The 28-kDa Protein Whose Phosphorylation Is Induced by Protein Kinase C Activators in MCF-7 Cells Belongs to the Family of Low Molecular Mass Heat Shock Proteins and Is the Estrogen-regulated 24-kDa Protein*

We have previously reported the presence of a 28- kDa protein in human mammary adenocarcinoma MCF-7 cells, whose phosphorylation by phorbol ester 12-0-tetradecanoylphorbol-13-acetate (TPA) and per- meant diacylglycerol 1,2-dioctanoyl-sn-glycerol was correlated to growth arrest induced by the protein kinase C (PKC) activators. We now investigate the possible identity of this protein with the estrogen-regulated “24-kDa” protein shown as related to the mammalian heat shock protein 27 (Fuqua, S. A. W., Blum-Salingaros, M., and McGuire, W. L. (1989) Can- cer Res 49, 4126-4129). phoprotein suggested identical sites of phosphorylation upon TPA and heat shock stimulation. Partial amino acid sequencing of the 28-kDa protein revealed iden- tity with both the 24-kDa protein and the mammalian HSP27. The fact that estrogens and PKC, respectively, regulate expression and phosphorylation of this 24128-kDa protein strongly argues for its key role in MCF-7 cell proliferation and differentiation. incomplete Freund's adjuvant, and antisera were then collected. Purification and Internal Sequencing of the 28-kDa Protein-The 28-kDa isoform a, isolated from two-dimensional IEF/SDS-PAGE, was digested in the gel matrix with porcine trypsin, and the resulting peptides were separated on a narrow bore C18 Altex column (25 X 0.2 cm, Beckman Instruments). Selected peptides were submitted to automatic amino-terminal sequencing on an Applied Biosystems se- quenator (model 470) coupled to a phenylthiohydantoin-derivative analyzer.


Department of Medicine, Division of Oncology, University of Texas Health Science Center, San Antonio, Texas 78284
We have previously reported the presence of a 28-kDa protein in human mammary adenocarcinoma MCF-7 cells, whose phosphorylation by phorbol ester 12-0-tetradecanoylphorbol-13-acetate (TPA) and permeant diacylglycerol 1,2-dioctanoyl-sn-glycerol was correlated to growth arrest induced by the protein kinase C (PKC) activators. We now investigate the possible identity of this protein with the estrogenregulated "24-kDa" protein shown as related to the mammalian heat shock protein 27 (Fuqua, S. A. W., Blum-Salingaros, M., and McGuire, W. L. (1989) Cancer Res 49,[4126][4127][4128][4129]. 32P-Labeled 28-kDa protein from TPA-treated MCF-7 cells was immunoprecipitated with a 24-kDa-specific monoclonal antibody. Immunoblots from cell extracts fractionated by two-dimensional isoelectric focusing/SDS-polyacrylamide gel electrophoresis demonstrated that TPA induced the conversion of a 28-kDa isoform "a" (PI 6.7) to a more acidic isoform "b" (PI 6.2). Two-dimensional gel analysis of [SH]leucine-labeled MCF-7 cell extracts demonstrated that conversely to TPA, which induced only phosphorylation of 28-kDa protein, heat shock induced both synthesis (increase of isoform a) and phosphorylation (conversion of isoforms a to b) of the protein. 32P labeling of MCF-7 cells allowed demonstration of the presence of an extra phosphoisoform "c" (PI 5.9) upon TPA as well as heat shock treatment. When cells were pretreated with the bisindolylmaleimide GF109203X, a selective inhibitor of PKC, the heat shock-induced phosphorylation was unchanged, while the TPA effect was almost abolished, suggesting that the heat shockactivated protein kinase was very likely different from PKC. However, peptide mapping of the 28-kDa phos-* This work was supported by Institut National de la Santi et de la Recherche Midicale and by Association de la Recherche contre le Cancer, France. A preliminary report of this study was presented at the 8th International Conference on Second Messengers and Phosphoproteins, Glasgow, Scotland, August 3-8, 1992. 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. phoprotein suggested identical sites of phosphorylation upon TPA and heat shock stimulation. Partial amino acid sequencing of the 28-kDa protein revealed identity with both the 24-kDa protein and the mammalian HSP27. The fact that estrogens and PKC, respectively, regulate expression and phosphorylation of this 24128-kDa protein strongly argues for its key role in MCF-7 cell proliferation and differentiation.

The nucleotide sequencefs) reported in thispaper has been submitted to the GenBankTM/EMBL Data
Heat shock proteins (HSPs)' consist of a number of highly conserved proteins that are synthesized by all pro-and eucaryotic organisms in response to environmental stress including hyperthermia (1-3). Although these proteins are thought to play primarily a protective role in cells subjected to high temperature and other stresses, several lines of evidence suggest that they are involved in a number of other cell functions (2,3). High molecular weight HSPs including HSPSO, HSP70, and HSP6O have been studied with the greatest details and demonstrated as molecular "chaperones" in protein-protein interactions (1, 3). For example, HSPSO has been shown to be associated with steroid receptor (4) or with tyrosine kinases encoded by oncogenes (5). HSP7O and HSP6O have been implicated in protein folding, unfolding, oligomerization, and translocation (2,3). The low molecular weight HSPs are much less understood. Like the other families of heat shock proteins, they are involved in thermotolerance (6) and very likely in other cell functions, including cell growth and differentiation, for the following reasons. (i) Although their synthesis is stress-induced, they have been shown as constitutive proteins that are expressed at specific stages of development at normal temperatures (1). (ii) They are phosphorylated in response to a wide variety of stimuli including growth factors (7,8).
Protein kinase C (PKC) is believed to play a key role in transmembrane signaling leading to cell differentiation and proliferation (9). Characterization and identification of endogenous proteins phosphorylated by PKC activators have received particular attention. We have previously demonstrated the presence of a 28-kDa protein in human mammary adenocarcinoma cell line MCF-7 (IO), whose phosphorylation by TPA and DiCa was closely correlated to growth arrest 'The abbreviatTons used are: HSP, heat shock protein; PKC, protein kinase C; TPA, 12-0-tetradecanoylphorbol-13-acetate; DiCs, 1,2-dioctanoyl-sn-glycerol; IEF, isoelectric focusing; PAGE, polyacrylamide gel electrophoresis; PBS, phosphate-buffered saline; HPLC, high pressure liquid chromatography. induced by the PKC activators (11). We further brought evidence indicating that this 28-kDa protein was very likely a member of the low molecular weight HSP family (12). Indeed, when proteins phosphorylated upon MCF-7 cell exposure to TPA and those synthesized after heat shock were fractionated on SDS-PAGE, the 32P-and [3H]leucine-labeled 28-kDa proteins showed the same electrophoretic mobility. Moreover, heat shock treatment of cells induced a clear-cut increase of 32P incorporation into the 28-kDa protein.
In the meantime, McGuire and co-workers demonstrated that a 27128-kDa estrogen-regulated protein from MCF-7 cells, originally termed "24-kDa" protein (13) and more recently "stress-responsive protein 27" (14), also belonged to the low molecular weight HSP family (14). Indeed, the carboxyl-terminal amino acid sequence deduced from the partial cDNA encoding for this protein contained striking homology with both low molecular weight HSPs of Drosophila and mammalian a-crystallin, now also reported as a HSP (15). Furthermore, the truncated cDNA of the 24-kDa protein was identical to the 3'-region of the human HSP27 cloned from a genomic library (16). The 24-kDa mRNA was significantly induced both by estrogen and heat shock treatment of . It was tempting to anticipate that the 28-kDa protein phosphorylated by PKC activators was similar to the 24-kDa estrogen-regulated protein. However, such a hypothesis was strikingly provocative since the former is believed to mediate cell growth arrest while the latter is supposed to be a marker for cell proliferation.
In this report, we further characterize the MCF-7 cell 28-kDa phosphoprotein as a HSP, and we bring evidence that this protein is the estrogen-regulated 24-kDa protein (or stress-responsive protein 27).
For protein phosphorylation studies, subconfluent cultures (0.5-1 X 10' cells/35-mm dish) were washed in phosphate-free Krebs-Ringer buffer containing 20 mM Hepes pH 7.3, 0.1% bovine serum albumin, and 0.2% glucose, then incubated in 1 ml of the same fresh buffer containing 50 pCi of [32P]phosphoric acid neutralized with 0.1 M Tris base. Incubation was performed either at 37 "C for 1 h followed by 1 h at 42 "C (heat shock-induced protein phosphorylation) or at 37 "C for 2 h, the phorbol ester TPA (100 ng/ml) being added the last 20 min (TPA-induced protein phosphorylation).
For protein synthesis studies, subconfluent cultures (0.5-1 X lo6 cells/35-mm dish) were or were not submitted to heat shock (1 h at 42 "C) and then incubated for 1 h at 37 "C in 1 ml of leucine-free RPMI 1640 medium containing 50 pCi of [3H]leucine. When indicated, TPA (100 ng/ml) was added the last 20 min.
Alternatively, blots were incubated with the polyclonal anti-HSP27 peptide antibody (dilution 1/200). In this case, the rabbit anti-mouse IgG step was omitted.
Immunoprecipitation-32P-Labeled MCF-7 cells were incubated for 20 min in the absence or in the presence of 100 ng/ml TPA. After cell washing in cold PBS, cells were rapidly harvested in PBS, and cell pellets were homogenized with Dounce in 0.1 ml of 20 mM Tris-HCI, pH 7.4, 10% (v/v) glycerol, 1 mM EDTA, 10 pg/ml leupeptin, and 5 mM (3-mercaptoethanol. Homogenates were centrifuged for 1 h at 105,000 X g and cytosols incubated for 2 h at 20 "C with protein A-Sepharose CL-4B previously coupled to rabbit anti-mouse IgG (incubation for 2 h at 20 "C; antibody dilution, 1/10) and 24-kDaspecific monoclonal antibody (incubation for 2 h at 20 "C; dilution, 1/50). The antigen-antibody complexes were then extracted using 0.1 ml of the electrophoresis sample buffer, heated at 60 "C for 10 min, and analyzed by monodimensional SDS-PAGE.
Peptide Mapping of 28-kDa Protein Isoforrm-One-dimensional peptide mapping was carried out according to Cleveland et al. (19) using protease V8 from S. aureus. After 32P-labeled MCF-7 cells were exposed to TPA or heat shock and fractionated on two-dimensional IEF/SDS-PAGE, the respective phosphoisoforms b and c of 28-kDa protein were excised from the gels and directly loaded on 4.5-15% SDS-PAGE and then overlaid with protease V8 (200 ng). Digestion proceeded in the stacking gel during the subsequent electrophoresis.
For two-dimensional peptide analysis, pieces of gel containing the required b and c phosphoisoforms were excised, minced, and incubated overnight with 12.5 pg of diphenylcarbamyl chloride-treated trypsin in 0.5 ml of 50 mM ammonium bicarbonate, pH 8. Lyophilized phosphopeptides were dissolved in 5% acetic acid and then applied to thin-layer cellulose plates and electrophoresed (500 V for 25 min) using a pH 4.4 buffer containing 15% acetone, 2% pyridine, 4% acetic acid, and 79% water. Ascending chromatography was performed in 37.5% butanol, 7.5% acetic acid, 25% pyridine, and 30% water.
Polyclonal Anti-HSP27 Peptide Antibody-An oligopeptide corresponding to the carboxyl-terminal end of human HSP27 (residues 184-193: TFESRAQLGG) was synthesized by the standard solid phase method using an Applied Biosystems 430A peptide synthesizer. The following side chain protecting groups were used on the tbutoxycarbonyl amino acids: tosyl (Arg), benzyl ether (Ser, Thr), and benzyl ester (Glu). Cleavage of the peptide from the resin and removal of side chain protecting groups were performed using the HF method. Peptide purity was checked by reverse phase HPLC analysis using an RP300 C8 column with a linear acetonitrile-0.1% trifluoroacetic acid gradient. Molecular mass was confirmed by fast atom bombardment mass spectrometry (MH+, 1065.4) using a ZAB-H5 double focusing spectrometer (VG analytical, Manchester, UK). The peptide (5 mg) was coupled to keyhole limpet hemocyanin (5 mg) in 5 ml of 0.1 M NaHC03, pH 8.6, and 0.05% glutaraldehyde, and the mixture was dialyzed against 0.1 M NaCl for 24 h. After mixing with complete Freund's adjuvant, the conjugate was then injected subcutaneously into rabbits (0.25 mg of peptide). After 4 weeks, animals were boosted every 2 weeks for 6 weeks with the same amount of peptide in Identification of 28-kDa Phosphoprotein in MCF-7 Cells incomplete Freund's adjuvant, and antisera were then collected.
Purification and Internal Sequencing of the 28-kDa Protein-The 28-kDa isoform a, isolated from two-dimensional IEF/SDS-PAGE, was digested in the gel matrix with porcine trypsin, and the resulting peptides were separated on a narrow bore C18 Altex column (25 X 0.2 cm, Beckman Instruments). Selected peptides were submitted to automatic amino-terminal sequencing on an Applied Biosystems sequenator (model 470) coupled to a phenylthiohydantoin-derivative analyzer.

RESULTS
Immunodetection of the 28-kDa Phosphoprotein in  Cells-Previous data indicated that the 28-kDa protein phosphorylated in MCF-7 cells under TPA or DiCs stimulation was very likely a member of the low molecular weight HSP family (12). We wondered whether this protein could be related to the estrogen-regulated 24-kDa protein reported by McGuire and co-workers (14) and further demonstrated as homologous to the mammalian HSP27. T o assess the possible identity of the respective 28-and 24-kDa proteins, we performed immunoprecipitation studies with a 24-kDa-specific monoclonal antibody (C11) following stimulation of '"P-labeled MCF-7 cells with the PKC activator TPA. Immunoprecipitates were fractionated on SDS-12% PAGE, and the gels were submitted to autoradiography. Fig. 1A shows that the C11 antibody immunoprecipitated the 28-kDa protein phosphorylated upon TPA stimulation of cells. The specificity of this immunoprecipitation was assessed by using a normal mouse serum instead of the 24-kDa-specific antibody. Immunodetection of the 28-kDa protein was also performed after Western blotting of unlabeled MCF-7 cell extracts fractionated by SDS-PAGE (Fig. 1 B ) or by two-dimensional IEF/ SDS-PAGE (Fig. 1C). While TPA stimulation of cells did not increase the amount of the specifically recognized 28-kDa protein (Fig. lB), it clearly induced its phosphorylation, leading to the conversion of the isoform a (PI = 6.7) to the more acidic isoform b (PI = 6.2).
Two-dimensional IEF/SDS-PAGEAnalysis of fH]Leucineand :i2P-Labeled 28-kDa Protein upon TPA or Heat Shock Treatment of MCF-7 Cells-Previous data suggested that heat shock treatment of MCF-7 cells could induce both synthesis and phosphorylation of the 28-kDa protein (12). To further characterize this phenomenon and to compare it to the TPA effect, we performed two-dimensional IEF/SDS-PAGE analysis of ['Hlleucine-and "P-labeled MCF-7 cells following TPA in the absence (Cont) or in the presence of 100 ng/ml TPA. Immunoprecipitates obtained from '"P-labeled cells with antibody C11 or normal mouse serum ( N M S ) were subjected to SDS-PAGE followed hy autoradiography ( A ) . Western blots from unlabeled cells fractionated by SDS-PAGE ( R ) or two-dimensional IEF/SDS-PAGE (C) were probed with antibody C11. Only portions of the respective autoradiographs corresponding to the recognized 28-kDa protein or isoforms a and b were shown. and heat shock treatment (Fig. 2).
["]Leucine labeling showed that TPA induced phosphorylation, i.e. conversion of isoform a (PI = 6.7) to isoform b (PI = 6.2), but not synthesis of the 28-kDa protein (no increase of isoforms a + b from TPA-treated cells uersus isoform a from control cells) while heat shock induced both phosphorylation (appearance of isoform b) and synthesis (increase of isoform a) of the 28-kDa protein. The sensitivity of :12P labeling was allowed to demonstrate the presence of two phosphoisoforms, b (PI = 6.2) and c (PI = 5.9), upon TPA as well as heat shock treatment. A small amount of phosphoprotein b was visible in the control confirming the two-dimensional pattern observed in Fig. 1C where isoform b was weakly present in the control.
To investigate the nature of the protein kinase involved in the heat shock-induced phosphorylation of the 28-kDa protein, we studied the protein phosphorylation pattern observed when cells were pretreated with staurosporine, a compound that is believed to be a potent PKC inhibitor. As shown in Fig. 3, in such staurosporine-treated cells, the TPA-induced 28-kDa protein phosphorylation was markedly inhibited with a total disappearance of the phosphoisoform c and a marked reduction of the "P labeling of isoform b. Staurosporine also decreased, although a t a lesser extent, the heat shock-induced 28-kDa protein phosphorylation. As staurosporine has been recently reported to inhibit other protein kinases than PKC (20,21), we performed identical studies in the presence of the bisindolylmaleimide GF109203X, a potent and more selective inhibitor of PKC (22). Fig. 4 shows that, in GF109203Xtreated cells, the effect of TPA on 28-kDa protein phosphorylation was almost abolished while heat shock-induced phosphorylation was unchanged, suggesting that the heat shockactivated protein kinase is very likely different from PKC. Such a hypothesis is further reinforced by the fact that TPA but not heat shock induced the phosphorylation of a 80-kDa/ PI 4.5 protein, the selectivity of GF109203X being demonstrated by the disappearance of this PKC-specific protein phosphorylation in cells treated with this compound.
T o investigate whether TPA and heat shock induced phosphorylation of the 28-kDa protein at the same sites, protease V8 peptide maps were performed from the individual b and c Normal cells GFX-treated cells phosphoisoforms obtained from two-dimensional IEF/SDS-PAGE. Fig. 5A shows identical phosphopeptide maps for both isoforms b and c from the 28-kDa protein phosphorylated upon TPA as well as heat shock treatment of MCF-7 cells.

H leu
To confirm this finding, we performed two-dimensional peptide analysis following trypsin digestion of b and c 28-kDa isoforms upon phosphorylation by TPA and heat shock. Fig.  5 B again shows similar patterns with two major spots observed in each case. Subcellular Localization of the 28-kDa Protein upon TPA and Heat Shock Treatment of MCF-7 Cells-Low molecular weight HSPs have been previously shown to translocate from cytosol to nuclear compartment during heat shock exposure of cells (23)(24)(25). To further compare the effects of TPA and heat shock on 28-kDa protein, we investigated the cellular localization of the protein upon the two distinct treatments. Fig. 6 illustrates an immunodetection of the 28-kDa protein in the respective cytosolic and nuclear fractions using a polyclonal antibody raised against a synthetic peptide derived from the HSP27 amino acid sequence. While heat shock induced the expected redistribution of 28-kDa protein from cytosol to mclear pellet, TPA did not change the initial cytoplasmic localization of the protein. The heat shock-induced translocation of the 28-kDa protein concerned, at least in part, the phosphorylated protein as the 32P labeling of 28-kDa protein followed the redistribution pattern observed when the total amount of the protein was measured (not shown).
Partial Sequencing of the 28-kDa Protein-Attempts to obtain protein sequence information of the 28-kDa isoform a from two-dimensional gel transferred to polyvinylidene difluoride Immobilon membranes were unsuccessful, very probably because the amino terminus of the protein was blocked. Thus, to obtain internal sequence information, we digested the protein in the gel matrix after isolation on two-dimensional gels. The resulting peptides were separated on HPLC and three peaks were selected for sequencing. The material eluting in two of those peaks gave unique sequences corresponding to the peptides 13-20 and 38-46 of the human HSP27 (16). The material eluting in the third peak showed that it was a mixture of three peptides. The deduced sequences could be assigned to peptides 97-110,141-154, and 172-186 of HSP27. In other words, the peptide sequences obtained were found to be identical to the amino acid sequences of 24-kDa protein (stressresponsive protein 27) and human HSP27 (Fig. 7). The horizontal arrow indicates the direction of electrophoresis to the cathode, while the vertical arrow indicates the ascending chromatography.

DISCUSSION
Growth arrest of MCF-7 cells by PKC activators TPA and Dice has been correlated previously to the phosphorylation of a 28-kDa endogenous protein (11). In the present study, we have identified definitely this protein as a member of the low molecular weight HSP family. ["HILeucine labeling of MCF-7 cells (Fig. 2) demonstrated that heat shock induced both synthesis (increase of isoform a) and phosphorylation (appearance of isoform b) of 28-kDa protein while TPA caused only its phosphorylation (conversion of isoform a to isoform b). j2P labeling of MCF-7 cells was allowed to detect a second phosphoisoform of the 28-kDa protein, isoform c in addition to isoform b, confirming the capability of both TPA and heat shock to induce the phosphorylation of 28-kDa protein.
In several but not all experiments, the labeling of isoform c was more pronounced after heat shock than upon TPA exposure (Figs. 3 and 4). However, it is difficult to make conclusions about this phenomenon because of the distinct procedures used. "P labeling of cells was either performed for 1 h a t 37 "C followed by 1 h a t 42 "C (heat shock-induced phosphorylation)  (bottom). Cells were homogenized with Dounce in 20 mM Tris-HCI, pH 7.4, 10% glycerol, 1 mM EDTA, 10 pg/ml leupeptin, and 5 mM B-mercaptoethanol. 2,000 X g pellets (P) and 105,000 X g supernatants ( S ) were subjected to SDS-PAGE fractionation. Western blots were probed with the polyclonal antibody raised against the HSP27 peptide.
or 2 h at 37 "C, TPA being added the last 20 min (TPAinduced phosphorylation). When 32P labeling of cells was carried out for 2 h a t 37 "C following 1 h of heat shock at 42 "C, isoform c was much less evident (not shown), suggesting that this isoform is likely subject to rapid dephosphorylation or degradation. Whether the poor labeling of isoform c following TPA treatment of cells reflects the stimulation by the phorbol ester of specific phosphatase or protease for 28-kDa c remains to be established. Alternatively, the heat shockdependent phosphorylation of the 28-kDa protein might involve a protein kinase distinct from PKC, inducing a more pronounced degree of phosphorylation of 28-kDa protein with the appearance of isoform c. In any case, the stoichiometry of phosphorylation of the 28-kDa protein was difficult to assess upon heat shock treatment as both synthesis and phosphorylation occurred. In the case of TPA stimulation, 30-50% of ['Hlleucine-labeled isoform a appeared converted to phosphoisoform b (Fig. 2). This was confirmed by immunological quantification of unphosphorylated (isoform a) and phosphorylated (isoform b) 28-kDa protein (Fig. IC). Experiments using the bisindolylmaleimide GF109203X strongly suggest that the protein kinase activated by heat shock is different from PKC. While the TPA-induced phosphorylation of 28-kDa protein was almost abolished in cells pretreated with this compound, no change was observed in the degree of the 28-kDa phosphorylation caused by heat shock (Fig. 4). Conversely to staurosporine, which has been shown to inhibit other protein kinases in addition to PKC (20,21), GF109203X has been demonstrated as a selective PKC inhibitor (22). Our data confirmed these findings, as staurosporine partly inhibited the heat shock-induced 28-kDa protein phosphorylation (Fig. 3). The activation by heat shock of a protein kinase distinct from PKC was further demonstrated by the fact that TPA but not heat shock induced the phosphorylation of a 80-kDa/pI 4.5 protein, which is likely related to the MARCKS protein shown as a specific PKC substrate in various cell systems (26-28). GF109203X totally abolished this TPAinduced phosphorylation (Fig. 4). However, our results indicated that the respective kinases involved in 28-kDa phosphorylation, i.e. PKC and heat shock-dependent kinase, phosphorylate the protein a t similar sites, as the phosphopeptide maps of the 28-kDa isoforms b and c obtained upon TPA as well as heat shock treatment were identical (Fig. 5 , A and B ) .
As two-dimensional peptide analysis following trypsin diges-  7. Comparison of the amino acid sequence of the peptides obtained from 28-kDa protein with the 24-kDa protein and human HSP27. Shown is the amino acid sequence of human HSP27 cloned from a human genomic library (16). The underline represents the peptides sequenced from MCF-7 cell 28-kDa protein (isoform a), while the ouerline corresponds to the amino acid sequence deduced from the partial cDNA encoding the 24-kDa protein (14).

I MTERRVPFSL LRGPSWDPFR DWYPHSRLFD QAFGLPRLPE EWsQWLGGS
tion of b showed two major spots, it is tempting to postulate the presence of two phosphorylation sites in this isoform. The identity of b and c peptide maps is rather intriguing as c is supposed to contain an additional phosphate with regard to b. Such a result could suggest that the putative extra phosphorylation site in c is very close to one of the sites phosphorylated in b. This point needs further investigation.
The other important feature of our study is the demonstration that the 28-kDa phosphoprotein was the estrogen-regulated 24-kDa protein (or stress-responsive protein 27). First, the specific monoclonal antibody C11 raised against this 24-kDa protein recognized the 28-kDa protein phosphorylated upon MCF-7 cell stimulation by TPA (Fig. 1). There was a striking coincidence in the respective PI values of the unphosphorylated a and phosphorylated b isoforms of the immunodetected 28-kDa protein when compared with those found for the 32Pand [3H]leucine-labeled protein (Figs. 2-4).
Second, the 28-kDa protein was also recognized by a specific polyclonal antibody raised against a synthetic peptide derived from the carboxyl-terminal amino acid sequence of HSP27. McGuire and co-workers have reported previously that the 91 carboxyl-terminal amino acids deduced from the partial cDNA encoding the 24-kDa protein showed total homology with HSP27 previously cloned from a human genomic library (16).
Finally, our partial sequencing of the 28-kDa protein demonstrated identity between the peptides sequenced and the corresponding sequences of both HSP27 and 24-kDa protein. Although discrete differences in the whole amino acid sequence of 28and 24-kDa proteins cannot be totally ruled out, it is more probable that both proteins are in fact the same molecule.
Our study also demonstrates that the 28-kDa protein is susceptible to nuclear targeting under heat shock treatment of MCF-7 cells, this subcellular redistribution concerning at least in part the phosphorylated form of the protein. This finding confirms similar results obtained in other cell systems showing translocation of low molecular weight HSPs from cytosol to nuclear compartment upon heat shock (23-25). Whether this phenomenon has a physiological significance remains to be established. However, the fact that TPA did not induce a similar targeting of the 28-kDa protein might suggest different cellular functions of the protein depending on the stimuli. Such a multifunctionality of this HSP is further indicated by the fact that the 24/28-kDa protein is induced both by heat shock (this paper and Ref. 14) and estrogen (14). Moreover, recent immunological evidence (29) suggests that the 24/28-kDa protein from breast cancer might be related to an estrogen-associated protein previously reported in human myometrium (30) and so far not identified. Thus, the function of the 24/28-kDa protein might depend not only on its cellular expression and subcellular localization but also on its possible implication in estrogen receptor machinery. Finally, one can postulate reasonably that the state of phosphorylation of the protein might account for its cellular function. Indeed, it is tempting to hypothesize that the unphosphorylated form of the protein may be related to the stimulation of cell proliferation, while its phosphorylated forms might, on the contrary, refer to cell growth arrest. In any case, the different levels of regulation of this 24/28-kDa protein, i.e. synthesis, phosphorylation, subcellular localization, and possible association with estrogen receptor, strongly argue for its key role in MCF-7 cell function.