Development 100, 463-469 (1987)Printed in Great Britain © T h e Company of Biologists Limited 1987

463

Metallothionein gene regulation in the preimpiantation rabbit blastocyst

G. K. ANDREWS1, Y. M. HUET2, L. D. LEHMAN3•• and S. K. DEY2

'Department of Biochemistry, ^Departments of Obstetrics - Gynecology and Physiology, Ralph L. Smith Research Center and ^Department ofPharmacology, University of Kansas Medical Center, 39th and Rainbow Boulevard, Kansas City, Kansas 66103, USA

•Current address: The Proctor and Gamble Company, Miami Valley Laboratories, PO Box 39175, Cincinnati, OH 45247, USA

Summary

Expression of metallothionein (MT) genes in thepreimpiantation rabbit blastocyst was analysed bydetermination of the levels of MT mRNA and relativerates of MT synthesis. MT was found to be constitut-ively expressed at low levels in the blastocyst. Ex-posure of the day-6 blastocyst to zinc ions in vitrorapidly increased the level of MT gene expression in adose-dependent manner, with a ten-fold induction inthe relative rate of synthesis at 400^M-Zn2+. Ion-exchange chromatography of pulse-labelled blastocystprotein showed that the relative rates of synthesis ofboth MT-I and MT-II were markedly increased fol-lowing zinc treatment, with MT-I being the predomi-

nant isometallothionein. Zinc induction of MT syn-thesis in the blastocyst was also detected on day 4 ofgestation just after the morula-torblastocyst tran-sition. In contrast to the zinc effects on MT, in vitroexposure to 10/iM-Cd2+ resulted in a large inductionof MT mRNA but only a modest increase in therelative rate of MT synthesis. Cadmium was found tobe toxic to the day-6 blastocyst, and 10/iM-Cd2+

induced an acute stress response as indicated by adramatic induction of heat-shock protein (HSP-70)gene expression.

Key words: metallothionein, rabbit blastocyst, metalions, heat shock, gene regulation, zinc, cadmium.

Introduction

Metallothioneins (MT) are small cysteine-rich pro-teins that bind to heavy metal ions with high affinity.Two isoforms of MT (MT-I and MT-II), which differslightly in amino acid sequence and net charge, arefound in mammals (for review see Hamer, 1986).Transcription of MT genes can be rapidly increasedby several agents including metal ions, glucocorti-coids and interleukin-1 (Searle et at. 1984).

MT has been suggested to play a role in zinc andcopper homeostasis, and to provide protection fromheavy metal toxicity (Karin, 1985; Hamer, 1986). MThas recently been shown to scavenge free hydroxylions (Thornalley & Vasak, 1985) and may be involvedin the animals' responses to stress. The importance ofMT to the developing embryo is suggested by thefinding that MT genes are expressed in a tissue-specific manner during development. In the mid-gestation mouse embryo, MT is actively synthesizedby parietal and visceral endoderm cells of the extra-embryonic membranes, and by the developing liver(Andrews, Adamson & Gedamu, 1984).

Although development of the preimpiantation em-bryo can be radically disturbed by exposure to metalions such as cadmium (Yu, Tarn & Chan, 1985) orconversely by deficiencies in metal ions such as zinc(Hurley & Shrader, 1975); nothing is known aboutexpression of MT genes during this important periodof embryonic growth and differentiation. We chose tostudy MT gene expression in the preimpiantationrabbit blastocyst (day 4 and day 6 of gestation)because rabbit embryos at these stages have manymore cells than do the preimpiantation rat or mouseblastocysts (Lutwak-Mann, 1966) which makes bio-chemical analysis more feasible. In addition, therabbit is an induced ovulator allowing for accuratetiming of embryo development. We report here thatMT genes are constitutively expressed at low levels inrabbit blastocysts and that the MT-I and MT-II genescan be coordinately induced by exposure of theblastocyst to metal ions. Furthermore, we show thatcadmium induces an acute stress response in theblastocyst leading to increased expression of MTgenes and heat-shock genes.

464 G. K. Andrews, Y. H. Huet, L. D. Lehman and S. K. Dey

Methods

Induction ofsuperovidation and collection of embryosAdult (7-8 kg, 4-4i month old) virgin female New Zealandwhite rabbits were induced to superovulate by injecting,subcutaneously, 0-15 mg follicle stimulating hormone por-cine (Sigma Chemical Co., St. Louis, MO) dissolved in1-5 ml of O15M-NaCl solution twice daily for 3 consecutivedays. On the morning of the 4th day, rabbits were mated insuccession with two fertile bucks and then injected intra-venously with 25i.u. human chorionic gonadotropin(Sigma) dissolved in 0-5 ml saline (Mukherjee et al. 1978).Day-4 and day-6 embryos (Day 1 = 24 h postcoitum) werecollected by flushing the uteri with RPMI-1640 (KC Bio-logicals, Lenexa, KS) fortified with 2 mM-L-glutamine. Em-bryos were washed twice and incubated in the same mediaunder a gas phase of 5 % CO2, 5 % O2 and 90 % N2 at 37°C(Pakrasi & Dey, 1983).

In order to determine the effects of metal ions onblastocyst gene expression, day-6 blastocysts were incu-bated in RPMI-1640 containing 0-3% bovine serum albu-min and 2 mM-L-glutamine for 4 h in the presence or absenceof various concentrations of ZnCl2 or CdCl2. If day-6blastocysts were to be pulse labelled with [35S]cysteine (seebelow), the BSA was omitted from the final incubationmedia.

Because of the sticky nature and the subsequent difficultyin handling day-4 embryos, they were always incubated inthe presence of 0-1 % BSA.

Hybridization probesA mouse MT-I cDNA clone was provided by Dr RichardPalmiter (University of Washington, Seattle, WA). Thecloned mouse heat-shock 70 (HSP-70) gene was from DrRichard Morimoto (Northwestern University, Evanston,IL). The MT-I and HSP-70 clones were inserted into theSP6 vectors (Promega Biotech, Madison, WI) and used astemplates for the synthesis of 32P-labelled RNA probes(Melton et al. 1984). Probes had specific activities of about2xl09disintsmin"Vg~'-

Isolation of total RNABlastocyst RNA was extracted by a modification of sodiumdodecyl sulphate (SDS)-phenol-chloroform proceduresdescribed by Andrews & Teng (1979). About 80 blastocystswere collected and flash frozen in liquid nitrogen. Sampleswere stored at —70°C until RNA was extracted. The frozenblastocysts (about 1-5 ml) were adjusted to 0-5% SDS,25mM-EDTA, 75mM-NaCl, pH80 (SDS-buffer) and anequal volume of phenol (saturated with SDS-buffer,pH8-0) was added. The mixture was vortexed vigorously,chilled to 4°C and centrifuged at 16 000 £ for 10min. Theresulting aqueous phase and protein interface was re-extracted at room temperature in 0-5 vol. of phenol (satu-rated with SDS-buffer) and 0-5 vol. of chloroform: isoamylalcohol (24:1 v/v). The extract was chilled to 4°C andcentrifuged at 16 000 g for 15 min. The aqueous phase,which contained nucleic acids, was retained and RNA wasprecipitated by the addition of 3 vol. of 4M-ammoniumacetate at room temperature (Palmiter, 1973). The mixturewas chilled for 45 min at 4°C and the RNA precipitate was

collected by centrifugation at 16000 g for 50min. RNA wasdissolved in 0-5 ml of 25mM-EDTA (pH80 at room tem-perature) and precipitated as described above in 3 M-ammonium acetate. The RNA pellet was dissolved in 0-5 mlsterile water, adjusted to 0-3M-ammonium acetate andprecipitated with 2-5 vol. of absolute ethanol at -20°C for18 h. RNA was collected by centrifugation at 10 000 g for10 min at 4°C and the RNA pellet was dissolved in 25 fi\ ofsterile water. The concentration of RNA was determinedby measuring absorbance at 260 nm. A yield of 30 to 50f/gof RNA was obtained from 80 day-6 blastocysts.

Northern blot hybridizationRNA (20ng) was denatured for 10 min at 60°Cin a solutionof 20 mM-phosphate buffer (18mM-sodium phosphate [di-basic] and 2mM-sodium phosphate [monobasic]; pH8-0)containing 50% formamide and 2-2M-foimaldehyde. De-natured RNA (6 ̂ g in 20/̂ 1) was separated by electrophor-esis in a 1-5% agarose gel (5x7-5cm) containing 20mM-phosphate buffer and 2-2M-formaldehyde (Lehrach et al.1977). During electrophoresis, gels were submerged inrunning buffer (2-2 M-formaldehyde and 20 mM-phosphatebuffer) that was recirculated constantly to prevent markedchanges in pH. Electrophoresis was carried out by applyinga constant voltage (45 V) across the gels. Following electro-phoresis, gels were soaked for 30 min in lOmM-sodiumphosphate (pH70) and transferred to nitrocellulose in thepresence of 20xSSC (3M-NaCl and 0-3M-sodium citrate,pH 7-4) as described by Thomas (1980). Following transfer,the filters were baked in a vacuum oven at 75°C for 5 h.

Northern blots were prehybridized and washed as de-scribed by Andrews et al. (1984). Filters were incubated insealed bags at 65°C in 3xSET (450 mM-NaCl, 6 mM-EDTA,90mM-Tris, pH8-0) containing 0-1% SDS for 30min,followed by 3xSET containing 10 x Denhardt's solution(0-2% each of bovine serum albumin, Ficoll and poly-vinylpyrrolidone) (Denhardt, 1966) for 2h, and finally3xSET containing 10 x Denhardt's solution and yeasttRNA (250^/gmr1) for 2h. Filters were hybridized withlabelled probe for 18 h at 65°C in a solution containing3xSET, 0-1 % SDS, 10 x Denhardt's solution, yeast tRNA(250/igmr1), 10% dextran sulphate, 20mM-sodium phos-phate (pH7-2) and approximately lxl06ctsmin~' ml"1

hybridization probe. Following hybridization, filters werewashed for lh each in lxSSC containing 0-1% SDSfollowed by 0-3xSSC containing 0 1 % SDS. Autoradio-graphs of the filters were obtained by exposing X-ray film(XAR-05, Kodak, Rochester, NY) for 6 to 24 h at -70°C inconjunction with a high-plus intensifying screen (Dupont,Wilmington, DE).

Determination of relative rates of MT synthesisDay-6 blastocysts (20—40 per point) were cultured for 4h inRPMI-1640 (5 ml) in the presence or absence of metal ions(CdCl2 or ZnCl2). The blastocysts were washed twice inbalanced salt solution and the incubation was continued for1 h in cysteine-free Minimal Essential Media (3 ml) contain-ing 100/iCiml~' [35S]cysteine (specific activity oflOOOCimM"1; New England Nuclear, Boston, MA) plusmetal ions as indicated. After labelling, the blastocysts

were washed twice in cysteine-free media, passed gentlythrough an 18-gauge needle and cells were collected by low-speed centrifugation. Cells were lysed by freeze thawingthree times in 25^1 of buffer (Hamer & Walling, 1982)containing 1 % Nonidet P40, 50 mM-Tris (pH7-4), 100 ITIM-NaCl, 5 mM-dithiothreitol (DTT). The soluble proteinswere recovered by centrifugation at 15 000 g for 15min at4°C, and the supernatant (30/il) was carboxymethylated byincubated with 15^1 of 0-3M-iodoacetic acid in lM-Tris(pH 8-4) for 45 min in the dark at room temperature. Then30fi\ of sample buffer (125 mM-Tris (pH6-8), 2-5% SDS;2-5% /S-mercaptoethanol, 25% glycerol, 0-5% bromo-phenol blue) were added and the sample was boiled for5min. Equal amounts (100000ctsmin"1) of trichloroacetic-acid-precipitable radioactivity were separated by 30%acrylamide-0-1 % SDS gel electrophoresis. Gels were fixedand stained with Coomassie blue, soaked for 1 h in Enlight-ening (New England Nuclear), dried and exposed to KodakXAR-5 film at -70°C.

Day-4 blastocysts (75 per point) were cultured as de-scribed above in the absence or presence of 400 j*M-ZnCl2.After pulse labelling for 1 h with [35S]cysteine the blasto-cysts were collected by low-speed centrifugation andwashed in balanced salt solution. Cells were lysed andproteins carboxymethylated as described above. Sampleswere then heated to 85°C for 6min, chilled to 4°C andcentrifuged at 15 000 g for 15 min. Heat-stable proteinswere precipitated with 5 vol. of cold ethanol for 18 h at-20°C. Protein precipitates were collected by centrifu-gation (15000g, 30 min, 4°C), dissolved in 15 jul of samplebuffer, boiled for 5 min and the entire sample subjected toelectrophoresis and fluorography as described above.

To analyse the relative rate of synthesis of MT-I andMT-II, [35S]cysteine-labelled day-6 blastocysts (80 em-bryos) were homogenized in 2 vol. of 10 mM-Tris (pH7-4)and centrifuged at 10 000 g for 10 min. The resultant super-natant was mixed with rat liver cystosol as a carrier in a ratioof 1:4. Rat liver cytosol was prepared 24 h after injection ofzinc chloride (20 mg kg"1). Liver was homogenized in 2 vol.of 10 mM-Tris (pH 7-4) and the homogenate was centrifugedat 100 000 g for 50 min at 4°C. The blastocyst-cytosolmixture was heat denatured at 100°C for 1 min and freecysteine was removed by applying the heat-denaturedsample to a Sephadex G-25 column previously equilibratedwith 10mM-Tris-acetate (pH7-4), 3mM-DTT and 002%sodium azide. Proteins were eluted with the same bufferand the eluate (about 2xl06ctsmin"1) was then applied toa DEAE-Sephadex A25 column (Wong & Klaassen, 1979).MT-I and MT-II were eluted with a linear gradient of10-240 mM-Tris-acetate (pH7-4; containing 3mM-DTTand 002% sodium azide) at a flow rate of SOmlh"1.Fractions were collected at 10 min intervals and the amountof [35S]cysteine in each fraction was determined by liquidscintillation counting. The MT-I peak eluted betweenfractions 25-35 (approximately 65 mM-Tris-acetate),whereas MT-II eluted between fractions 90-100 (approxi-mately 165 mM-Tris-acetate).

Blastocyst metallothionein gene expression 465

Results

Expression of MT genes in the rabbit blastocyst wasanalysed by Northern blot hybridization with a comp-lementary RNA (cRNA) probe synthesized frommouse MT-I cDNA (Durnam, Perrin, Gannon &Palmiter, 1980). The day-6 rabbit blastocysts con-tained readily detectable levels of MT-I mRNA(Fig. 1; tracks 2 and 4). To ascertain the effects ofmetal ions on blastocyst MT gene expression, day-6blastocysts were incubated in vitro with 40 or 400 /XM-Zn2+ for 4h before Northern blot analysis of RNA.Zinc (400 HM) significantly increased (about fivefold)the level of MT-I mRNA in the blastocyst (Fig. 1;track 5), whereas 40 ̂ xM-Zn2+ had little effect on MT-ImRNA levels (Fig. 1; track 3). Incubation of theblastocysts with 10/iM-Cd2+ for 4h also significantlyincreased MT-I mRNA levels (Fig. 1; track 6). Thelevels of MT-I mRNA in the blastocyst after incu-bation for 4h with 400/zM-Zn2+ or 10^M-Cd2+ ap-proached those found in the zinc-induced adult liver(Fig. 1; track 1). In the course of these experiments,we noted that cadmium at concentrations greaterthan 10 pm rapidly caused the day-6 blastocyst tocollapse and severely reduced the incorporation of[35S]cysteine into protein (data not shown). Northernblot analysis of RNA recovered following incubationof blastocysts with 10/zM-Cd2+ for 4h showed that alarge induction of HSP-70 mRNA had occurred(Fig. 1; track 9), whereas 400jiM-Zn2+ had littleeffect on HSP-70 mRNA levels (Fig. 1; track 8)which were very low in the control blastocyst (Fig. 1;track 7).

To determine the effects of metal ions on therelative rate of synthesis of MT, blastocysts werecultured for 4 h with various concentrations of zinc orcadmium ions and proteins were subsequently pulselabelled for 1 h with [35S]cysteine. Newly synthesizedMT was preferentially labelled under these cultureconditions due to its high cysteine content (20 of 61amino acids) (Hamer, 1986). Labelled cytosolic pro-teins were carboxymethylated, separated on a 30 %polyacrylamide gel (Hamer & Walling, 1982) anddetected by fluorography (Fig. 2). Incubation of day-6 blastocysts with zinc ions resulted in a concen-tration-dependent increase in the relative rate of MTsynthesis with a maximal response at between 300 and600/zM-Zn2+ (Fig. 2A; tracks 2, 3 and 6). The relativerate of MT synthesis was also increased followingincubation with 10/*M-Cd2+ (Fig. 2A; track 5), how-ever the magnitude of this induction was muchreduced relative to the maximal induction of MTsynthesis by zinc (Fig. 2A; compare tracks 5 and 6).Incubation of day-4 embryos with 400 /XM-Zn2+ for 4 halso increased the relative rate of MT synthesis(Fig. 2B). About 30 % of the day-4 embryos were at

466 G. K. Andrews, Y. H. Huet, L. D. Lehman and S. K. Dey

1 2 3 4 5 6 7 8

- HSP-70

MT-

LiveriC

I40/zMZn2 +

IC

I I400,uM 10 jUMZn 2 + Cd 2 +

IC 400 juM 10 /

Z n 2 + C d 2 +

Fig. 1. Northern blot detection of MT and HSP-70 in day-6 rabbit blastocysts. RNA (6^ig) was separated on 1-5 %agarose-formaldehyde gels (Lehrach et al. 1977) and Northern blotted to nitrocellulose filters (Thomas, 1980). Filterswere hybridized, under conditions described previously (Andrews et al. 1984), with a 32P-labelled MT-I cRNA probe orwith a mouse HSP-70 cRNA probe. After hybridization the filters were washed at 66°C under stringent conditions andthe hybrids were detected by autoradiography. Samples were as follows: tracks 2, 4 and 7 (C), control day-6 blastocysts;track 3, blastocysts incubated for 4h with 40/iM-zinc chloride; tracks 5 and 8, 400/iM-zinc chloride; tracks 6 and 9, 10/^M-cadmium chloride; track 1, adult liver after injection of zinc chloride (25mgkg~', injected 2 times in 24h).

1

- M T -MT

1

c1

100/JviZ n 2 +

r600 fMZn2 +

iC

#1

10/^MCd2 +

I300 juMZn 2 +

B

C 400//IVZ n 2 +

Fig. 2. Effects of metal ions on the relative rate of MT synthesis in rabbit blastocysts. (A) Day-6 blastocysts wereincubated with ZnCl2 or CdCl2 for 5 h and proteins were pulse labelled with [35S]cysteine during the last hour of culture.Labelled proteins (lOOOOOctsmin"1) were carboxymethylated, separated by 30% SDS-polyacrylamide gelelectrophoresis and detected by fluorography (Hamer & Walling, 1982). Samples were as follows: tracks 1, 4 (C),controls; track 2, 100/iM-Zn2+; track 3, 600/m-Zn2+; track 5, 10/iM-Cd2+; and track 6, 300/JM-Zn?+. (B) Day-4 rabbitblastocysts were incubated with 400fiM-Zn2+ for 4h and the relative rate of synthesis of MT analysed as detailed in theMethods section. The electrophoretic migration of purified rabbit MT is indicated. The isoforms of MT comigrated inthese gels.

Blastocyst metallothionein gene expression 467

the preblastocyst stage of development and 70%were blastocysts.

The above results established that in vitro exposureof blastocysts to zinc increases the abundance of MTmRNA leading to an increase in the relative rate ofMT synthesis. However, the isoforms of MT (MT-Iand MT-II) were not resolved by SDS-polyacryl-amide gel electrophoresis. Therefore, to examine theeffects of zinc on the relative rates of synthesis of theisoforms of MT, [35S]cysteine-labelled heat-stablecytosolic proteins were subjected to anion-exchangechromatography (Wong & Klaasen, 1979) (Fig. 3).Exposure of blastocysts to Zn2+ (400 JIM for 4h)increased the relative rates of synthesis of bothisoforms of MT about tenfold. The cytosol fromcontrol blastocysts contained 2-8xl06ctsmin~1 ofwhich 1-6% eluted in the MT peaks. In contrast,cytosol from zinc-treated blastocysts contained l-8x

^ 1 of which 16% eluted in MT peaks

50

25*

15

_ 1 4

2 13X

I 1 2

t rac

t

g. 10

?c 9

e

(Ct!

1 61/5

i. 5

4

3

2

1

U o o Control

•

•

.

• c

•

•

•

•

• • Zn treated

b

Ii/

y

''' ii/ 1• 111 • ' l i

l y ^ i

0-2

- 0-1

10 20 30 40 50 60 70 80 90 100 110Fraction number

Fig. 3. Synthesis of isometallothioneins by rabbitblastocysts following exposure to zinc. Day-6 rabbitblastocysts were incubated for 5h in the absence (O) orpresence (•) of 400/YM-ZnG2 and proteins were pulselabelled with [35S]cysteine for the last 2h of incubation.Incorporation of [35S]cysteine into MT-I and MT-II wasdetermined by anion-exchange chromatography (Wong &Klaassen, 1979).

(Fig. 3). These results establish that the M r genes inthe rabbit blastocyst are coordinately induced by zinc.

Discussion

The finding that the MT genes are constitutivelyexpressed, albeit at low levels, in the preimplantationblastocyst is consistent with the hypothesis that MTplays an important role in development. Presumably,MT could provide zinc or copper ions to appropriateenzymes in the rapidly growing blastocyst. In ad-dition, a recent report indicates that MT can scavengefree radicals (Thornalley & Vasak, 1985) and maytherefore serve to protect cells from the injuriouseffect of these molecules. In this regard, it is interest-ing to note that, during blastocyst formation in therabbit, the consumption of oxygen increases three-to fourfold (Brinster, 1974), which could lead toincreased formation of free radicals. Both isoforms ofMT (MT-I and MT-II) are synthesized by the blasto-cysts and the relative rate of synthesis of each isoformincreased in response to increased levels of metal ion.Thus, from the studies reported here, as well as fromour earlier studies of mouse development (Andrewset al. 1984), there is no evidence for the exclusiveexpression of an isometallothionein in a tissue-specific or temporal manner during early embryonicdevelopment. This suggests that the isoforms of MThave a common physiological function in developingtissues.

Mechanisms that normally regulate MT expressionin the blastocyst are unknown. Uptake of zinc by theblastocyst is slow (Lutwak-Mann & Mclntosh, 1969)which could account for the high levels of zincrequired for rapid increase of MT gene expression inour experiments. However, it has been shown thatduring the progestational phase, the rabbit endo-metrium is rich in zinc (Mclntosh & Lutwak-Mann,1972). This is consistent with the idea that metal ionsmay be the physiological inducer of MT gene ex-pression in the blastocyst. Our results clearly estab-lish that the blastocyst can respond to alterations inenvironmental metal ion concentration by increasedexpression of MT genes. In preliminary studies wehave noted that glucocorticoids, which are able toinduce hepatic MT, have little effect on blastocyst MTgene expression (unpublished data). However, ourresults do not exclude the possibility that the consti-tutive expression of MT genes in the blastocyst is apreprogrammed developmental event. A recentstudy of sea urchin development showed that MT is amaternal mRNA with embryonic MT niRNA syn-thesis beginning at the 8-cell cleavage stage (Nemer etal. 1984). The cell type(s) expressing the AfTgene inthe rabbit blastocyst has not been identified directly;however, at this stage of development over 90 % of

468 G. K. Andrews, Y. H. Huet, L. D. Lehman and S. K. Dey

the cells are trophoblast cells (Lutwak-Mann, 1966).This makes it likely that these cells are the majorsource of blastocyst's MT in the rabbit. In situhybridization experiments will be required in order todetermine in which cells and when during preimplan-tation development the MT genes are expressed.

Yu et al. (1985) recently showed that in vitroexposure of preimplantation mouse embryos to10^M-Cd2+ was embryotoxic, leading to necrosis ofmorulae within 24 h. We show here that cadmium hasa direct effect on blastocyst gene expression, rapidlyenhancing expression of MT as well as HSP-70 genes.Although cadmium enhanced transcription of the MTgenes in the blastocyst, we noted only a modestenhancement in the relative rate of MT synthesis.This probably reflects a toxic effect of this metal. Incontrast, a 40-fold higher dose of zinc greatlyenhanced the relative rate of MT synthesis in theblastocyst without such dramatic effects on HSP-70gene expression. These results tempt us to predictthat pretreatment of the blastocyst with zinc might beable to provide protection from subsequent toxiceffects of cadmium exposure, through enhanced syn-thesis of MT. It has been shown that zinc can preventthe teratogenic and embryotoxic effects of cadmiumon mouse embryos in whole embryo culture in vitro(Warner et al. 1984). However, no measurements ofMT synthesis were reported in that study. Theresponse of the HSP-70 gene to a variety of physio-logical stresses has been well documented (Wu et al.1986). It has been shown previously that HSP-70 geneexpression can be induced by elevated temperature atthe blastocyst stage of development in the mouseand rabbit, but not in the cleavage-stage embryos(Heikkila et al. 1985). As reported here, cadmium canalso induce HSP-70 in the rabbit blastocyst. We donot yet know the stage of rabbit embryonic develop-ment at which either the MT or the HSP-70 genesbecome competent to respond to metal ions. It hasbeen shown, however, that Mr-fusion genes microin-jected into the fertilized mouse egg are responsive tometal ions (Brinster et al. 1982). This established thatthe factors necessary for metal ion enhancement ofMT gene promoter activity are also functional at thisstage of development. Since the MT and HSP-70genes each have a conserved sequence, termed ametal-regulatory element (Wu et al. 1986), in thepromoter region, it seems likely that the HSP-70 genewill also be metal-responsive in the fertilized egg.

In summary, we have shown that the MT genes areconstitutively expressed in the rabbit blastocyst andthat MT-I and MT-II can be coordinately induced inresponse to increased concentrations of metal ions(Zn2+, Cd2+) in the environment. Furthermore, wefound that exposure of the blastocyst to cadmium

induced a marked stress response as measured byinduction of HSP-70 gene expression.

This work was supported (in part) by BRSG S07RR05373 awarded by the Biomedical Research SupportGrant Program, Division of Research Resources and byHD 12122 from NIH, and USPHS grant ES-01142 fromNIEHS. LDL was supported by USPHS gTant ES-07079and a Proctor and Gamble Fellowship. YMH was sup-ported by an NSF Minority Graduate Fellowship.

References

ANDREWS, G. K., ADAMSON, E. D. & GEDAMU, L.(1984). The ontogeny of expression of murinemetallothionein: Comparison with the a'-fetoproteingene. Devi Biol. 103, 294-303.

ANDREWS, G. K. & TENG, C. S. (1979). Studies on sex-organ development: Prenatal effect of estrogenichormone on tubular gland cell morphogenesis andovalbumin gene expression in chick mullerian duct.Biochem. J. 182, 272-286.

BRINSTER, R. L. (1974). Embryo development. /. AnimalSci. 38, 1003-1012.

BRINSTER, R. L., CHEN, H. Y., WARREN, R., SARTHY, A.& PALMITER, R. D. (1982). Regulation ofmetallothionein-thymidine kinase fusion plasmidsinjected into mouse eggs. Nature, Lond. 296, 39-42.

DENHART, D. T. (1966). A membrane-filter technique fordetection of complementary DNA. Biochem. Biophys.Res. Commun. 23, 641-646.

DURNAM, D. M., PERRJN, F., GANNON, F. & PALMITER, R.D. (1980). Isolation and characterization of the mousemetallothionein-I gene. Proc. natn. Acad. Sci. U.S.A.77, 6511-6515.

HAMER, D. H. (1986). Metallothionein. A. Rev. Biochem.55, 913-951.

HAMER, D. J. & WALLING, M. (1982). Regulation in vivoof a cloned mammalian gene: Cadmium inducestranscription of a mouse metallothionein gene in SV40vectors: J. molec. appl. Genet. 1, 273-288.

HEIKKILA, J. J., MILLER, J. O., SCHULTZ, G. A., KLOC,M. & BROWDER, L. (1985). Heat shock expressionduring early animal development. In Changes in GeneExpression in Response to Environmental Stress (ed. B.G. Atkinson & D. B. Walden), pp. 135-158. NewYork: Academic Press.

HURLEY, L. S. & SHRADER, R. E. (1975). Abnormaldevelopment of preimplantation rat eggs after threedays of maternal dietary zinc deficiency. Nature, Lond.254, 427-429.

KARIN, M. (1985). Metallothioneins: Proteins in search offunction. Cell 41, 9-10.

LEHRACH, H., DIAMOND, D., WOZNEY, J. M. &BOEDTKER, H. (1977). RNA molecular weightdetermination by gel electrophoresis under denaturingconditions, a critical reexamination. Biochemistry 16,4743-4751.

LUTWAK-MANN, C. (1966). Some physiological andbiochemical properties of the mammalian blastocyst.Bull. Swiss. Acad. Med. Sci. 22, 101-122.

Blastocyst metallothionein gene expression 469

LUTWAK-MANN, C. & MCINTOSH, J. E. A. (1969). Zincand carbonic anhydrase in the rabbit uterus. Nature,Lond. 221, 1111-1114.

MELTON, D. A., KRIEG, P. A., REBAGLIATI, M. R.,

MANIATIS, T., KINN, K. & GREEN, M. R. (1984).Efficient in vitro synthesis of biologically active RNAand RNA hybridization probes from plasmidscontaining a bacteriophage SP6 promoter. Nucl. AcidsRes. 12, 7035-7057.

MCINTOSH, J. E. A. & LUTWAK-MANN, C. (1972). Zinctransport in rabbit tissues. Some hormonal aspects ofthe turnover of zinc in female reproductive organs,liver and body fluids. Biochem. J. 126, 869-876.

MUKHERJEE, A., DEY, S. K., SENGUPTA, J., RAMADOSS, C.

S. & DICKMANN, Z. (1978). Regulatory enzymes ofcarbohydrate and energy metabolism in the rabbitblastocyst. /. Reprod. Pert. 53, 77-80.

NEMER, M., TRAVAGLINI, E. C , RONDINELLI, E. &

D'ALONZO, J. (1984). Developmental regulation,induction, and embryonic tissue specificity of sea urchinmetallothionein gene expression. Devi Biol. 102,471-482.

PAKRASI, P. L. & DEY, S. K. (1983). Catechol estrogensstimulate synthesis of prostaglandins in thepreimplantation rabbit blastocyst and endometrium.Biol. Reprod. 29, 347-354.

PALMITER, R. D. (1973). Ovalbumin messengerribonucleic acid translation. /. biol. Chem. 248,2095-2106.

SEARLE, P. F., DAVISON, B. L., STUART, G. W., WILKIE,T. M., NORSTEDT, G. & PALMITER, R. D. (1984).Regulation, linkage and sequence of mousemetallothionein-I and II genes. Molec. cell. Biol. 4,1221-1230.

THOMAS, P. S. (1980). Hybridization of RNA and smallDNA fragments transferred to nitrocellulose. Proc.natn. Acad. Sci. U.S.A. 77, 5201-5205.

THORNALLEY, P. J. & VASAK, M. (1985). Possible role formetallothionein in protection against radiation-inducedoxidative stress. Kinetics and mechanism of its reactionwith superoxide and hydroxyl radicals. Biochim.Biophys. Ada 827, 36-44.

WARNER, C. W., SADLER, T. W., TULIS, S. A. & SMITH,M. K. (1984). Zinc amelioration of cadmium-inducedteratogenesis in vitro. Teratology 30, 47-53.

WONG, K.-L. & KLAASSEN, C. D. (1979). Isolation andcharacterization of metallothionein which is highlyconcentrated in newborn rat liver. /. biol. Chem. 254,12399-12403.

Wu, B. J., KINGSTON, R. E. & MORIMOTO, R. I. (1986).Human HSP70 promoter contains at least two distinctregulatory domains. Proc. natn. Acad. Sci. U.S.A. 83,629-633.

Yu, H. S., TAM, P. P. L. & CHAN, S. T. H. (1985).Effects of cadmium on preimplantation mouse embryosin vitro with special reference to their implantationcapacity and subsequent development. Teratology 32,347-353.

(Accepted 11 March 1987)