Expression of the constitutive and inducible forms of heat shock protein 70 in human proximal tubule cells exposed to heat, sodium arsenite, and CdCl(2).

We determined the expression of the constitutive (hsc 70) and inducible (hsp 70) forms of heat shock protein 70 mRNA and protein in human proximal tubule (HPT) cells exposed to lethal and sublethal concentrations of Cd(+2) under both acute and extended conditions of exposure. The HPT cells exhibited the classic heat shock response when subjected to a physical (heat) or chemical stress (sodium arsenite); hsc 70 mRNA and protein levels were constant or slightly increased, whereas hsp 70 mRNA and protein were greatly elevated. Acute exposure to 53.4 microM CdCl(2) for 4 hr failed to increase either hsc 70 mRNA or protein, a finding similar to that observed under classic conditions of stress. However, under identical conditions of acute exposure to Cd(2+), the expected increase in hsp 70 protein level was suppressed as compared to that found under classic conditions of physical or chemical stress. The decrease in hsp 70 protein level correlated to the reduced expression of mRNA from the hsp 70B gene. The expression of mRNA from the hsp 70A and hsp 70C genes was similar to that found when the cells were treated with heat shock or sodium arsenite. We modeled an extended exposure to Cd(2+) by treating the cells continuously with Cd(2+) at both lethal and sublethal levels over a 16-day time course. Chronic exposure to Cd(2+) failed to increase either hsc 70 mRNA or protein levels in the HPT cells at a nonlethal dosage level and decreased hsc 70 mRNA and protein levels late in the time course of lethal exposure. Under identical conditions, the expression of hsp 70 protein remained at basal levels that were only marginally detectable throughout the time course. Hsp 70A and hsp 70C mRNA levels were unaltered by extended exposure to Cd(2+), and hsp 70B mRNA was not detected during the 16-day time course. Cd(2+) is a poor inducer of hsc 70 and hsp 70 in the proximal tubule under both acute and long-term exposure. These results reinforce the fact that the expression of hsp 70 protein does not result from the transcription of a single gene, but is derived from what may be a complex interplay of several underlying genes. ImagesFigure 1Figure 2Figure 3Figure 4Figure 5Figure 6

The heat shock response (stress response) is widely recognized and accepted as a major weapon in the cell's armamentarium for protection against and recovery from environmental insult, both physical and chemical [for review (1)(2)(3)(4)]. The most obvious feature of the heat shock response consists of alterations in gene expression leading to increased synthesis of heat shock proteins (hsps) and a cessation of other cellular protein synthesis when cells are exposed to stress. The hsps are a large superfamily of proteins with molecular weights ranging from 8 to 170 kD; members are referred to as hsp 27, hsp 60, hsc 70 (constitutive form), hsp 70 (inducible form), and hsp 90, the proteins classically identified to be induced as a result of heat treatment of mammalian cells.
The current study was undertaken for two reasons. First, a goal of this laboratory is to define the role of the stress response in mediating toxicity that the heavy metal pollutant cadmium elicits on the human proximal tubule (HPT) cell. The working hypothesis is that identification of members of the stress response superfamily that do not respond to Cd2+ exposure by induction of the corresponding protein might identify cellular responsibilities particularly susceptible to Cd2+-induced damage. The first test of this hypothesis involved determining the expression of metallothionein (MT), a protein thought to mediate heavy metal toxicity by metal sequestration in cultures of HPT cells exposed to both lethal and sublethal concentrations of Cd2+ (5)(6)(7). Garrett et al. (5,6) and Hoey et al. (7) demonstrated that the HPT cells responded to both lethal and sublethal exposures to Cd2+ with a large, rapid, and sustained increase in the level of MT protein, providing strong evidence that the MT stress response was active and not compromised in these cells.
A second test of this hypothesis involved the determination of hsp 27 in HPT cells identically exposed to Cd2+ (8). In this instance, the results were equivocal because hsp 27 was increased early in the time course of Cd2+ exposure, but returned rapidly to values below control despite continued presence of the metal. The finding that hsp 27 expression was reduced in HPT cells chronically exposed to Cd2+ suggests that a consequence of chronic Cd2+exposure might involve a loss in the maintenance of microfilament dynamics. This is based on many studies which show that hsp 27 exerts its protective effects on the cell, at least in part, through a chaperone action that stabilizes microfilament dynamics (9)(10)(11). Studies in the rodent model have shown that enhanced expression of hsp 27 correlates to the protection of the proximal tubule from brief periods of ischemia and the corresponding disruption of the actin filament network (12)(13)(14).
The present study examines the expression of the hsp 70 family in HPT cells exposed to Cd2+ in a manner identical to that described above for MT and hsp 27 (5,6,8). The goals of this study are to determine if the dassic induction of hsp 70 that occurs ubiquitously for mammalian cells exposed to heat shock also occurs when HPT cells are exposed to sodium arsenite and Cd2+, and if the hsp 70 isoform genes have distinct patterns of expression when exposed to these agents. The hsp 70 chaperones are the most studied heat shock proteins and, with their co-chaperones, comprise a set of abundant cellular machines that assist a large variety of protein-folding processes in almost all cellular compartments (15). Under conditions of stress, they have been shown to prevent aggregation of denatured proteins and assist in refolding of misfolded proteins. Under normal conditions, heat shock proteins serve a number of essential roles: they assist in folding of selected newly translated proteins; they assist in the translocation of proteins across organellar membranes; they disassemble oligomeric protein structures; they facilitate proteolytic degradation of unstable proteins; and they control the biologic activity of folded regulatory proteins (15). Materials and Methods Cell culture. Stock cultures of HPT cells used in experimental protocols were grown in 75-cm2 T-flasks (Corning, Corning, NY) using procedures previously described by this laboratory (16,17). The growth medium was a serum-free formulation consisting of a 1:1 mixture of Dulbecco's modified Eagles' medium (DME) and Ham's F-12 growth medium supplemented with selenium (5 ng/mL), insulin (5 pglmL), transferrin (5 pg/mL), hydrocortisone (36 ng/mL), triiodothyronine (4 pg/mL), and epidermal growth factor (10 ng/mL). The growth surface was treated with a collagen matrix to promote cell attachment and subculture. The cells were fed fresh growth medium every 3 days and, at confluence (normally 3-6 days after subculture), were subcultured using trypsin-EDTA (0.05% trypsin, 0.02% EDTA). For use in experimental protocols, the cells were subcultured at a 1:2 ratio and allowed to reach confluence (6 days after subculture) before initiation of experimental protocols. The cells were fed every 3 days. Three isolates of HPT cells were used; these isolates were derived from normal cortical tissue obtained from kidneys removed for renal cell carcinoma. The kidneys were from a 72-year-old female, a 63year-old male, and a 58-year-old female. We used HPT cells between passages 5 and 7 in the present study. The total protein and RNA samples used in the present analysis of hsc 70 and hsp 70 expression were obtained from a previous study that examined hsp 27 expression in HPT cells exposed to heat shock, sodium arsenite, and lethal and sublethal concentrations of CdCI2 (8).
The effect of chemical stress on the expression of hsc 70 and hsp 70 was determined by exposing confluent HPT cells from three independent cell isolates to 100 ,uM sodium arsenite for 4 hr, followed by a recovery period of 48 hr.
We determined the effect of heat shock on HPT cells by exposing the cells to heat shock (42.50C) for 1 hr. The cells were then returned to 37°C for a 48 hr recovery period.
The effect of acute Cd2+ exposure on the expression of hsc 70 and hsp 70 in HPT cells was determined by exposing confluent cell monolayers to 53.4 pM Cd 2+ for 4 hr; this was followed by a recovery period of 48 hr in Cd2+-free growth media.
We determined the expression of hsc 70 and hsp 70 mRNA and protein after exposing the HPT cells to Cd2+ for an extended time. We used three concentrations of Cd2+: 9.0 pM, which produces no cell death over the 16-day time course; 27.0 pM, which produces cell death late in the 16-day time course; and 45.0 pM, which produces cell death early in the 16-day time course (5).
Cell viability, isolation of total RNA, and RT-PCR. The effects of various treatments on the viability of confluent cell monolayers were determined by the automated counting of 4',6-diamidino-2-phenylindole (DAPI)-stained nuclei of cells (8). Total RNA was isolated according to the protocol supplied with TRI REAGENT (Molecular Research Center Inc., Cincinnati, OH) as described previously (5). We determined the concentration and purity of samples using spectrophotometer scan in the ultra violet (UV) region and ethidium bromide visualization of intact 18S and 28S RNA bands following agarose gel electrophoresis.
For reverse transcriptase-polymerase chain reaction (RT-PCR), 500 ng total RNA from cultured HPT cells was reverse transcribed in a 20-jiL reaction mixture using 50 units Murine Leukemia Virus Reverse Transcriptase (N808-0143; Perkin Elmer, Foster City, CA) in IX PCR buffer (50 mM KCI, 10 mM Tris-HCl, pH 8.3), 5 mM MgCl2, 20 units RNase inhibitor, 1 mM each of the dNTPs, and 2.5 pM random hexanucleotide primers. The samples were reverse transcribed for 20 min at 42°C, followed by a 5 min denaturation step at 99°C using a DNA thermocycler (Perkin Elmer 9600). The resulting cDNA was amplified in 100 1L reaction mixture containing 2 mM MgCI2, IX PCR buffer, 2.5 units of AmpliTaq DNA polymerase (Perkin Elmer), and the specific upstream and downstream primers at a concentration of 0.15 pM each. For the amplification of hsc 70, primers were (upper and lower, respectively) 5'TGTGGCTTC-CTTCGTTATTGG3' and 5'GCCAG-CATCATTCACCACCAT3' (StressGen, STM-505, Victoria, British Columbia, Canada). For the amplification of hsp 70A, B, and C, primers were (upper and lower, respectively) 5'TGTTCCGTTTCCAGCC-CCCAA3' and 5'GGGCTTGTCTCC-GTCGTTGAT3' (STM-506) for 70A; 5 'CTCCAGCATCCGACAAGAAGC3' and 5'ACGGTGTTGTGGGGGTTC-AGG3' (STM-507) for 70B; and 5'TTG-AGGAGGTGGATTAGGGGC3' and 5 'AGCCTTTGTAGTGTTTTCGCC3' (STM-508) for 70C. Primers for the determination of glyceraldehyde 3-phosphate dehydrogenase (g3pdh) gene were obtained commercially (Clontech, Palo Alto, CA). The thermocycler was programmed to cycle at 95°C for a 2-min initial step, at 95°C for 30 sec, and at 50°C for 30 sec, with a final elongation step at 50°C for 7 min. Controls for each PCR induded a no-template control in which water was added instead of RNA and a no-reverse transcriptase control in which water was added instead of the enzyme. Samples were removed at 30, 35, and 40 PCR cycles to ensure that the reaction remained in the linear region. The final PCR products were electrophoresed on 2% agarose gels containing ethidium bromide along with DNA markers.
For reactions in the linear region of the cycle, integrated optical densities (IOD) of the samples were obtained by input of the ethidium bromide fluorescent image into a Roche Pathology Workstation (AutoCyte, Burlington, NC) configured with KS400 software (Zeiss, Thornwood, NY) using a Kodak DCS 420 CCD camera (Kodak, Rochester, NY). The resulting IOD values were used to generate a relative IOD that is the ratio of the hsc 70 and hsp 70 reaction products to products for the housekeeping gene glyceraldehyde 3-phosphate dehydrogenase.
Western analysis. Cells were washed two times with phosphate-buffered saline (PBS) and lysed directly in the flask by addition of 400 pL hot (85°C) IX SDS buffer [2% sodium dodecyl sulfate (SDS) and 50mM Tris-HCI, pH 6.8]. The cell lysate was heated in a boiling water bath for 10 min. DNA was sheared by repeated passage through a 23gauge needle. The samples were centrifuged at 10,000g for 10 min at room temperature, and the supernatant was transferred to a new tube. The concentration of protein in the samples was determined by the BCA protein assay (Pierce Chemical Co., Rockford, IL).
Total cellular protein (10 pg) was separated on 12% SDS-containing polyacrylamide gels and electrophoretically transferred to polyvinylidine difluoride membrane (Bio-Rad, Hercules, CA). Hsc 70 was visualized using primary rat monoclonal antibody (SPA-815; StressGen), whereas hsp 70 was visualized using primary mouse monoclonal antibody (SPA-8 10; StressGen). Membranes were blocked with 10% (w/v) nonfat dry milk in PBS and incubated with rat monoclonal antibody specific for hsc 70 diluted 1:2000 in PBS containing 0.1% (w/v) bovine serum albumin as carrier or mouse monoclonal antibody specific for hsp 70 diluted 1:2000 in PBS containing 5% (w/v) nonfat dry milk as carrier and 0.05% Tween-20. This was followed by incubation with a goat anti-rat or goat anti-mouse alkaline phosphatase conjugated secondary antibody (Promega, Madison, WI). Colorimetric detection employed an alkaline phosphatase Vectastain ABC-AP kit (Vector, Burlingame, CA). For color reactions in the linear range of analysis, IOD values of the samples were obtained by input of the image into a Roche Pathology Workstation configured with KS400 software using a Kodak DCS 420 CCD camera.
The IOD for an individual band is the sum of OD values x pixels2. The relative IOD was obtained by dividing the IOD for each hsp 70 and hsc 70 band by the IOD for g3pdh at the appropriate time point.
Nomenclature. In this study, we define the term hsc 70 as the constitutive member(s) of the hsp 70 gene family (18). The exact number of hsc 70 genes present in the genome is not known because many processed hsc 70 pseudogenes are present, which complicates analysis. A genomic structure for one active hsc 70 gene has been determined; this gene is characterized by the presence of eight introns (18). The hsc 70 protein has been purified and an antibody has been generated that does not cross-react Volume 107, Number 1 1, November 1999 * Environmental Health Perspectives Articles * Hsc and hsp 70 expression and CdC 12 toxicity with the inducible forms of hsp 70. We used this commercially available antibody and primers for RT-PCR that span an intron/exon junction of hsc 70 in the present study to define the expression of hsc 70 mRNA and protein. In the present study we use the term hsp 70 to indicate the inducible members of the hsp 70 gene family (18). The exact number of hsp 70 genes present in the genome is also unknown because there are numerous active and pseudogene sequences, but commercially available RT-PCR primers can specifically determine the mRNA originating from three distinct hsp 70 genes (19). The complete genomic structures of two of these genes have been determined; they are intron-less genes that encode an identical protein product of 641 amino acids and map in the major histocompatability complex class III region (20). Originally referred to as HSP70-1 and HSP70-2, in this study we refer to them as hsp 70A and hsp 70C, respectively, according to commercial nomenclature. A third hsp 70 gene has been identified based on a partial sequence that maps to chromosome 1 whose mRNA can be determined by RT-PCR (18). We refer to this gene, originally designated HSP70B, as hsp 70B. While the primers developed for RT-PCR may also monitor the mRNA from other unidentified hsp 70 genes, they are specific in that they individually distinguish the hsp 70A, hsp 70B, and hsp 70C isoforms from one another. The hsp 70 protein has been purified, and an antibody that does not cross-react with the constitutive hsc 70 has been generated and is commercially available.

Results
Hsc 70 and Hsp 70 expression in HPT cells exposed to sodium arsenite and heat shock.
We anaylzed the effect of chemical stress on the expression of hsc 70 and hsp 70 in confluent HPT cells from three independent cell isolates exposed to 100 pM sodium arsenite for 4 hr, followed by a recovery period of 48 hr. The results for the expression of mRNA originating from the hsc 70 and hsp 70 A, B, and C genes are presented as native gel data for one isolate (Figure IA) and graphically as the mean relative IOD values for all three isolates when expression is compared to mRNA from the g3pdh housekeeping gene ( Figure 1B). This analysis demonstrated that each mRNA species had a distinct pattern of expression under basal conditions and when the HPT cells were exposed to sodium arsenite. For hsc 70, a high level of basal mRNA was demonstrated that increased approximately 2-fold as a result of exposure to sodium arsenite. This elevation in hsc 70 mRNA occurred between 2 and 4 hr of treatment and returned to control values within 16 hr of recovery. The hsp 70A isoform was the only inducible isoform that had a detectable level of basal mRNA that was consistently present in control cells, and this was approximately 4-fold less than hsc 70 expression.   Assuming equal RT-PCR efficiencies and mRNA degradation rates, the following induction pattern can be proposed based on both absolute peak values and when a given nonbasally expressed mRNA species first appears following sodium arsenite exposure: hsc 70 < hsp 70A < hsp 70B < hsp 70C. The levels of hsc 70 and hsp 70 protein were determined by Western analysis of protein extracts prepared simultaneously with the total RNA samples analyzed above (Figure 2A, B). Western analysis disclosed an easily determined basal level of hsc 70 protein expression; however, this level did not increase during or after exposure of the HPT cells to sodium arsenite. This indicates that the 2-fold elevated levels of hsc 70 mRNA demonstrated above do not translate into increased levels of hsc 70 protein. Using Western analysis, we demonstrated that the basal level of hsp 70 protein was at the limit of detection, with convincing bands representing hsp 70 protein being visualized in approximately half the basal samples analyzed throughout the present study. During the 4 hr that the HPT cells were exposed to 100 pM sodium arsenite, hsp 70 protein remained at the limit of detection. However, within 4 hr of sodium arsenite removal, hsp 70 protein was easily detectable, and levels approximately equaled levels for hsc 70. The level of hsp 70 remained elevated for at least 48 hr after removal of sodium arsenite. The antibody used for the Western analysis does not distinguish between the hsp 70 isoforms.
For heat-shocked HPT cells, the expression patterns of hsc 70 and hsp 70 mRNA and protein were very similar to those obtained for the sodium arsenite-treated cells ( Figure 3A, B). Hsc 70 mRNA was increased only moderately by heat shock, with a return to control values after removal of heat stress ( Figure 3A). Similar to sodium arsenite-treated cells, the increase in hsc 70 mRNA did not translate into an increase in hsc 70 protein ( Figure 3B). The mRNA for the hsp 70A, hsp 70B, and hsp 70C isoforms increased rapidly after heat shock, peaked at 2-4 hr into the recovery period, and then returned to control values ( Figure  3A). Identical to that for arsenite-exposed cells, the pattern of induction of mRNA was hsc 70 < hsp 70A < hsp 70B < hsp 70C. A low basal expression of hsp 70 protein was demonstrated in control HPT cells. This basal level increased 1-2 hr after heat shock and reached a peak 12-16 hr into the recovery period, with peak values being approximately twice those attained by hsc 70 ( Figure 3B). Thereafter, the expression of hsp 70 decreased but remained elevated over control at the end of the 48-hr recovery period. in Cd2P-free growth media. This represents a level of Cd2+ exposure previously shown to be lethal to 15-25% of the cells by the end of the recovery period (8). In contrast to the results obtained previously for HPT cells exposed to both heat and sodium arsenite, hsc 70 mRNA did not increase as a consequence of acute exposure to Cd2+ ( Figure 4A). However, identical to that described previously for heat shock and sodium arsenite, hsc 70 protein levels were not altered as a consequence of acute Cd2P exposure ( Figure 4B). The expression patterns and levels of mRNA for the hsp 70A and hsp 70C isoforms were also very similar to those obtained when HPT cells were exposed to heat and chemical stress ( Figure  4A). Hsp 70A and hsp 70C mRNA increased (3-  the expression of hsp 70B mRNA was very different from that obtained when HPT cells were exposed to heat or chemical stress ( Figure 4A). Whereas hsp 70B mRNA levels increased to IOD values of 7-9 for heatand arsenite-exposed HPT cells, they peaked at < 4 relative IOD units for the Cd2+-exposed cells (compare Figure iB, Figure 3A, and Figure 4A). The reduction in hsp 70B mRNA noted for Cd2+-exposed cells is given increased validity by the fact that identical total RNA samples were used for the determination of the three respective hsp 70 isoforms. Even more striking than the reduced induction of hsp 70B mRNA accumulation in the Cd2+-exposed cells was the decreased accumulation of hsp 70 protein in the HPT cells acutely exposed to Cd2+ ( Figure 4B). In contrast to heat-shocked or sodium arsenite-treated cells, hsp 70 protein was only marginally increased in the HPT cells by acute Cd2+ exposure (compare Figure 2A, Figure 3B, and Figure 4B). In cells acutely exposed to Cd2+, hsp 70 protein reached a maximum of 1 The expression of hsc 70 and hsp 70 mRNA and protein was also determined when the HPT cells were exposed to Cd2+ for an extended time, rather than the acute exposure classically used to define the stress  of Cd2+ had no effect on the expression of either hsc 70 mRNA or protein ( Figure 5B, D). This was most apparent for HPT cells exposed to 9.0 iIM Cd +, a level of exposure that elicited no cell death over the entire 16day time course. For cells exposed to 27.0 PM Cd , there was no change in the expression of hsc 70 mRNA and protein in the early days of the time course. However, once cell death was appreciable (days 13 and 16) there was a marked reduction in the levels of both hsc 70 mRNA and protein.
For the inducible isoform, the expression of hsp 70A and hsp 70C mRNA was constant | a hsp70mNA| *A.  bands normalized to control in HPT cells exposed to CdCI2. (C) Mean (± SE) relative IOD of bands representing hsp 70C mRNA in control cells. (D) IOD of hsp 70C bands normalized to control in HPT cells exposed to CdCI2 *There was only one viable sample. **There were only two viable samples. over the 16-day time course for HPT cells unexposed to Cd2+ ( Figure 6A, C). Likewise, exposure to Cd2+ had no effect on the level of hsp 70A or hsp 70C mRNA over the 16-day time course (Figure 6B, D). No expression of mRNA for the hsp 70B isoform was demonstrated in either control or Cd2+-exposed cells at any time during the time course. As demonstrated by Western analysis, there was no detectable (or only marginally detectable) hsp 70 protein in the control or Cd2+-treated cells at any point in the time course (data not shown).

Discussion
The first goal of the present study was to define the response of hsc 70 and hsp 70 when HPT cells were acutely exposed to Cd2+. The response expected from such an exposure was defined by both comparison to that reported in the literature and to the response demonstrated when HPT cells were exposed to the classic stimuli of heat shock and sodium arsenite. For hsc 70, acute exposure of HPT cells to Cd> had no effect on the level of hsc 70 mRNA or protein, either during exposure to Cd> or during the subsequent postexposure recovery period. That hsc 70 was not induced in HPT cells by Cd> exposure was not unexpected, being in agreement with the finding that hsc 70 was also not induced in the HPT cells by either heat shock or sodium arsenite treatment. Furthermore, these findings are in agreement with the initial study that described hsc 70 as a constitutive protein with a high level of expression, attaining a level that can comprise 1% of total cell protein under nonstressed conditions (21). While the majority of studies on the heat shock response are cell culture based, one study has examined hsc 70 expression in the rat kidney after administration of Cd> (22). In agreement with the present findings, this study demonstrated that Cd2+, when administered intravenously (iv) as CdCl2 (2 mg/kg), had no effect on the level of expression of hsc 70 protein in the kidney 4 hr after exposure. However, this study also modeled chronic exposure to Cd2+ by exposing the rats to an equivalent dose of Cd2+ for 4 (24,25), 9L rat brain tumor cells (26), Reuber H35 hepatoma cells (27,28), CEM-C12 human T cells (29), HepG2 hepatoblastoma cells (30), and HeLa cells (31). The only in vivo data in the kidney detailing hsp 70 expression was the previously detailed study in the rat kidney that examined hsc 70 expression after administration of Cd2+ as both the CdCl2 salt and as the Cd-Cys complex (22). In this study it was also demonstrated that acute exposure of rats to iv-administered CdCl2 had no effect on the expression of renal hsp 70, whereas identical exposure to the Cd-Cys complex resulted in a modest (2-fold) elevation of hsp 70. This finding suggested that a chronic exposure to Cd may induce a modest increase in the accumulation of hsp 70. This possibility was examined in HPT cells using a 16-day exposure to both lethal and sublethal levels of Cd2+; however, there was no alteration in the level of hsp 70. The difference between the two observations may simply reflect the fact that the in vivo analysis measured new synthesis of hsp 70 and the HPT cell culture analysis measured total accumulation. Alternatively, the increased hsp 70 noted in the in vivo studies could be derived from a cell type other than the proximal tubule. However, even considering the small increase in hsp 70 found Volume 107, Number 1 1, November 1999 * Environmental Health Perspectives 892 Articles * Hsc and hsp 70 expression and CdC 1 2 toxic in vivo, overall, the findings support the concept that Cd2+ is a poor inducer of hsp 70 accumulation in the proximal tubule under both acute and long-term conditions of exposure.
An additional goal of the present study was to determine if the individual genes responsible for the expression of hsp 70 have distinct patterns of regulation when HPT cells were acutely exposed to heat shock, sodium, or Cd2+. At present, three genes (hsp 70A, hsp 70B, and hsp 70C) underlying the expression of hsp 70 can be specifically identified using RT-PCR technology. The pattern of hsp 70 isoformspecific mRNA expression was shown to be very similar when the HPT cells were exposed to either heat shock or sodium arsenite. This similarity included the important finding that mRNA for all three hsp 70 isoforms was not only increased, but increased to a similar maximum value of peak expression and a period of sustained expression. For HPT cells exposed acutely to Cd2 , the patterns of expression for the hsp 70A and hsp 70C isoforms were also indistinguishable from the heat-shocked and sodium arsenite-treated cells. In contrast, the increase in the expression of mRNA representing the hsp 70B isoform was blunted for Cd2+-treated cells as compared to that shown for the HPT cells exposed to heat shock or sodium arsenite. Furthermore, no hsp 70B mRNA was expressed when HPT cells were exposed to either lethal or sublethal levels of Cd2P for 16 days. The potential significance of the decreased expression of hsp 70B mRNA is heightened by the fact that it correlated with the marked reduction in the expression of hsp 70 protein noted in Cd2+-exposed cells as compared to those treated with heat or sodium arsenite. This finding suggests that Cd2+ exposure can inhibit the induction of hsp 70 protein by affecting the expression of only one of the hsp 70 isoform genes. This observation also provides initial evidence that hsp 70 protein accumulation in stressed cells is derived preferentially from the hsp 70B isoformspecific mRNA as compared to those transcripts originating from the hsp 70A and hsp 70C genes. Overall, these observations highlight the fact that the expression of the hsp 70 protein does not result from the transcription of a single gene, but is derived from what may be a complex interplay of several underlying genes.