HSP27 is a member of the small HSPs family that acts as an ATP-independent chaperone and mainly localized in the cytosol. They are potent mediator of protein folding and also involved in architecture of cytoskeleton, cell migration, cell growth/differentiation, and tumor progression [15,21,24,25]. HSP27 also has antiapoptotic property [26].

High levels of HSP27 have been observed in many cancer cells including breast carcinoma [27,28], compared to normal cells in which expression is undetectable or relatively low [15]. Moreover, its aberrant expression in cancer is associated with aggressive tumor behavior, increased resistance to chemotherapy, and poor prognosis for the patients.

HSP27 is activated in various stress conditions both by transcriptional activation and posttranslational modification (phosphorylation). HSP27 can be phosphorylated at three serine residues 15, 78, and 82, and its phosphorylation is mediated by the p38 MAPK stress kinase pathway [29]. This phosphorylation is a reversible event that modulates the oligomerization of HSP27. Different cellular functions of HSP27 seem to be related to its oligomerization state. Moreover, various different phosphorylation patterns of HSP27 have been found to be associated with the aggressiveness of different tumor types. A recent report demonstrated a twofold increased phophorylation of HSP27 at serine 78, but not serine 15 and 82, in HER2 positive breast cancer samples compared to HER2 negative tumors. However, the exact role of HSP27 phosphorylation in the physiology of cancer remains incompletely understood.

Recently, several studies demonstrated that the overexpression of HSP27 seems to be correlated with increased resistance to chemotherapeutic drug-induced apoptosis in cancer cells [30,31]. Hansen et al. [32] reported the inhibition of doxorubicin induced apoptosis in the HSP27 overexpressing breast cancer cell, demonstrating a protective role of HSP27 against apoptosis. In addition, a recent report presented that upregulation of HSP27 in breast cancer cells reduces trastuzumab susceptibility by increasing HER2 protein stability [33]. These recent studies suggest possibility of HSP27 inhibition as molecular target for cancer therapy. However, unlike other HSPs, the small HSPs do not bind ATP, and it makes this molecule problematic for targeting with small compounds.

Human cells contain several HSP70 family members including the stress-inducible HSP70 (also called HSP72 or HSPA1) and the constitutively expressed heat shock cognate 70 (HSC70, HSP73 or HSPA8) in the cytosol and nucleus, mitochondrial HSP70 (Grp75, Mortalin or HSPA9), and glucose regulated protein 78 (Grp78, HSPA5) in the endoplasmic reticulum [34-36]. HSP70, like other HSPs, is a molecular chaperone expressed in response to stress. Under normal conditions, HSP70 also plays multiple roles, including the folding of newly synthesized proteins [37,38], the transport of proteins and vesicles [39], the assembly and dissociation of multi-protein complexes [40], and the degradation of denature proteins [41,42].

HSP70 is also powerful anti-apoptotic protein that acts at different key points, affecting both the extrinsic and intrinsic pathway of apoptosis. For example, HSP70 was reported to inhibit the important apoptotic mediator, Bax translocation, thus preventing mitochondrial membrane permeabilization. Together with its co-chaperone HSP40, HSP70 also blocks TNF-induced apoptosis. Moreover, HSP70 directly interacts with apoptosis protease activating factor-1 (Apaf-1), thereby inhibiting recruitment of procaspase-9 to the apoptosome and the consequent caspase-3 activation. HSP70 can also block caspase-independent signaling through inhibition of apoptosis-inducing factor (AIF)-induced chromatin condensation and cathepsins release. In conclusion, HSP70 regulates apoptosis by inhibiting stress-induced signals, by preventing mitochondrial membrane permeabilization, and by suppressing caspase activation and DNA fragmentation [26,43]. HSP70 also plays role in senescence through effects on the p53-p21 pathway [44].

HSP70 is expressed at high levels in a wide spectrum of cancer cells and HSP 70 expression has been routinely associated with poor prognosis [43]. The exact role of the HSP70 in cancer remains to be elucidated. However, in cancer cell, HSP70 overexpression is thought to provide a survival advantage due to its ability to inhibit apoptosis and senescence [26,45]. Moreover, the HSP70 also acts as co-chaperones for HSP90 by its essential role in the substrate-loading phase of the HSP90 molecular chaperone cycle, key to the stability and function of multiple oncoproteins.

Many studies reported that HSP70 overexpression has also been correlated with therapeutic resistance [46]. Although the detailed mechanisms of resistance remain to be elucidated, recent evidence suggests that reduced activation of ERK, NF-κB, and JNK pathways may be responsible [47]. Moreover, pharmacological inhibition of HSP90 has been found to induce a compensatory expression of HSP70 [48]. This might be because HSP70 is a highly protective protein that may strongly reduce the cell death effect induced by HSP90 inhibition. In this view, the potential therapeutic benefit of modulating HSP70 activity has become attractive, especially dual therapy against both HSP90 and HSP90 [49]. The search for inhibitors of HSP70 has dramatically increased over the last 2 years; however, there are limited small molecule inhibitors of HSP 70 available.

HSP90 is a highly abundant and evolutionarily conserved protein in the all eukaryotic cell. Five HSP90 isoforms have been identified to date, including the two major cytoplasmic isoforms, HPS90α and HPS90β (also called HSPC1 and HSPC3, respectively), endoplasmic reticulum localized glucose regulated protein 94 (Grp94), mitochondrial tumor necrosis factor receptor-associated protein 1 (TRAP1), and membrane-associated HSP90N [16]. Despite their different cellular localization, these isoforms have a similar overall structure and function as chaperone by a common mechanism involving the cyclic conformational change.

HSP90 exists as a homodimer within the cell, and each subunit is composed of three domains: N-terminal ATPase domain, middle domain implicated in client protein binding, and C-terminal domain containing protein-protein interaction and dimerization motif [50]. The N-terminal domain contains ATP-binding pocket, and the chaperoning activity of HSP90 requires both the binding and hydrolysis of ATP at this site [51]. Besides the role of C-terminal domain in dimerization, it was suggested that the C-terminal domain contains a second ATP-binding site of HSP90 [52-54]. The contribution of this second site to the overall regulation of the chaperone is still unknown. The C-terminal domain also recruits co-chaperones through a conserved tetratricopeptide repeat (TPR)-binding motif, EEVD [55]. Co-chaperones, containing TPR domains such as HOP, and the non-TPR co-chaperones, CDC37, p23, and Aha1, play an important role in client protein maturation and modulation of ATPase activity [39,19]. Some of these co-chaperones such as Aha1 and HSP70 have been proposed for independent molecular targets [56].

HSP90 functions as a part of a multichaperone complex via association with a variety co-chaperons and client proteins that rely on the complex for maturation and stability. The HSP90 complex appears to cycle between an ADP-bound and ATP-bound state [57]. In an ATP-bound state, HSP90 undergoes a conformational change and becomes a mature complex that is essential for it to perform its function of client protein folding and stabilization. The hydrolysis of ATP to ADP facilitates release of these client proteins, and then they are degraded by ubiquitin proteasome pathway [58].

HSP90 is important molecular chaperone that regulates the stability and activity of numerous client proteins covering almost all cellular processes. More than 200 client proteins have been identified so far, and the list is constantly being updated [59]. Its client proteins include BCL-ABL, SRC, HER2, EGFR, CRAF, BRAF, AKT, MET, VEGFR, FLT3, androgen and estrogen receptor, hypoxia-inducible factor (HIF)-1α, and telomerase that are directly involved in malignancy and mutated oncogenic proteins that are required for the transformed phenotype. These include proteins important in breast cancer progression such as HER2 and c-SRC. Indeed, HSP90 overexpression has been observed in a variety of human malignancies including the breast cancer [59].

There are several principal functions of HSP90 in malignant cells. As mentioned above, HSP90 stabilizes many oncogenic proteins in cancer cell. HSP90 may inhibit apoptosis through several interactions. For example, it has been reported that HSP90 binds directly to apoptotic protease activating factor 1 (Apaf-1), and inhibits its oligomerization, recruitment of procaspase-9, thus blocking the assembly of a apoptosome [60]. Moreover, increased expression of HSP90 has been implicated in resistance to senescence due to its essential role of telomerase stability [61]. HSP90 also has a role in angiogenesis owing to its stabilizing properties on the transcription factor HIF-1α, and VEGF and nitric oxide synthase, two basic players in angiogenesis, are HSP90 client proteins. Finally, HSP90 may play a role in tumor invasion and metastasis. Interestingly, HSP90 inhibitors have been implicated in bone metastasis in breast cancer, but its mechanism is not fully explained [62].

Many kinds of HSP90 inhibitors have been identified so far. The majority of HSP90 inhibitors bind to the N-terminal ATP-binding site of HSP90 and inhibit the ATPase cycle which is essential for the HSP90 chaperone activity [48]. Therefore, HSP90 inhibition results in degradation of important oncogenic client proteins by the ubiquitin-proteasome pathway, inhibition of tumor growth and activation of apoptosis in cancer cells. Although blockade of N-terminal ATP binding of HSP90 has been focus of drug development, distinct modes of inhibition are being considered and include disruption of co-chaperone-HSP90 interactions, inhibition of the C-terminal ATP binding site, or inhibition of client-HSP90 interaction.

The natural products, geldanamycin (GA) and radicicol were first inhibitors discovered. Geldanamycin, an ansamycin antibiotic, was first isolated from Streptomyces hygroscopicus and noted to have inhibitory activity against HSP90 [63,64]. Undesirable properties associated with GA, such as hepatotoxicity and poor solubility [65], led to a necessary round of compound optimization. Therefore, the less toxic and more effective GA derivatives, 17-allylamino-17-demethogeldanamycin (17-AAG, tanespimycin, KOS-953), 17-dimethylaminoethylamino-17-demethoxygeldanamycin hydrochloride (17-DMAG, alvespimycin, KOS-1022), and IPI-504 (retaspimycin) have been developed as potential therapeutics in a variety of clinical trials [66]. More recently, purine-scaffold derivates such as PU-H71, PU-DZ8, and CNF2024 (BIIB021) developed based on the structure of the nucleotide ligand [67,68]. Because of the potential toxicity of GA derivatives, specific small molecular weight HSP90 inhibitors may be more effective clinical agents. Several small molecular weight HSP90 inhibitors, including SNX-5422, CNF2024, STA 9090, and AUY 922, are currently in clinical trials in various tumor types [69-72]. Current phase I and II clinical trials with HSP90 inhibitors in breast cancer are seen in Table 1. Interestingly, HSP90 isolated from tumor cells has a binding affinity for the inhibitors between 20 and 200 times higher than does HSP90 isolated from normal cells. This might be due to the fact that tumor cells, as compared to their counterparts, might exhibit a stressed phenotype, with an enhanced dependency on the cytoprotective action of HSP90. This 'addiction' of cancer cells to HSP 90 client proteins have been proposed as rationale for selectivity of HSP90 inhibitors for cancer versus normal cells [73].

Breast cancer is good target of HSP90 inhibitor for several reasons. HER2 is among the most sensitive client proteins of HSP90, demonstrating degradation within 2 hours of HSP90 inhibition in cell culture experiments [74] and HSP90 inhibitors have shown activity in HER2-deriven xenograft model [75]. Modulation of estrogen and progesterone receptor has been long-standing target of breast cancer and both receptors are also client proteins of HSP90. Moreover, resistance of breast cancer cells to chemotherapy is known to involve the phosphatidylinositol 3-kinase (PI3K) pathway [76], and its key signaling protein Akt is modulated by HSP90. In addition, the expression of HSP90 has been shown to be correlated to adverse clinical prognosis.

We previously investigated the HSP70/90 expression and the effect of HSPs inhibitor in breast cancer [77]. We found that more prominent HSP90 expression in breast cancer tissue than in benign tissue by immunohistochemistry staining and it was related to more aggressive tumor type; positive lymph node status and poor differentiated histologic type. However, we could not find the difference of HSP70 expression between two lesions (Table 2). We also demonstrated the expression of HSP70/90 in breast cancer cell lines using the Western blot (Figure 1). In this study, we investigated the effect of HSP90 inhibitor (GA) on cell growth of the human breast cancer cell lines: MDA-MB231, MDA-MB 435, MCF-7, and T47D cell line. GA markedly inhibited the cell growth of these cell lines in a dose-dependent manner (Figure 2).

HER2 overexpression is observed in 20-25% of breast cancer patients and it predicts for a poor clinical outcome. HSP90 expression is associated with HER2 expression [78]. Preclinical studies have demonstrated the notable sensitivity of HER2-overexpressing breast tumors to HSP90 inhibitors [69,70]. 17-AAG (tanespimycin) is being developed in the clinic for HER2 positive breast cancer and demonstrates a moderate clinical response in combination with trastuzumab in patients with trastuzumab-refractory HER2-positive metastatic breast cancer in recent report [79]. In phase II study, 31 patients (metastatic HER2-positive breast cancer progressing on trastuzumab) received weekly tanespimycin at 450 mg/m² intravenously and trastuzumab. The most common toxicities were diarrhea, fatigue, nausea, and headache. The overall response rate was 22% and the clinical benefit including complete response, partial response, and stable disease was 59%. IPI-504 (retaspimycin), a 17-AAG analogue, has improved water solubility properties thereby facilitating formulation for parental administration. In breast cancer, Phase I/II trials are currently underway to evaluate the dosing schedules [80]. Moreover, a HSP90 inhibitor (SNX-5422) inhibits p95-HER2, the truncated form of HER2 associated with trastuzumab resistance, and suppress their growth [81]. Similarly, 17-DMAG can overcome resistance of HER2 positive breast cancer cells to aromatase inhibitors [82].

Triple negative breast cancer (TNBC; defined by the lack of expression of estrogen, progesterone, and HER2) patients have poor prognosis and survival outcomes, but there are currently no specific targeted therapies. Clinical studies have been shown the EGFR overexpression and activation of PI3K pathway in TNBCs and it has been associated with poor prognosis. Hence, HSP90 inhibitors may provide an opportunity to inhibit tumor progression of TNBCs because the many of HSP90 client proteins are oncoproteins including EGFR and involved in multiple oncogenic signaling pathways. Interestingly, PU-H71 (purine based synthetic HSP90 inhibitor) induces tumor regression in a xenograft model of TNBCs and that are not candidate for 17-AAG treatment [83].