Oncogene (2010) 29, 5204–5213 & 2010 Macmillan Publishers Limited All rights reserved 0950-9232/10 www.nature.com/onc ORIGINAL ARTICLE Heat-shock HSF1 has a critical role in human epidermal growth factor -2-induced cellular transformation and tumorigenesis

L Meng, VL Gabai and MY Sherman

Department of Biochemistry, Boston University School of Medicine, Boston, MA, USA

The heat-shock transcription factor HSF1 was recently Introduction shown to have a key role in the development of tumors associated with activation of Ras or inactivation of . Induction of heat-shock (Hsp)72, Hsp27 and Here, we show that HSF1 is required for the cell other Hsps represents a highly conserved protective transformation and tumorigenesis induced by the human mechanism against environmental insults, which is epidermal growth factor receptor-2 (HER2) oncogene predominantly regulated by the heat-shock transcription responsible for aggressive breast tumors. Upon expression factor HSF1 (Lindquist and Craig, 1988; Sorger, 1991; of HER2, untransformed human mammary epithelial Morimoto et al., 1992). Recent reports showed that MCF-10A cells underwent neoplastic transformation, HSF1 has an essential role in the development of formed foci in culture and tumors in nude mouse lymphomas in p53-deficient mice and of carcinomas in xenografts. However, expression of HER2 in MCF-10A Ras tumor model (Dai et al., 2007; Min et al., 2007). cells with knockdown of HSF1 did not cause either foci Moreover, HSF1 was shown to support proliferation of formation or tumor growth in xenografts. The antitumori- several cell lines (Dai et al., 2007), but specific genic effect of downregulation of HSF1 was associated mechanisms of these effects have not been explored. with HER2-induced accumulation of the cyclin-dependent Normal tissues or non-transformed cells usually kinase inhibitor p21 and decrease in the mitotic regulator express relatively low levels of Hsp72 and Hsp27, and survivin, which resulted in growth inhibition and cell these are strongly upregulated following stress- senescence. In fact, either knockout of p21 or over- ful treatments in an HSF1-dependent manner (Jolly and expression of survivin alleviated these effects of HSF1 Morimoto, 2000). On the other hand, many human knockdown. The proliferation of certain human HER2- tumors and cancer cell lines express Hsp72 or/and positive breast cancer lines also requires HSF1, as its Hsp27 at elevated levels constitutively (Jaattela, 1999; knockdown led to upregulation of p21 and/or decrease in Jolly and Morimoto, 2000; Mosser and Morimoto, survivin, precipitating growth arrest. Similar effects were 2004; Calderwood et al., 2006). High levels of Hsp72 observed with a small-molecular-weight inhibitor of the and Hsp27 correlate with metastasis, resistance to heat-shock response NZ28. The effects of HSF1 knock- anticancer drugs and poor prognosis in many human down on the growth arrest and senescence of HER2- (Ciocca and Calderwood, 2005). Previous results expressing cells were associated with downregulation of from this laboratory indicated that certain cancer cells heat-shock protein (Hsp)72 and Hsp27. Therefore, HSF1 become ‘addicted’ to high levels of Hsp72, as specific is critical for proliferation of HER2-expressing cells, most depletion of Hsp72 rapidly suppressed cell proliferation likely because it maintains the levels of HSPs, which in and precipitated senescence in several cancer lines, but not turn control regulators of senescence p21 and survivin. in untransformed epithelial cells (Yaglom et al.,2007; Oncogene (2010) 29, 5204–5213; doi:10.1038/onc.2010.277; Gabai et al., 2009). Interestingly, in different cancer cells published online 12 July 2010 distinct senescence signaling pathways were activated upon depletion of Hsp72, depending on the presence of Keywords: HSF1; HER2; growth arrest; senescence; either phosphatidylinositol 3-kinase, catalytic, alpha p21; survivin polypeptide or Ras oncogenes (Gabai et al., 2009). Constitutively elevated expression of Hsp72 and/or Hsp27 in cancers is often due to the HSF1-dependent mechanisms (although sometimes it is HSF1-indepen- dent) (Cai and Zhu, 2003; Wu et al., 2005). However, the role of HSF1 in cancers may not be limited to regulating the expression of HSPs, as this transcription factor also Correspondence: Dr MY Sherman or Dr VL Gabai, Department of controls expression of many other proteins (Cahill et al., Biochemistry, Boston University School of Medicine, 72 East Concord 1996, 1997; Singh et al., 2000, 2002; Xie et al., 2002a, b). Street K-323, Boston, MA 02118, USA. E-mails: [email protected] or [email protected] Human epidermal growth factor receptor-2 (HER2) is Received 13 August 2009; revised 25 May 2010; accepted 9 June 2010; a member of the epidermal growth factor receptor published online 12 July 2010 tyrosine kinase family, which is overexpressed in HSF1 is essential for HER2-induced transformation L Meng et al 5205 25–30% of human breast and ovarian cancers (Moasser, led to merely mild upregulation of vimentin, and foci 2007). Besides activating mitogenic pathways, HER2 formation was inhibited (Figure 1b and c), indicating can upregulate growth-inhibitory/senescence signaling that HSF1 is required for HER2-induced transforma- (Trost et al., 2005), suggesting that malignant transfor- tion in vitro. Of note, in contrast to late-passage cells, mation needs not only oncogene activation but also a infection with HER2-expressing retrovirus of early- shutdown of senescence pathways. Here, in addressing passage MCF-10A cells did not cause transformation the role of HSF1 in HER2-induced tumorigenesis, we and triggered senescence even without HSF1 depletion discovered that HSF1 controls senescence signaling (not shown). initiated by HER2. If infection with shHSF1 retrovirus followed infection with HER2 retrovirus, downregulation of HSF1 still triggered senescence even in fully transformed MCF- Results 10A cells (Supplementary Figure S1), indicating that HSF1 is necessary for the maintenance of growth of Depletion of HSF1 inhibits the transformation ability HER2-transformed cells. of HER2 To address the role of HSF1 in tumorigenesis in vivo, To investigate the role of HSF1 in neoplastic transfor- HER2-expressing control and shHSF1 cells were tested mation, we depleted HSF1 from breast epithelial MCF- for their ability to form tumors in xenografts. Control 10A cells, and then introduced the HER2 oncogene. HER2-expressing cells formed noticeable tumors at Briefly, late-passage MCF-10A cells were infected with the injection site by day 3 post-injection (Figure 1d). control retrovirus or retrovirus expressing the small However, HER2-expressing cells depleted of HSF1 hairpin RNA (shRNA) against HSF1 (shHSF1), fol- could not form tumors at least until day 25. Two lowed by puromycin selection, which resulted in more tumors appeared at days 26 and 35 post-injection, and than 95% depletion of HSF1 (Figure 1a). These these tumors were much smaller. Importantly, in both treatments did not affect cell viability, and only slightly tumors control of HSF1 expression by shRNA was lost, diminished growth (see below in Figure 2a), further and levels of HSF1 partially restored (Supplementary indicating that HSF1 is dispensable for growth of Figure S2), further supporting that in the absence of untransformed cells. Then, cells were infected with HSF1, HER2-induced tumors cannot be formed. retrovirus expressing the constitutively active HER2 mutant V664E, which led to increased expression of HER2 triggers growth arrest and senescence in cells with vimentin and formation of foci (Figures 1b and c), knockdown of HSF1 indicating that cells underwent neoplastic transforma- We investigated the effects of HSF1 on the growth rates tion. In contrast, expression of HER2 in shHSF1 cells of HER2-expressing cells. Control and shHSF1 cells

Cont. shHSF1 Cont. shHSF1 EV HER2EV HER2 HSF1 vimentin

β-actin β-actin

Cont. EV shHSF1 EV 1000 Cont.-1 Cont.-2 900 Cont.-3 Cont.-4 ) 800 3 shHSF1-1 700 shHSF1-2 600 shHSF1-3 shHSF1-4 500 Cont. HER2 shHSF1 HER2 400 300 Tumor volume (mm volume Tumor 200 100 0 0 5 10 15 20 25 30 35 Days after injection Figure 1 Human epidermal growth factor receptor-2 (HER2)-induced MCF-10A transformation requires HSF1. (a) HSF1 depletion by small hairpin RNA against HSF1 (shHSF1) virus. MCF-10A cells were infected with control (Cont.) or shHSF1 retroviruses, and following 3 days of selection, HSF1 levels were measured by immunoblotting. (b) Effect of HSF1 depletion on the HER2-induced vimentin expression. Control (Cont. and shHSF1 MCF-10A cells were infected with control empty vector (EV) or HER2- overexpressing (HER2) retroviruses, and vimentin levels were measured on day 4 post-infection. (c) Effect of HSF1 depletion on the HER2-induced foci formation. Cells described in b) were stained with hematoxylin to visualize the foci. Dark masses in the bottom left panel represent the foci. (d) HSF1 depletion suppresses tumor emergence in xenografts. Growth curves of individual tumor in xenografts of HER2-expressing Cont. and shHSF1 MCF-10A cells.

Oncogene HSF1 is essential for HER2-induced transformation L Meng et al 5206

1000000 Cont. EV Cont. EV shHSF1 EV Cont. HER2 shHSF1 EV shHSF1 HER2 100000

10000 Cont. HER2 shHSF1 HER2 Cell number per well Cell number

1000 2468 Days after infection

Cont. EV shHSF1 EV ** 80 70 60 50 40

Cont. HER2 shHSF1 HER2 cells -gal-positive 30 β 20 % of 10 0

Cont. EV Cont. HER2shHSF1 EV shHSF1 HER2 Figure 2 Human epidermal growth factor receptor-2 (HER2) triggers senescence in HSF1-deficient MCF-10A cells. (a) Growth curves of control (Cont.) and small hairpin RNA against HSF1 (shHSF1) MCF-10A cells with (HER2) or without (empty vector, EV) HER2 overexpression. (b) Representative morphology of MCF-10A cultures described in (a). Please note that the mass seen in the bottom left panel represents the focus. (c) b-galactosidase (b-gal) activity staining of MCF-10A populations described in (a) and its quantification.

were infected with HER2-expressing retrovirus or p21 is a mediator in HER2-induced senescence upon control retrovirus. On day 2 post-infection cells were depletion of HSF1 plated, and the cell number was counted every 48 h. p21Cip is the major regulator of senescence in cancer cells Knockdown of HSF1 caused slightly slower growth in (Roninson, 2003). Expression of HER2 leads to normal MCF-10A cells (Figure 2a). Similarly, a minor upregulation of p21 in MCF7 cells (Trost et al., 2005). reduction of growth rate was seen upon expression of Therefore, we tested for the role of p21 in senescence HER2. Importantly, when we expressed HER2 in caused by expression of HER2 in shHSF1 cells. p21 was HSF1-depleted cells, a significant growth inhibition mildly upregulated by HER2 expression in control cells, was observed (Figure 2a), associated with a significant and knockdown of HSF1 on its own also led to a similar change in cells’ appearance, including enlarged, flat- minor increase in p21 levels (Figure 3a). However, there tened morphology and extensive vacuolization reminis- was a marked induction of p21 upon expression of cent of senescence (Figure 2b). To assess senescence, HER2 in shHSF1 cells (Figure 3a). cells were stained for the acidic b-galactosidase (b-gal) To investigate the role of p21 in the onset of activity (Figure 2c). In control population, there was senescence under these conditions, wild-type and p21 only 1% of b-gal-positive cells, and in shHSF1 cells, this knockout MCF-10A cells (kind gift from Dr Park; fraction raised to 4%. HER2 expression in control Bachman et al., 2004) were infected with shHSF1 and MCF-10A cells led to a significant increase in b-gal- control retroviruses, followed by brief puromycin positive population (B30%). Importantly, expression of selection. Then, cells were infected with a virus expres- HER2 in shHSF1 MCF10A cells resulted in about 70% sing HER2, and plated to monitor senescence. In of b-gal-positive cells (Figure 2c). Therefore, HSF1 contrast to wild-type cells, in p21 knockout cells, the seems to be critical for growth upon expression of combination of shHSF1 and HER2 expression did not HER2 because it keeps HER2-induced growth arrest cause significant growth inhibition or senescence, as and senescence under control. judged by cell morphology (Figure 3b).

Oncogene HSF1 is essential for HER2-induced transformation L Meng et al 5207

Figure 3 p21 is involved in human epidermal growth factor receptor-2 (HER2)-induced senescence in HSF1-deficient cells. (a) p21 levels in control (Cont.) and small hairpin RNA against HSF1 (shHSF1) MCF-10A cells with (HER2) or without (empty vector, EV) HER2 overexpression. (b) Cell morphology in wild-type (WT) and p21 knockout (p21KO) MCF-10A cells with same retrovirus infections as in (a). Pictures were taken at day 4 post-infection. Whitish masses represent the foci. (c) WT and p21 knockout (KO) MCF-10A cells were treated with Cont. or shHSF1 retroviruses, followed by HER2 overexpression. Growth curves of xenograft tumors from these HER2-expressing cells were shown.

Further, we tested whether suppression of senescence in out was sufficient to restore tumor growth, indicating that the p21 knockout cells can restore tumor development in HER2-triggered senescence of HSF1-depleted cells repre- HSF1 knockdown cells. Control and p21 knockout cells sents the major barrier for tumor development. were depleted of HSF1, and then infected with HER2 retrovirus and injected into nude mice to establish Survivin is involved in HER2-induced senescence upon xenograft tumors. As seen in Figure 3c, in contrast to depletion of HSF1 wild-type cells, depletion of HSF1 had only a mild We noticed a decrease in survivin levels upon expression inhibitory effect on tumor development by p21 knockout of HER2 in shHSF1 cells (see below). Survivin, a cells. Therefore, suppression of senescence by p21 knock- member of the inhibitors of apoptosis (IAPs) family, is

Oncogene HSF1 is essential for HER2-induced transformation L Meng et al 5208 a critical regulator of proliferation (Altieri, 2003). Upon To test the role of survivin in HER2-induced HER2 expression, survivin levels were dramatically senescence, we infected control and shHSF1 cells with decreased in shHSF1 cells, but not in control cells a retrovirus expressing survivin, and following brief (Figure 4a). Knockout of p21 suppressed the decrease in selection, infected the cells with HER2-expressing virus. survivin after expression of HER2 in shHSF1 cells, Survivin overexpression restored growth and rescued indicating that downregulation of survivin is at least senescence phenotype in HSF1-knockdown cells upon partially p21-dependent (Figure 4a). On the other hand, HER2 expression, as judged by cell morphology and in other cell lines, the decrease in survivin levels was p21- b-gal activity staining (Figures 4b and c), indicating that independent (see below), indicating an alternative survivin is important for regulation of growth inhibition mechanism of the survivin control. and senescence under these conditions.

Figure 4 Survivin is another mediator in human epidermal growth factor receptor-2 (HER2)-induced senescence in HSF1-deficient cells. (a) Survivin levels in control or small hairpin RNA against HSF1 (shHSF1) wild-type (WT, upper panel) and p21 knockout (p21KO) (lower panel) MCF-10A cells with (HER2) or without (empty vector, EV) HER2 overexpression. (The numbers below each panel are the relative percentage of survivin levels). (b) Control and shHSF1 cells were infected with retrovirus expressing survivin (surv), and 2 days later, these cells were infected with HER2-expressing (HER2) retrovirus. EV1 – control empty vector for survivin; EV2 – control empty vector for HER2. Cells were stained for b-galactosidase (b-Gal) activity, and (c) quantification of b-gal staining was performed. ** indicates Po0.01.

Oncogene HSF1 is essential for HER2-induced transformation L Meng et al 5209 HSF1 is critical for proliferation of HER2-positive human Effects of HSF1 on HER2-induced transformation cancer cells may be related to expression of Hsps We further investigated the effects of HSF1 depletion on , one of the major HSF1 targets, has an important HER2-positive human cancer cell lines SK-BR-3 (having role in the activity and stability of cancer-related mutant p53) and MDA-MB-453 (having wild-type p53). signaling proteins (Pratt and Toft, 2003). Therefore, HSF1 was depleted in these cell lines using a retroviral the effects of HSF1 on tumorigenesis could be related to vector, as described above. In both cell lines, infection the control of Hsp90. However, we did not observe with shHSF1 retrovirus led to more than 80% depletion downregulation of major Hsp90 client proteins Akt, of HSF1 (Figures 5a and b). As with MCF-10A cells, a-MET and glycogen synthase kinase-3b, and activation HSF1 depletion in MDA-MB-453, which has normal p53 of ERK1/2 was not reduced in shHSF1 cells (Figure 6b). (Lacroix et al., 2006), led to induction of p21, as well as Therefore, the effects of HSF1 knockdown on HER2- slight downregulation of survivin (Figure 5a). induced growth inhibition and senescence are unrelated Knockdown of HSF1 in SK-BR-3 cells led to strong to downregulation of Hsp90. downregulation of survivin (Figure 5b), but no upregu- The control of senescence could be specifically linked lation of p21 (not shown). As SK-BR-3 cells have to the expression of Hsp72, as this Hsp regulates mutant p53 (Lacroix et al., 2006), these data indicate senescence signaling pathways (Yaglom et al., 2007; that the effect of HSF1 on survivin could be indepen- Gabai et al., 2009). Indeed, we observed significant dent of p53, whereas its effect on p21 appears to be p53- upregulation of both Hsp72 and Hsp27 following HER2 dependent. As expected, in both MDA-MB-453 and SK- expression, which was fully dependent on HSF1 BR-3 cells, alterations of survivin and/or p21 upon (Figure 6c). This finding is in line with previous study HSF1 depletion were associated with growth inhibition that demonstrated that HER2/HER3 ligand heregulin (Figures 5a and b, lower panels). Of note, these cell lines induces Hsp72 in an Hsf1-dependent manner (Khaleque did not show signs of senescence in spite of growth et al., 2005). Notably, there is no change in the levels of arrest. This observation suggests that in certain cancer -2 (Supplementary Figure S3). lines at least some components of the senescence We tested whether prevention of induction of Hsp72 program are defective. Overall, these data indicate that and/or Hsp27 is involved in the HER2-triggered in HER2-positive human cell lines, HSF1 serves an senescence. Infection with shHsp72 or shHsp27 retro- important function in supporting cell proliferation by viruses brought about depletion of these chaperones by regulation of p21 and/or survivin. more than 70% (Figure 6d). No significant effect on either cell morphology or growth was seen under these conditions in untransformed MCF-10A cells Heat-shock response inhibitor NZ28 suppresses (Figure 6e). However, subsequent expression of HER2 HER2-induced transformation rapidly triggered senescence (Figure 6e). Therefore, lack NZ28, a small-molecular-weight inhibitor of heat-shock of Hsp72 or Hsp27 is sufficient for senescence triggered response that we recently developed (Zaarur et al., by HER2, which may explain the effects of depletion of 2006), was used to further address the role of HSF1 in HSF1 on HER2-triggered senescence. HER2-induced cell transformation. MCF-10A cells were infected with HER2-expressing retrovirus, and NZ28 was administered to cells 2 days after infection. Incubation with 1 mM of NZ28 for 3 days led to Discussion reduction of the Hsp72 levels in HER2-expressing cells (Figure 5c), which were associated with growth inhibi- Here we discovered that HSF1 has a critical role in tion (see below). NZ28 treatment also significantly HER2-induced tumorigenesis in mammary epithelial suppressed foci formation in HER2-overexpressing cells. In fact, depletion of HSF1 strongly suppressed MCF-10A cells (Figure 5c). In fact, in control untrans- both transformation judged by foci formation in culture formed cells, 1 mM of NZ28 caused only minor growth and tumor development in xenografts. The suppression inhibition, whereas in HER2-expressing cells, the of transformation was associated with a paradoxical inhibition was more than 70% (Figure 5c, right panel). effect of HER2 on cells that lacked HSF1. Upon Importantly, treatment with NZ28 led to a strong expression of HER2, instead of transformation these downregulation of survivin (Figure 5c), whereas the p21 cells stopped proliferating and underwent senescence. levels were not significantly changed (not shown), also These data strongly suggest that HER2 activates both pointing toward additional effects of this drug. mitogenic and senescence signals, and HSF1 serves in In HER2-positive human cancer lines, the effects of suppression of senescence. Accordingly, in cells that lack incubation with NZ28 (5 mM) for 3 days, in general, were HSF1, senescence signals prevail and proliferation similar to the effects of HSF1 depletion. Indeed, we stops. observed strong downregulation of Hsp72, a decrease of In part, growth inhibition signaling was owing to the HSF1 levels, downregulation of survivin and growth cell cycle inhibitor p21. HER2 was able to mildly inhibition (Figure 5d). Overall, these results indicate upregulate p21 even in naı¨ve cells. However, p21 that inhibition of HSF1 or its knockdown can prevent induction was much stronger in cells with depleted HER2-induced cell transformation and suppress HSF1. Furthermore, p21 knockout suppressed the proliferation of HER2-expressing tumor cells. senescence phenotype triggered by HER2.

Oncogene HSF1 is essential for HER2-induced transformation L Meng et al 5210

Figure 5 HSF1 is critical for proliferation of human epidermal growth factor receptor-2 (HER2)-positive human cancer cells. MDA- MB-453 (a) and SK-BR3 (b) cells were infected with small hairpin RNA against HSF1 (shHSF1) retrovirus, and selected with puromycin. The levels of HSF1, p21 and survivin were assessed by immunoblotting on day 10 post-infection (a, b, upper panels). Cell number was assessed starting on day 7 post-infection (a, b, lower panels). (c) MCF-10A cells were infected with HER2-expressing retrovirus, selected for 2 days and incubated with NZ28 for additional 2 days. Then, the levels of heat-shock protein (Hsp72) and survivin, and foci formation ability were assessed. Dark masses represent foci stained with hematoxylin. Effects of NZ28 on growth rate of MCF-10A cells were also determined (right panel). (d) MDA-MB-453 and SK-BR3 cells were incubated with 2 and 5 mM NZ28 for 3 days and the levels of HSF1, Hsp72 and survivin were measured by immunoblotting (upper panels). Effect of NZ28 on growth of SK-BR3 cells was determined by cell counting with hemocytometer (lower panel).

Interestingly, in addition to p21, which is a known suppress apoptosis (Altieri, 2003a). In addition, survivin player in the oncogene-induced senescence, we uncov- regulates mitosis by modulation of aurora B kinase ered a novel role of survivin in this process. Survivin is (Altieri, 2008). Here, we found that survivin was homologous to IAPs, and originally was found to downregulated in cells with HSF1 knockdown upon

Oncogene HSF1 is essential for HER2-induced transformation L Meng et al 5211

Figure 6 Effects of HSF1 on human epidermal growth factor receptor-2 (HER2)-induced transformation may be related to the expression of heat shock proteins (Hsps). (a) Control and small hairpin RNA against HSF1 (shHSF1) MCF-10A cells were infected with control (empty vector, EV) or HER2-expressing retroviruses. The protein level of Hsp90 was measured. (b) Hsp90 client proteins and pERK levels were monitored in cells described in (a). (c) The levels of Hsp72 and Hsp27 were examined by immunoblotting in cells described in (a). (d) Efficiency of knockdown of Hsp27 or Hsp72 by shHsp27 or shHsp72 retroviruses in MCF-10A cells, and (e) the development of senescence in shHsp27 or shHsp72 cells following HER2 overexpression was assessed by b-galactosidase activity staining. ** indicates Po0.01. expression of HER2, and, importantly, ectopic expres- knockout cells. On the other hand, the survivin decrease sion of survivin suppressed senescence under these was partially p21-independent, indicating that a distinct conditions. Expression of survivin was previously shown signaling pathway controlled by HSF1 regulates survi- to be regulated by p21 (Altieri, 2003a), and indeed the vin levels. Analyses of HER2-positive human cancer survivin decrease was partially suppressed in p21 cells suggested that even in SK-BR-3 line with mutant

Oncogene HSF1 is essential for HER2-induced transformation L Meng et al 5212 p53 that has undetectable levels of p21, HSF1 depletion reported before (Yaglom et al., 2007; Gabai et al., 2009). led to survivin decrease and precipitated senescence, Briefly, HEK293T cells were co-transfected with plasmids indicating a p21-independent pathway of survivin expressing retroviral proteins Gag-Pol, vesicular stomatitis regulation. At least in part, HSF1-mediated control of virus glycoprotein pseudotype and enhanced green fluorescent cancer cell senescence was due to the regulation of protein or our constructs using Lipofectamine 2000 (Invitrogen, expression of Hsp72 and Hsp27. In fact, if either Hsp72 Carlsbad, CA, USA). At 48 h after transfection, supernatants containing the retroviral particles were collected and frozen at or Hsp27 was depleted, expression of HER2 also À70 1C until use. MCF-10A cells were infected with diluted brought about senescence instead of transformation, supernatant in the presence of 10 mg/ml Polybrene overnight, indicating that downregulation of any of these Hsps is and were selected with puromycin (0.75 mg/ml) 48 h after sufficient for the paradoxical effects of HER2. In line infection. Retroviral vectors expressing enhanced green with this notion, overexpression of Hsp72 alone was not fluorescent protein was used as infection efficiency indicator: sufficient for reversing the effect of depletion of HSF1 usually B90% of cells were fluorescent 2 days after infection. on senescence. Furthermore, analyses of various human cancer cell lines indicated that there was an association Mouse xenografts between the senescence (or growth inhibition) in Animal maintenance and experiments were conducted in response to HSF1 and depletion of Hsp72. There is a compliance with the guidelines of the Institutional Animal possibility that a general decline of protein homeostasis Care and Use Committee. Briefly, HER2-infected MCF-10A contributes to the oncogene-induced senescence. On the cells with or without shHSF1 were trypsinized, mixed at 1:1 ratio with matrigel and 0.25 million cells of each culture were other hand, it is currently clear that HSF1 may have injected subcutaneously into 6-week-old female NCR nude other effects on tumorigenesis unrelated to the expres- mice (Taconic, Hudson, NY, USA). Tumor growth was sion of Hsp72. In fact, HSF1 was found to regulate pro- monitored weekly. inflammatory cytokines (Cahill et al., 1997; Xie et al., 2002b), translational machinery, glucose metabolism Immunoblot analysis (Dai et al., 2007) and cooperate with the invasion- Cells were washed twice with phosphate-buffered saline and associated protein MTA (Khaleque et al., 2008), which lysed in lysis buffer (40 mM 4-(2-hydroxyethyl)-1-piperazine- obviously may contribute to tumor development. ethanesulfonic acid (pH 7.5), 50 mM KCl, 1% Triton X-100, Very importantly, we observed that HER2 expression 1mM Na3VO4,50mM glycerophosphate, 50 mM NaF, 5 mM triggers growth arrest in MCF-10A cells treated with a EDTA, 5 mM ethylene glycol tetraacetic acid, 1 mM phenyl- small-molecular-weight inhibitor of heat-shock re- methylsulfonyl fluoride, 1 mM benzamidine and 5 mg/ml each sponse, NZ28. Originally, we developed this inhibitor leupeptin, pepstatin A and aprotinin). The protein concentra- to enhance anticancer activities of the proteasome and tion of the lysates was measured with the Bio-Rad protein assay reagent, after which they were diluted with lysis buffer to Hsp90 inhibitors (Zaarur et al., 2006). Here, we found achieve equal protein concentrations. The antibodies used for that it inhibits growth of MCF-10A cells that express this study were listed as follows: anti-AKT, anti-pERK, anti- HER2, as well as HER2-positive cancer lines. These ERK, anti-a-MET and anti-glycogen synthase kinase-3b from findings support the idea that HSF1 could be an Cell Signaling (Danvers, MA, USA); anti-HSF1, anti-Hsp90, interesting drug target for treatment of HER2-positive anti-Hsp72 and anti-Hsp27 from Stressgen (Ann Arbor, MI, herceptin-resistant cancers. USA); anti-p21 and anti-PARP from BD PharMingen (San Jose, CA, USA); anti-vimentin from LabVision (Thermo Scientific, Fremont, CA, USA); anti-survivin from Santa Cruz (Santa Cruz, CA, USA); and anti-b-actin from Sigma Materials and methods (St Louis, MO, USA).

Cell cultures and reagents b-Galactosidase assay À/À MCF-10A wild-type and p21 cells were cultured in b-Galactosidase assay was carried out using X-gal (5-bromo-4- Dulbecco’s modified Eagle’s medium/F12 medium supplemen- chloro-3-indolyl b-D-galactopyranoside) (pH 6.0) as described ted with 5% horse serum, 20 ng/ml epidermal growth factor, previously (Yaglom et al., 2007; Gabai et al., 2009). Cells were 0.5 mg/ml hydrocortisone, 10 mg/ml human insulin and 100 ng/ plated at a low density and fixed with 2% formaldehyde and ml cholera toxin. HEK293T, SK-BR3 and MDA-MB-453 cells 0.2% glutaraldehyde. The b-gal activity was determined by were from ATCC (Manassas, VA, USA). HEK293T cells were incubation with 1 mg/ml of solution of X-gal in 40 mM sodium cultivated in Dulbecco’s modified Eagle’s medium supplemen- citrate, 5 mM K3FeCN6,5mM K4FeCN6, 150 mM NaCl and ted with 10% heat-inactivated fetal bovine serum. All other 2mM MgCl2 diluted in phosphate-buffered saline (pH 6.0). cells were cultivated in Dulbecco’s modified Eagle’s medium The number of stained cells was counted under a microscope supplemented with 10% fetal bovine serum. from five different fields and the proportion of stained cells was calculated. The results were expressed as mean value±s.d. Retroviral vectors and infection on three independent experiments. RNAi-Ready pSIREN-RetroQ vector from BD Biosciences (San Jose, CA, USA) retroviral delivery system was used for Cell growth and foci formation knockdown of HSF1. The sequence of human HSF1 was Cell growth curve was performed by seeding 5000 cells per well selected as reported before (Zaarur et al., 2006): 50-TATGGA in 12-well plate and cell number was monitored every 2 days. CTCCAACCTGGATAA-30. HER2 and control (pBABE) To study foci formation, cells were seeded at 3 Â 105 per well in retroviral vectors were a kind gift from C Spangenberg (Trost a six-well plate and were let to grow for 2 days until the foci et al., 2005). This version of HER2 carries the activating could be observed under the microscope. Then, cells were V664E mutation (NeuT). Retroviruses were produced as washed twice with phosphate-buffered saline, fixed with 2%

Oncogene HSF1 is essential for HER2-induced transformation L Meng et al 5213 formaldehyde/0.2% glutaraldehyde in phosphate-buffered Acknowledgements saline for 10 min and stained with hematoxylin. This work was supported by grants from the National Institute Conflict of interest of Health CA081244. We thank Dr B Park and Dr C Spangenberg for their kind supply of MCF-10A cells and The authors declare no conflict of interest. HER2-overexpressing retroviral vector.

References

Altieri DC. (2003). Survivin, versatile modulation of cell division and p53-deficient mouse model for spontaneous tumors. Oncogene 26: apoptosis in cancer. Oncogene 22: 8581–8589. 5086–5097. Altieri DC. (2008). Survivin, cancer networks and pathway-directed Moasser MM. (2007). The oncogene HER2: its signaling and drug discovery. Nat Rev Cancer 8: 61–70. transforming functions and its role in human cancer pathogenesis. Bachman KE, Blair BG, Brenner K, Bardelli A, Arena S, Zhou S et al. Oncogene 26: 6469–6487. (2004). p21(WAF1/CIP1) mediates the growth response to TGF- Morimoto RI, Sarge KD, Abravaya K. (1992). Transcriptional beta in human epithelial cells. Cancer Biol Ther 3: 221–225. regulation of heat shock . A paradigm for inducible genomic Cahill CM, Lin HS, Price BD, Bruce JL, Calderwood SK. (1997). responses. J Biol Chem 267: 21987–21990. Potential role of heat shock transcription factor in the expression of Mosser DD, Morimoto RI. (2004). Molecular chaperones and the inflammatory cytokines. Adv Exp Med Biol 400B: 625–630. stress of oncogenesis. Oncogene 23: 2907–2918. Cahill CM, Waterman WR, Xie Y, Auron PE, Calderwood SK. Pratt WB, Toft DO. (2003). Regulation of signaling protein function (1996). Transcriptional repression of the prointerleukin 1beta gene and trafficking by the hsp90/hsp70-based machinery. by 1. J Biol Chem 271: 24874–24879. Exp Biol Med (Maywood) 228: 111–133. Cai L, Zhu JD. (2003). The tumor-selective over-expression of the Roninson IB. (2003). Tumor cell senescence in cancer treatment. human Hsp70 gene is attributed to the aberrant controls at Cancer Res 63: 2705–2715. both initiation and elongation levels of transcription. Cell Res 13: Singh IS, He JR, Calderwood S, Hasday JD. (2002). A high affinity 93–109. HSF-1 binding site in the 50-untranslated region of the murine Calderwood SK, Khaleque MA, Sawyer DB, Ciocca DR. (2006). Heat tumor necrosis factor-alpha gene is a transcriptional repressor. J shock proteins in cancer: chaperones of tumorigenesis. Trends Biol Chem 277: 4981–4988. Biochem Sci 31: 164–172. Singh IS, Viscardi RM, Kalvakolanu I, Calderwood S, Hasday JD. Ciocca DR, Calderwood SK. (2005). Heat shock proteins in cancer: (2000). Inhibition of tumor necrosis factor-alpha transcription in diagnostic, prognostic, predictive, and treatment implications. Cell macrophages exposed to febrile range temperature. A possible role Stress Chaperones 10: 86–103. for heat shock factor-1 as a negative transcriptional regulator. J Biol Dai C, Whitesell L, Rogers AB, Lindquist S. (2007). Heat shock factor Chem 275: 9841–9848. 1 is a powerful multifaceted modifier of carcinogenesis. Cell 130: Sorger PK. (1991). Heat shock factor and the . Cell 1005–1018. 65: 363–366. Gabai VL, Yaglom JA, Waldman T, Sherman MY. (2009). Heat shock Trost TM, Lausch EU, Fees SA, Schmitt S, Enklaar T, Reutzel D et al. protein Hsp72 controls oncogene-induced senescence pathways in (2005). Premature senescence is a primary fail-safe mechanism of cancer cells. Mol Cell Biol 29: 559–569. ERBB2-driven tumorigenesis in breast carcinoma cells. Cancer Res Jaattela M. (1999). Escaping cell death: survival proteins in cancer. 65: 840–849. Exp Cell Res 248: 30–43. Wu G, Osada M, Guo Z, Fomenkov A, Begum S, Zhao M et al. Jolly C, Morimoto RI. (2000). Role of the heat shock response and (2005). DeltaNp63alpha up-regulates the Hsp70 gene in human molecular chaperones in oncogenesis and cell death. J Natl Cancer cancer. Cancer Res 65: 758–766. Inst 92: 1564–1572. Xie Y, Chen C, Stevenson MA, Auron PE, Calderwood SK. (2002a). Khaleque MA, Bharti A, Gong J, Gray PJ, Sachdev V, Ciocca DR Heat shock factor 1 represses transcription of the IL-1beta gene et al. (2008). Heat shock factor 1 represses estrogen-dependent through physical interaction with the nuclear factor of interleukin 6. transcriptionthroughassociationwithMTA1.Oncogene 27: 1886–1893. J Biol Chem 277: 11802–11810. Khaleque MA, Bharti A, Sawyer D, Gong J, Benjamin IJ, Stevenson Xie Y, Chen C, Stevenson MA, Hume DA, Auron PE, Calderwood MA et al. (2005 ). Induction of heat shock proteins by heregulin SK. (2002b). NF-IL6 and HSF1 have mutually antagonistic effects beta1 leads to protection from apoptosis and anchorage-indepen- on transcription in monocytic cells. Biochem Biophys Res Commun dent growth. Oncogene 24: 6564–6573. 291: 1071–1080. Lacroix M, Toillon RA, Leclercq G. (2006). p53 and breast cancer, an Yaglom JA, Gabai VL, Sherman MY. (2007). High levels of heat update. Endocr Relat Cancer 13: 293–325. shock protein Hsp72 in cancer cells suppress default senescence Lindquist S, Craig EA. (1988). The heat-shock proteins. Annu Rev pathways. Cancer Res 67: 2373–2381. Genet 22: 631–677. Zaarur N, Gabai VL, Porco Jr JA., Calderwood S, Sherman MY. Min JN, Huang L, Zimonjic DB, Moskophidis D, Mivechi NF. (2007). (2006). Targeting heat shock response to sensitize cancer cells to Selective suppression of lymphomas by functional loss of Hsf1 in a proteasome and Hsp90 inhibitors. Cancer Res 66: 1783–1791.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Oncogene