Heat Shock Protein-70 Expressed on the Surface of Cancer Cells Binds Parathyroid Hormone-Related Protein in Vitro
http://www.100md.com
内分泌学杂志 2005年第8期
Veterans Affairs San Diego Healthcare System and the Departments of Surgery (J.J.G., C.T., M.B.) and Medicine (Endocrinology) (K.C.S., C.C., D.W.B., L.J.D.), University of California, San Diego, San Diego, California 92161
Address all correspondence and requests for reprints to: Michael Bouvet, M.D., Department of Surgery (112-E), University of California, Veterans Affairs San Diego Healthcare System, 3350 La Jolla Village Drive, San Diego, California 92161. E-mail: mbouvet@ucsd.edu.
Abstract
Recent studies have shown that the functions of PTH-related protein (PTHrP) and its derived peptides cannot be attributed solely to PTH/PTHrP receptor binding. The present study focused on the identification of other proteins that might bind PTHrP at the cell surface. Using affinity chromatography, we applied extracts of cell-surface biotinylated proteins from cancer and normal cell lines over Sepharose beads coupled with different PTHrP-derived peptides. Elution with the corresponding free peptide revealed a major protein of about 70 kDa that was present in all of the PTHrP peptide eluates from cancer cell extracts but not from normal breast cell extracts. Mass spectroscopy analysis and immunoblotting identified this PTHrP-binding protein as heat shock protein-70 (HSP70). Using a recently published algorithm that predicts HSP70 binding sites within proteins, we found that all four PTHrP peptides used in these studies contain amino acid motifs with high probabilities for HSP70 binding in vivo. Cell culture studies in the presence of a polyclonal anti-HSP70 antibody demonstrated increased PTHrP secretion, decreased total cellular protein, and differentially regulated proliferation. Taken together, these studies demonstrate a novel and biologically relevant interaction between cell surface-expressed HSP70 and PTHrP in cancer.
Introduction
PTH-RELATED PROTEIN (PTHrP) is an oncofetal protein originally identified as a product of breast and lung cancer cells that produced humoral hypercalcemia of malignancy (1). Since its initial discovery, PTHrP has been shown to be produced by many normal and malignant cells, including those not typically associated with hypercalcemia of malignancy (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13). PTHrP is known to be a critical regulator of cellular and organ growth, development, migration, angiogenesis, differentiation and survival, and epithelial calcium ion transport (4, 14). In particular, regulation of apoptosis appears to be a key mechanism whereby PTHrP mediates growth (15).
The PTHrP gene results in three mature isoforms generated through alternative mRNA splicing in humans: PTHrP 1–141, a truncated 1-to 139-residue form, and a human-specific 1- to 173-residue form (7). In addition to mRNA splicing, proteolytic processing of PTHrP into peptides appears to be an important regulatory mechanism that could potentially result in the generation of greater than 90 unique peptides (7).
The biological effects of PTHrP are mediated, at least in part, by a member of the seven membrane-spanning and G protein-coupled cell surface receptors, the PTH/PTHrP receptor (16, 17, 18, 19). The PTH/PTHrP receptor has been shown to exert most of its biological effects by binding to the 1–34 amino-terminal fragment of PTHrP (1). In addition to its amino terminus, there is accumulating and compelling evidence that mid-region and carboxyl-terminal forms of PTHrP, generated through processing of the native isoforms, also exert biological effects (1, 16, 17, 20). For example, it has been shown, using PTHrP and PTH/PTHrP receptor knockout and double-knockout mice, that PTHrP gene products inhibit vascular invasion during endochondral bone formation in mice lacking the PTH/PTHrP receptor (21); PTHrP (38–94) stimulates ATP-dependent calcium transport in the basal plasma membrane of the human syncytiotrophoblast (22); synthetic PTHrP (107–139) has been shown to inhibit bone resorption, stimulate osteoblast proliferation, and stimulate IL-6 expression in osteoblasts (23, 24, 25); PTHrP (140–173) has been shown to exert a protective effect from apoptosis in BEN lung cancer cells (26); and finally, the tetrabasic KKKK, PTHrP (147–150) motif regulates collagen synthesis (27). These studies suggest the existence of receptors to these other regions of PTHrP, although none have been identified.
In the present study, we identified surface-expressed heat shock protein-70 (HSP70) as a major PTHrP binding protein from the cancer cell lines. A recently published algorithm that predicts HSP70 binding sites within proteins demonstrates that all four PTHrP peptides used in these studies contain amino acid motifs with high probabilities for HSP70 binding in vivo. Along with HSP70 antibody inhibition studies that resulted in increased PTHrP secretion, decreased total cell protein, and differentially regulated proliferation, our findings demonstrate a novel interaction between PTHrP and surface-expressed HSP70 in cancer cells with cytoprotective consequences.
Materials and Methods
Cell surface biotinylation
Saos-2 osteosarcoma, FG and MiaPaCa2 pancreatic adenocarcinoma, and DU-145 prostate cancer cell lines were cultured in DMEM or RPMI 1640 medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (Invitrogen). All of these cell lines have been previously shown to be PTH/PTHrP receptor positive (3, 28) and responsive to the PTHrP peptides used in these studies (2, 3, 29, 30, 31). MCF10 spontaneously immortalized normal breast epithelial cells were cultured in DMEM and Ham’s F-12 mixture (1:1), containing 5% equine serum and epidermal growth factor as previously described (32). All cell lines were cultured in a humidified atmosphere of 5% CO2 at 37 C. Cultures were harvested at approximately 70% confluency by the addition of 4 mM EDTA in PBS and incubation at 37 C for about 15 min. Detached cells were collected by pipetting and washed three times with PBS. Cells were biotinylated by adding 0.5 mg EZ-Link Sulfo-NHS-LC-Biotin (Pierce Biotechnology, Rockford, IL)/25 x 106 cells, suspended in PBS, and incubated 30 min at 37 C followed by extensive washing with PBS. Biotinylated cell pellets were frozen at –20 C until ready for further processing. Cell viability after the biotinylation procedure was greater than 91% as determined by trypan blue exclusion.
In some experiments, membrane fractions were harvested after biotinylation using Mem-PER eukaryotic membrane protein extraction kit according to the manufacturer’s instructions (Pierce Biotechnology, Inc., Rockford, IL). This method uses aqueous two-phase partitioning as previously described (33).
Extract preparation
Biotinylated cells were extracted in the following ice-cold buffer: lactated Ringer’s buffer (each 100 ml contains 600 mg NaCl, 310 mg sodium lactate, 30 mg KCl, and 20 mg CaCl2, pH 6.5) supplemented with 100 mM n-octyl-?-D-glucopyranoside (Calbiochem, La Jolla, CA), 1 mM MgCl2, and 100 μM phenylmethylsulfonyl fluoride. One milliliter of lysis buffer was added to 75 x 106 biotinylated cells, the pellet was gently resuspended by pipetting, and the lysate was incubated for 30 min on ice with occasional pipetting. The cell lysate was centrifuged 30 min at 16,100 x g and the cleared lysate applied over the appropriate equilibrated column.
Peptides coupling to CnBr-activated Sepharose 4B
Synthetic PTHrP peptides (Bachem, King of Prussia, PA) were coupled to Sepharose according to the manufacturer’s instructions (Sigma-Aldrich, St. Louis, MO). Coupling concentrations were as follows: PTHrP (1–34)-Sepharose, 2.2 mg/ml of packed Sepharose beads; PTHrP (37–86)-Sepharose, 5.3 mg/ml; PTHrP (109–141)-Sepharose, 5.5 mg/ml; PTHrP (140–173)-Sepharose, 7.1 mg/ml; and control BSA-Sepharose, 6.8 mg/ml. The coupling procedure was also conducted according to the manufacturer’s instructions in the absence of peptide as another control.
Affinity chromatography
Cell lysates were incubated with 0.6–0.8 ml PTHrP peptide-Sepharose that was previously equilibrated with wash buffer containing lactated Ringer’s buffer (Baxter Healthcare Corp., Deerfield, IL) supplemented with 50 mM n-octyl-?-D-glucopyranoside and 1 mM MgCl2. Incubations ranged from 2 h at room temperature to overnight at 4 C. Flow-through extract was collected, and the columns were washed with at least 10 column-volumes of wash buffer. Elution was conducted with wash buffer supplemented with 1 mg/ml PTHrP peptide followed by 8 M urea. In some studies, an elution sequence was conducted using 20 mM EDTA in wash buffer, followed by 1 M NaCl, followed by 8 M urea.
Cell culture assay for PTHrP expression
Cells were seeded into six-well tissue culture plates at a density of 2.5 x 105/well in DMEM supplemented with 5% fetal bovine serum and 25 μg/ml anti-HSP70 polyclonal antibody (Chemicon, Temecula, CA). Control cultures were incubated in media described above with irrelevant rabbit IgG, also at 25 μg/ml. Cultures were grown for 72 h, and the conditioned media and cell lysates were analyzed for PTHrP as described below.
Proliferation assays
Cells were seeded into 96-well tissue culture plates at a density of 5 x 103/well under the same conditions described above. After 72 h, proliferation was determined using CellTiter 96 AQueous One Solution cell proliferation assay according to the manufacturer’s instructions (Promega, Madison, WI). This reagent is composed of a novel tetrazolium compound (methyl-p-tolyl sulfide), and an electron coupling reagent, phenazine ethosulfate. Values obtained at initial cell seeding were subtracted from the 72-h values.
RIAs
At the indicated time points, media were harvested and cell extracts were prepared by sonication of cells in lysis buffer containing 0.25 M Tris (pH 7.4), 0.25% Nonidet P-40, and 2 mM EDTA. The insoluble cell fractions were pelleted by centrifugation at 16,000 x g for 15 min, and the soluble fractions were transferred to a fresh tube. Insoluble fractions were solubilized using 8 M urea. Total cell protein from both the soluble and insoluble fractions was then measured using a modified Bradford protein assay with BSA as the standard (Bio-Rad Laboratories, Inc., Hercules, CA). PTHrP was measured by RIA, using modifications as previously described (9, 34). All samples were assayed in multiple dilutions. Intra- and interassay variations were 7 and 12%, respectively (35).
Immunoblotting
Cell lysates and affinity chromatography eluates were analyzed on 12% Bis/Tris NuPAGE gels in MOPS buffer under reducing conditions and transferred to nitrocellulose, and the membranes were incubated with either horseradish peroxidase (HRP)-conjugated streptavidin (1:25,000) or polyclonal anti-HSP70 antibody (1:200) (Chemicon) and monoclonal anti-?-actin antibody (1:1000) (Sigma Chemical Co., St. Louis, MO), followed by HRP-conjugated goat antirabbit or antimouse IgG (1:1000). Peroxidase activity was detected using chemiluminescence according to manufacturer’s instructions (Amersham Biosciences, Little Chalfont, UK).
Immunoprecipitation (IP)
To perform IP, 5 μg of monoclonal antibody directed against either the ?1-integrin subunit, clone P5D2 (Chemicon), or ?-actin, clone AC-15 (Sigma-Aldrich), were absorbed overnight at 4 C onto 25 μl packed antimouse IgG-agarose (Sigma) in IP wash buffer [50 mM Tris-HCl (pH 7.5), 0.5 M NaCl, 1 mM CaCl2, 1 mM MgCl2, and 0.1%Tween 20]. After washing, the antibody-bound agarose beads were incubated with 200 μl of the unbound fractions (flow-through), from cell-surface biotinylated cell lysates after PTHrP affinity chromatography as follows: FG cell lysates over PTHrP (140–173)-Sepharose; MiaPaCa-2 cell lysates over PTHrP (37–86)-Sepharose; MCF10 cell lysates over PTHrP (109–141)-Sepharose; DU-145 cell lysates over PTHrP (140–173)-Sepharose; or Saos-2 cell lysates over PTHrP (1–34)-Sepharose. After 3 h incubation at room temperature, immunoprecipitates were washed extensively, SDS-PAGE sample buffer was added, and samples were boiled and analyzed on 12% Bis/Tris NuPAGE gels in MOPS buffer under nonreducing (?1-integrin subunit) or reducing (?-actin) conditions, transferred to nitrocellulose, and the membranes incubated with either HRP-conjugated streptavidin (1:25,000) (Pierce, Rockford IL) to identify surface-labeled proteins or monoclonal anti-?-actin antibody (1:2000) (Sigma), followed by HRP-conjugated goat antimouse IgG (1:1000). Peroxidase activity was detected using chemiluminescence according to manufacturer’s instructions (Amersham).
Two-dimensional (2-D) electrophoresis
The 2-D electrophoresis was performed as previously described (36). Briefly, samples were microdialyzed for 4 h against 5 mM Tris (pH 7.0) using 6000–8000 molecular weight cutoff membranes at 4 C. Samples were then lyophilized, redissolved to 1 mg/ml total protein in SDS sample buffer, and heated in a boiling water bath for 5 min before loading 50 μl (50 μg) of each sample. Isoelectric focusing was carried out in a glass tube with 2 mm inner diameter using 2% ampholines (pH 3.5–10) (Amersham Pharmacia Biotech, Piscataway, NJ), for about 14 h at 700 V. One microgram of an isoelectric focusing internal standard, tropomyosin, was also added to each sample. This protein migrates as a doublet with a lower polypeptide spot of 33 kDa and pI 5.2. After equilibration for 10 min in buffer containing 10% glycerol, 50 mM dithiothreitol, 2.3% SDS, and 0.0625 M Tris (pH 6.8), the tube gels were sealed to the top of stacking gels on 10% acrylamide slab gels (0.75 mm thick), and SDS slab gel electrophoresis was carried out for about 4 h at 12.5 mA/gel. After slab gel electrophoresis, the gels were either dried onto filter paper or used for blotting. Gels were placed into transfer buffer [12.5 mM Tris (pH 8.8), 86 mM glycine, and 10% MeOH] and proteins transferred onto polyvinylidene difluoride membranes overnight at 200 mA and about 100 V/two gels. The membranes were then incubated for 2 h in 5% nonfat dry milk in Tris-buffered saline/0.1% Tween 20 (TBS/Tween), rinsed in TBS/Tween, and incubated with HRP-conjugated streptavidin, diluted 1:25,000 in 2% nonfat dry milk in TBS/Tween for 2 h followed by three washes in TBS/Tween. Bound HRP was detected by chemiluminescence (Amersham) and autoradiography (Eastman-Kodak, Rochester, NY).
Mass spectroscopy
Coomassie-stained protein bands were cut out and subjected to in-gel proteolytic digestion on a ProGest workstation. Briefly, ammonium carbonate was added to each sample, reduction was performed with dithiothreitol, and the samples were denatured by boiling for 5 min. Samples were allowed to cool to room temperature, and alkylation was performed with iodoacetamide. Samples were incubated at 37 C overnight in the presence of trypsin, and formic acid was added to stop the reaction. A portion of the resulting hydrolysate was used for matrix-assisted laser desorption ionization mass spectrometry (MALDI/MS) analysis as follows: samples were spotted onto a MALDI target robotically (ProMS) using ZipTips; peptides were eluted from the C18 (Zip Tip) material with matrix (-cyano-4-hydroxy cinnamic acid) prepared in 60% acetonitrile and 0.2% trifluoroacetic acid; MALDI/MS data were acquired on an Applied Biosystems (Foster City, CA) Voyager DE-STR instrument and the observed m/z values were submitted to ProFound, a proteometrics software package, for peptide mass fingerprint searching from a locally stored copy of the NCBInr database. Samples that proved inconclusive after MALDI/MS were analyzed by nano-liquid chromatography/tandem mass spectrometry on a Micromass Q-T of 2 as follows: 15 μl of hydrolysate were processed on a 75-μm C18 column at a flow-rate of 200 ml/min; nano-tandem mass spectrometry data were searched using a local copy of MASCOT (www.matrixscience.com). Detailed protocols for each of these methods can be found in the technical information section at http://www.prsproteomics.com/.
Results
Affinity chromatography of cell surface-labeled proteins from Saos-2 osteosarcoma fibroblasts with PTHrP (1–34)
Affinity chromatography has been previously used to identify transmembrane receptor proteins, such as integrins (37). We reasoned that this technique might also be suitable for the elucidation of novel binding proteins that interact with PTHrP and its various fragments. We coupled the classical PTHrP (1–34) peptide to Sepharose and incubated extracts of PTH/PTHrP receptor-positive cell-surface biotinylated proteins from Saos-2 cells with the PTHrP (1–34)-bound Sepharose. After washing, the column was eluted with 1 mg/ml PTHrP (1–34). Figure 1 demonstrates that major protein bands of about 70 and 30 kDa as well as other minor protein bands were apparent in the peptide eluates. The 8 M urea eluates, which sequentially followed those of the free peptide, contained large amounts of many proteins, including more of the 70- and 30-kDa species. Similar results were obtained with FG cell extracts (not shown). Immunoblotting of the eluates for PTH/PTHrP receptor with the antihuman antibody (clone 3D1.1, Santa Cruz Biotechnology, Santa Cruz, CA) were unsuccessful, possibly because the PTH/PTHrP receptor is known to be proteolytically cleaved when removed from the cell membrane, making the receptor epitope unrecognizable (38, 39).
FIG. 1. Interaction of cell-surface biotinylated proteins from Saos-2 extracts on PTHrP (1–34)-Sepharose. Extracts of cell-surface biotinylated proteins from Saos-2 cells were applied onto PTHrP (1–34)-Sepharose 4B, and the column was washed and eluted with 1 mg/ml PTHrP (1–34) followed by 8 M urea. The eluted fractions were analyzed on a 12% Bis/Tris NuPAGE gel in MOPS buffer under reducing conditions, transferred to nitrocellulose, and incubated with streptavidin-HRP, and proteins were visualized using chemiluminescence and autoradiography. Arrows indicate protein bands of 70 and 30 kDa. Molecular mass markers are indicated.
Affinity chromatography of surface-labeled proteins from MiaPaCa-2 pancreatic cancer cells with mid-region PTHrP (37–86)-Sepharose
After coupling PTHrP (37–86) to Sepharose, extracts of cell-surface biotinylated MiaPaCa-2 cells were chromatographed, and bound proteins were eluted with either 1 mg/ml free peptide, followed by 8 M urea (Fig. 2A), or with 20 mM EDTA, 1 M NaCl, and 8 M urea (Fig. 2B). Like PTHrP (1–34), a major protein band of about 70 kDa was also eluted with the free peptide from PTHrP (37–86)-Sepharose. The 30-kDa protein bound to PTHrP (1–34) did not appear to be eluted with free peptide on PTHrP (37–86)-Sepharose but did appear subsequently in the 8 M urea eluates. To determine the nature of the interaction between PTHrP (37–86) and the 70-kDa protein, we chromatographed extracts of cell-surface biotinylated MiaPaCa-2 proteins over PTHrP (37–86)-Sepharose and eluted with 20 mM EDTA, 1 M NaCl, and 8 M urea. This elution sequence was used to identify protein-protein interactions that were divalent cation dependent (EDTA), ionic (NaCl), and hydrophobic (urea). Figure 2B demonstrates that most of the 70-kDa protein bound PTHrP (37–86) in a divalent cation-dependent manner. Bands of 30- and 15-kDa as well as more of the 70-kDa protein were eluted with 1 M NaCl, suggesting ionic interactions. These results collectively suggest that a 70-kDa protein expressed on the cell surface of MiaPaCa-2 pancreatic cancer cells binds to PTHrP (37–86) reversibly and that the protein-protein interaction is divalent cation dependent. Mass spectroscopy analyses indicated that the 70-kDa protein bound to PTHrP (1–34) and (37–86)-Sepharose was HSP70 (not shown).
FIG. 2. Interaction of cell-surface biotinylated proteins from MiaPaCa-2 extracts on PTHrP (37–86)-Sepharose. A, Extracts of cell-surface biotinylated proteins from MiaPaCa-2 cells were applied onto PTHrP (37–86)-Sepharose 4B, and the column was washed and eluted with 1 mg/ml PTHrP (37–86) followed by 8 M urea. The proteins were visualized as described in Fig. 1. The arrow indicates a 70-kDa protein band. B, Extracts of cell-surface biotinylated proteins from MiaPaCa2 cells were chromatographed over PTHrP (37–86)-Sepharose and eluted with wash buffer containing 20 mM EDTA followed by 1 M NaCl and then 8 M urea and analyzed as described in Fig. 1.
The SPECTRA-70 algorithm predicts regions of high HSP70 binding specificity in PTHrP
To understand the nature of the interaction between HSP70 and the two different regions of PTHrP, we relied upon a recent report where HSP70 binding proteins were identified in vivo using chromatographic techniques (40). The authors proposed an algorithm (SPECTRA-70) based on a preferred nine-amino-acid sequence used for predicting proteins likely to bind HSP70. Applying that algorithm to all nine amino acid sequences in PTHrP (1–34) and (37–86), in both the forward and reverse directions, we found several amino acid sequences within each peptide that were favored with more than 95% confidence (Table 1). We also applied the SPECTRA-70 algorithm to PTHrP (109–141) and (140–173) and found that those fragments also contain amino acid sequences predicted to bind HSP70 (Table 1). PTHrP (109–141), in particular, contains a nine-amino-acid stretch that yields a predictive value nearly two times greater than the highest value observed in the studies used to create the SPECTRA-70 algorithm (40). It is also worth noting the percentage of positive hits (those yielding matrix scores > 1) present in each peptide; for PTHrP (1–34), there were six significant hits out of 52 possible peptides (11.5%); for PTHrP (37–86), there were 16 significant hits out of 84 peptides (19%); for PTHrP (109–141), there were eight significant hits out of 50 peptides (16%); and for PTHrP (140–173), there were seven significant hits out of 52 peptides (13.5%).
TABLE 1. Evaluation of the PTHrP peptide sequences used in affinity chromatography experiments using the PRESTO-70 matrix (40 ) to identify predictive HSP70 binding motifs
Membrane-associated HSP70 binds PTHrP (109–141) and (140–173) in cancer cell lines
Based on the SPECTRA-70 predictions, we coupled PTHrP (109–141) and (140–173) to Sepharose and chromatographed surface-labeled extracts of FG and DU-145 cells, respectively, over the peptide columns, eluting with either free peptide or sequentially with EDTA, NaCl, and urea. Figure 3A demonstrates that HSP70 binds PTHrP (109–141), the majority of which is elutable with free peptide. Figure 3B also shows that HSP70 interacts with PTHrP (140–173) in a divalent cation-dependent manner. Similar results were obtained with FG cancer cells and PTHrP (140–173)-Sepharose (not shown). These results are consistent with SPECTRA-70 predictions that HSP70, expressed on the cell surface in cancer cells, should bind to all four PTHrP peptides used in these studies.
FIG. 3. Interaction of cell-surface biotinylated proteins from FG cell extracts on PTHrP (109–141)-Sepharose. A, Extracts of cell-surface biotinylated proteins from FG cells were applied onto PTHrP (109–141)-Sepharose 4B, and the column was washed and eluted with 1 mg/ml PTHrP (109–141) followed by 8 M urea. The eluted fractions were analyzed as described in Fig. 1. B, Extracts of cell-surface biotinylated DU-145 cells were chromatographed over PTHrP (140–173)-Sepharose and eluted with wash buffer containing 20 mM EDTA followed by 1 M NaCl and then 8 M urea and analyzed as described in Fig. 1.
HSP70 expressed on the cell surface of cancer cells binds PTHrP
Because HSP70 is generally not expressed on the surface of normal cells (41), we reasoned that PTHrP should not bind to surface-labeled HSP70 from a spontaneously immortalized normal breast epithelial cell line, MCF10 (32). We therefore chromatographed cell-surface biotinylated proteins from MCF10 extracts over PTHrP (109–141)-Sepharose, eluting directly with 8 M urea to collect all bound proteins. Figure 4 demonstrates that very little, if any, surface-expressed (biotinylated) HSP70 was bound to the PTHrP peptide from MCF10 cells. There were several protein bands in the 70-kDA range, but none of them were present at nearly the density observed in the cancer cell line eluates (Figs. 1–3). It is important to note that all proteins were eluted directly with urea, indicating that the overall protein content obtained with MCF10 cell extracts was far below that observed with the cancer cell extracts. BSA-Sepharose incubated with biotinylated proteins from FG cell extracts and mock-coupled Sepharose incubated with DU-145 cell extracts also yielded essentially no surface-expressed HSP70 (Fig. 4).
FIG. 4. Interaction of cell-surface biotinylated MCF10 normal breast epithelial cell extracts on PTHrP (109–141)-Sepharose, DU-145 extracts on uncoupled Sepharose 4B, and FG cell extracts on BSA-Sepharose. MCF10 extracts were applied onto PTHrP (109–141)-Sepharose and eluted with 8 M urea. Extracts of cell-surface biotinylated DU-145 cells were applied onto a mock-coupled Sepharose 4B, and the column was washed and eluted as described above. Extracts of cell-surface biotinylated FG cells were applied onto BSA-Sepharose, and the column was washed and eluted as described above. The eluted fractions from each column were analyzed as described in Fig. 1.
Immunoblotting of peak fractions from all of the affinity chromatography experiments with a polyclonal antibody raised against HSP70 confirmed the presence of HSP70 in all peak fractions from the PTHrP peptides tested but not from the BSA-Sepharose column (Fig. 5A). Interestingly, HSP70 from MCF10 cell extracts is abundantly present in eluates from PTHrP (109–141) affinity chromatography experiments (lane 7), although very little is actually surface expressed (Fig. 4). Differences in levels of HSP70 from lane to lane can be attributed to different elution regimes. HSP70 is present at high levels in the peak fractions from FG cells on PTHrP (1–34) (lane 1), MiaPaCa-2 cells on PTHrP (37–86) (lane 3), FG cells on PTHrP (140–173) (lane 6), and MCF10 cells on PTHrP (109–141), which were eluted directly with 8 M urea in six to 10 fractions. HSP70 is present in relatively lower amounts in peak fractions using longer elution regimes of 18 fractions, using free peptide followed by 8 M urea as shown for Saos-2 cells on PTHrP (1–34) (lane 1) and FG cells on PTHrP (109–141) (lane 5) or EDTA followed by NaCl followed by 8 M urea as shown for MiaPaCa-2 cells on PTHrP (37–86) (lane 3). To determine whether the increased yield of surface-expressed HSP70 found in the eluates from our cancer cell extracts on PTHrP peptides relative to MCF10 cells was a result of lower overall expression of HSP70 in the normal cell line, we conducted immunoblotting studies with both the soluble and insoluble fractions from each of the cell lines used in these studies. Figure 5B demonstrates that HSP70 is expressed essentially equally in all the cell lines.
FIG. 5. HSP70 immunoblotting of peak fractions from affinity chromatography experiments and HSP70 expression in soluble and insoluble fractions from cell lysates. A, Peak fractions from affinity chromatography experiments were loaded volumetrically (16.5 μl/well) and electrophoresed on 12% Bis/Tris gels under reducing conditions in MOPS buffer and transferred to nitrocellulose, and the membranes were subsequently immunoblotted with polyclonal anti-HSP70 as described in Materials and Methods. Lane 1, FG cell extracts on PTHrP (1–34)-Sepharose, eluted with 8 M urea; lane 2, Saos-2 cell extracts on PTHrP (1–34)-Sepharose, eluted with free peptide; lane 3, MiaPaCa-2 cell extracts on PTHrP (37–86)-Sepharose, eluted with 8 M urea; lane 4, MiaPaCa-2 cell extracts on PTHrP (37–86)-Sepharose, eluted with 20 mM EDTA; lane 5, FG cell extracts on PTHrP (109–141)-Sepharose, eluted with free peptide; lane 6, FG cell extracts on PTHrP (140–173)-Sepharose, eluted with 8 M urea; lane 7, MCF10 cell extracts on PTHrP (109–141)-Sepharose, eluted with 8 M urea; lane 8, FG cell extracts on BSA-Sepharose, eluted with 8 M urea. B, Whole-cell lysates were prepared as described in Materials and Methods for each cell line. The soluble (S) and insoluble (I) fractions (5 μg protein/lane) were analyzed by immunoblotting for HSP70 and ?-actin expression as described in Materials and Methods. C, IP analysis of unbound fractions of cell lysates after PTHrP affinity chromatography were conducted as described in Materials and Methods. After IP with clone P5D2, directed against the surface-expressed ?1-integrin subunit, streptavidin-HRP identified the ?1-integrin subunit with its associated -subunits as surface labeled. After IP with clone AC-15, directed against ?-actin, a known intracellular protein, streptavidin-HRP revealed no biotinylated protein, even though subsequent immunoblotting with the ?-actin monoclonal antibody revealed abundant ?-actin in the immunoprecipitates. IB, Immunoblot.
To confirm the specificity of the biotinylation procedure for cell surface-expressed proteins, we conducted IP analyses with the unbound fractions of cell lysates after affinity chromatography for known surface-expressed and intracellular proteins. As shown in Fig. 5C, the ?1-integrin subunit, along with its associated -subunits, were identified by streptavidin-HRP in all cell lysates, indicating surface expression, which is well established (42). IP of the unbound fractions of chromatographed cell lysates for ?-actin, a well-known and abundant intracellular protein, demonstrated essentially no reaction after streptavidin-HRP treatment, even after prolonged exposure of the treated blot to film, indicating that no ?-actin is biotinylated. Subsequent immunoblot analyses of the ?-actin IP with the anti-?-actin antibody demonstrated abundant protein in the immunoprecipitates from all cell lines. Taken together with the affinity chromatography and HSP70 immunoblotting results, these results collectively indicate that the biotinylation procedure effectively labels cell-surface and not intracellular proteins and that, in cancer cell lines, HSP70 is surface expressed.
The 2-D electrophoresis of 8 M urea eluates from PTHrP (1-34, 109-141, and 140-173) and mock-coupled Sepharose confirms the presence of HSP70 in the PTHrP peptide eluates and its absence in the mock-coupled Sepharose eluates (Fig. 6, A and B). Because these 2-D gels were loaded quantitatively, Fig. 6B demonstrates further that PTHrP (109–141) eluates contain the largest amount of surface-labeled proteins. As an additional control, we isolated plasma membranes from MiaPaCa2 cells after biotinylation by phase separation and chromatographed protein extracts from these membrane preparations over PTHrP (37–86)-Sepharose. The 8 M urea eluates were indistinguishable from elution profiles obtained using surface-labeled total cell lysates (not shown). Taken together, these results confirm the specificity of the interaction between surface-expressed HSP70 and PTHrP in cancer but not in normal cells.
FIG. 6. Two-dimensional electrophoresis of total protein (A) and cell-surface-expressed protein (B) eluted from the indicated PTHrP fragments using extracts from the indicated cancer cell lines. Fifty micrograms total protein from 8 M urea eluates from the indicated affinity chromatography experiments were quantitatively loaded and electrophoresed in two dimensions. Gels were then stained with Coomassie blue to visualize total protein or transferred to polyvinylidene difluoride membranes, incubated with HRP-conjugated streptavidin at 1:25,000, and developed with enhanced chemiluminescence to visualize biotinylated, cell-surface-expressed proteins. Arrowheads indicate the internal standard, tropomyosin, pI 5.2 and molecular mass 32.7 kDa. Arrows indicate the location of the 70-kDa HSP70 protein.
Blocking HSP70 increases PTHrP secretion and decreases proliferation in cancer cell lines
To begin to understand the potential significance of the interaction between PTHrP and HSP70 at the cancer cell surface, we conducted experiments with all the cell lines used in affinity chromatography, grown in the presence of a polyclonal antibody raised against HSP70 or irrelevant rabbit IgG. Figure 7A demonstrates that blocking the HSP70 interaction with PTHrP resulted in statistically significant increases in the expression of secreted PTHrP in all cell lines tested except MiaPaCa-2, which expresses very low amounts of PTHrP (12). This increase in PTHrP secretion was associated with a trend toward decreased total protein from 72-h cultures grown in the presence of anti-HSP70 in all the cell lines compared with the irrelevant control antibody (Fig. 7B), although statistical significance was achieved only in the pancreatic cancer cell lines, MiaPaCa-2 and FG. Studies conducted in the presence of anti-HSP70 or rabbit IgG also demonstrated a differential regulation of proliferation, with three cell lines (MiaPaCa-2, FG, and MCF10 cells) trending toward decreased proliferation and two cell lines (Saos-2 and DU-145) trending toward increased proliferation (Fig. 7C). The FG and Saos-2 results were statistically significant. Together, these data are consistent with the hypothesis that the interaction of PTHrP with HSP70 at the cancer cell surface is biologically relevant.
FIG. 7. The effect of anti-HSP70 polyclonal antibody on PTHrP secretion, total protein, and proliferation after 72 h in culture in the presence of 5% serum. In each well of a six-well tissue culture plate, 2.5 x 105 cells were cultured in the presence of polyclonal anti-HSP70 antibody at 25 μg/ml (black bars) or an irrelevant rabbit IgG at 25 μg/ml (gray bars) for 72 h. A, Media from triplicate wells were harvested and assayed for PTHrP (1–34) by RIA to assess secretion. Data presented represent the mean ± SEM from three independent experiments; 100% = 2.4 pg/μg protein. P values were derived from two-tailed Student’s t test. *, P < 0.05. B, Cell lysates were prepared from the experiments described in A and assayed for total protein using Bio-Rad protein assay according to the manufacturer’s instructions. Black bars, anti-HSP70; gray bars, irrelevant rabbit IgG. Data presented represent the mean ± SEM from three independent experiments. *, P < 0.05. C, Proliferation assays were conducted in the presence of polyclonal anti-HSP70 antibody at 25 μg/ml or irrelevant rabbit IgG at 25 μg/ml using the CellTiter 96 AQueous One Solution cell proliferation assay according to the manufacturer’s instructions (Promega). The absorbance at 450 nm at time zero is subtracted from the absorbance at 450 nm at 72 h and expressed as percent maximum (%Max) considering all five cell lines. Data presented represent the mean ± SEM from three independent experiments. *, P < 0.05.
Discussion
In the present study, we demonstrate, using affinity chromatography, that HSP70 is a major cell-surface protein expressed by cancer cells that binds to PTHrP, regardless of the peptide fragment used. This interaction is specific to cancer cells that express HSP70 on their cell surface, because essentially no surface-expressed HSP70 was found bound to PTHrP peptides from MCF-10 normal breast epithelial cell extracts. This interaction appears to be partially reversible, because free peptide can elute HSP70. This protein-protein interaction also appears to be partially divalent cation dependent, as evidenced by elution of HSP70 with EDTA. Some of the PTHrP-bound HSP70 also consists of other ionic as well as hydrophobic interactions, as evidenced by its appearance in the high salt and urea eluates. This is not surprising in light of our results, which demonstrated that an average of 15% of the PTHrP sequences used in these studies, which span the entire PTHrP peptide without the nuclear localization sequence from amino acids 88–106 (43), contain high predictive SPECTRA-70 matrix scores for HSP70 binding. It is reasonable to expect that some of these binding interactions between PTHrP and HSP70 could be divalent cation dependent, whereas others could be ionic or hydrophobic.
HSP70 has recently been shown in a number of studies to be among a set of chaperone proteins expressed on the cell surface of cancer cells, including neuroblastoma, lung, and colon adenocarcinoma, acute lymphoblastic leukemia, and ovarian tumor cell lines (44, 45, 46, 47, 48, 49) as well as freshly isolated human biopsy material of colorectal, lung, neuronal, and pancreatic carcinomas, liver metastases, and leukemic blasts of patients with acute myelogenous leukemia (50). Those observations are extended by our studies, where cell surface-expressed HSP70 from Saos-2 osteosarcoma, FG and MiaPaCa-2 pancreatic adenocarcinoma, and DU-145 prostate cancer cells bound PTHrP. Even when membrane fractions were isolated before affinity chromatography, HSP70 was still the major cell-surface biotinylated protein found bound to PTHrP. And even though MCF10 cells express significant amounts of HSP70 that binds to PTHrP peptides, essentially no surface-expressed HSP70 was recovered in the eluates. The 2-D electrophoresis and immunoblotting confirmed the presence of surface-labeled HSP70 in the PTHrP peptide eluates compared with BSA-Sepharose eluates, where HSP70 was essentially absent. IP analyses further validated the cell-surface specificity of the biotinylation procedure and our HSP70 observations, identifying known surface-expressed proteins, the integrins, as biotinylated and identifying ?-actin as not biotinylated and thus not surface-expressed, which is well established.
This novel interaction between PTHrP and HSP70 at the cell surface in cancer cell lines appears to be biologically relevant, as evidenced by increased PTHrP secretion and decreased cancer cell growth in the presence of polyclonal anti-HSP70 antibody. The function of HSP70 inside of normal cells, and at the cancer cell surface in particular, appears to be related to growth and, in particular, protection from apoptosis (51). PTHrP shares in these growth-regulating and antiapoptotic functions (15). Although speculative, these results suggest that HSP70 expressed at the cell surface is somehow involved in the internalization of secreted PTHrP, which results in an overall growth-promoting or cytoprotective effect. Although not usually found on the cell surface, the complexing of HSP70 with other proteins intracellularly, leading to cytoprotection in cancer, has been shown (51, 52). For example, inducible HSP70 forms complexes with Kruppel-like factor 6 and inducible nitric oxide synthase, resulting in cytoprotection in male SwissWebster mice subjected to 40% hemorrhage.
It is possible that HSP70, binding at the cell surface, acts secondarily in the presentation of PTHrP to its cognate receptor or to an as yet undefined receptor. Alternatively, HSP70 could act independently in PTHrP peptide transport from the cell surface into the cell. Mechanisms are obviously in place to traffic HSP70 and its bound peptides to the cell surface. It seems reasonable to expect that those same mechanisms could work in reverse, transporting HSP70 and its bound peptides into the cell as well. A relationship between PTHrP and stress-related genes, including HSP90, which is found bound to PTHrP in our studies, is not unprecedented (53). In those studies, the authors demonstrated that PTHrP (67–86) regulates the expression of stress proteins in breast cancer cells, inducing modifications in urokinase-plasminogen activator and matrix metalloproteinase-1 expression, and resulting in increased tumor cell invasiveness. The mechanism by which HSP70 and PTHrP interact at the cell surface to achieve these growth-promoting effects is currently under investigation.
It is interesting to note that HSP70 and G protein-coupled receptors, of which the PTH/PTHrP receptor is a member, have been shown to be enriched in lipid rafts (54, 55). Lipid raft theory suggests that cholesterol and sphingolipids partition away from other membrane lipids and that these lipid rafts are important in processes such as signaling and vesicle traffic (56). With that in mind, it is well established that HSP70 has the capacity for multiple protein binding. Some of the proteins identified by mass spectroscopy in vivo that bind to HSP70 include keratin, actin, HSP60, and elongation factors 1 and 2 (40). Although not surface expressed, it is remarkable that mass spectroscopy analyses of PTHrP (140–173) urea eluates from DU-145 cell extracts identified a complex mixture of proteins, including keratin, actin, elongation factors 1 and 2, and HSP60 and -90 as well as other proteins (data not shown). In support of these observations, PTHrP has been previously shown to bind to keratin, possibly affecting keratinocyte growth and differentiation (57). In addition, HSP70 and peroxiredoxin-2, both found bound to PTHrP in our studies (not shown), are abundantly expressed in human ductal carcinomas of the breast (58).
The divalent cation-dependent nature of the interaction of HSP70 with PTHrP is also consistent with the known binding of divalent cation binding proteins such as zinc finger proteins and calpain to HSP70 in vivo (40). Although it seems probable that some of these proteins are actually bound to PTHrP by association with HSP70 or another protein, these studies collectively suggest that in cancer, a complex of proteins, including HSP70 and PTHrP, could be enriched in lipid rafts at the cell surface and that this complex is important in the regulation of cell growth.
The HSP70 binding region in various proteins is generally at least nine amino acids in length and contains both hydrophobic and basic amino acids (40). It is also interspersed with at least two hydrophilic amino acids and usually contains acidic residues. It often contains a five-amino-acid core, with hydrophobic residues, a proline or glycine at positions 2, 3, or 4, and at least one hydrophilic residue. The five-amino-acid core generally does not contain basic amino acids but often does contain one or more acidic amino acids. Our studies indicate that PTHrP contains multiple HSP70 binding regions as predicted by the SPECTRA-70 matrix scores and as demonstrated by our affinity chromatography results. It is worth emphasizing that a strong correlation exists between the very large SPECTRA-70 matrix score for PTHrP (109–141) and the amount of HSP70 and associated cell surface-expressed proteins present in our quantitative 2-D gel analysis compared with the other PTHrP peptides.
In summary, our studies demonstrate novel protein-protein interactions between HSP70, expressed on the surface of cancer cells, and PTHrP. This binding is partially reversible and consists of divalent cation-dependent, ionic, and hydrophobic interactions. Our affinity chromatography studies also identify other PTHrP binding proteins previously shown to bind HSP70 in vivo, and include proteins shown to be enriched in lipid rafts. Functional studies demonstrating growth-related effects as a result of the interaction of PTHrP and HSP70 suggest a biological relevance to these protein-protein interactions. Taken together, these results have important implications for our understanding of the functions of PTHrP in both malignant and normal cells that are not currently attributable to the PTH/PTHrP receptor.
Acknowledgments
We thank Drs. Erkki Ruoslahti and Eva Engvall for critical readings of the manuscript; Kendricks Labs, Inc. (Madison, WI) for expert 2-D electrophoresis services; and Proteomic Research Services, Inc. (Ann Arbor, MI) and Scripps Institute (La Jolla, CA) for expert mass spectroscopy services.
References
Nissenson RA 2000 Parathyroid hormone-related protein. Rev Endocr Metab Disord 1:343–352
Grzesiak JJ, Smith KC, Burton DW, Deftos LJ, Bouvet M 2005 GSK3 and PKB/Akt are associated with integrin-mediated regulation of PTHrP, IL-6 and IL-8 expression in FG pancreatic cancer cells. Int J Cancer 114:522–530
Grzesiak JJ, Clopton P, Chalberg C, Smith K, Burton DW, Silletti S, Moossa AR, Deftos LJ, Bouvet M 2004 The extracellular matrix differentially regulates the expression of PTHrP and the PTH/PTHrP receptor in FG pancreatic cancer cells. Pancreas 29:85–92
Philbrick WM, Wysolmerski JJ, Galbraith S, Holt E, Orloff J J, Yang KH, Vasavada RC, Weir EC, Broadus AE, Stewart AF 1996 Defining the roles of parathyroid hormone-related protein in normal physiology. Physiol Rev 76:127–173
Tachibana I, Nakano S, Akiyama T, Ogami Y, Tabaru A, Otsuki M 1994 Parathyroid hormone-related protein mediates hypercalcemia in an exocrine pancreatic cancer. Am J Gastroenterol 89:1580–1581
Brandt DW, Wachsman W, Deftos LJ 1994 Parathyroid hormone-like protein: alternative messenger RNA splicing pathways in human cancer cell lines. Cancer Res 54:850–853
Bucht E, Rong H, Pernow Y, Nordqvist AC, Eriksson E, Rankin W, von Schoultz E, Burtis WJ, Granberg B, Falkmer UG, Burton DW, Deftos LJ 1998 Parathyroid hormone-related protein in patients with primary breast cancer and eucalcemia. Cancer Res 58:4113–4116
Deftos LJ, Gazdar AF, Ikeda K, Broadus AE 1989 The parathyroid hormone-related protein associated with malignancy is secreted by neuroendocrine tumors. Mol Endocrinol 3:503–508
Deftos LJ, Burton DW, Brandt DW 1993 Parathyroid hormone-like protein is a secretory product of atrial myocytes. J Clin Invest 92:727–735
Drucker DJ, Asa SL, Henderson J, Goltzman D 1989 The parathyroid hormone-like peptide gene is expressed in the normal and neoplastic human endocrine pancreas. Mol Endocrinol 3:1589–1595
Alipov GK, Ito M, Nakashima M, Ikeda Y, Nakayama T, Ohtsuru A, Yamashita S, Sekine I 1997 Expression of parathyroid hormone-related peptide (PTHrP) in gastric tumours. J Pathol 182:174–179
Bouvet M, Nardin SR, Burton DW, Behling C, Carethers JM, Moossa AR, Deftos LJ 2001 Human pancreatic adenocarcinomas express parathyroid hormone-related protein. J Clin Endocrinol Metab 86:310–316
Bouvet M, Nardin SR, Burton DW, Lee NC, Yang M, Wang X, Baranov E, Behling C, Moossa AR, Hoffman RM, Deftos LJ 2002 Parathyroid hormone-related protein as a novel tumor marker in pancreatic adenocarcinoma. Pancreas 24:284–290
Bakre MM, Zhu Y, Yin H, Burton DW, Terkeltaub R, Deftos LJ, Varner JA 2002 Parathyroid hormone-related peptide is a naturally occurring, protein kinase A-dependent angiogenesis inhibitor. Nat Med 8:995–1003
Henderson JE, Amizuka N, Warshawsky H, Biasotto D, Lanske BM, Goltzman D, Karaplis AC 1995 Nucleolar localization of parathyroid hormone-related peptide enhances survival of chondrocytes under conditions that promote apoptotic cell death. Mol Cell Biol 15:4064–4075
Goltzman D 1999 Interactions of PTH and PTHrP with the PTH/PTHrP receptor and with downstream signaling pathways: exceptions that provide the rules. J Bone Miner Res 14:173–177
Strewler GJ 2000 The physiology of parathyroid hormone-related protein. N Engl J Med 342:177–185[Free Full Text]
Kronenberg HM, Lanske B, Kovacs CS, Chung UI, Lee K, Segre GV, Schipani E, Juppner H 1998 Functional analysis of the PTH/PTHrP network of ligands and receptors. Recent Prog Horm Res 53:283–301; discussion 301–303
Segre GV 1996 Receptors for parathyroid hormone (PTH) and PTH-related peptide. In: Bilezikian JP, Raisy LG, Rodan FA, eds. Principles of bone biology. New York: Academic Press; 389–406
Clemens TL, Cormier S, Eichinger A, Endlich K, Fiaschi-Taesch N, Fischer E, Friedman PA, Karaplis AC, Massfelder T, Rossert J, Schluter KD, Silve C, Stewart AF, Takane K, Helwig JJ 2001 Parathyroid hormone-related protein and its receptors: nuclear functions and roles in the renal and cardiovascular systems, the placental trophoblasts and the pancreatic islets. Br J Pharmacol 134:1113–1136
Lanske B, Amling M, Neff L, Guiducci J, Baron R, Kronenberg HM 1999 Ablation of the PTHrP gene or the PTH/PTHrP receptor gene leads to distinct abnormalities in bone development. J Clin Invest 104:399–407
Strid H, Care A, Jansson T, Powell T 2002 Parathyroid hormone-related peptide (38–94) amide stimulates ATP-dependent calcium transport in the basal plasma membrane of the human syncytiotrophoblast. J Endocrinol 175:517–524
Cornish J, Callon KE, Nicholson GC, Reid IR 1997 Parathyroid hormone-related protein-(107–139) inhibits bone resorption in vivo. Endocrinology 138:1299–1304
Cornish J, Callon KE, Lin C, Xiao C, Moseley JM, Reid IR 1999 Stimulation of osteoblast proliferation by C-terminal fragments of parathyroid hormone-related protein. J Bone Miner Res 14:915–922
De Miguel F, Martinez-Fernandez P, Guillen C, Valin A, Rodrigo A, Martinez ME, Esbrit P 1999 Parathyroid hormone-related protein (107–139) stimulates interleukin-6 expression in human osteoblastic cells. J Am Soc Nephrol 10:796–803
Hastings RH, Araiza F, Burton DW, Zhang L, Bedley M, Deftos LJ 2003 Parathyroid hormone-related protein ameliorates death receptor-mediated apoptosis in lung cancer cells. Am J Physiol Cell Physiol 285:C1429–C1436
Goomer RS, Johnson KA, Burton DW, Amiel D, Maris TM, Gurjal A, Deftos LJ, Terkeltaub R 2000 The tetrabasic KKKK(147–150) motif determines intracrine regulatory effects of PthrP 1–173 on chondrocyte PPi metabolism and matrix synthesis. Endocrinology 141:4613–4622
Jonsson KB, John MR, Gensure RC, Gardella TJ, Juppner H 2001 Tuberoinfundibular peptide 39 binds to the parathyroid hormone (PTH)/PTH-related peptide receptor, but functions as an antagonist. Endocrinology 142:704–709
Ahmadpour OA, Burton DW, Tu S, Deftos LJ 2002 Regulation of OPG/RANK/RANKL by PTHrP peptides in human prostate cancer cells. J Bone Miner Res 17(S):286
Burton DW, Tu S, Gujral A, Patel JA, Terkeltaub RA, Rhim JS, Deftos LJ 1999 Intracellular (intracrine), non-PTH/PTHrP receptor mediated effects of PTHrP on prostate cancer cell growth. J Bone Miner Res 14(S):290
Maheshwari VV, Burton DW, Hastings RH, Aguilera JA, Pardo FS, Deftos LJ 2003 Apoptotic effects of parathyroid hormone related protein in prostate cancer cells. J Bone Miner Res 18(S):119 (Abstract)
Sloane BF, Moin K, Sameni M, Tait LR, Rozhin J, Ziegler G 1994 Membrane association of cathepsin B can be induced by transfection of human breast epithelial cells with c-Ha-ras oncogene. J Cell Sci 107:373–384
Lenstra JA, Bloemendal H 1983 Topography of the total protein population from cultured cells upon fractionation by chemical extractions. Eur J Biochem 135:413–423
Yeung SC, Eton O, Burton DW, Deftos LJ, Vassilopoulou-Sellin R, Gagel RF 1998 Hypercalcemia due to parathyroid hormone-related protein secretion by melanoma. Horm Res 49:288–291
Carron JA, Fraser WD, Gallagher JA 1997 PTHrP and the PTH/PTHrP receptor are co-expressed in human breast and colon tumours. Br J Cancer 76:1095–1098
O’Farrell PH 1975 High resolution two-dimensional electrophoresis of proteins. J Biol Chem 250:4007–4021
Ruoslahti E, Pierschbacher MD 1987 New perspectives in cell adhesion: RGD and integrins. Science 238:491–497
Gentili C, Morelli S, de Boland AR 2003 Characterization of PTH/PTHrP receptor in rat duodenum: effects of ageing. J Cell Biochem 88:1157–1167
Karpf DB, Arnaud CD, King K, Bambino T, Winer J, Nyiredy K, Nissenson RA 1987 The canine renal parathyroid hormone receptor is a glycoprotein: characterization and partial purification. Biochemistry 26:7825–7833
Grossmann ME, Madden BJ, Gao F, Pang YP, Carpenter JE, McCormick D, Young CY 2004 Proteomics shows Hsp70 does not bind peptide sequences indiscriminately in vivo. Exp Cell Res 297:108–117
Bhattacharyya T, Karnezis AN, Murphy SP, Hoang T, Freeman BC, Phillips B, Morimoto RI 1995 Cloning and subcellular localization of human mitochondrial hsp70. J Biol Chem 270:1705–1710
Ruoslahti E, Noble NA, Kagami S, Border WA 1994 Integrins. Kidney Int Suppl 44:S17–S22
Fiaschi-Taesch NM, Stewart AF 2003 Parathyroid hormone-related protein as an intracrine factor: trafficking mechanisms and functional consequences. Endocrinology 144:407–411
Shin BK, Wang H, Yim AM, Le Naour F, Brichory F, Jang JH, Zhao R, Puravs E, Tra J, Michael CW, Misek DE, Hanash SM 2003 Global profiling of the cell surface proteome of cancer cells uncovers an abundance of proteins with chaperone function. J Biol Chem 278:7607–7616
Multhoff G, Botzler C, Wiesnet M, Muller E, Meier T, Wilmanns W, Issels RD 1995 A stress-inducible 72-kDa heat-shock protein (HSP72) is expressed on the surface of human tumor cells, but not on normal cells. Int J Cancer 61:272–279
Botzler C, Schmidt J, Luz A, Jennen L, Issels R, Multhoff G 1998 Differential Hsp70 plasma-membrane expression on primary human tumors and metastases in mice with severe combined immunodeficiency. Int J Cancer 77:942–948
Okazaki Y, Ohno H, Takase K, Ochiai T, Saito T 2000 Cell surface expression of calnexin, a molecular chaperone in the endoplasmic reticulum. J Biol Chem 275:35751–35758
White TK, Zhu Q, Tanzer ML 1995 Cell surface calreticulin is a putative mannoside lectin which triggers mouse melanoma cell spreading. J Biol Chem 270:15926–15929
Xiao G, Chung TF, Fine RE, Johnson RJ 1999 Calreticulin is transported to the surface of NG108–15 cells where it forms surface patches and is partially degraded in an acidic compartment. J Neurosci Res 58:652–662
Hantschel M, Pfister K, Jordan A, Scholz R, Andreesen R, Schmitz G, Schmetzer H, Hiddemann W, Multhoff G 2000 Hsp70 plasma membrane expression on primary tumor biopsy material and bone marrow of leukemic patients. Cell Stress Chaperones 5:438–442
Mosser DD, Morimoto RI 2004 Molecular chaperones and the stress of oncogenesis. Oncogene 23:2907–2918
Kiang JG, Bowman PD, Wu BW, Hampton N, Kiang AG, Zhao B, Juang YT, Atkins JL, Tsokos GC 2004 Geldanamycin treatment inhibits hemorrhage-induced increases in KLF6 and iNOS expression in unresuscitated mouse organs: role of inducible HSP70. J Appl Physiol 97:564–569
Luparello C, Sirchia R, Pupello D 2003 PTHrP [67–86] regulates the expression of stress proteins in breast cancer cells inducing modifications in urokinase-plasminogen activator and MMP-1 expression. J Cell Sci 116:2421–2430
Chini B, Parenti M 2004 G-protein coupled receptors in lipid rafts and caveolae: how, when and why do they go there? J Mol Endocrinol 32:325–338
Broquet AH, Thomas G, Masliah J, Trugnan G, Bachelet M 2003 Expression of the molecular chaperone Hsp70 in detergent-resistant microdomains correlates with its membrane delivery and release. J Biol Chem 278:21601–21606
Foster LJ, De Hoog CL, Mann M 2003 Unbiased quantitative proteomics of lipid rafts reveals high specificity for signaling factors. Proc Natl Acad Sci USA 100:5813–5818
Seidel J, Zabel M, Kasprzak A 2002 PTHrP and cytokeratins in human epidermis. Folia Histochem Cytobiol 40:209–210
Somiari RI, Sullivan A, Russell S, Somiari S, Hu H, Jordan R, George A, Katenhusen R, Buchowiecka A, Arciero C, Brzeski H, Hooke J, Shriver C 2003 High-throughput proteomic analysis of human infiltrating ductal carcinoma of the breast. Proteomics 3:1863–1873(John J. Grzesiak, Kathy C)
Address all correspondence and requests for reprints to: Michael Bouvet, M.D., Department of Surgery (112-E), University of California, Veterans Affairs San Diego Healthcare System, 3350 La Jolla Village Drive, San Diego, California 92161. E-mail: mbouvet@ucsd.edu.
Abstract
Recent studies have shown that the functions of PTH-related protein (PTHrP) and its derived peptides cannot be attributed solely to PTH/PTHrP receptor binding. The present study focused on the identification of other proteins that might bind PTHrP at the cell surface. Using affinity chromatography, we applied extracts of cell-surface biotinylated proteins from cancer and normal cell lines over Sepharose beads coupled with different PTHrP-derived peptides. Elution with the corresponding free peptide revealed a major protein of about 70 kDa that was present in all of the PTHrP peptide eluates from cancer cell extracts but not from normal breast cell extracts. Mass spectroscopy analysis and immunoblotting identified this PTHrP-binding protein as heat shock protein-70 (HSP70). Using a recently published algorithm that predicts HSP70 binding sites within proteins, we found that all four PTHrP peptides used in these studies contain amino acid motifs with high probabilities for HSP70 binding in vivo. Cell culture studies in the presence of a polyclonal anti-HSP70 antibody demonstrated increased PTHrP secretion, decreased total cellular protein, and differentially regulated proliferation. Taken together, these studies demonstrate a novel and biologically relevant interaction between cell surface-expressed HSP70 and PTHrP in cancer.
Introduction
PTH-RELATED PROTEIN (PTHrP) is an oncofetal protein originally identified as a product of breast and lung cancer cells that produced humoral hypercalcemia of malignancy (1). Since its initial discovery, PTHrP has been shown to be produced by many normal and malignant cells, including those not typically associated with hypercalcemia of malignancy (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13). PTHrP is known to be a critical regulator of cellular and organ growth, development, migration, angiogenesis, differentiation and survival, and epithelial calcium ion transport (4, 14). In particular, regulation of apoptosis appears to be a key mechanism whereby PTHrP mediates growth (15).
The PTHrP gene results in three mature isoforms generated through alternative mRNA splicing in humans: PTHrP 1–141, a truncated 1-to 139-residue form, and a human-specific 1- to 173-residue form (7). In addition to mRNA splicing, proteolytic processing of PTHrP into peptides appears to be an important regulatory mechanism that could potentially result in the generation of greater than 90 unique peptides (7).
The biological effects of PTHrP are mediated, at least in part, by a member of the seven membrane-spanning and G protein-coupled cell surface receptors, the PTH/PTHrP receptor (16, 17, 18, 19). The PTH/PTHrP receptor has been shown to exert most of its biological effects by binding to the 1–34 amino-terminal fragment of PTHrP (1). In addition to its amino terminus, there is accumulating and compelling evidence that mid-region and carboxyl-terminal forms of PTHrP, generated through processing of the native isoforms, also exert biological effects (1, 16, 17, 20). For example, it has been shown, using PTHrP and PTH/PTHrP receptor knockout and double-knockout mice, that PTHrP gene products inhibit vascular invasion during endochondral bone formation in mice lacking the PTH/PTHrP receptor (21); PTHrP (38–94) stimulates ATP-dependent calcium transport in the basal plasma membrane of the human syncytiotrophoblast (22); synthetic PTHrP (107–139) has been shown to inhibit bone resorption, stimulate osteoblast proliferation, and stimulate IL-6 expression in osteoblasts (23, 24, 25); PTHrP (140–173) has been shown to exert a protective effect from apoptosis in BEN lung cancer cells (26); and finally, the tetrabasic KKKK, PTHrP (147–150) motif regulates collagen synthesis (27). These studies suggest the existence of receptors to these other regions of PTHrP, although none have been identified.
In the present study, we identified surface-expressed heat shock protein-70 (HSP70) as a major PTHrP binding protein from the cancer cell lines. A recently published algorithm that predicts HSP70 binding sites within proteins demonstrates that all four PTHrP peptides used in these studies contain amino acid motifs with high probabilities for HSP70 binding in vivo. Along with HSP70 antibody inhibition studies that resulted in increased PTHrP secretion, decreased total cell protein, and differentially regulated proliferation, our findings demonstrate a novel interaction between PTHrP and surface-expressed HSP70 in cancer cells with cytoprotective consequences.
Materials and Methods
Cell surface biotinylation
Saos-2 osteosarcoma, FG and MiaPaCa2 pancreatic adenocarcinoma, and DU-145 prostate cancer cell lines were cultured in DMEM or RPMI 1640 medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (Invitrogen). All of these cell lines have been previously shown to be PTH/PTHrP receptor positive (3, 28) and responsive to the PTHrP peptides used in these studies (2, 3, 29, 30, 31). MCF10 spontaneously immortalized normal breast epithelial cells were cultured in DMEM and Ham’s F-12 mixture (1:1), containing 5% equine serum and epidermal growth factor as previously described (32). All cell lines were cultured in a humidified atmosphere of 5% CO2 at 37 C. Cultures were harvested at approximately 70% confluency by the addition of 4 mM EDTA in PBS and incubation at 37 C for about 15 min. Detached cells were collected by pipetting and washed three times with PBS. Cells were biotinylated by adding 0.5 mg EZ-Link Sulfo-NHS-LC-Biotin (Pierce Biotechnology, Rockford, IL)/25 x 106 cells, suspended in PBS, and incubated 30 min at 37 C followed by extensive washing with PBS. Biotinylated cell pellets were frozen at –20 C until ready for further processing. Cell viability after the biotinylation procedure was greater than 91% as determined by trypan blue exclusion.
In some experiments, membrane fractions were harvested after biotinylation using Mem-PER eukaryotic membrane protein extraction kit according to the manufacturer’s instructions (Pierce Biotechnology, Inc., Rockford, IL). This method uses aqueous two-phase partitioning as previously described (33).
Extract preparation
Biotinylated cells were extracted in the following ice-cold buffer: lactated Ringer’s buffer (each 100 ml contains 600 mg NaCl, 310 mg sodium lactate, 30 mg KCl, and 20 mg CaCl2, pH 6.5) supplemented with 100 mM n-octyl-?-D-glucopyranoside (Calbiochem, La Jolla, CA), 1 mM MgCl2, and 100 μM phenylmethylsulfonyl fluoride. One milliliter of lysis buffer was added to 75 x 106 biotinylated cells, the pellet was gently resuspended by pipetting, and the lysate was incubated for 30 min on ice with occasional pipetting. The cell lysate was centrifuged 30 min at 16,100 x g and the cleared lysate applied over the appropriate equilibrated column.
Peptides coupling to CnBr-activated Sepharose 4B
Synthetic PTHrP peptides (Bachem, King of Prussia, PA) were coupled to Sepharose according to the manufacturer’s instructions (Sigma-Aldrich, St. Louis, MO). Coupling concentrations were as follows: PTHrP (1–34)-Sepharose, 2.2 mg/ml of packed Sepharose beads; PTHrP (37–86)-Sepharose, 5.3 mg/ml; PTHrP (109–141)-Sepharose, 5.5 mg/ml; PTHrP (140–173)-Sepharose, 7.1 mg/ml; and control BSA-Sepharose, 6.8 mg/ml. The coupling procedure was also conducted according to the manufacturer’s instructions in the absence of peptide as another control.
Affinity chromatography
Cell lysates were incubated with 0.6–0.8 ml PTHrP peptide-Sepharose that was previously equilibrated with wash buffer containing lactated Ringer’s buffer (Baxter Healthcare Corp., Deerfield, IL) supplemented with 50 mM n-octyl-?-D-glucopyranoside and 1 mM MgCl2. Incubations ranged from 2 h at room temperature to overnight at 4 C. Flow-through extract was collected, and the columns were washed with at least 10 column-volumes of wash buffer. Elution was conducted with wash buffer supplemented with 1 mg/ml PTHrP peptide followed by 8 M urea. In some studies, an elution sequence was conducted using 20 mM EDTA in wash buffer, followed by 1 M NaCl, followed by 8 M urea.
Cell culture assay for PTHrP expression
Cells were seeded into six-well tissue culture plates at a density of 2.5 x 105/well in DMEM supplemented with 5% fetal bovine serum and 25 μg/ml anti-HSP70 polyclonal antibody (Chemicon, Temecula, CA). Control cultures were incubated in media described above with irrelevant rabbit IgG, also at 25 μg/ml. Cultures were grown for 72 h, and the conditioned media and cell lysates were analyzed for PTHrP as described below.
Proliferation assays
Cells were seeded into 96-well tissue culture plates at a density of 5 x 103/well under the same conditions described above. After 72 h, proliferation was determined using CellTiter 96 AQueous One Solution cell proliferation assay according to the manufacturer’s instructions (Promega, Madison, WI). This reagent is composed of a novel tetrazolium compound (methyl-p-tolyl sulfide), and an electron coupling reagent, phenazine ethosulfate. Values obtained at initial cell seeding were subtracted from the 72-h values.
RIAs
At the indicated time points, media were harvested and cell extracts were prepared by sonication of cells in lysis buffer containing 0.25 M Tris (pH 7.4), 0.25% Nonidet P-40, and 2 mM EDTA. The insoluble cell fractions were pelleted by centrifugation at 16,000 x g for 15 min, and the soluble fractions were transferred to a fresh tube. Insoluble fractions were solubilized using 8 M urea. Total cell protein from both the soluble and insoluble fractions was then measured using a modified Bradford protein assay with BSA as the standard (Bio-Rad Laboratories, Inc., Hercules, CA). PTHrP was measured by RIA, using modifications as previously described (9, 34). All samples were assayed in multiple dilutions. Intra- and interassay variations were 7 and 12%, respectively (35).
Immunoblotting
Cell lysates and affinity chromatography eluates were analyzed on 12% Bis/Tris NuPAGE gels in MOPS buffer under reducing conditions and transferred to nitrocellulose, and the membranes were incubated with either horseradish peroxidase (HRP)-conjugated streptavidin (1:25,000) or polyclonal anti-HSP70 antibody (1:200) (Chemicon) and monoclonal anti-?-actin antibody (1:1000) (Sigma Chemical Co., St. Louis, MO), followed by HRP-conjugated goat antirabbit or antimouse IgG (1:1000). Peroxidase activity was detected using chemiluminescence according to manufacturer’s instructions (Amersham Biosciences, Little Chalfont, UK).
Immunoprecipitation (IP)
To perform IP, 5 μg of monoclonal antibody directed against either the ?1-integrin subunit, clone P5D2 (Chemicon), or ?-actin, clone AC-15 (Sigma-Aldrich), were absorbed overnight at 4 C onto 25 μl packed antimouse IgG-agarose (Sigma) in IP wash buffer [50 mM Tris-HCl (pH 7.5), 0.5 M NaCl, 1 mM CaCl2, 1 mM MgCl2, and 0.1%Tween 20]. After washing, the antibody-bound agarose beads were incubated with 200 μl of the unbound fractions (flow-through), from cell-surface biotinylated cell lysates after PTHrP affinity chromatography as follows: FG cell lysates over PTHrP (140–173)-Sepharose; MiaPaCa-2 cell lysates over PTHrP (37–86)-Sepharose; MCF10 cell lysates over PTHrP (109–141)-Sepharose; DU-145 cell lysates over PTHrP (140–173)-Sepharose; or Saos-2 cell lysates over PTHrP (1–34)-Sepharose. After 3 h incubation at room temperature, immunoprecipitates were washed extensively, SDS-PAGE sample buffer was added, and samples were boiled and analyzed on 12% Bis/Tris NuPAGE gels in MOPS buffer under nonreducing (?1-integrin subunit) or reducing (?-actin) conditions, transferred to nitrocellulose, and the membranes incubated with either HRP-conjugated streptavidin (1:25,000) (Pierce, Rockford IL) to identify surface-labeled proteins or monoclonal anti-?-actin antibody (1:2000) (Sigma), followed by HRP-conjugated goat antimouse IgG (1:1000). Peroxidase activity was detected using chemiluminescence according to manufacturer’s instructions (Amersham).
Two-dimensional (2-D) electrophoresis
The 2-D electrophoresis was performed as previously described (36). Briefly, samples were microdialyzed for 4 h against 5 mM Tris (pH 7.0) using 6000–8000 molecular weight cutoff membranes at 4 C. Samples were then lyophilized, redissolved to 1 mg/ml total protein in SDS sample buffer, and heated in a boiling water bath for 5 min before loading 50 μl (50 μg) of each sample. Isoelectric focusing was carried out in a glass tube with 2 mm inner diameter using 2% ampholines (pH 3.5–10) (Amersham Pharmacia Biotech, Piscataway, NJ), for about 14 h at 700 V. One microgram of an isoelectric focusing internal standard, tropomyosin, was also added to each sample. This protein migrates as a doublet with a lower polypeptide spot of 33 kDa and pI 5.2. After equilibration for 10 min in buffer containing 10% glycerol, 50 mM dithiothreitol, 2.3% SDS, and 0.0625 M Tris (pH 6.8), the tube gels were sealed to the top of stacking gels on 10% acrylamide slab gels (0.75 mm thick), and SDS slab gel electrophoresis was carried out for about 4 h at 12.5 mA/gel. After slab gel electrophoresis, the gels were either dried onto filter paper or used for blotting. Gels were placed into transfer buffer [12.5 mM Tris (pH 8.8), 86 mM glycine, and 10% MeOH] and proteins transferred onto polyvinylidene difluoride membranes overnight at 200 mA and about 100 V/two gels. The membranes were then incubated for 2 h in 5% nonfat dry milk in Tris-buffered saline/0.1% Tween 20 (TBS/Tween), rinsed in TBS/Tween, and incubated with HRP-conjugated streptavidin, diluted 1:25,000 in 2% nonfat dry milk in TBS/Tween for 2 h followed by three washes in TBS/Tween. Bound HRP was detected by chemiluminescence (Amersham) and autoradiography (Eastman-Kodak, Rochester, NY).
Mass spectroscopy
Coomassie-stained protein bands were cut out and subjected to in-gel proteolytic digestion on a ProGest workstation. Briefly, ammonium carbonate was added to each sample, reduction was performed with dithiothreitol, and the samples were denatured by boiling for 5 min. Samples were allowed to cool to room temperature, and alkylation was performed with iodoacetamide. Samples were incubated at 37 C overnight in the presence of trypsin, and formic acid was added to stop the reaction. A portion of the resulting hydrolysate was used for matrix-assisted laser desorption ionization mass spectrometry (MALDI/MS) analysis as follows: samples were spotted onto a MALDI target robotically (ProMS) using ZipTips; peptides were eluted from the C18 (Zip Tip) material with matrix (-cyano-4-hydroxy cinnamic acid) prepared in 60% acetonitrile and 0.2% trifluoroacetic acid; MALDI/MS data were acquired on an Applied Biosystems (Foster City, CA) Voyager DE-STR instrument and the observed m/z values were submitted to ProFound, a proteometrics software package, for peptide mass fingerprint searching from a locally stored copy of the NCBInr database. Samples that proved inconclusive after MALDI/MS were analyzed by nano-liquid chromatography/tandem mass spectrometry on a Micromass Q-T of 2 as follows: 15 μl of hydrolysate were processed on a 75-μm C18 column at a flow-rate of 200 ml/min; nano-tandem mass spectrometry data were searched using a local copy of MASCOT (www.matrixscience.com). Detailed protocols for each of these methods can be found in the technical information section at http://www.prsproteomics.com/.
Results
Affinity chromatography of cell surface-labeled proteins from Saos-2 osteosarcoma fibroblasts with PTHrP (1–34)
Affinity chromatography has been previously used to identify transmembrane receptor proteins, such as integrins (37). We reasoned that this technique might also be suitable for the elucidation of novel binding proteins that interact with PTHrP and its various fragments. We coupled the classical PTHrP (1–34) peptide to Sepharose and incubated extracts of PTH/PTHrP receptor-positive cell-surface biotinylated proteins from Saos-2 cells with the PTHrP (1–34)-bound Sepharose. After washing, the column was eluted with 1 mg/ml PTHrP (1–34). Figure 1 demonstrates that major protein bands of about 70 and 30 kDa as well as other minor protein bands were apparent in the peptide eluates. The 8 M urea eluates, which sequentially followed those of the free peptide, contained large amounts of many proteins, including more of the 70- and 30-kDa species. Similar results were obtained with FG cell extracts (not shown). Immunoblotting of the eluates for PTH/PTHrP receptor with the antihuman antibody (clone 3D1.1, Santa Cruz Biotechnology, Santa Cruz, CA) were unsuccessful, possibly because the PTH/PTHrP receptor is known to be proteolytically cleaved when removed from the cell membrane, making the receptor epitope unrecognizable (38, 39).
FIG. 1. Interaction of cell-surface biotinylated proteins from Saos-2 extracts on PTHrP (1–34)-Sepharose. Extracts of cell-surface biotinylated proteins from Saos-2 cells were applied onto PTHrP (1–34)-Sepharose 4B, and the column was washed and eluted with 1 mg/ml PTHrP (1–34) followed by 8 M urea. The eluted fractions were analyzed on a 12% Bis/Tris NuPAGE gel in MOPS buffer under reducing conditions, transferred to nitrocellulose, and incubated with streptavidin-HRP, and proteins were visualized using chemiluminescence and autoradiography. Arrows indicate protein bands of 70 and 30 kDa. Molecular mass markers are indicated.
Affinity chromatography of surface-labeled proteins from MiaPaCa-2 pancreatic cancer cells with mid-region PTHrP (37–86)-Sepharose
After coupling PTHrP (37–86) to Sepharose, extracts of cell-surface biotinylated MiaPaCa-2 cells were chromatographed, and bound proteins were eluted with either 1 mg/ml free peptide, followed by 8 M urea (Fig. 2A), or with 20 mM EDTA, 1 M NaCl, and 8 M urea (Fig. 2B). Like PTHrP (1–34), a major protein band of about 70 kDa was also eluted with the free peptide from PTHrP (37–86)-Sepharose. The 30-kDa protein bound to PTHrP (1–34) did not appear to be eluted with free peptide on PTHrP (37–86)-Sepharose but did appear subsequently in the 8 M urea eluates. To determine the nature of the interaction between PTHrP (37–86) and the 70-kDa protein, we chromatographed extracts of cell-surface biotinylated MiaPaCa-2 proteins over PTHrP (37–86)-Sepharose and eluted with 20 mM EDTA, 1 M NaCl, and 8 M urea. This elution sequence was used to identify protein-protein interactions that were divalent cation dependent (EDTA), ionic (NaCl), and hydrophobic (urea). Figure 2B demonstrates that most of the 70-kDa protein bound PTHrP (37–86) in a divalent cation-dependent manner. Bands of 30- and 15-kDa as well as more of the 70-kDa protein were eluted with 1 M NaCl, suggesting ionic interactions. These results collectively suggest that a 70-kDa protein expressed on the cell surface of MiaPaCa-2 pancreatic cancer cells binds to PTHrP (37–86) reversibly and that the protein-protein interaction is divalent cation dependent. Mass spectroscopy analyses indicated that the 70-kDa protein bound to PTHrP (1–34) and (37–86)-Sepharose was HSP70 (not shown).
FIG. 2. Interaction of cell-surface biotinylated proteins from MiaPaCa-2 extracts on PTHrP (37–86)-Sepharose. A, Extracts of cell-surface biotinylated proteins from MiaPaCa-2 cells were applied onto PTHrP (37–86)-Sepharose 4B, and the column was washed and eluted with 1 mg/ml PTHrP (37–86) followed by 8 M urea. The proteins were visualized as described in Fig. 1. The arrow indicates a 70-kDa protein band. B, Extracts of cell-surface biotinylated proteins from MiaPaCa2 cells were chromatographed over PTHrP (37–86)-Sepharose and eluted with wash buffer containing 20 mM EDTA followed by 1 M NaCl and then 8 M urea and analyzed as described in Fig. 1.
The SPECTRA-70 algorithm predicts regions of high HSP70 binding specificity in PTHrP
To understand the nature of the interaction between HSP70 and the two different regions of PTHrP, we relied upon a recent report where HSP70 binding proteins were identified in vivo using chromatographic techniques (40). The authors proposed an algorithm (SPECTRA-70) based on a preferred nine-amino-acid sequence used for predicting proteins likely to bind HSP70. Applying that algorithm to all nine amino acid sequences in PTHrP (1–34) and (37–86), in both the forward and reverse directions, we found several amino acid sequences within each peptide that were favored with more than 95% confidence (Table 1). We also applied the SPECTRA-70 algorithm to PTHrP (109–141) and (140–173) and found that those fragments also contain amino acid sequences predicted to bind HSP70 (Table 1). PTHrP (109–141), in particular, contains a nine-amino-acid stretch that yields a predictive value nearly two times greater than the highest value observed in the studies used to create the SPECTRA-70 algorithm (40). It is also worth noting the percentage of positive hits (those yielding matrix scores > 1) present in each peptide; for PTHrP (1–34), there were six significant hits out of 52 possible peptides (11.5%); for PTHrP (37–86), there were 16 significant hits out of 84 peptides (19%); for PTHrP (109–141), there were eight significant hits out of 50 peptides (16%); and for PTHrP (140–173), there were seven significant hits out of 52 peptides (13.5%).
TABLE 1. Evaluation of the PTHrP peptide sequences used in affinity chromatography experiments using the PRESTO-70 matrix (40 ) to identify predictive HSP70 binding motifs
Membrane-associated HSP70 binds PTHrP (109–141) and (140–173) in cancer cell lines
Based on the SPECTRA-70 predictions, we coupled PTHrP (109–141) and (140–173) to Sepharose and chromatographed surface-labeled extracts of FG and DU-145 cells, respectively, over the peptide columns, eluting with either free peptide or sequentially with EDTA, NaCl, and urea. Figure 3A demonstrates that HSP70 binds PTHrP (109–141), the majority of which is elutable with free peptide. Figure 3B also shows that HSP70 interacts with PTHrP (140–173) in a divalent cation-dependent manner. Similar results were obtained with FG cancer cells and PTHrP (140–173)-Sepharose (not shown). These results are consistent with SPECTRA-70 predictions that HSP70, expressed on the cell surface in cancer cells, should bind to all four PTHrP peptides used in these studies.
FIG. 3. Interaction of cell-surface biotinylated proteins from FG cell extracts on PTHrP (109–141)-Sepharose. A, Extracts of cell-surface biotinylated proteins from FG cells were applied onto PTHrP (109–141)-Sepharose 4B, and the column was washed and eluted with 1 mg/ml PTHrP (109–141) followed by 8 M urea. The eluted fractions were analyzed as described in Fig. 1. B, Extracts of cell-surface biotinylated DU-145 cells were chromatographed over PTHrP (140–173)-Sepharose and eluted with wash buffer containing 20 mM EDTA followed by 1 M NaCl and then 8 M urea and analyzed as described in Fig. 1.
HSP70 expressed on the cell surface of cancer cells binds PTHrP
Because HSP70 is generally not expressed on the surface of normal cells (41), we reasoned that PTHrP should not bind to surface-labeled HSP70 from a spontaneously immortalized normal breast epithelial cell line, MCF10 (32). We therefore chromatographed cell-surface biotinylated proteins from MCF10 extracts over PTHrP (109–141)-Sepharose, eluting directly with 8 M urea to collect all bound proteins. Figure 4 demonstrates that very little, if any, surface-expressed (biotinylated) HSP70 was bound to the PTHrP peptide from MCF10 cells. There were several protein bands in the 70-kDA range, but none of them were present at nearly the density observed in the cancer cell line eluates (Figs. 1–3). It is important to note that all proteins were eluted directly with urea, indicating that the overall protein content obtained with MCF10 cell extracts was far below that observed with the cancer cell extracts. BSA-Sepharose incubated with biotinylated proteins from FG cell extracts and mock-coupled Sepharose incubated with DU-145 cell extracts also yielded essentially no surface-expressed HSP70 (Fig. 4).
FIG. 4. Interaction of cell-surface biotinylated MCF10 normal breast epithelial cell extracts on PTHrP (109–141)-Sepharose, DU-145 extracts on uncoupled Sepharose 4B, and FG cell extracts on BSA-Sepharose. MCF10 extracts were applied onto PTHrP (109–141)-Sepharose and eluted with 8 M urea. Extracts of cell-surface biotinylated DU-145 cells were applied onto a mock-coupled Sepharose 4B, and the column was washed and eluted as described above. Extracts of cell-surface biotinylated FG cells were applied onto BSA-Sepharose, and the column was washed and eluted as described above. The eluted fractions from each column were analyzed as described in Fig. 1.
Immunoblotting of peak fractions from all of the affinity chromatography experiments with a polyclonal antibody raised against HSP70 confirmed the presence of HSP70 in all peak fractions from the PTHrP peptides tested but not from the BSA-Sepharose column (Fig. 5A). Interestingly, HSP70 from MCF10 cell extracts is abundantly present in eluates from PTHrP (109–141) affinity chromatography experiments (lane 7), although very little is actually surface expressed (Fig. 4). Differences in levels of HSP70 from lane to lane can be attributed to different elution regimes. HSP70 is present at high levels in the peak fractions from FG cells on PTHrP (1–34) (lane 1), MiaPaCa-2 cells on PTHrP (37–86) (lane 3), FG cells on PTHrP (140–173) (lane 6), and MCF10 cells on PTHrP (109–141), which were eluted directly with 8 M urea in six to 10 fractions. HSP70 is present in relatively lower amounts in peak fractions using longer elution regimes of 18 fractions, using free peptide followed by 8 M urea as shown for Saos-2 cells on PTHrP (1–34) (lane 1) and FG cells on PTHrP (109–141) (lane 5) or EDTA followed by NaCl followed by 8 M urea as shown for MiaPaCa-2 cells on PTHrP (37–86) (lane 3). To determine whether the increased yield of surface-expressed HSP70 found in the eluates from our cancer cell extracts on PTHrP peptides relative to MCF10 cells was a result of lower overall expression of HSP70 in the normal cell line, we conducted immunoblotting studies with both the soluble and insoluble fractions from each of the cell lines used in these studies. Figure 5B demonstrates that HSP70 is expressed essentially equally in all the cell lines.
FIG. 5. HSP70 immunoblotting of peak fractions from affinity chromatography experiments and HSP70 expression in soluble and insoluble fractions from cell lysates. A, Peak fractions from affinity chromatography experiments were loaded volumetrically (16.5 μl/well) and electrophoresed on 12% Bis/Tris gels under reducing conditions in MOPS buffer and transferred to nitrocellulose, and the membranes were subsequently immunoblotted with polyclonal anti-HSP70 as described in Materials and Methods. Lane 1, FG cell extracts on PTHrP (1–34)-Sepharose, eluted with 8 M urea; lane 2, Saos-2 cell extracts on PTHrP (1–34)-Sepharose, eluted with free peptide; lane 3, MiaPaCa-2 cell extracts on PTHrP (37–86)-Sepharose, eluted with 8 M urea; lane 4, MiaPaCa-2 cell extracts on PTHrP (37–86)-Sepharose, eluted with 20 mM EDTA; lane 5, FG cell extracts on PTHrP (109–141)-Sepharose, eluted with free peptide; lane 6, FG cell extracts on PTHrP (140–173)-Sepharose, eluted with 8 M urea; lane 7, MCF10 cell extracts on PTHrP (109–141)-Sepharose, eluted with 8 M urea; lane 8, FG cell extracts on BSA-Sepharose, eluted with 8 M urea. B, Whole-cell lysates were prepared as described in Materials and Methods for each cell line. The soluble (S) and insoluble (I) fractions (5 μg protein/lane) were analyzed by immunoblotting for HSP70 and ?-actin expression as described in Materials and Methods. C, IP analysis of unbound fractions of cell lysates after PTHrP affinity chromatography were conducted as described in Materials and Methods. After IP with clone P5D2, directed against the surface-expressed ?1-integrin subunit, streptavidin-HRP identified the ?1-integrin subunit with its associated -subunits as surface labeled. After IP with clone AC-15, directed against ?-actin, a known intracellular protein, streptavidin-HRP revealed no biotinylated protein, even though subsequent immunoblotting with the ?-actin monoclonal antibody revealed abundant ?-actin in the immunoprecipitates. IB, Immunoblot.
To confirm the specificity of the biotinylation procedure for cell surface-expressed proteins, we conducted IP analyses with the unbound fractions of cell lysates after affinity chromatography for known surface-expressed and intracellular proteins. As shown in Fig. 5C, the ?1-integrin subunit, along with its associated -subunits, were identified by streptavidin-HRP in all cell lysates, indicating surface expression, which is well established (42). IP of the unbound fractions of chromatographed cell lysates for ?-actin, a well-known and abundant intracellular protein, demonstrated essentially no reaction after streptavidin-HRP treatment, even after prolonged exposure of the treated blot to film, indicating that no ?-actin is biotinylated. Subsequent immunoblot analyses of the ?-actin IP with the anti-?-actin antibody demonstrated abundant protein in the immunoprecipitates from all cell lines. Taken together with the affinity chromatography and HSP70 immunoblotting results, these results collectively indicate that the biotinylation procedure effectively labels cell-surface and not intracellular proteins and that, in cancer cell lines, HSP70 is surface expressed.
The 2-D electrophoresis of 8 M urea eluates from PTHrP (1-34, 109-141, and 140-173) and mock-coupled Sepharose confirms the presence of HSP70 in the PTHrP peptide eluates and its absence in the mock-coupled Sepharose eluates (Fig. 6, A and B). Because these 2-D gels were loaded quantitatively, Fig. 6B demonstrates further that PTHrP (109–141) eluates contain the largest amount of surface-labeled proteins. As an additional control, we isolated plasma membranes from MiaPaCa2 cells after biotinylation by phase separation and chromatographed protein extracts from these membrane preparations over PTHrP (37–86)-Sepharose. The 8 M urea eluates were indistinguishable from elution profiles obtained using surface-labeled total cell lysates (not shown). Taken together, these results confirm the specificity of the interaction between surface-expressed HSP70 and PTHrP in cancer but not in normal cells.
FIG. 6. Two-dimensional electrophoresis of total protein (A) and cell-surface-expressed protein (B) eluted from the indicated PTHrP fragments using extracts from the indicated cancer cell lines. Fifty micrograms total protein from 8 M urea eluates from the indicated affinity chromatography experiments were quantitatively loaded and electrophoresed in two dimensions. Gels were then stained with Coomassie blue to visualize total protein or transferred to polyvinylidene difluoride membranes, incubated with HRP-conjugated streptavidin at 1:25,000, and developed with enhanced chemiluminescence to visualize biotinylated, cell-surface-expressed proteins. Arrowheads indicate the internal standard, tropomyosin, pI 5.2 and molecular mass 32.7 kDa. Arrows indicate the location of the 70-kDa HSP70 protein.
Blocking HSP70 increases PTHrP secretion and decreases proliferation in cancer cell lines
To begin to understand the potential significance of the interaction between PTHrP and HSP70 at the cancer cell surface, we conducted experiments with all the cell lines used in affinity chromatography, grown in the presence of a polyclonal antibody raised against HSP70 or irrelevant rabbit IgG. Figure 7A demonstrates that blocking the HSP70 interaction with PTHrP resulted in statistically significant increases in the expression of secreted PTHrP in all cell lines tested except MiaPaCa-2, which expresses very low amounts of PTHrP (12). This increase in PTHrP secretion was associated with a trend toward decreased total protein from 72-h cultures grown in the presence of anti-HSP70 in all the cell lines compared with the irrelevant control antibody (Fig. 7B), although statistical significance was achieved only in the pancreatic cancer cell lines, MiaPaCa-2 and FG. Studies conducted in the presence of anti-HSP70 or rabbit IgG also demonstrated a differential regulation of proliferation, with three cell lines (MiaPaCa-2, FG, and MCF10 cells) trending toward decreased proliferation and two cell lines (Saos-2 and DU-145) trending toward increased proliferation (Fig. 7C). The FG and Saos-2 results were statistically significant. Together, these data are consistent with the hypothesis that the interaction of PTHrP with HSP70 at the cancer cell surface is biologically relevant.
FIG. 7. The effect of anti-HSP70 polyclonal antibody on PTHrP secretion, total protein, and proliferation after 72 h in culture in the presence of 5% serum. In each well of a six-well tissue culture plate, 2.5 x 105 cells were cultured in the presence of polyclonal anti-HSP70 antibody at 25 μg/ml (black bars) or an irrelevant rabbit IgG at 25 μg/ml (gray bars) for 72 h. A, Media from triplicate wells were harvested and assayed for PTHrP (1–34) by RIA to assess secretion. Data presented represent the mean ± SEM from three independent experiments; 100% = 2.4 pg/μg protein. P values were derived from two-tailed Student’s t test. *, P < 0.05. B, Cell lysates were prepared from the experiments described in A and assayed for total protein using Bio-Rad protein assay according to the manufacturer’s instructions. Black bars, anti-HSP70; gray bars, irrelevant rabbit IgG. Data presented represent the mean ± SEM from three independent experiments. *, P < 0.05. C, Proliferation assays were conducted in the presence of polyclonal anti-HSP70 antibody at 25 μg/ml or irrelevant rabbit IgG at 25 μg/ml using the CellTiter 96 AQueous One Solution cell proliferation assay according to the manufacturer’s instructions (Promega). The absorbance at 450 nm at time zero is subtracted from the absorbance at 450 nm at 72 h and expressed as percent maximum (%Max) considering all five cell lines. Data presented represent the mean ± SEM from three independent experiments. *, P < 0.05.
Discussion
In the present study, we demonstrate, using affinity chromatography, that HSP70 is a major cell-surface protein expressed by cancer cells that binds to PTHrP, regardless of the peptide fragment used. This interaction is specific to cancer cells that express HSP70 on their cell surface, because essentially no surface-expressed HSP70 was found bound to PTHrP peptides from MCF-10 normal breast epithelial cell extracts. This interaction appears to be partially reversible, because free peptide can elute HSP70. This protein-protein interaction also appears to be partially divalent cation dependent, as evidenced by elution of HSP70 with EDTA. Some of the PTHrP-bound HSP70 also consists of other ionic as well as hydrophobic interactions, as evidenced by its appearance in the high salt and urea eluates. This is not surprising in light of our results, which demonstrated that an average of 15% of the PTHrP sequences used in these studies, which span the entire PTHrP peptide without the nuclear localization sequence from amino acids 88–106 (43), contain high predictive SPECTRA-70 matrix scores for HSP70 binding. It is reasonable to expect that some of these binding interactions between PTHrP and HSP70 could be divalent cation dependent, whereas others could be ionic or hydrophobic.
HSP70 has recently been shown in a number of studies to be among a set of chaperone proteins expressed on the cell surface of cancer cells, including neuroblastoma, lung, and colon adenocarcinoma, acute lymphoblastic leukemia, and ovarian tumor cell lines (44, 45, 46, 47, 48, 49) as well as freshly isolated human biopsy material of colorectal, lung, neuronal, and pancreatic carcinomas, liver metastases, and leukemic blasts of patients with acute myelogenous leukemia (50). Those observations are extended by our studies, where cell surface-expressed HSP70 from Saos-2 osteosarcoma, FG and MiaPaCa-2 pancreatic adenocarcinoma, and DU-145 prostate cancer cells bound PTHrP. Even when membrane fractions were isolated before affinity chromatography, HSP70 was still the major cell-surface biotinylated protein found bound to PTHrP. And even though MCF10 cells express significant amounts of HSP70 that binds to PTHrP peptides, essentially no surface-expressed HSP70 was recovered in the eluates. The 2-D electrophoresis and immunoblotting confirmed the presence of surface-labeled HSP70 in the PTHrP peptide eluates compared with BSA-Sepharose eluates, where HSP70 was essentially absent. IP analyses further validated the cell-surface specificity of the biotinylation procedure and our HSP70 observations, identifying known surface-expressed proteins, the integrins, as biotinylated and identifying ?-actin as not biotinylated and thus not surface-expressed, which is well established.
This novel interaction between PTHrP and HSP70 at the cell surface in cancer cell lines appears to be biologically relevant, as evidenced by increased PTHrP secretion and decreased cancer cell growth in the presence of polyclonal anti-HSP70 antibody. The function of HSP70 inside of normal cells, and at the cancer cell surface in particular, appears to be related to growth and, in particular, protection from apoptosis (51). PTHrP shares in these growth-regulating and antiapoptotic functions (15). Although speculative, these results suggest that HSP70 expressed at the cell surface is somehow involved in the internalization of secreted PTHrP, which results in an overall growth-promoting or cytoprotective effect. Although not usually found on the cell surface, the complexing of HSP70 with other proteins intracellularly, leading to cytoprotection in cancer, has been shown (51, 52). For example, inducible HSP70 forms complexes with Kruppel-like factor 6 and inducible nitric oxide synthase, resulting in cytoprotection in male SwissWebster mice subjected to 40% hemorrhage.
It is possible that HSP70, binding at the cell surface, acts secondarily in the presentation of PTHrP to its cognate receptor or to an as yet undefined receptor. Alternatively, HSP70 could act independently in PTHrP peptide transport from the cell surface into the cell. Mechanisms are obviously in place to traffic HSP70 and its bound peptides to the cell surface. It seems reasonable to expect that those same mechanisms could work in reverse, transporting HSP70 and its bound peptides into the cell as well. A relationship between PTHrP and stress-related genes, including HSP90, which is found bound to PTHrP in our studies, is not unprecedented (53). In those studies, the authors demonstrated that PTHrP (67–86) regulates the expression of stress proteins in breast cancer cells, inducing modifications in urokinase-plasminogen activator and matrix metalloproteinase-1 expression, and resulting in increased tumor cell invasiveness. The mechanism by which HSP70 and PTHrP interact at the cell surface to achieve these growth-promoting effects is currently under investigation.
It is interesting to note that HSP70 and G protein-coupled receptors, of which the PTH/PTHrP receptor is a member, have been shown to be enriched in lipid rafts (54, 55). Lipid raft theory suggests that cholesterol and sphingolipids partition away from other membrane lipids and that these lipid rafts are important in processes such as signaling and vesicle traffic (56). With that in mind, it is well established that HSP70 has the capacity for multiple protein binding. Some of the proteins identified by mass spectroscopy in vivo that bind to HSP70 include keratin, actin, HSP60, and elongation factors 1 and 2 (40). Although not surface expressed, it is remarkable that mass spectroscopy analyses of PTHrP (140–173) urea eluates from DU-145 cell extracts identified a complex mixture of proteins, including keratin, actin, elongation factors 1 and 2, and HSP60 and -90 as well as other proteins (data not shown). In support of these observations, PTHrP has been previously shown to bind to keratin, possibly affecting keratinocyte growth and differentiation (57). In addition, HSP70 and peroxiredoxin-2, both found bound to PTHrP in our studies (not shown), are abundantly expressed in human ductal carcinomas of the breast (58).
The divalent cation-dependent nature of the interaction of HSP70 with PTHrP is also consistent with the known binding of divalent cation binding proteins such as zinc finger proteins and calpain to HSP70 in vivo (40). Although it seems probable that some of these proteins are actually bound to PTHrP by association with HSP70 or another protein, these studies collectively suggest that in cancer, a complex of proteins, including HSP70 and PTHrP, could be enriched in lipid rafts at the cell surface and that this complex is important in the regulation of cell growth.
The HSP70 binding region in various proteins is generally at least nine amino acids in length and contains both hydrophobic and basic amino acids (40). It is also interspersed with at least two hydrophilic amino acids and usually contains acidic residues. It often contains a five-amino-acid core, with hydrophobic residues, a proline or glycine at positions 2, 3, or 4, and at least one hydrophilic residue. The five-amino-acid core generally does not contain basic amino acids but often does contain one or more acidic amino acids. Our studies indicate that PTHrP contains multiple HSP70 binding regions as predicted by the SPECTRA-70 matrix scores and as demonstrated by our affinity chromatography results. It is worth emphasizing that a strong correlation exists between the very large SPECTRA-70 matrix score for PTHrP (109–141) and the amount of HSP70 and associated cell surface-expressed proteins present in our quantitative 2-D gel analysis compared with the other PTHrP peptides.
In summary, our studies demonstrate novel protein-protein interactions between HSP70, expressed on the surface of cancer cells, and PTHrP. This binding is partially reversible and consists of divalent cation-dependent, ionic, and hydrophobic interactions. Our affinity chromatography studies also identify other PTHrP binding proteins previously shown to bind HSP70 in vivo, and include proteins shown to be enriched in lipid rafts. Functional studies demonstrating growth-related effects as a result of the interaction of PTHrP and HSP70 suggest a biological relevance to these protein-protein interactions. Taken together, these results have important implications for our understanding of the functions of PTHrP in both malignant and normal cells that are not currently attributable to the PTH/PTHrP receptor.
Acknowledgments
We thank Drs. Erkki Ruoslahti and Eva Engvall for critical readings of the manuscript; Kendricks Labs, Inc. (Madison, WI) for expert 2-D electrophoresis services; and Proteomic Research Services, Inc. (Ann Arbor, MI) and Scripps Institute (La Jolla, CA) for expert mass spectroscopy services.
References
Nissenson RA 2000 Parathyroid hormone-related protein. Rev Endocr Metab Disord 1:343–352
Grzesiak JJ, Smith KC, Burton DW, Deftos LJ, Bouvet M 2005 GSK3 and PKB/Akt are associated with integrin-mediated regulation of PTHrP, IL-6 and IL-8 expression in FG pancreatic cancer cells. Int J Cancer 114:522–530
Grzesiak JJ, Clopton P, Chalberg C, Smith K, Burton DW, Silletti S, Moossa AR, Deftos LJ, Bouvet M 2004 The extracellular matrix differentially regulates the expression of PTHrP and the PTH/PTHrP receptor in FG pancreatic cancer cells. Pancreas 29:85–92
Philbrick WM, Wysolmerski JJ, Galbraith S, Holt E, Orloff J J, Yang KH, Vasavada RC, Weir EC, Broadus AE, Stewart AF 1996 Defining the roles of parathyroid hormone-related protein in normal physiology. Physiol Rev 76:127–173
Tachibana I, Nakano S, Akiyama T, Ogami Y, Tabaru A, Otsuki M 1994 Parathyroid hormone-related protein mediates hypercalcemia in an exocrine pancreatic cancer. Am J Gastroenterol 89:1580–1581
Brandt DW, Wachsman W, Deftos LJ 1994 Parathyroid hormone-like protein: alternative messenger RNA splicing pathways in human cancer cell lines. Cancer Res 54:850–853
Bucht E, Rong H, Pernow Y, Nordqvist AC, Eriksson E, Rankin W, von Schoultz E, Burtis WJ, Granberg B, Falkmer UG, Burton DW, Deftos LJ 1998 Parathyroid hormone-related protein in patients with primary breast cancer and eucalcemia. Cancer Res 58:4113–4116
Deftos LJ, Gazdar AF, Ikeda K, Broadus AE 1989 The parathyroid hormone-related protein associated with malignancy is secreted by neuroendocrine tumors. Mol Endocrinol 3:503–508
Deftos LJ, Burton DW, Brandt DW 1993 Parathyroid hormone-like protein is a secretory product of atrial myocytes. J Clin Invest 92:727–735
Drucker DJ, Asa SL, Henderson J, Goltzman D 1989 The parathyroid hormone-like peptide gene is expressed in the normal and neoplastic human endocrine pancreas. Mol Endocrinol 3:1589–1595
Alipov GK, Ito M, Nakashima M, Ikeda Y, Nakayama T, Ohtsuru A, Yamashita S, Sekine I 1997 Expression of parathyroid hormone-related peptide (PTHrP) in gastric tumours. J Pathol 182:174–179
Bouvet M, Nardin SR, Burton DW, Behling C, Carethers JM, Moossa AR, Deftos LJ 2001 Human pancreatic adenocarcinomas express parathyroid hormone-related protein. J Clin Endocrinol Metab 86:310–316
Bouvet M, Nardin SR, Burton DW, Lee NC, Yang M, Wang X, Baranov E, Behling C, Moossa AR, Hoffman RM, Deftos LJ 2002 Parathyroid hormone-related protein as a novel tumor marker in pancreatic adenocarcinoma. Pancreas 24:284–290
Bakre MM, Zhu Y, Yin H, Burton DW, Terkeltaub R, Deftos LJ, Varner JA 2002 Parathyroid hormone-related peptide is a naturally occurring, protein kinase A-dependent angiogenesis inhibitor. Nat Med 8:995–1003
Henderson JE, Amizuka N, Warshawsky H, Biasotto D, Lanske BM, Goltzman D, Karaplis AC 1995 Nucleolar localization of parathyroid hormone-related peptide enhances survival of chondrocytes under conditions that promote apoptotic cell death. Mol Cell Biol 15:4064–4075
Goltzman D 1999 Interactions of PTH and PTHrP with the PTH/PTHrP receptor and with downstream signaling pathways: exceptions that provide the rules. J Bone Miner Res 14:173–177
Strewler GJ 2000 The physiology of parathyroid hormone-related protein. N Engl J Med 342:177–185[Free Full Text]
Kronenberg HM, Lanske B, Kovacs CS, Chung UI, Lee K, Segre GV, Schipani E, Juppner H 1998 Functional analysis of the PTH/PTHrP network of ligands and receptors. Recent Prog Horm Res 53:283–301; discussion 301–303
Segre GV 1996 Receptors for parathyroid hormone (PTH) and PTH-related peptide. In: Bilezikian JP, Raisy LG, Rodan FA, eds. Principles of bone biology. New York: Academic Press; 389–406
Clemens TL, Cormier S, Eichinger A, Endlich K, Fiaschi-Taesch N, Fischer E, Friedman PA, Karaplis AC, Massfelder T, Rossert J, Schluter KD, Silve C, Stewart AF, Takane K, Helwig JJ 2001 Parathyroid hormone-related protein and its receptors: nuclear functions and roles in the renal and cardiovascular systems, the placental trophoblasts and the pancreatic islets. Br J Pharmacol 134:1113–1136
Lanske B, Amling M, Neff L, Guiducci J, Baron R, Kronenberg HM 1999 Ablation of the PTHrP gene or the PTH/PTHrP receptor gene leads to distinct abnormalities in bone development. J Clin Invest 104:399–407
Strid H, Care A, Jansson T, Powell T 2002 Parathyroid hormone-related peptide (38–94) amide stimulates ATP-dependent calcium transport in the basal plasma membrane of the human syncytiotrophoblast. J Endocrinol 175:517–524
Cornish J, Callon KE, Nicholson GC, Reid IR 1997 Parathyroid hormone-related protein-(107–139) inhibits bone resorption in vivo. Endocrinology 138:1299–1304
Cornish J, Callon KE, Lin C, Xiao C, Moseley JM, Reid IR 1999 Stimulation of osteoblast proliferation by C-terminal fragments of parathyroid hormone-related protein. J Bone Miner Res 14:915–922
De Miguel F, Martinez-Fernandez P, Guillen C, Valin A, Rodrigo A, Martinez ME, Esbrit P 1999 Parathyroid hormone-related protein (107–139) stimulates interleukin-6 expression in human osteoblastic cells. J Am Soc Nephrol 10:796–803
Hastings RH, Araiza F, Burton DW, Zhang L, Bedley M, Deftos LJ 2003 Parathyroid hormone-related protein ameliorates death receptor-mediated apoptosis in lung cancer cells. Am J Physiol Cell Physiol 285:C1429–C1436
Goomer RS, Johnson KA, Burton DW, Amiel D, Maris TM, Gurjal A, Deftos LJ, Terkeltaub R 2000 The tetrabasic KKKK(147–150) motif determines intracrine regulatory effects of PthrP 1–173 on chondrocyte PPi metabolism and matrix synthesis. Endocrinology 141:4613–4622
Jonsson KB, John MR, Gensure RC, Gardella TJ, Juppner H 2001 Tuberoinfundibular peptide 39 binds to the parathyroid hormone (PTH)/PTH-related peptide receptor, but functions as an antagonist. Endocrinology 142:704–709
Ahmadpour OA, Burton DW, Tu S, Deftos LJ 2002 Regulation of OPG/RANK/RANKL by PTHrP peptides in human prostate cancer cells. J Bone Miner Res 17(S):286
Burton DW, Tu S, Gujral A, Patel JA, Terkeltaub RA, Rhim JS, Deftos LJ 1999 Intracellular (intracrine), non-PTH/PTHrP receptor mediated effects of PTHrP on prostate cancer cell growth. J Bone Miner Res 14(S):290
Maheshwari VV, Burton DW, Hastings RH, Aguilera JA, Pardo FS, Deftos LJ 2003 Apoptotic effects of parathyroid hormone related protein in prostate cancer cells. J Bone Miner Res 18(S):119 (Abstract)
Sloane BF, Moin K, Sameni M, Tait LR, Rozhin J, Ziegler G 1994 Membrane association of cathepsin B can be induced by transfection of human breast epithelial cells with c-Ha-ras oncogene. J Cell Sci 107:373–384
Lenstra JA, Bloemendal H 1983 Topography of the total protein population from cultured cells upon fractionation by chemical extractions. Eur J Biochem 135:413–423
Yeung SC, Eton O, Burton DW, Deftos LJ, Vassilopoulou-Sellin R, Gagel RF 1998 Hypercalcemia due to parathyroid hormone-related protein secretion by melanoma. Horm Res 49:288–291
Carron JA, Fraser WD, Gallagher JA 1997 PTHrP and the PTH/PTHrP receptor are co-expressed in human breast and colon tumours. Br J Cancer 76:1095–1098
O’Farrell PH 1975 High resolution two-dimensional electrophoresis of proteins. J Biol Chem 250:4007–4021
Ruoslahti E, Pierschbacher MD 1987 New perspectives in cell adhesion: RGD and integrins. Science 238:491–497
Gentili C, Morelli S, de Boland AR 2003 Characterization of PTH/PTHrP receptor in rat duodenum: effects of ageing. J Cell Biochem 88:1157–1167
Karpf DB, Arnaud CD, King K, Bambino T, Winer J, Nyiredy K, Nissenson RA 1987 The canine renal parathyroid hormone receptor is a glycoprotein: characterization and partial purification. Biochemistry 26:7825–7833
Grossmann ME, Madden BJ, Gao F, Pang YP, Carpenter JE, McCormick D, Young CY 2004 Proteomics shows Hsp70 does not bind peptide sequences indiscriminately in vivo. Exp Cell Res 297:108–117
Bhattacharyya T, Karnezis AN, Murphy SP, Hoang T, Freeman BC, Phillips B, Morimoto RI 1995 Cloning and subcellular localization of human mitochondrial hsp70. J Biol Chem 270:1705–1710
Ruoslahti E, Noble NA, Kagami S, Border WA 1994 Integrins. Kidney Int Suppl 44:S17–S22
Fiaschi-Taesch NM, Stewart AF 2003 Parathyroid hormone-related protein as an intracrine factor: trafficking mechanisms and functional consequences. Endocrinology 144:407–411
Shin BK, Wang H, Yim AM, Le Naour F, Brichory F, Jang JH, Zhao R, Puravs E, Tra J, Michael CW, Misek DE, Hanash SM 2003 Global profiling of the cell surface proteome of cancer cells uncovers an abundance of proteins with chaperone function. J Biol Chem 278:7607–7616
Multhoff G, Botzler C, Wiesnet M, Muller E, Meier T, Wilmanns W, Issels RD 1995 A stress-inducible 72-kDa heat-shock protein (HSP72) is expressed on the surface of human tumor cells, but not on normal cells. Int J Cancer 61:272–279
Botzler C, Schmidt J, Luz A, Jennen L, Issels R, Multhoff G 1998 Differential Hsp70 plasma-membrane expression on primary human tumors and metastases in mice with severe combined immunodeficiency. Int J Cancer 77:942–948
Okazaki Y, Ohno H, Takase K, Ochiai T, Saito T 2000 Cell surface expression of calnexin, a molecular chaperone in the endoplasmic reticulum. J Biol Chem 275:35751–35758
White TK, Zhu Q, Tanzer ML 1995 Cell surface calreticulin is a putative mannoside lectin which triggers mouse melanoma cell spreading. J Biol Chem 270:15926–15929
Xiao G, Chung TF, Fine RE, Johnson RJ 1999 Calreticulin is transported to the surface of NG108–15 cells where it forms surface patches and is partially degraded in an acidic compartment. J Neurosci Res 58:652–662
Hantschel M, Pfister K, Jordan A, Scholz R, Andreesen R, Schmitz G, Schmetzer H, Hiddemann W, Multhoff G 2000 Hsp70 plasma membrane expression on primary tumor biopsy material and bone marrow of leukemic patients. Cell Stress Chaperones 5:438–442
Mosser DD, Morimoto RI 2004 Molecular chaperones and the stress of oncogenesis. Oncogene 23:2907–2918
Kiang JG, Bowman PD, Wu BW, Hampton N, Kiang AG, Zhao B, Juang YT, Atkins JL, Tsokos GC 2004 Geldanamycin treatment inhibits hemorrhage-induced increases in KLF6 and iNOS expression in unresuscitated mouse organs: role of inducible HSP70. J Appl Physiol 97:564–569
Luparello C, Sirchia R, Pupello D 2003 PTHrP [67–86] regulates the expression of stress proteins in breast cancer cells inducing modifications in urokinase-plasminogen activator and MMP-1 expression. J Cell Sci 116:2421–2430
Chini B, Parenti M 2004 G-protein coupled receptors in lipid rafts and caveolae: how, when and why do they go there? J Mol Endocrinol 32:325–338
Broquet AH, Thomas G, Masliah J, Trugnan G, Bachelet M 2003 Expression of the molecular chaperone Hsp70 in detergent-resistant microdomains correlates with its membrane delivery and release. J Biol Chem 278:21601–21606
Foster LJ, De Hoog CL, Mann M 2003 Unbiased quantitative proteomics of lipid rafts reveals high specificity for signaling factors. Proc Natl Acad Sci USA 100:5813–5818
Seidel J, Zabel M, Kasprzak A 2002 PTHrP and cytokeratins in human epidermis. Folia Histochem Cytobiol 40:209–210
Somiari RI, Sullivan A, Russell S, Somiari S, Hu H, Jordan R, George A, Katenhusen R, Buchowiecka A, Arciero C, Brzeski H, Hooke J, Shriver C 2003 High-throughput proteomic analysis of human infiltrating ductal carcinoma of the breast. Proteomics 3:1863–1873(John J. Grzesiak, Kathy C)