Cadmium-induced accumulation of metallothionein messenger RNA in rat liver.

Multiple injections of nontoxic levels of cadmium to a rat result in much higher level of metallothionein (MT) production in the liver than does the single injection. In order to understand the underlying mechanisms we have quantitated and compared the metallothionein-specific messenger RNA contents in the livers following the two induction regimens. Cell-free translation assays coupled with specific immunoprecipitation of MT revealed that MT-mRNA activity in livers of animals multiply injected with Cd is 7- to 10-fold higher than that in livers 4 h after a single Cd-induction. By oligo(dT)-cellulose chromatography, sucrose density gradient centrifugation, and methylmercuric hydroxide-agarose gel electrophoresis this mRNA has been enriched approximately 100-fold from the total RNA. The size of the mRNA is about 400 nucleotides. Hybridization assays with a complementary DNA probe synthesized against the enriched MT-mRNA showed a 4-fold difference in the level of MT-mRNA between the two induction regimens in agreement with the results obtained by the cell-free translation assays. The possible mechanisms for these observations in consideration of the short lived nature of MT-mRNA are discussed.

The transition class I1 B metal cadmium (Cd) is known to induce an accumulation in the rat liver of thionein protein which binds strongly to this and other metals of this class (1, 2). The exact mechanism for such an induction is not clear, although increases in the level of mRNA have been suspected. Induction with a single injection of Cd resulted in a rapid appearance of mRNA for Cd-thionein. The level of this mRNA peaked 4-5 h after the induction and decreased thereafter, indicating that this RNA has a definable and relatively short half-life (3)(4)(5).
Cd-thionein accumulates after the appearance of its mRNA ( 3 , 5 ) . The half-life of metallothionein in the liver is 3% days (6-8). Cd released from the degraded protein may further induce a new round of transcription.
Animals which are chronically exposed to nontoxic doses of Cd develop a tolerance toward subsequent exposure to toxic levels of this metal. Repetitive injections of Cd resulted in elevated levels of Cd-thionein in the rat liver as compared to a single injection (9,10). To date, however, little information is available as to the transcriptional state elicited by the * This study was supported by National Institutes of Health Grants KO1 ES01596, and TO1 ES00034, and PO1 ES00454. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $7'0 whom all correspondence should be addressed. multiple induction. For example, extrapolating the above observation made with the single dose injection, 24-h interval, multiple induction should not be expected to result in a high level of mRNA because of the short lived nature of thionein mRNA despite more accumulation of thionein.
In this study, we have given single and repeated injections of Cd to rats and compared the amount of thionein mRNA in the livers. Our results show, contrary to the expectation mentioned above, that the thionein mRNA is amplified with the repeated induction regimen.

EXPERIMENTAL PROCEDURES'
with as libitum feeding andiwatering. For this study rats seven to eight Laboratory animals. Mccqllvm rats were nalntalned as a pure colony months 0 -used. Males welghed 4 5 0 to 500 g and fenale8 weighed 250 to 300 9. Spraque-Daiwley rats were also Used and maintained similarly.
HT-protein and mRNA rats were injecte su utaneously with a CdCl Solurlon Induction of metallothloneln IMTI and MT-mRNA. For the inductlo" of The rnrtial dose of CdCl was 0.25 mg per kg of body weight. Subsequent four times at 2 4 hour Intervals, rest2d f$ 4 8 hours and then Sac?lfIced.
wlth ether prior to inlectian. Control rata Inon-Induced1 were not handled xn~ectrons were 1.0 mg pzr kg of body weight. The rats were anaesthetized until the day Of the experiment. MT and mRNA were a180 prepared after four hours following a slngle subcutaneous Lnjectmn Of cdc12 at a dose of 1.8 mg per kg of body weight. of Sephadex G-75 equilibrated wlth 50 mM TrrS-HC1. pH 7.6/5 mM 2-mercaptoethanol/O.02~ NaN and eluted wrth the Same buffer. The MT peak, detected by the blndlng Of 'b9Cd and A254 (Ve/Va = 1.8-2.01, was pmled, dialyzed agalnst 1 mM ammonlum formate buffer. pH 6.8, and lyophillzed. The lyophilized material was dissolved ~n 10 M Tris-HC1, pH 8. 6 and cleared by centrifugation. The supernatant was applled to a c o l m Of DEm-SephadeX A-50 which was equllibraeed with the Same buffer. Two 1~0bpecles Of metallothlonein. MT-A and MT-E, Were separated by elution vlth a llnear gradient of 5 mM to 400 mM TriS-HC1, pH 8.6 (300 rnl eachl. pooled correspondingly and dlalyred agalnst 50 mM Tris-HC1, pH 8 . 6 . CdCl2 ' Portions of this paper (including "Experimental Procedures,"     Fig.  1 shows that ""Cd binds specifically to protein fractions of M, -10, OOO and the peak of these fractions runs as a major band on SDS-gel electrophoresis (Fig. lB, lane 4) with a mobility also corresponding to this molecular weight. These peak fractions were further separated into two isospecies of metallothionein, MT-A and MT-B by DEAE-Sephadex A-50 column chromatography (Fig. 2). Upon electrophoresis on non-SDS gels, MT-A and MT-B show different mobilities (Fig. 2B).

study, tissue homogenates were centrifuged and the heatstable MT was isolated from the supernatants and fractionated on a Sephadex G-75 column. The profile depicted in
Thus, our preparations of MT were identified and quantitated on the basis of absorbance (e?:* = 1.78 X lo4 liter mol" cm"), ability to bind ""Cd, mobility on molecular sieve chromatography, as well as by SDS and non-SDS-polyacrylamide gel electrophoreses.

Accumulation of Metallothionein in Rat Liver in Response to Dose and Frequency of Induction
The ability of the liver to accumulate the Cd-binding protein, metallothionein, in response to the exposure to a nontoxic level of cadmium has been well recognized (2). Our results, included in Table 1  native MT-A and MT-B. As a control it is shown that rat serum albumin did not react with the antiserum (Fig. 3).

P i g . 28 is the electrophoretogram of UT-A and UT-8 in a non-SDS
("""-). Table 1 kcunulation Of atallothionein in rat liver as

Quantitation of MT-mRNA Induced by Single and Multiple Induction Regimens by Cell-free Translation Assays Preparation of Anti-MT Antibody-
In order to characterize and quantitate metallothionein synthesis in cell-free translation systems, anti-MT antibody against an equimolar mixture of purified rat liver MT-A and MT-B (see Fig. 2) was elicited in rabbits. From the Ouchterlony assay shown in Fig.  3, it is apparent that the antisera reacted specifically not only

Cell-free Synthesis of MT and Quantitation of MT-mRNA
Actiuity-The anti-MT antisera were used to react with the in vitro translation products in order to identify MT synthesis.
With this assay method, MT-mRNA activities of the rat livers induced by two different induction regimens were examined and compared. Using the same amount of template (50 pg/ ml) and within their linear response range, it can be seen in Fig. 4 that the mRNA from rat liver after 5 days of multiple Cd-induction directed the synthesis of a metallothionein much more strongly than did the mRNA from 4-h Cd-induced liver. By densitometric tracing, we estimate that the MT-mRNA activity of multiply Cd-induced liver preparation is 6.6-and 10.7-fold higher in wheat germ extract and reticulocyte lysate systems, respectively, than that of 4-h Cd-induction. The products banded at the same position as the authentic MT, and the banding patterns were essentially the same between the wheat germ extract and reticulocyte lysate systems. Purification of MT-mRNA Total cytoplasmic RNA of Cd-induced rat livers was separated into poly(A)-containing and -lacking (deficient) RNA species using oligo(dT)-cellulose affinity chromatography. The [A'ImRNA was then fractionated by size in a sucrose gradient and the MT-mRNA-enriched fractions were localized by the cell-free translation assay coupled with immunoprecipitation (Fig. 5, A and B ) . Applying the relationship derived by Burgi and Hershey (22), the relative sedimentation coefficient of metallothionein mRNA is calculated to be 9.4 S, equivalent to an average molecular weight of 139,000 or about 400 nucleotides.
Methylmercuric hydroxide-agarose gel electrophoresis further enriched the MT-mRNA (Fig. 6). The enriched fractions I-ml refers to the translation products with the initial lA+lmF34A but also measured about 9 S (Fig. 6B) and were active in directing in titre synthesis of metallothionein. Shown also in Fig. 6 is the degree of purification of MT-mRNA in each purification step. Comparing lanes 2 through 5 in Fig. 6, it is apparent that MT-mRNA was progressively enriched among the 9 S mRNA species.

Comparative Quantitation of MT-mRNA with cDNA Probe
MT-mRNA enriched by sucrose gradient fractionation was used as a template for complementary DNA (cDNA) synthesis in the presence of [a-:"P]dATP. This radiolabeled DNA was used as a probe to titrate the MT-mRNA content of RNA preparations from samples taken after 4-h single dose induction and after 5-day, multiple dose induction by Cd. The results are shown in Fig. 7 and Table 11. It can be seen, first, that fractionation of total poly(A)' RNA by sucrose density gradient yielded 9 S RNA which hybridized with the cDNA probe 10-fold more than the unfractionated poly(A)' RNA and approximately 100-fold more than the total RNA (Table  11). RNA without purification or poly(A)-RNA alone contained insufficient quantities of MT-mRNA sequences to show hybrid reactivity as detected in this autoradiograph (Fig. 7). Secondly, since the titration was made in the linear range of the hybridization reaction (compare ZX to 2X), the results also show that in RNA prepared from 5-day multiple induction there is an approximately 4-fold higher level of MT-mRNA compared with the RNA from 4-h-induced rat liver.

DISCUSSION
In this study we have shown that poly(A)-containing 9.4 S RNA from rat liver with an apparent molecular weight of 1.39 X 10" directed the in vitro synthesis of metallothionein. The product was identified by specific immunoprecipitation. While other studies also reported a similar observation (14), this is the first to provide a more positive identification of the translation product. We have produced antibody to the metallothionein and used it to specifically precipitate the in vitro translation products. Metallothionein, a small protein of only 61 residues, is a poor antigen unless polymerized to elicit antibody formation (11). The immunoassay used here, having not been previously available, is of special value to this study.
We have also taken advantage of the more recently developed Northern hybridization technology (21) to immobilize the RNA in question on diazotized papers for solid state DNA-RNA hybridization, thus allowing both visual and quantitative examination of the various stages of RNA preparation for their enrichment in MT-mRNA. This procedure enabled us to estimate that the specific activity of MT messenger is at least 4-fold higher in the liver RNA preparations after multiple induction with Cd over a 5-day period than that in RNA obtained after 4 h with a single induction shot, which corroborates results obtained by the cell-free translation assays (Fig.  4). It is significant to note this difference since previous studies have shown that MT-mRNA is short lived, having a half-life of only a few hours (3).
The high level of MT-mRNA in the multiply induced liver may be explained in several ways: Greater numbers of the liver cells may be recruited to respond to the multiple induction. Preliminary studies with indirect immunoassay of liver sections show, however, the increase in reactivity with MTspecific antibody is uniformly distributed among most, if not all, cells. On the other hand, repetitive induction by Cd may, 1) stabilize this mRNA; 2) enhance the rate of specific MT