PKCϵ Is a Unique Regulator for hsp90β Gene in Heat Shock Response*

An early event in cellular heat shock response is the transmittance of stress signals from the cell surface into the nuclei, resulting in the induction of heat shock proteins (Hsps). Protein kinase C (PKC) has been implicated as a key player in transducing stress signals. However, mechanism(s) by which PKC regulates heat shock-induced events remains largely unknown. Here we present data that pan-PKC inhibitor GF109203X, but not classic PKC inhibitor Gö6976, specifically repressed heat shock-induced accumulation of mRNA as well as promoter activity of hsp90β, but not hsp90α, in Jurkat cells. Subcellular fractionation studies revealed that heat shock exclusively induced PKC-ϵ membrane translocation. Consistently, expression of a constitutively active PKC-ϵ(A159E) resulted in an enhanced promoter activity of hsp90β upon heat shock, whereas a dominant-negative PKC-ϵ(K437R) abolished this effect. In contrast, constitutively active-PKC-α or dominant-negative-PKC-α had no effects on heat shock induction of the gene. The effect of PKC-ϵ on hsp90β expression seems to be stimuli-specific, as phorbol myristate acetate-mediated hsp90β expression was PKC-ϵ-independent. We conclude that PKC-ϵ is specifically required in the signaling pathway leading to the induction of hsp90β gene in response to heat shock.

In a mammalian cell stress response, the most remarkable early events are the transmittance of stress signals from cell surface into the nucleus, resulting in the induction of a set of proteins, dominantly the heat shock protein (HSP) 1 family. Hsps, with or without the partnerships of co-chaperones, may interact with various signaling molecules to form active components in signaling pathways that regulate cell growth, development, and protect the cells from extreme stress through the induction of stress-tolerance (1)(2)(3). Over-or underexpression of the molecular chaperone Hsps may result in aberrant growth control, developmental malformations, diseases, or even cell death (4 -6).
As one of the most abundant cytoplasmic heat shock proteins and a member of the molecular chaperone family, Hsp90 is highly conserved during evolution and is ubiquitously expressed for cell survival in animals and plants (1). Hsp90 is capable of interacting with a wide variety of proteins both native and denatured to affect their turnover, cellular localization, and activity (7,8). The specificity of Hsp90 may reside in its cellular "client" selection for players in transcription regulation and signal transduction pathways, such as the steroid hormone receptors, basic helix-loop-helix transcription factors, oncogenic tyrosine kinase, or cellular serine/threonine kinases, and has been defined as a specific molecular chaperone (9,10). There are two cytoplasmic versions of the Hsp90 subfamily in mammalian cells encoded by two distinct genes, the hsp90␣ and hsp90␤ (11,12). Although Hsp90␣ and Hsp90␤ proteins share very high homology, we have demonstrated that the regulation mechanisms of the two genes are quite different (13,14): hsp90␣ is more sensitive to heat shock induction, whereas the hsp90␤ gene is the major cellular counterpart responding to mitogenic stimulation (15). We reported elsewhere (16) that Hsp90 is not only induced in stress response but also uniquely expressed in mitogen-activated T lymphoid cells, indicating its potential importance in cell growth. Additionally, recent studies demonstrated that in heat-shocked or other stressed cells, Hsp90 takes part in protecting cell survival and counteracts apoptosis (17)(18)(19).
PKC is a family of serine/threonine kinases that participates as a central component in signal transduction pathways responding to multiple external stimuli (2,20). It plays a pivotal role in the regulation of many biological functions, including proliferation, differentiation and apoptosis (21). Although PKC isoforms share conserved domains, they exhibit different tissue distribution, subcellular localization, substrate specificity, and cofactor requirements (22,23). The PKC isoforms consist of three subgroups based on their structure and cofactor requirements, the conventional isoforms (cPKC-␣, -␤ I , -␤ II , and -␥) which require both Ca 2ϩ and diacylglycerol/phorbol ester for activation; the novel isoforms (nPKC-␦, -⑀, -, and -) which are Ca 2ϩ -independent but diacylglycerol/phorbol ester-dependent; and the atypical isoforms (aPKC-, -, and -/) which are Ca 2ϩ -and diacylglycerol/phorbol ester-independent (2,24). These differences suggest that individual PKC isoforms may play a distinct role in regulation of cellular functions. In some cases, PKC isoforms may counteract each other in regulating cell growth, transformation, or apoptosis (21,24). For example, PKC-⑀ functions as an oncogene upon overexpression, whereas PKC-␦ usually promotes differentiation or apoptosis (25). * This work was supported by National Natural Sciences Foundation of China Grant 39930050 (to Y.-F. S.) and in part by National Institutes of Health Grant R01-CA88815 (to L. X.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
¶ To whom correspondence should be addressed: Univ. Our early findings showed that PKC was involved in hsp90 gene expression in heat-shocked Jurkat cells (15). However, it is unclear which isoforms of the PKC family are specifically involved in the heat shock-induced responses, such as the induction of hsp90 expression. We demonstrate here that PKC-⑀ plays a critical role in the induction of the hsp90␤ gene but not the hsp90␣ gene in response to heat shock in Jurkat cells. The results suggest a link between the Hsp family and the PKC family in heat shock response. It is possible that cooperation between these two families of proteins may play an important role in cell growth control in mammalian cells.
For PKC inhibitor studies, Jurkat cells were pretreated with PKC inhibitor GF109203X or Gö6976 at the indicated concentrations at 37°C for 2 h, followed by a wash of the cells with phosphate-buffered saline. Cells in each condition were then separated into two groups and incubated at either 42°C or 37°C for 1 h. Reagents-PKC inhibitors (GF109203X and Gö6976) and phorbol 12-myristate 13-acetate (PMA) were purchased from Sigma. Anti-␤actin antibody and rabbit polyclonal antibodies that recognize PKC-␦, -, -, and -␣ were obtained from Santa Cruz Biotechnology. Mouse IgG2a against PKC-⑀ was obtained from BD Transduction Laboratories. Antibody against HSP90␤ and anti-FLAG antibody were obtained from Stressgen and Sigma, respectively.
DNA Constructs-A full-length cDNA clone for human PKC-⑀ isoform was obtained from Dr. A. P. Fields (University of Texas Medical Branch, Galveston, TX). The human PKC-⑀ cDNA was subcloned into pBluescript II-SK(ϩ) (Stratagene, La Jolla, CA), and a FLAG-epitope tag was inserted to the NH 2 terminus of the PKC-⑀ in frame. The FLAG-PKC-⑀ cDNA was then cloned into an expression vector pCI-neo (Promega, Madison, WI). Constructs expressing constitutively active (CA) mutant PKC-⑀ (A159E) and dominant-negative (DN) mutant PKC-⑀ (K437R) were described previously (26). Plasmids expressing For quantifying the promoter activity of human hsp90 genes, the Ϫ1756/ϩ37 fragment of the hsp90␣ gene and the Ϫ1039/ϩ1531 fragment of the hsp90␤ gene were independently fused to the upstream of a CAT reporter gene in pBLCAT3 to form reporter plasmids of p90␣ 1 -CAT and p90␤ 3.1 -CAT, respectively (13,14). A transfection control plasmid was constructed in which the ϩ698/ϩ1003-bp fragment of the CAT gene was deleted to express a mutant CAT and was designated as pM-CAT (28).
DNA Transfection-Transient transfection of Jurkat cells was carried out by electroporation (as described in Ref. 29). For promoter activity assays, cells (1.5 ϫ 10 7 ) were transfected with a mixture of 27 g of reporter plasmid hsp90␣ 1 -CAT or hsp90␤ 3.1 -CAT and 3 g of control plasmid pM-CAT. For assessing PKC isoform-specific effects, constructs expressing wild-type or mutants of PKC-⑀ or PKC-␣ were individually co-transfected with reporter plasmid hsp90␤ 3.1 -CAT along with a control plasmid of pM-CAT into Jurkat cells. At 48-h after transfection, cells were separated into two groups and incubated at either 42°C or 37°C for 1 h. Total cellular RNA was extracted according to the method described by Chomezynski and Sacchi (30) and used for detecting mRNA expression and promoter activity of hsp90 genes.
Detection of mRNA Expression and Promoter Activity of hsp90 Genes-The mRNA levels of hsp90 genes were determined based on a competitive RT-PCR assay as described by Wang et al. (31). Briefly, an internal control RNA (icRNA) was first transcribed in vitro from pHS-YL3 plasmid, which contains the same 5Ј and 3Ј fragments that existed in the cDNAs of the hsp90␣ and hsp90␤ genes. An equal amount of icRNA was then mixed with each aliquot of cellular RNA, reversetranscribed, and amplified in the competitive RT-PCR system. The sizes of amplified fragments for hsp90␣ mRNA and icRNA were 564 and 841 bp, respectively, and the amplified fragments for hsp90␤ mRNA and icRNA were 337 and 625 bp, respectively.
For promoter activity assays, control plasmid pM-CAT was co-transfected into Jurkat cells with either hsp90␣ 1 -CAT or hsp90␤ 3.1 -CAT in an appropriate ratio. Total cellular RNA was then isolated for competitive RT-PCR using paired 5Ј and 3Ј primers corresponding to nucleotides ϩ554/ϩ573 and ϩ1122/ϩ1141 of pBLCAT3, respectively. A 588-bp fragment for wild-type CAT driven by the hsp90 promoter and a smaller fragment of 286 bp generated from control plasmid pM-CAT were coamplified and separated by agarose gel electrophoresis.
Fluorescence intensity of each band stained with ethidium bromide was analyzed with an UltroScan XL laser densitometer (Amersham Biosciences). The ratio of the fluorescence intensity of two bands in each RT-PCR (hsp90 mRNA to icRNA or hsp90-CAT to pM-CAT) was defined as the relative mRNA expression or the relative promoter activity of the hsp90 gene, respectively.
Subcellular Fractionation-All of the experimental procedures in subcellular fractionation were carried out at 4°C with all of the solutions pre-chilled. The fractionation was performed as described by Ding et al. (32) with minor modifications. Briefly, Jurkat cells (2 ϫ 10 7 ) harvested by low speed centrifugation at 4°C were washed twice with phosphate-buffered saline and suspended in 1.5 ml of hypotonic Hepes buffer (HHB: 10 mM Hepes, pH 7.4, 5 mM MgCl 2 , 40 mM KCl, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml aprotinin, 10 g/ml leupeptin). After incubation on ice for 15 min, cell suspensions were homogenized by using a Dounce homogenizer for 30 -50 strokes and then spun at 200 ϫ g for 10 min. The supernatants were further centrifuged at 10,000 ϫ g for 15 min at 4°C. The resulting pellets were washed with HHB, suspended in 200 l of HHB, and sonicated for 5 s, which was designated as HM for heavy membrane fraction. The resulting supernatants were further centrifuged at 150,000 ϫ g for 90 min at 4°C (Beckman SW-60 rotor) and were designated as the cytosolic fraction (Cyto).
Western Blot Analysis-Western blotting was performed as described previously (33) with minor modifications. Aliquots of subcellular fractions were separated by SDS-PAGE and electroblotted to nitrocellulose filters in a Trans-Blot Cell (Bio-Rad). Membranes were blocked for 1 h in a blocking solution (for PKC-␦, -e, -, -, and HSP90␤: 5% nonfat milk, 0.05% Tween 20 in phosphate-buffered saline; for PKC-⑀, 5% nonfat milk, 0.1% Tween 20, 10 mM Tris-Cl, pH 7.5, and 100 mM NaCl) and then incubated 1 h at 37°C or overnight at 4°C with specific antibodies in blocking solution. Protein bands were visualized with the ECL detection system (Amersham Pharmacia). The intensity of each protein band was scanned and analyzed on an UltroScan XL laser densitometer (Amersham Biosciences).

Novel PKCs Are Involved in Heat Shock-induced hsp90␤
Gene Expression-Our previous findings indicated that PKC was involved in hsp90 gene expression in heat-shocked Jurkat cells (15). To identify the specific PKC isoform(s) involved, we assessed the effects of two isoform-selective PKC inhibitors, GF109203X and Gö6976, on heat shock-induced expression of hsp90 genes (34,35). Jurkat cells were pretreated with indicated PKC inhibitors and subsequently heat shocked at 42°C for 1 h. Controls were incubated at 37°C for 1 h. The mRNA levels of hsp90 genes were determined by competitive RT-PCR. As shown in Fig. 1A, GF109203X, which inhibits the activity of both classical (cPKC-␣ and -␤) and novel (nPKC-␦ and -⑀) PKC isoforms, specifically repressed the heat shock-induced accumulation of hsp90␤ mRNA in a dose-dependent manner, which was initially observed at the concentration of 0.02 M. In contrast, GF109203X showed no effect on the induction of hsp90␣ mRNA by heat shock. GF109203X elicited similar effects on the promoter activity of hsp90 genes; the inhibitor repressed only the heat shock-induced promoter activity of the hsp90␤ gene but not the hsp90␣ gene (Fig. 1B). At the concentration of 2.0 M, GF109203X abolished the heat shock-induced elevation of the hsp90␤ promoter activity as well as mRNA induction. The results indicated that PKC plays a specific role in mediating the induction of the hsp90␤ gene expression in heat-shocked Jurkat cells.
To further define the identity of PKC isoform(s) involved, a cPKC-specific inhibitor Gö6976 was used. As shown in Fig. 2, at various concentrations, Gö6976 exerted little inhibition on the heat shock-induced mRNA accumulation (Fig. 2, left panel) and promoter activity (Fig. 2, right panel) of the hsp90␤ gene. Together, the data exclude the participation of cPKC in the regulation of heat shock-induced hsp90␤ gene expression and suggest a role for nPKCs in this event.
Heat Shock-induced Membrane Translocation of nPKC-⑀ in Jurkat Cells-PKCs have been shown to undergo activationdependent intracellular redistribution. In particular, a membrane translocation upon stimulation has been linked to the PKC activation. We investigated whether any nPKC isoform can be induced to translocate from the cytosol to the membrane in Jurkat cells by heat shock at 42°C for 1 h. Cells were either maintained at 37°C (mock controls) or heat shocked at 42°C (experimental groups). As a positive control, cells were also stimulated with a PKC activator, PMA. Cells harvested from different treatments were subjected to subcellular fractionation followed by Western blot analysis for detecting cytosolic and membrane levels of nPKC isoforms. Results (Fig. 3) showed that Jurkat cells expressed all four novel PKCs, PKC-␦, -⑀, -, and -, as determined by Western blot analysis. PKC-⑀ was detected as doublet with molecular sizes around 90 kDa, whereas the other three novel PKCs were around 80 kDa, which is consistent with data from others (36,37). We found that in the unstimulated state all nPKCs existed predominantly in the cytosol. Upon heat shock at 42°C or PMA treatment, only PKC-⑀ was substantially enriched in the membrane fraction (Fig. 3A, row 2, lanes 3 and 5). In contrast, little difference was observed in the distribution of the other three nPKC (Fig. 3A, rows 1 (PKC␦), 3 (PKC), and 4 (PKC)). The redistribution of PKC-⑀ in response to heat shock may be indicative of its activation, suggesting that this isoform may be involved in heat shock-induced hsp90␤ expression in Jurkat cells. PKC-⑀ Mediates Heat Shock-induced Expression of the hsp90␤ Gene-Because PKC-⑀ showed a distinct change in its intracellular location during heat shock, we further examined whether PKC-⑀ is a major player in regulating hsp90␤ expression and whether PKC-⑀ could act independently. Jurkat cells were transfected with wild type (WT), constitutively activated PKC-⑀ (A159) (CA-PKC⑀), or dominant-negative PKC-⑀ (K437R) (DN-PKC⑀); effects of PKC-⑀ expression on hsp90␤ expression were assayed by mRNA level and promoter activity in response to heat shock. Western blot analysis (Fig. 4A, bottom panel) demonstrated that transfected FLAG-tagged WT PKC-⑀ and its mutants were efficiently expressed in Jurkat cells. As shown in Fig. 4A, none of PKC-⑀ constructs notably affected the promoter activity of the hsp90␤ gene in the absence of heat shock (white bars). Upon heat shock, expression of CA-PKC⑀ remarkably induced the hsp90␤ promoter activity (a 4-fold induction), whereas expression of DN-PKC⑀ blocked heat shock-induced induction of the hsp90␤ promoter activity (Fig.  4A, lane 8 versus lanes 2, 4, and 6). Of note, expression of WT-PKC-⑀ showed no significant effect on the basal and heat shock-induced promoter activity (Fig. 4A, WT versus PCI).
To confirm the specificity of PKC-⑀, we investigated the effect of PKC-␣ on hsp90␤ expression by heat shock (Fig. 4B). Consistent with PKC inhibitor studies (Fig. 2), which exclude the involvement of cPKCs in the induction of hsp90␤ expression upon heat shock, we found that expression of PKC-␣ (wild type, CA-PKC-␣, or DN-PKC-␣) had no effect on the promoter activity of the hsp90␤ gene in response to heat shock. Together, these data clearly suggest an exclusive role for PKC-⑀ in regulating the expression of the hsp90␤ gene by heat shock.
PKC-⑀ Is Not Involved in the Regulation of hsp90␤ Expression by PMA-Several lines of evidence indicate that PMA is capable of inducing the expression of hsp90 genes in a cell type-specific manner, which may also depend on the dosage and length of treatment adopted (38, 39). In particular, we showed previously that PMA efficiently induced hsp90␤ mRNA but was less effective in the induction of hsp90␣ mRNA in Jurkat cells (39). Because both PMA and heat shock can lead to PKC-⑀ translocation to the membrane in Jurkat cells (Fig. 3), we investigated whether PKC-⑀ also plays a role in the PMAinduced hsp90␤ expression. Jurkat cells were transfected with DN-PKC-⑀ (Fig. 5A, black bars) or an empty vector (Fig. 5A, white bars) and then treated with heat shock (42°C, 1 h), PMA (100 nM, 30 min), or a combination of heat shock and PMA. The promoter activity of the hsp90␤ gene was assessed by competitive RT-PCR. As shown in Fig. 5A, when transfected cells were heat shocked in the absence of PMA (Fig. 5A, left panel), the heat shock-induced promoter activity of the hsp90␤ gene was specifically blocked by the expression of DN-PKC-⑀ (Fig. 5A,  bar 4 versus bar 3). When Jurkat cells were treated with PMA alone, the promoter activity of the hsp90␤ gene in vectortransfected cells (bar 5) was about 2-fold over that of the untreated control (bar 1), indicating that PMA induced hsp90␤ gene expression. Interestingly, this PMA-induced promoter activity was not affected by the expression of DN-PKC-⑀ (bar 6 versus bar 5). Additionally, we found that when cells were treated with heat shock in addition to the PMA stimulation, there was an additive effect on the induction of hsp90␤ gene expression (bar 7), and DN-PKC-⑀ partially blocked this induction (bar 8 versus bar 7). These data suggest that DN-PKC-⑀ abrogated the heat shock but not PMA-induced promoter activity of the hsp90␤ gene. In contrast to DN-PKC-⑀, expression of DN-PKC-␣ showed no obvious effects on heat shock as well as PMA-mediated induction of the hsp90␤ promoter activity (Fig 5B, bar 4 versus bar 3 and bar 8 versus bar 7). These results further confirm our notion that PKC-⑀ is specifically involved in the heat shock-induced expression of the hsp90␤ gene. It also suggests that heat shock-induced and PMA-induced expression of the hsp90␤ gene are regulated by distinct mechanisms in a PKC-⑀-dependent and -independent manner. DISCUSSION PKC has been implicated as a potential mediator of inflammatory stimuli-induced gene expression (22). It was shown that  (C, control). Promoter activities of the hsp90␤ gene were detected and presented in a manner similar to that described in Fig. 1B. Expression of wild-type and mutant PKC-⑀ was determined by Western blotting with an anti-FLAG antibody (bottom panel). B, effects of PKC-␣ on heat shock-induced promoter activity. PKC-␣ constructs (WT, CA, or DN) or vector (pCDNA3) was individually co-transfected with hsp90␤3.1-CAT and pM-CAT into Jurkat cells following the procedure similar to that described above. Equal amounts of cell lysates from the vector or PKC-␣ construct transfected cells were subjected to Western blotting with an anti-PKC-␣ antibody to detect PKC-␣ expression. Actin expression was detected for loading controls. the nuclear PKC activity was stimulated by heat shock prior to the induction of hsp70 mRNA in rat liver (40,41). Furthermore, we have shown previously that PKC inhibitor GF109203X attenuated the heat shock-induced hsp90␤ expression in Jurkat cells (15). These data indicate a link between PKC stimulation and heat shock response and suggest that a role for PKC in heat shock response is to modulate the expression of hsp genes.
The present study determined the mechanism by which PKC regulates the heat shock response, in particular the induction of hsp90 expression. We showed that pan-PKC inhibitor GF109203X, which inhibits both cPKCs (-␣ and -␤) and nPKCs (-␦ and -⑀), preferentially affected heat shock-induced mRNA expression (Fig. 1A) and the promoter activity (Fig. 1B) of the ␤ version of the hsp90 gene but had no effect on that of the hsp90␣ gene. This finding suggests that the repression of hsp90␤ expression is mediated by either nPKCs or cPKCs, or alternatively, by the combination of the two PKC subfamilies. To disclose which one of the two PKC subfamily members exerts the enhancing function to hsp90␤, we employed a second PKC inhibitor, Gö6976, that is specific to the cPKC subfamily and found that no changes occurred on heat shock-induced mRNA expression and promoter activity of hsp90␤ at concentrations up to 2 M (Fig. 2). Thus we excluded the involvement of the cPKC subfamily and suggest that members of the nPKC subfamily are responsible for mediating the induction of hsp90␤ expression in response to heat shock.
PKC isoforms exhibit distinct enzymological properties, differential subcellular localization, and different modes of cellular regulation. Because translocation from the cytosol to the phospholipid-rich environment such as membrane is one of the essential steps that activate the enzyme (42), we investigated the subcellular distribution of nPKC in response to heat shock and PMA treatment. Subcellular fractionation experiments indicated that Jurkat cells expressed four novel PKCs, PKC-␦, -⑀, -, and -. However, only PKC-⑀ demonstrated a heat shock or PMA-mediated translocation from the cytosol to the membrane, evidenced by a decreased level in the cytosol and an increased level in the membrane-associated fractions, indicating that PKC-⑀ may be activated by heat shock in the manner similar to PMA treatment. In contradiction to our results, Holmberg et al. (40) reported that PKC-⑀ was incapable of translocation from the cytosol to the cellular particulate fraction upon heat shock (42°C for 15 min) in K562 cells. It is possible that heat shock at 42°C for 15 min (versus 42°C for 60 min in our study) may not be sufficient to activate the enzyme (membrane translocation). Alternatively, the heat shock-induced translocation of PKC-⑀ is cell-type specific.
Expression of wild-type, constitutive active, or dominantnegative PKC-⑀ in Jurkat cells did not lead to an obvious change in basal hsp90␤ expression in the absence of heat shock (Fig. 4A). Upon heat shock, however, the constitutively active PKC-⑀ exerted a marked induction (a 4-fold increase) of the hsp90␤ gene, whereas dominant-negative PKC-⑀ completely blocked this effect, suggesting that PKC-⑀ is required only for heat shock-mediated events. It should be noted that expression of WT-PKC-⑀ seems to have little effect on heat shock-mediated induction of hsp90␤ expression, which was only slightly higher than that in the vector-transfected cells. This suggests that the endogenous PKC-⑀ may be sufficient for the induction of hsp90␤ by heat shock. In the case of expression of CA-PKC-⑀, it is possible that activated PKC-⑀ may interact directly with Hsp90, a molecular chaperone, in a functional complex or module leading to the phosphorylation of Hsp90, which in turn stabilizes and keeps PKC-⑀ in a conformation amenable to a full activation upon heat shock. This possibility is supported, at least in part, by our findings that, in Jurkat cells, Hsp90␤ translocated simultaneously with PKC-⑀ to the membrane upon heat shock, and after the recovery from heat shock, the membrane fractions of both hsp90␤ and PKC-⑀ were reduced. 2 The specificity of PKC-⑀ on heat shock-induced expression of the hsp90␤ gene is further supported by the fact that neither 2 J. M. Wu and Y. F. Shen, unpublished data. CA-PKC-␣ nor DN-PKC-␣ affected the promoter activity of the hsp90␤ gene in response to heat shock, suggesting that heat shock-induced expression of the hsp90␤ gene is regulated by a PKC-⑀-dependent mechanism.
Despite the fact that Hsp90␣ and Hsp90␤ are highly homologous at the protein level, we have demonstrated earlier that the regulatory sequences and mechanisms that control the expression of these two genes are completely different (13,14). Studies from others and ours suggest that distinct signaling mechanisms are involved in the precise regulation of each hsp90 gene in a heat shock response. For example, we found previously that the intronic heat shock element of hsp90␤ preferentially binds to the heat shock factor (HSF1) and plays a critical role on heat shock induction of the gene (14). Additionally, it was reported that heat shock-induced Hsp70 expression was controlled primarily at the transcriptional level through the activation of HSFs, whereas the PMA-induced HSP70 expression resulted from a stabilization of HSP70 mRNA through the post-transcriptional-dependent mechanism(s) (38). These data imply that mechanism(s) that control the expression of hsp genes may also be stimuli-specific. Consistent with this notion, we found that PKC-⑀ is required for heat shock-but not PMA-induced induction of hsp90␤ expression in Jurkat cells. Because PKC has been reported to mediate the phosphorylation of HSF1 and to regulate Hsp70 expression in human cells (41), it is therefore interesting to explore whether PKC-⑀-mediated heat shock response requires the participation of HSF1 in our system.
In summary, we demonstrate that PKC-⑀ can be activated by heat shock and is required for the induction of hsp90␤ gene expression in heat-shocked Jurkat cells. It is conceivable that PKC-⑀ and Hsp90␤ may participate in a functional active module to protect cells from external damage, particularly in heat shock-mediated events.