, R. D. (1985). Identification of multiple metal regulatory elements in mouse metallothionein-I promoter by assaying synthetic sequences. Nuture 317, 828831. WEBB, M. (1987). Mctallothionein in regeneration, reproduction and development. E.rperiwtiu 52(Suppl.), 483-498. WHITTEN, W. K. (1971). Nutrient requirements for the culture of preimplantation embryos irr t,itro. Ad?,. Biol. Sci. 6, 129-141. WILKINSON, D. G., and NEMER, M. (1987). Metallothionein genes MT-cu and MT-8 expressed under distinct quantitative and tissue-specific regulation in sea urchin embryos. Mol. Cell. Biol. 7, 48-58. Yu, H. S., and CHAN, S. T. H. (1986). Effects of cadmium on lactate dehydrogenase activities in mouse preimplantation embryos at various stages. Terutdo{]y34, 323-329. Yu, H. S.. TAM, P. P. L., and CHAN, S. T. H. (1985). Effects of cadmium on prcimplantation mouse embryos ilc vitro with special reference to their implantation capacity and subsequent development. TentO/O(Jy 32, 347-353. MORRIS,