Interaction between HSP60 and -catenin promotes metastasis
Ya-Ping Tsai,
1 Muh-Hwa Yang,2,3,5Chi-Hung Huang,
6Shyue-Yi h Chang,
4,5Po-Min Chen,
3,5Chung-Ji Liu,
7Shu-Chun Teng,
8and Kou-Juey Wu
1,2,5,*1
Institutes of Biochemistry & Molecular Biology and 2Clinical Medicine, National Yang-Ming
University, Taipei 112; 3Division of Oncology, Departments of Medicine, 4Otolaryngology, and 5Genomic Medicine Center, Taipei Veterans General Hospital, Taipei 112; 6Institute of
Biochemistry, National Chung-Hsing University, Taichung 402; 7Department of Dentistry,
Taipei Mackay Memorial Hospital, Taipei 104; 8Graduate Institute of Microbiology, College of
Medicine, National Taiwan University, Taipei 100, Taiwan
*: corresponding author
Abbreviations: HSP60, heat shock protein 60
Running title: Interaction between HSP60 and -catenin Keywords: HSP60,-catenin, metastasis
*: To whom reprint requests should be addressed. Email:[email protected]
Tel: 011886-228267328Abstract
Heat shock protein 60 (HSP60) plays an essential role in assisting many newly synthesized proteins to reach their native forms. Increased HSP60 expression is observed in different types of human cancers with metastasis (e.g. pancreatic cancer, large bowel carcinoma). However, the role of HSP60 in metastasis remains little known. Aberrant activation of -catenin plays a key role in tumorigenesis and
metastasis. Here we show that overexpression of HSP60 induces metastatic phenotypes in vitro and in vivo. HSP60 interacts with -catenin, increases -catenin
protein levels through the apical domain, and enhances its transcriptional activity. Short-interference RNA (siRNA) mediated repression of -catenin reverts
metastatic activity caused by HSP60 overexpression. Proteosomal activity is not required for the induction of -catenin by HSP60. Co-expression of HSP60 and
nuclear -catenin predicts a worse prognosis of metastatic head and neck cancer
patients. These results implicate a novel role of HSP60 in metastasis.
Introduction
Heat shock protein 60 (HSP60) is the major component of the chaperonins complex which weighs nearly a million Daltons and is composed of back-to-back rings of identical or closely related rotationally symmetric subunits (1, 2). HSP60 plays an essential role in assisting a large variety of newly synthesized proteins to reach their native forms by binding and facilitating their folding into native status inside a large central channel within each ring, as demonstrated in the bacterial GroEL-GroES-ATP complex (1-3). However, folding of native proteins by a single domain was demonstrated in the mammalian HSP60 protein (4), which is different from the bacterial model (1-3). HSP60 was mostly shown to be involved in innate immunity and cardiac diseases (5, 6). The proteins identified to bind to HSP60 are relatively few, with integrins 31 and NPM-ALK (nucleophosmin-anaplastic lymphoma kinase) being the examples (7, 8). Increased HSP60 expression is observed in different types of human cancers (prostate cancer, ovarian cancer, pancreatic cancer, large bowel carcinoma, etc) (9-12). In addition, HSP60 was shown to be a putative c-MYC target gene (13-16), suggesting its possible role in c-MYC mediated transformation and metastasis. However, the role of HSP60 in tumorigenesis and metastasis remains relatively unknown.
The multifunctional protein -catenin plays a dual role in cells as a major structural component of cell-cell adherens junctions and as a signaling molecule in the Wnt pathway (17). The nonjunctional pool of -catenin is normally degraded by the ubiquitin-proteasome system (18, 19). However, under the activation of Wnt signaling, -catenin is not degraded but stabilized, and it then activates transcription of downstream targets (e.g. c-MYC, cyclin D1) by complexing with lymphoid enhancer factor (Lef)/T cell factor
(Tcf) transcription factors (20, 21). -catenin also associates with a wide variety of protein partners, including the destabilization partner (GSK-3, AXIN1, APC) or interacts with tyrosine kinase receptors at the cell periphery (18, 19, 22). Regulation of the functions of -catenin lies at least at three different levels: location of -catenin (membrane vs. cytoplasm/nucleus), stability of -catenin (through complexing with its destabilization partner), and transactivation (through complexing with Tcf to activate downstream target gene expression) (23, 24). -catenin activation was shown to be related to different types of cancer with metastasis (colorectal, lung, prostate, ovarian, etc) (25-28). In addition, -catenin activation caused invasion (29) and was related to epithelial-mesenchymal transition, a mechanism contributing to metastasis (30). Some of the major targets of -catenin are responsible for mediating metastasis (31). These results demonstrated the critical role of -catenin in metastasis. Other than the classical Wnt signaling pathway to activate -catenin, the alternative mechanisms to activate-catenin remain little known. Proteins interacting with -catenin to modulate its functions were very few with 14-3-3as the example to activate -catenin (32) and other proteins (Sox9, HIF-1) as inhibitors (33, 34). Understanding the molecular mechanism of -catenin regulation by its interacting proteins is important to delineate its functions related to tumorigenesis and metastasis.
In this report, we demonstrated that overexpression of HSP60 induces metastasis. HSP60 interacts with -catenin using a specific domain to activate its downstream targets and -catenin is critical for the metastatic activity induced by HSP60. Co-expresssion of HSP60 and nuclear -catenin predicts a worse prognosis in metastastic head and neck
Materials and Methods
Cell lines, plasmids, and reagents. The FADU, RatCMV, and RatMyc cell lines were
previously described (35-38). The construction of different plasmids including the vector backbones, restriction sites, inserted fragments, and the oligonucleotides used in the construction of plasmids were described (Supplementary Tables S1 & S2 are available at Carcinogenesis online). All the stable clones were established by transfection of plasmids using calcium phosphate method (37) and the clones were designated by the name of the plasmids. The constructions of pSUPER-siRNA plasmids were performed using the oligonucleotides (Supplementary Table S3 is available at Carcinogenesis online). All the siRNA stables clones were designated by the suffix “-si”. The HA-Akt, FlagCBP, and HA-ubiquitin plasmids were obtained from J.H. Lin (Bristol-Meyers-Squibb), H.M. Shih and H. Chen (Academia Sinica, Taiwan), respectively. Both proteosome inhibitors, LLnL (N-Acetyl-Leu-Leu-Norleu-al) and MG-132 (Z-Leu-Leu-Leu-al), were purchased from Sigma (St. Louis, MO).
RNA purification, quantitative real-time PCR, protein extraction, Western blot analysis and co-immunoprecipitation assays. We prepared RNA from cell lines using
an RNAeasy kit (Qiagen, Germany). Quantitative real time PCR was performed in a PRISM7700 Sequence Detection System (ABI) with the preset PCR program, and GAPDH was selected as an internal control. 0.1 μg of cDNA and 80 μM of primers were used for reaction in 1x SYBR Green Mixture (ABI) in a total volume of 50 μl (35, 36). The sequences of primers used in the real-time PCR experiments were shown (Supplementary Table S3 is available at Carcinogenesis online). For extraction of proteins, cultured cells were put in 500 l of cell lysis buffer (50 mM Tris, pH 7.5, 30
mM MgSO4, 8 mM EDTA, 2 mM DTT, and 2% Triton X-100) containing protease
inhibitors. Cell lysates were clarified by centrifugation at 13000 rpm, 4°C for 10 minutes. The protein content was determined by Bradford method (Bio-Rad Laboratories, Hercules, CA). For Western blot analysis, 50 g of protein extracts from different cell lines were loaded to 10% SDS-PAGE gels and transferred to nitrocellulose filters (37, 39). The characteristics of the antibodies used in Western blot analysis were listed (Supplementary Table S4 is available at Carcinogenesis online). Data shown here are representative of two or more experimentsfrom independent cell cultures.
Transient transfection and luciferase assays. A human embryonic kidney cell line,
293T, was chosen to perform transient transfection. The reporter construct (TOPFalsh) was co-transfected into 293T cells with different -catenin expression vectors together with HSP60 or its deletion mutants. The TOPFlash reporter construct contains four wild type Tcf responsive sites linked to a pGL3-OT luciferase construct as described (20, 24). A pCMV--gal plasmid was used as an internal control. The luciferase activities were assayed as described (37-39).
Immunofluorescence. For immunofluorescence microscopy, cells on glass coverslips
were fixed in 4% formaldehyde,washed with PBS, and permeated with 0.01% Triton X-100. After washing with PBS, sampleswere incubated with blocking solution for 1 hour, followed by incubation with an anti-HSP60 antibody or an anti--catenin antibody. FITC-conjugated goat anti-rabbit IgG and rhodamine-conjugated goat anti-mouse IgG were used to visualize the location of -catenin and HSP60, respectively. Fluorescence images were captured using a Leica Laser scanning confocal microscope (35, 36). The
characteristics of the antibodies used were listed (Supplementary Table S4 is available at Carcinogenesis online).
In vitro migration/invasion assays and in vivo tail vein metastatic assays. Eight-µm
pore size Boyden chamber was used for in vitro migration and invasionassays. Cells (1 x 105) in 0.5% serum-containing RPMI were plated in the upper chamber and 15% fetal
bovine serum was added to RPMI1640 in the lower chamber as a chemoattractant. For invasion assay, the upper side of the filter was covered with Matrigel (Collaborative Research Inc.,Boston, MA)(1:3 dilution with RPMI). After 12 hours formigration assay or 24 hours for invasion assay, cells on theupper side of the filter were removed, and cells that remained adherent to the underside of membrane were fixed in 4% formaldehyde and stainedwith Hoechst 33342 dye. The number of migrated cells was counted using a fluorescence microscope. Ten contiguous fields of each sample were examined using a 40x objective to obtain a representative number of cells which migrated/invaded across the membrane (35, 36). The results performed in Boyden chambers were designated as “migration” and “invasion” in the text. Six-week old female nonobese diabetic-severe combined immunodeficiency (NOD-SCID) mice received injectionof 4 X 106 cells of different cell lines in 0.1 ml of PBS via the tail vein (6 mice
for each group). Six weeks after injection, mice were examined grossly at necropsyfor the presence of metastases in internal organs. Microscopic examination of metastases was performed on the cross-sections of formalin-fixed, paraffin-embedded lung tissuesstained with H&E. The counting of metastatic lesions in the internal organ of each mouse was evaluated by gross and microscopic examination. The results of in vivo tail vein injection assays were designated as “metastasis” in the text. The tumor size was measured with
calipers and the tumor volume was calculated according to the formula (Length×Width2)/2 (40). This study was approved by the Ethics Committee of the Taipei
Veterans General Hospital.
Synthesis and purification of different GST-HSP60 fragments and His-tagged
-catenin proteins as well as GST pull down assays. Purification of GST proteins was
previously described (41). Briefly, 192-287 and pGEX-4T-HSP60-288-383 constructs were transformed respectively into E. coli BL21(DE3)pLysS (Novagen, Gibbstown, NJ). The GST fusion proteins were induced by 0.5 mM IPTG at 30 ℃ for 4 h, harvested, and lysed by French press (a device using differential pressure to gently lyse the cells without the damage caused by mechanical or chemical methods). The cell lysates were suspended in binding buffer (PBS with 150 mM NaCl), clarified by centrifugation (20,000 rpm. for 30 min at 4℃) and loaded on GST-FF column (GE Healthcare, UK). Bound proteins were washed with binding buffer and eluted with elution buffer (50 mM Tris-HCl, 10 mM reduced glutathione, pH8.0). Purification of His-tagged-catenin proteins was performed by transforming pET32a--catenin-400-530 and pET32a--catenin-531-781 plasmids respectively into E. coli Bl21(DE3)pLysS followed by induction for 2 h with IPTG and lysis by French press. The cell lysates were suspended in Buffer A (PBS with 150 mM NaCl), clarified by centrifugation (20,000 rpm. for 30 min at 4℃), adjusted to 10 mM imidazole and loaded on HisTrap chelating columns (GE Healthcare, UK). Bound proteins were washed with 50, 100 mM imidazole in Buffer A and eluted with 300 mM imidazole in buffer A. For GST pull down assays, different His-tagged -catenin protein was incubated with different GST-HSP60 proteins (or GST only as a control) and Glutathione-Agarose (Sigma, St. Louis, MI) for 3 h at 4
℃ in 600 l of GST buffer (142.5 mM NaCl, 10 mM Hepes pH 7.6, 5 mM MgCl2, 1 mM
EDTA and 2.5 mg/ml BSA). The beads were washed five times with 1 ml GST buffer without BSA. Finally, the bound complexes were eluted and analyzed by Western blot.
In vitro dephosphorylation assays. Proteins overexpressed by transfection in 293T cells
were immunoprecipitated using appropriate antibodies. The immunoprecipitates were washed three times with 1 ml IP buffer (without EDTA and EGTA) and once with CIP buffer (100 mM NaCl, 50 mM Tris-HCl pH 7.9, 10 mM MgCl2, 1 mM dithiothreitol).
After washing, the immunoprecipitates were resuspended in 200 μl of CIP buffer with 100 U of calf intestinal alkaline phosphatase (CIP, New England Biolabs, Beverly, MA) at 37 °C for 30 min followed by Western blot analysis.
Study population and sample collection. Fifty-eight patients with metastatic head and
neck squamous cell carcinoma (HNSCC) who underwent treatment at Taipei Mackay Memorial Hospital and Taipei Veterans General Hospital between January 2001 and December 2006 were enrolled in this study. This study has been approved by the Institutional Review Board of both Taipei Veterans General Hospital and Taipei Mackay Memorial Hospital. All patientswithhistologically proven HNSCCs received surgery or surgery plus post-operative radiotherapy, and lymph node/distant metastasis developed after completion of primary treatment. Primary tumor samples and the corresponding non-cancerous matched tissue (NCMT) were obtained during primary surgery; whereas metastatic tumor samples (including 10 cases with lung metastasis, 40 cases with cervical lymph nodes metastasis and 8 cases with both lung and neck nodes metastasis) were obtained from biopsy/salvage neck dissection when metastasis occurred. Salvage surgery plus cisplatin-based chemotherapy was given for patients only with neck nodes
metastasis; whereas cisplatin-based chemotherapy was used for patients with visceral metastasis. The median follow-up duration was 8 months (range, 1 ~ 26 months). The clinical characteristics of 58 metastatic HNSCC patients are illustrated (Supplementary Table S5 is available at Carcinogenesis online).
Tissue microarray construction, Immunohistochemstry (IHC) and scoring. A
high-density tissue microarray (TMA) was constructed using formalin-fixed, paraffin-embedded specimens of primary HNSCC, metastatic tumor and NCMT as previously described (36). Six m thick sections of tumor and neighboring normal tissue were cut from the frozen specimens for immunohistochemistry (IHC) analysis. The samples were fixed in acetone, air-dried, and subsequently bathed in Tris buffered saline (TBS) solution (pH 7.6). The endogenous peroxidase activity was blocked with 3% hydrogen peroxide. The antibodies used in IHC were described (Supplementary Table S4 is available at Carcinogenesis online). After reacting with a biotinylated secondary antibody for 30 minutes, antigen-antibody reactions were visualized using streptavidin-horseradish peroxidase conjugate (DAKO LSAB kit; DAKO, Los Angeles, CA), with 3-amino-9-ethylcarbazole as the chromogen. All slides were counterstained with hematoxylin (35, 36). The staining results were interpreted under microscopic fields of 200-fold magnification by two independent specialists and were scored as described: positive HSP60 expression was defined when the proportion of moderately or intensely stained of cytoplasmic and/or nucleus was identified in at least 20% of cancer cells (9,42). For interpretation of -catenin IHC result, nuclear -catenin expression in more than 10% of cancer cells was defined as a positive result (43). Co-expression of HSP60 and nuclear -catenin was defined when tumor samples showed overexpression of both HSP60 and
nuclear -catenin. Positive WNT-1 IHC staining was interpreted when more than 50% of cancer cells were stained (44).
Statistical analysis. Pearson Chi-square or Fisher's exact tests were used for
comparison of dichotomous variables between groups, and the independent Student’s t-test was used to compare the continuous variables between two groups. The Kaplan-Meier estimate was used for overall survival analysis, and the log-rank test was selected to compare the cumulative survival durations in different patient groups. The control and experimental groups of all the statistical analyses were indicated in the figure legends. The level of statistical significance was set at 0.05 for all tests.
Results
Overexpression of HSP60 induces metastasis
Since increased HSP60 expression is observed in different types of human cancers with metastasis (e.g. pancreatic cancer, large bowel carcinoma) (10, 12), we tested whether overexpression of HSP60 was involved in metastasis. The migration and invasion activity of FADU cells (a head and neck cancer cell line) (35, 36) overexpressing HSP60 was tested (Figure 1A). The results showed that HSP60 overexpression increased the migration and invasion of FADU cells in vitro (Figure 1B and C). To test whether HSP60 overexpression could increase the metastatic ability in vivo, a tail-vein metastasis assay was performed. After six weeks of injection, the lungs of the mice receiving FADUHSP60 cells had significantly more and larger metastatic nodules compared with the mice receiving FADUcDNA cells (Figure 1D-F). Gross observation of metastatic nodules in lungs was confirmed by histological examination (Fig. 1D, lower panels). All the results demonstrated that HSP60 overexpression in FADU cells increases migration/invasion in vitro and metastasis in vivo. Migration/invasion assays of RatCMV control versus RatMyc cells (37, 38) were performed and the results showed that RatMyc cells had increased migration and invasion (~2 to 2.5 fold increase), whereas knockdown of endogenous HSP60 in RatMyc cells by siRNA significantly decreased migration and invasion (reduced to ~40 to 50 % of the activity of RatMyc control) (Supplementary Figure S1 is available at Carcinogenesis online). These results demonstrated the critical role of HSP60 in inducing metastatic phenotypes.
Interaction between HSP60 and -catenin
To delineate the mechanism of HSP60 overexpression mediated metastatic phenotypes, we used the yeast two-hybrid approach to search for HSP60-interacting proteins whose functions could be regulated by HSP60. Among numerous candidates, we focused on -catenin since --catenin is involved in metastasis (25-31). Co-immunoprecipitation experiments using the anti-HSP60 antibody to co-precipitate endogenous -catenin in FADUHSP60 clones showed that HSP60 interacted with -catenin (Figure 2A). Co-immunoprecipitation experiments using the anti--catenin antibody (or anti-HSP60 antibody) to co-precipitate 293T extracts overexpressing HSP60 and -catenin also showed their interaction (Figure 2B and C). Domain mapping experiments showed that the apical domain (a.a. 288-383; A-domain) of HSP60 interacted with -catenin (Figure 2D). The domain in -catenin interacting with HSP60 was mapped to the armadillo 7-9 repeats (Figure 2E). Indirect immunofluorescence experiments showed the cytoplasmic expression of HSP60 and predominant membranous expression of -catenin in 293T control cells (Figure 2F). Overexpression of HSP60 in 293T cells resulted in the translocation of HSP60 and -catenin into nucleus and their nuclear co-localization (Figure 2F). In order to test whether HSP60 directly interacted with -catenin, GST pull down assays were performed using GST control, HSP60-192-287, and GST-HSP60-288-383 fragments to pull down purified His-tagged -catenin proteins (His--catenin-400-530 or His--catenin-531-781). The results showed that only the GST-HSP60-288-383 fragment was able to pull down the -catenin domain (His--catenin-400-530) which could interact with HSP60, and it was unable to pull down the negative
control His--catenin-531-781 protein (Figure 2G). These results demonstrated the direct interaction between HSP60 and -catenin.
Activation of -catenin and its downstream targets by HSP60 overexpression
To examine the effects of interaction between HSP60 and -catenin, Western blot analysis showed that the -catenin protein levels increased in FADUHSP60 clones (Figure 2A). Increasing dosage of HSP60 increased the exogenous or endogenous -catenin protein levels in 293T cells under transient co-transfection (Supplementary Figures S2A and S2B are available at Carcinogenesis online). We tested whether HSP60 increased -catenin levels through any specific domain. The increase in -catenin protein levels was specific through interaction with the apical domain (a.a. 288-383) of HSP60, whereas a non-interacting HSP60 domain (a.a. 384-573) cannot increase the -catenin protein levels (Figure 3A and Supplementary Figure S2C are available at Carcinogenesis online). In addition, HSP60 overexpression either did not increase the protein levels of Akt, CBP or Heme oxygenase-1 (HO-1) or used the C-terminal domain (a.a. 384-573; E-domain) to increase the protein levels of 14-3-3 (Supplementary Figures S2D-G are available at Carcinogenesis online). Finally, the truncated HSP60 protein (a.a. 215-401) containing the apical domain alone was able to increase the -catenin protein levels (Supplementary Figure S2H is available at Carcinogenesis online), demonstrating the specific activation of -catenin by the apical domain of HSP60. We tested whether the downstream target genes of -catenin were also activated in FADUHSP60 clones. Figure 3B showed that the downstream targets of -catenin such as c-MYC, cyclin D1, laminin-2, and Membrane-type Matrix Metalloproteinase-1 (MT1-MMP) were activated (20, 21,
31), demonstrating the activation of -catenin pathway by HSP60 overexpression. The ability of HSP60 to enhance the transcriptional activity of -catenin was tested and the results showed that co-transfection of HSP60 increased the transcriptional activity of -catenin using a TOPFlash reporter system (24), whereas the non-interacting HSP60 mutant did not show enhancement effect (Figure 3C). All the results demonstrated that HSP60 activates -catenin through a specific domain.
The critical role of-catenin in in vitro and in vivo metastatic activity
We tested whether -catenin is critical for metastatic activity induced by HSP60 overexpression. In vitro metastatic assays showed that siRNA mediated repression of endogenous -catenin in FADUHSP60--catenin-si clones decreased migration and invasion compared to the control FADUHSP60 clones (Figure 4B and C). In vivo metastatic assays also showed the decreased metastatic activity in FADUHSP60--catenin-si clones compared to the wild type FADUHSP60 clone (Figure 4D). Using siRNA to repress the endogenous expression of-catenin in FADU parental cells also decreased the migration and invasion activity (Supplementary Figure S3 is available at Carcinogenesis online). These results demonstrated that -catenin is critical for HSP60 mediated in vitro and in vivo metastatic activity.
Induction of -catenin levels by HSP60 does not require proteosomal activity
To investigate the mechanism of -catenin activation by HSP60, co-immunoprecipitation experiments showed that overexpression of HSP60 did not decrease the binding affinity
of HSP60 into 293T cells in the presence or absence of proteasome inhibitors were also performed. The results showed that HSP60 overexpression was not able to prevent ubiquitination of endogenous -catenin in the presence of proteasome inhibitors (MG-132 or LLnL) (Figure 5B and D). HSP60 actually increased the HA-ubiquitinated-catenin levels when HA-ubiquitin was co-transfected with HSP60 in the presence of proteasome inhibitors (MG-132 or LLnL) (Figure 5C and E). To test whether the putative ubiquitinated -catenin (upper bands in Figure 5B and D) was not due to phosphorylation induced by HSP60, an in vitro dephosphorylation assay was performed under the similar conditions of Figure 5B. The results showed that the intensity of the putative ubiquitinated -catenin in Figure 5F (lower panel) was not decreased in the presence of phosphatase, in contrast to the positive control in which phosphorylated Akt proteins were effectively dephosphorylated under the same condition (upper panel, Figure 5F). All the results suggested that the induction of -catenin by HSP60 does not require proteosomal activity and HSP60 may increase the levels of ubiquitinated -catenin.
Co-expression of HSP60 and nuclear -catenin predicts a worse prognosis in
HNSCC patients
Due to the involvement of HSP60 and -catenin in metastasis (10, 12, 25-31), we evaluated the clinical significance of HSP60 overexpression and its association with nuclear -catenin expression in head and neck squamous cell carcinoma (HNSCC) patients. Tissue microarray-immunohistochemistry (TMA-IHC) analysis was performed in 58 sets of primary and metastatic HNSCC samples (for clinical information, Supplementary Table S5 is available at Carcinogenesis online). Increased HSP60
expression was identified in 21 (36.2%) of primary and 45 (77.6%) of metastatic HNSCC samples, whereas nuclear -catenin expression was found in 6 (10.3%) of primary and 38 (65.5%) of metastatic tumors (Figure 6A and B). The correlation of HSP60 overexpression and nuclear -catenin expression was not significant in primary tumors (P=0.657). However, a significant correlation between increased HSP60 expression and nuclear -catenin expression was identified in metatsatic tumor samples (P=0.003) (Supplementary Table S6 is available at Carcinogenesis online). We therefore investigated the clinical significance of co-expression of HSP60 and nuclear -catenin in the metastatic samples of HNSCC. Univariate survival analysis demonstrated that female gender, tumor location, overexpression of HSP60, and co-expression of HSP60/nuclear -catenin were significant prognostic factors (P = 0.03, 0.013, 0.013, 0.001, respectively; Supplementary Table S5 is available at Carcinogenesis online). Multivariate analysis showed that co-expression of HSP60 and nuclear -catenin was an independent factor for predicting prognosis of metastatic HNSCCs (Supplementary Table S7 is available at Carcinogenesis online). Co-expression of HSP60 and nuclear -catenin was associated with an advanced stage and visceral metastasis (i.e., lung metastasis in these cases) significantly (Supplementary Table S8 is available at Carcinogenesis online). These results indicated that co-expression of HSP60 and nuclear -catenin was associated with a more aggressive tumor behavior and predicted a worse prognosis of metastatic HNSCC patients. Activation of canonical WNT/-catenin signaling is reported in certain HNSCC cases with WNT-1 as the most frequent WNT family protein involved (43). To investigate whether HSP60 overexpression represents an alternative pathway to activate
metastatic HNSCC samples as described above. The result showed that WNT-1 was overexpressed in 26 (44.8%) of primary and 31 (53.4%) of metastatic HNSCCs and increased WNT-1 expression was significantly associated with nuclear -catenin expression in both primary and metastatic samples (P=0.006, <0.001, Supplementary Table S6 is available at Carcinogenesis online). (For staining of WNT-1 in a representative case, Supplementary Figure S4A is available at Carcinogenesis online). However, there was no correlation between the expression levels of HSP60 and WNT-1 in the primary or metastatic HNSCCs (P=0.820, 0.549; Supplementary Figure S4B is available at Carcinogenesis online). These results indicated that HSP60 overexpression may be an alternative pathway to achieve -catenin activation, which is consistent with the results from cell lines overexpressing HSP60.
Discussion
Collectively, our results demonstrated that overexpression of HSP60 promotes metastasis, which is consistent with the observation of increased HSP60 expression in certain types of human cancers with metastasis (10, 12). The possible mechanism involves the activation of -catenin through interacting with its armadillo 7-9 repeats followed by enhancement of -catenin transcriptional activity and activation of its downstream targets (Figures 2 and 3). The mechanism of activation did not require proteosomal activity since HSP60 overexpression was not able to prevent ubiquitination of -catenin in the presence of proteasome inhibitors or to disrupt the binding of destabilization partners to catenin (Figure 5). Co-expression of HSP60 and nuclear -catenin predicts a worse prognosis in metastatic HNSCC patients, which is consistent with the role of -catenin and its downstream targets (e.g. MT1-MMP, etc) to promote metastasis (20, 21, 31). Activation of c-MYC in the transcriptional level by -catenin activation (Figure 3B) could be induced by overexpression of HSP60, which was shown to be a putative c-MYC target gene (13-16). Our recent results also showed that c-MYC directly activates HSP60 expression through an E-box site located in the proximal promoter of the HSP60 gene (45). In this scenario, HSP60 activation by c-MYC could provide positive autoregulation of c-MYC activity.
Our data suggest a novel role of HSP60 in metastasis (through activating -catenin) since HSP60 was mostly shown to be involved in innate immunity and cardiac diseases (5, 6). In addition, these results provide an alternative mechanism of activating
increased -catenin levels without mutations in AXIN1, APC, or GSK-3 in certain types of human cancers (e.g. abnormal accumulation of -catenin in oral squamous cell carcinoma and pituitary adenoma) (46, 47). However, the possibility of mutated -catenin in these cancers (including HNSCCs) was not tested. Several types of metastatic cancers (colorectal, lung, prostate, ovarian, etc) showed increased catenin levels (25-28). -catenin activation was shown to cause invasion (29) and was related to epithelial-mesenchymal transition, a mechanism contributing to metastasis (30). Some of the major targets of -catenin are responsible for inducing metastasis (31). Our results supported the the critical role of -catenin in metastasis as shown by published results (25-31). Co-expression pattern of HSP60 and-catenin in the HNSCC samples also supported the alternative mechanism of -catenin activation by HSP60 overexpression (Figure 6). Taken together, our results implicate the novel role of HSP60 in metastasis and provide an alternative mechanism of -catenin activation.
Acknowledgements
We are grateful to R. Parsons and N. Raab-Traub for the generous gifts of -catenin and TOPFlash plasmids. We greatly appreciate T.Y. Chou and W.Y. Li of the Department of Pathology, Taipei Veterans General Hospital for providing expert opinions on pathology reading. We thank S.H. Chiou and K.W. Chang of theInstitutes of Clinical Medicine and Oral Biology, National Yang-Ming University for technical assistance. This work was supported in part by National Science Council (NSC97-2311-B-010-007, NSC97-2320-B-010-029)(K.J.W.) & (NSC95-2314-B-075-083)(M.H.Y.), National Research Program for Genomic Medicine (DOH97-TD-G-111-038, DOH98-TD-G-111-027)(K.J.W.), Taipei Veterans General Hospital (V97ER2-008)(M.H.Y.), (V97ER2-013, V98ER2-009)(S.Y.C.), a grant from Ministry of Education, Aim for the Top University Plan (97A-C-T509, 98A-C-T509)(K.J.W.) and National Health Research Institutes (NHRI-EX97-9611BI, NHRI-EX98-9611BI) (K.J.W.).
References
1. Bukau, B. and Horwich, AL. (1998) The Hsp70 and Hsp60 chaperone machines. Cell, 92, 351-366.
2. Sigler, P.B., Xu, Z., Rye, H.S., Burston, S.G., Fenton, W.A. and Horwich, A.L. (1998) Structure and function in GroEL-mediated protein folding. Annu. Rev. Biochem., 67, 581-608.
3. Richardson, A., Landry, S.J. and Georgopoulos, C. (1998) The ins and outs of a molecular chaperone machine. Trends Biochem. Sci., 23, 138-143.
4. Nielson, K.L. and Cowan, N.J. (1998) A single ring is sufficient for productive chaperonin-mediated folding in vivo. Mol. Cell, 2, 93-99.
5. Hibach, C. and Burkart,V. (2007) Heat shock protein 60: regulatory role on innate immune cells. Cell. Mol. Life Sci. 64, 742-751.
6. Shamaei-Tousi, A., Halcox, J.P. and Henderson, B. (2007) Stressing the obvious? Cell stress and cell stress proteins in cardiovascular disease. Cardiovascular Res.,
74, 19-28.
7. Barazi, H.O., Zhou, L., Templeton, N.S., Krutzsch, H.C. and Roberts, D.D. (2002) Identification of heat shock protein 60 as a molecular mediator of alpha 3 beta 1 integrin activation. Cancer Res., 62, 1541-1548.
8. Crockett, D.K., Lin, Z., Elenitoba-Johnson, K.S. and Lim, M.S. (2004) Identification of NPM-ALK interacting proteins by tandem mass spectrometry. Oncogene, 23, 2617-2629.
(2004) Immunohistochemical detection of HSP60-expression in human ovarian cancer. Correlation with survival in a series of 247 patients. Anticancer Res., 19, 2141-2146.
10. Piselli, P., Vendetti, S., Vismara, D., Cicconi, R., Poccia, F., Colizzi, V, and Delpino, A. (2000) Different expression of CD44, ICAM-1, and HSP60 on primary tumor and metastases of a human pancreatic carcinoma growing in scid mice. Anticancer Res., 20, 825-831.
11. Cappello, F., Rappa, F., David, S., Anzalone, R. and Zummo, G. (2003) Immunohistochemical evaluation of PCNA, p53, HSP60, HSP10 and MUC-2 presence and expression in prostate carcinogenesis. Anticancer Res., 23, 1325-1331. 12. Cappello, F., David, S., Rappa, F., Bucchieri, F., Marasa, L., Bartolotta, T.E.,
Farina, F. and Zummo, G. (2005) The expression of HSP60 and HSP10 in large bowel carcinomas with lymph node metastases. BMC Cancer, 5, 139.
13. Coller, A.H., Grandori, C., Tamayo, P., Colbert, T., Lander, E.S., Eisenman, R.N. and Goloub, T.R. (2000) Expression analysis with oligonucleotide microarrays reveals that MYC regulates genes involved in growth, cell cycle, signaling, and adhesion. Proc. Natl. Acad. Sci. USA, 97, 3260-3265.
14. Menssen, A. and Hermeking, H. (2002) Characterization of the c-MYC-regulated transcriptome by SAGE: identification and analysis of c-MYC target genes. Proc. Natl. Acad. Sci. USA, 99, 6274-6279.
15. O’Connell, B.C., Cheung, A.F., Simkevich, C.P., Tam, W., Ren, X., Mateyak, M.K., and Sedivy, J,M. (2003) A large scale genetic analysis of c-myc-regulated gene expression patterns. J. Biol. Chem. 278, 12563-12573.
16. Fernandez, P.C., Frank, S.R., Wang, L., Schroeder, M., Liu, S., Greene, J., Cocito, A. and Amati, B. (2003) Genomic targets of the human c-Myc protein. Genes & Dev., 17, 1115-1129.
17. Morin, P.J. (1999) beta-catenin signaling and cancer. Bioessays, 21, 1021-1030. 18. Moon, R.T., Kohn, A.D., De Ferrari, G.V. and Kaykas, A. WNT and beta-catenin
signalling: diseases and therapies. Nat. Rev. Genet., 5, 691-701.
19. Clevers, H. (2006) Wnt/beta-catenin signaling in development and disease. Cell,
127, 469-80.
20. He, T.C., Sparks, A.B., Rago, C., Hermeking, H., Zawel, L., da Costa, L.T., Morin, P.J., Vogelstein, B. and Kinzler, K.W. (1998) Identification of c-MYC as a target of the APC pathway. Science, 281, 1509-1512.
21. Tetsu, O. and McCormick, F. (1999) Beta-catenin regulates expression of cyclin D1 in colon carcinoma cells. Nature, 398, 422-426.
22. Hoschuetzky, H., Aberle, H. and Kemler, R. (1994) Beta-catenin mediates the interaction of the cadherin-catenin complex with epidermal growth factor receptor. J. Cell Biol., 127, 1375-1380.
23. Playford, M.P., Bicknell, D., Bodmer, W.F. and Macaulay, V.M. (2000) Insulin-like growth factor 1 regulates the location, stability, and transcriptional activity of beta-catenin. Proc. Natl. Acad. Sci. USA, 97, 12103-12108.
24. Morrison, J.A., Klingelhutz, A.J. and Raab-Traub, N. (2003) Epstein-Barr virus latent membrane protein 2A activates beta-catenin signaling in epithelial cells. J. Virol., 77, 12276-12284.
Cytoplasmic and/or nuclear staining of beta-catenin is associated with lung metastasis. Clin. Exp. Metastasis, 20, 525-529.
26. Baldus, S.E., Monig, S.P., Huxel, S., Landsberg, S., Hanisch, F.G., Engelmann, K., Schneider, P.M., Thiele, J., Holscher, A.H. and Dienes, H.P. (2004) MUC1 and nuclear beta-catenin are coexpressed at the invasion front of colorectal carcinomas and are both correlated with tumor progression. Clin. Cancer Res., 10, 2790-2796. 27. Chen, G., Shukeir, N., Potti, A., Sircar, K., Aprikian, A., Goltzman, D. and Rabbani,
S.A. (2004) Up-regulation of Wnt-1 and beta-catenin production in patients with advanced metastatic prostate carcinoma: potential pathogenetic and prognostic implications. Cancer, 101, 1345-1356.
28. Imai, T., Horiuchi, A., Shiozawa, T., Osada, R., Kikuchi, N., Ohira, S., Oka, K. and Konishi, I. (2004) Elevated expression of E-cadherin and alpha-, beta-, and gamma-catenins in metastatic lesions compared with primary epithelial ovarian carcinomas. Hum. Pathol., 35, 1469-1476.
29. Cheon, S.S., Cheah, A.Y., Turley, S., Nadesan, P., Poon, R., Clevers, H. and Alman, B.A. (2002) b-catenin stabilization dysregulates mesenchymal cell proliferation, motility, and invasiveness and causes aggressive fibromatosis and hyperplastic cutaneous wounds. Proc. Natl. Acad. Sci. USA 99, 6973-6978.
30. Medici, D., Hay, E.D. and Olsen, B.R. (2008) Snail and Slug promote epithelial-mesenchymal transition through beta-catenin-T-cell fctor-4-dependent expression of transforming growth factor-beta3. Mol. Biol. Cell 19, 4875-4887.
31. Hlubek, F., Spaderna, S., Jung, A., Kirchner, T. and Brabletz, T. (2004) -catenin activates a coordinated expression of the proinvasive factors laminin-5 2 chain and
MT1-MMP in colorectal carcinomas. Int. J. Cancer, 108, 321-326.
32. Tian, Q., Feetham, M.C., Tao, W.A., He, X.C., Li, L., Aebersold, R. and Hood, L. (2004) Proteomic analysis identifies that 14-3-3z interacts with beta-catenin and facilitates its activation by Akt. Proc. Natl. Acad. Sci. USA, 101, 15370-15375. 33. Akiyama, H., Lyons, J.P., Mori-Akiyama, Y., Yang, X., Zhang, R., Zhang, Z., Deng,
J.M., Taketo, M.M., Nakamura, T., Behringer, R.R., McCrea, P.D. and de Crombrugghe, B. (2004) Interactions between Sox9 and beta-catenin control chondrocyte differentiation. Genes & Dev., 18, 1072-1087.
34. Kaidi, A., Williams, A.C. and Paraskeva, C. (2007) Interaction between beta-catenin and HIF-1 promotes cellular adaption to hypoxia. Nat. Cell Biol., 9, 210-217.
35. Yang, M.H., Chang, S.Y., Chiou, S.H., Liu, C.J., Chi, C.W., Chen, P.M., Teng, S.C. and Wu, K.J. (2007) Overexpression of NBS1 induces epithelial-mesenchymal transition and co-expression of NBS1 and Snail predicts metastasis of head and neck cancer. Oncogene, 26, 1459-1467.
36. Yang, M.H., Wu, M.Z., Chiou, S.H., Chang, S.Y., Chen, P.M., Liu, C.J., Teng, S.C. and Wu, K.J. (2008) Direct regulation of TWIST by hypoxia-inducible factor-1 (HIF-1) promotes metastasis. Nat. Cell Biol., 10, 295-305.
37. Wu, K.J., Polack, A. and Dalla-Favera, R. (1999) Coordinated regulation of iron controlling genes, H-ferritin and IRP2, by c-MYC. Science, 283, 676-679.
38. Teng, S.C., Chen, Y.Y., Su, Y.N., Chou, P.C., Chiang, Y.C., Tseng, S.F. and Wu, K.J. (2004) Direct activation of HSP90A transcription by MYC contributes to c-MYC mediated transformation. J. Biol. Chem., 279, 14649-14655.
39. Chiang, Y.C., Teng, S.C., Su, Y.N., Hsieh, F.C. and Wu, K.J. (2003) c-Myc directly regulates the transcription of the NBS1 gene involved in DNA double-strand break repair. J. Biol. Chem., 278, 19286-19291.
40. Chen, Y.C., Hsu, H.S., Chen, Y.W., Tsai, T.H., How, C.K., Wang, C.Y., Hung, S.C., Chang, Y.L., Tsai, M.L., Lee, Y.Y., Ku, H.H. and Chiou, S.H. (2008) Oct-4 expression maintained cancer stem-like properties in lung cancer-derived CD133-positive cells. PLoS ONE 3, e2637.
41. Chen, Y.C., Chiang, H.Y., Yang, M.H., Chang, S.Y., Teng, S.C., Vanhaesebroeck, B, and Wu, K.J. (2008) Activation of phosphoinositide 3-kinase by the NBS1 DNA repair protein. J. Mol. Med., 86, 401-412.
42. Urushibara, M., Kageyama, Y., Akashi, T., Otsuka, Y., Takizawa, T., Koike, M. and Kihara, K. (2007) HSP60 may predict good pathological response to neoadjuvant chemoradiotherapy in bladder cancer. Jpn. J. Clin. Oncol., 37, 56-61
43. Ishida, K., Ito, S., Wada, N., Deguchi, H., Hata, T., Hosoda, M. and Nohno, T. (2007) Nuclear localization of beta-catenin involved in precancerous change in oral leukoplakia. Mol. Cancer, 6, 62.
44. Zhang, W.M., Lo Muzio, L., Rubini, G. and Yang, G. (2005) Effect of WNT-1 on beta-catenin expression and its relation to Ki-67 and tumor differentiation in oral squamous cell carcinoma. Oncol. Rep., 13, 1095-1099.
45. Tsai, Y.P., Teng, S.C. and Wu, K.J. (2008) Direct regulation of HSP60 expression by c-MYC induces transformation. FEBS Letters, 582, 4083-4088.
46. Iwai, S., Katagiri, W., Kong, C., Amekawa, S., Nakazawa, M. and Yura, Y. (2006) Mutations of the APC, beta-catenin, and axin 1 genes and cytoplasmic
accumulations of beta-catenin in oral squamous cell carcinoma. J. Cancer Res. Clin. Oncol., 131, 773-782.
47. Sun, C., Yamato, T., Kondo, E., Furukawa, T., Ikeda, H. and Horii, A. (2005) Infrequent mutation of APC, AXIN1, and GSK3beta in human pituitary adenoma with abnormal accumulation of CTNNB1. J. Neurooncol., 73, 131-134.
Figure legends
Fig. 1. HSP60 overexpression increases migration, invasion and metastasis of FADU cells. (A) Western blot analysis of stable clones overexpressing HSP60. (B) & (C) The
number of FADUHSP60 vs. FADU control cells which migrated (B) or invaded (C) across the membrane counted per high power field. Representative pictures of different clones that migrated or invaded across the Transwell membrane to the other side of the filter were shown. (D) Photographic pictures of the lungs of NOD-SCID mice 6 weeks after tail-vein injection of FADUHSP60 or FADU control cells (upper panels) and H&E staining of lung tissues of mice (lower panels). Black arrows indicated the metastatic nodules. The photographs were taken at the magnification of 40x. (E) Quantification of the average numbers of metastatic foci in the lungs of mice. (F) Quantification of the tumor volume of the metastatic nodules in the lungs of mice. The asterisk (*) indicated statistical significance (p<0.05) between control (FADUcDNA3) and experimental (FADUHSP60-1, 2) clones.
Fig. 2. Interaction between -catenin and HSP60 in FADUHSP60 cells as well as in
293T cells overexpressing both proteins, domain mapping, immunofluorescence co-localization experiments, and GST pull down assays. (A) Co-immunoprecipitation
using extracts from FADU control or FADUHSP60 cells showed the interaction between HSP60 and -catenin. The panel also showed that overexpression of HSP60 in FADU cells increased the protein levels of catenin. (B) & (C) Co-immunoprecipitation experiments showed the interaction between HSP60 and -catenin in 293T cells
overexpressing both proteins using an anti--catenin or an anti-HSP60 antibody. (D) Domain mapping experiments showed that interaction of the apical domain (a.a. 288-383) of HSP60 with -catenin. (E) Domain mapping experiments showed that the 7-9 armadillo repeats (a.a. 400-530) of -catenin interacted with HSP60. (F) Indirect immunofluorescence experiments showed the translocation of -catenin into nucleus and nuclear co-localization of HSP60 and -catenin in 293T cells overexpressing HSP60. Green fluorescence: -catenin; red fluorescence: HSP60. (G) GST pull down assay using the control GST, GST-HSP60-192-287, GST-HSP60-288-383 fragments to pull down the purified -catenin proteins (His--catenin-400-530 or His--catenin-531-781). The lower panel (Coomassie blue staining) showed the purified GST, HSP60-192-287, GST-HSP60-288-383 fragments. His--catenin-400-530 represented the -catenin domain which could interact with HSP60, whereas His--catenin-531-781 protein was used as a control.
Fig. 3. Activation of -catenin and its downstream targets, and enhancement of the
transcriptional activity of -catenin by HSP60. (A) Domain mapping experiments showed that the apical domain (a.a. 288-383) of HSP60 was responsible for the activation of-catenin. (B) Activation of -catenin downstream targets (cyclin D1, c-MYC, laminin-2, MT1-MMP) in FADUHSP60 clones. (C) Enhanced transcriptional activity of -catenin by HSP60 but not by a non-interacting HSP60 mutant in a TOPFlash system.
against -catenin versus the control FADUHSP60 clone. (B) & (C) Migration and invasion activity of FADUHSP60--catenin-si clones vs. FADUHSP60 control clones. Upper subpanels indicated representative pictures of different clones that migrated (B) or invaded (C) across the Transwell membrane to the other side of the filter. (D) In vivo tail vein metastasis assays showed that FADUHSP60--catenin-si clones had decreased metastatic activity compared to the FADUHSP60 control clone. The asterisk (*) indicated statistical significance (p<0.05) between control (FADUHSP60-SUPER) and experimental (FADUHSP60--catenin-si) clones.
Fig. 5. The ability of HSP60 to increase -catenin protein levels does not require
proteosomal activity and HSP60 increases ubiquitinated -catenin levels to enhance
transactivation activity. (A) Co-immunoprecipitation assays using an anti-AXIN1
antibody to pull down -catenin and associated proteins showed that the HSP60 did not decrease the binding affinity of AXIN1 and GSK-3 to -catenin (right subpanel). The left subpanel showed the levels of different proteins. (B) HSP60 overexpression in 293T cells did not decrease the ubiquitination levels of endogenous -catenin in the presence of proteasome inhibitors (MG-132). The longer exposure picture was put below the upper lane with shorter exposure. (C) Overexpression of HA-ubiquitin together with HSP60 in 293T cells increased the levels of endogenous HA-ubiquitinated -catenin in the presence of MG-132 by co-immunoprecipitation assays. The longer exposure picture was put below the upper lane with shorter exposure. (D) HSP60 overexpression in 293T cells did not decrease the ubiquitination levels of endogenous -catenin in the presence of
with shorter exposure. (E) Overexpression of HA-ubiquitin together with HSP60 in 293T cells increased the levels of endogenous HA-ubiquitinated -catenin in the presence of LLnL by co-immunoprecipitation assays. The longer exposure picture was put below the upper lane with shorter exposure. (F) In vitro dephosphorylation assays showed that the upper bands of -catenin induced by HSP60 were not due to phosphoryaltion. Upper panel: control experiment using phosphorylated Akt; lower panel: immunoprecipitated -catenin in the absence or presence of HSP60. CIP: calf intestinal phosphatase.
Fig. 6. Co-expression of HSP60 and nuclear -catenin predicts a worse prognosis of
metastatic HNSCC patients. Overexpression of HSP60 is associated with expression of
nuclear -catenin in metastatic HNSCC. (A) Immunohistochemistry analysis of HSP60 and -catenin expression in corresponding NCMT (N), primary tumor (T) and metastatic tumor (M) of a representative HNSCC case. The samples prepared for co-expression analysis were cut and examined at the same region. The photographs were taken at the magnification of 200x. No detectable HSP60 expression was noted in N; whereas cytoplasmic expression of HSP60 (red arrows) was noted in T, and nucleo-cytoplasmic expression of HSP60 was shown in M (red arrows, cytoplasmic expression; black arrows, nuclear expression). Membranous expression of -catenin (blue arrows) was identified in N & T; whereas nuclear expression of HSP60 and -catenin (black arrows) was noted in M. (B) Percentage of overexpression of HSP60 and nuclear -catenin in primary (T) and metastatic tumors (M). (C) Kaplan-Meier analysis of overall survival in 58 metastatic HNSCC cases. Co-expression of HSP60 and nuclear -catenin is associated with a worse prognosis. The scale bars represent 100 m in each panel (white bars).
