Introduction
Heat shock protein (Hsp) 90 is a highly conserved chaperone protein and among the most abundant proteins found in eukaryotic cells [
1‐
3]. Hsp90 exists as a homodimeric structure in which individual monomers are each characterized by three domains: an N-terminal nucleotide binding domain (NBD), the site of ATP binding; the middle domain (MD), involved in ATP hydrolysis and the site of co-chaperone and client protein binding; and a C-terminal dimerization domain (CDD), the site of dimerization [
4]. In addition to protecting cells by correcting misfolded proteins, Hsp90 also plays a key role in regulating the stability, maturation, and activation of a wide range of client substrates, including kinases, hormone receptors, and transcription factors [
5‐
8]. Most Hsp90 client proteins, such as epidermal growth factor receptor 2 (HER2), Akt, Raf-1, Cdk4, Bcr-Abl, and p53, are essential for cancer cell survival and proliferation [
9]. The chaperoning of these client proteins is regulated by a dynamic cycle driven by ATP binding to Hsp90 and subsequent hydrolysis of the protein [
10]. Hsp90 requires a series of co-chaperones to form a complex in order to function. These co-chaperones, including cell division cycle protein 37 (Cdc37), Hsp70, Hsp40, Hop, Hip, p23, pp5, and immunophilins, bind to the super-chaperone complex and are released at various time points to regulate the folding, assembly, and maturation of Hsp90 client proteins [
11]. To date, the mechanisms of developed Hsp90 inhibitors have greatly expanded, ranging from the Hsp90 protein function inhibitor to agents targeting the function of nucleotides and co-chaperones crucially involved in regulating the Hsp90 cycle [
4].
We adopted chemoproteomics-based drug screening [
12,
13] to identify clinical Hsp90 inhibitor candidates among a series of natural product, extracted from plants, Fungus, actinomycetes secondary metabolites and so on. Specifically, the histidine-tagged yeast Hsp90 was loaded onto an affinity column [
12] and was subsequently tested with these natural product. Mass spectrum analysis of the eluted solutions of proteins resulted in the identification of the compounds bound to Hsp90. This primary screening effort led to the discovery of FW-04-806 as one of the potential Hsp90 inhibitors. Secondary screening was conducted in parallel across multiple targets. The FW-04-806-loaded affinity columns were incubated with the histidine-tagged NBD, MD, and CDD of yeast Hsp90 to provide substantial binding information and relative binding affinities.
FW-04-806, extracted from the China-native
Streptomyces FIM-04-806 [
14], is identical in structure to Conglobatin [
15] according to ultraviolet spectra, infrared spectra, and NMR (
1H and
13C) data and single-crystal X-ray diffraction data [
16]. Cell proliferation assays have shown that FW-04-806 inhibits the growth of a human chronic myelocytic leukemia K562 cell line with an IC
50 of 6.66 μg/mL (almost 10 μM) [
16]. Conglobatin has been reported to be non-toxic at doses up to 1 g/kg when administered to mice either orally or interperitoneally [
15]. In addition, our acute toxicity test showed that mice survived after oral administration of 900 mg/kg of FW-04-806. In the present study, we investigated the effects of FW-04-806 on SKBR3 and MCF-7, HER2-overexpressed and HER2-underexpressed human breast cancer cell lines, respectively. Chemoproteomics and computational approaches together confirmed that FW-04-806 bound to the N-terminus of Hsp90. A colorimetric assay for inorganic phosphates and ATP pull-down assay showed that FW-04-806 had little effect on Hsp90 ATPase activity compared with 17AAG and did not affect ATP-binding of Hsp90. Indeed, immunoprecipitation confirmed that FW-04-806 disrupted Hsp90/Cdc37 chaperone/co-chaperone interactions, leading to enhanced tumor-arresting activity--and caused the degradation of Hsp90 client proteins. In addition, FW-04-806 exhibited anticancer activity in an
in vivo breast cancer xenograft model, and no major toxicity was observed in the animals. These data suggest that FW-04-806 is a potent Hsp90 inhibitor against human breast cancer cells.
Materials and methods
Cell lines and reagents
SKBR3 and MCF-7 breast cancer cells were originally obtained from American type culture collection. SKBR3 cells were cultured in Roswell Memorial Park Institute-1640 medium and MCF-7 cells were grown in Dulbecco’s modified Eagle medium. All media were supplemented with 10% fetal bovine serum. The cells were maintained under standard cell culture conditions at 37°C and 5% CO2 in a humid environment.
FW-04-806 (purity ≥98.5%) was produced by Fujian Institute of Microbiology, China [
14,
16]. Recombinant human Cdc37 was obtained from Sino Biological Inc. MG132 was obtained from Sigma Aldrich. 17AAG (Tanespimycin) was purchased from Selleckchem. MTS was obtained from Promega. Primary antibodies against Hsp90, Neu, Akt, Raf-1, His-probe and β-actin were purchased from Santa Cruz Biotechnology. Primary antibodies against phospho-Akt (Thr308), apoptosis and phospho-HER2/ErbB2 antibody sampler kits containing cleaved caspase-3 (Asp175), caspase-3, poly (ADP-ribose) polymerase (PARP), cleaved PARP (Asp214), caspase-9, cleaved caspase-9 (Asp330), caspase-7, cleaved caspase-7 (Asp 198), HER2/ErbB2 (D8F12), and phospho-HER2/ErbB2 (Tyr1221/1222) were obtained from Cell Signaling Technology. An Annexin V: fluorescein isothiocyanate (FITC) Apoptosis Detection Kit I was purchased from BD Biosciences.
Preparation of Hsp90 protein
Recombinant vectors were constructed for histidine-tagged full-length (1–732, 90 kDa), NBD (1–236, 25 kDa), MD (272–617, 40 kDa), and CDD (629–732, 15 KDa) of yeast Hsp90. The fusion proteins were expressed in BL21(DE3) and purified via Ni-NTA column and gel filtration [
17].
Resin synthesis
CNBr-activated Sepharose™4B (GE Healthcare) was swelled in 1 mM HCl and washed with coupling buffer (0.1 M NaHCO3, 0.5 M NaCl, pH = 8.3).
For the Hsp90-loaded affinity column, 10 mg of protein per mL of medium was added to the resin, the mixture was rotated overnight at 4°C, and then washed with coupling buffer. Any remaining active groups were blocked with capping solution (1 M ethanolamine) at room temperature for 2 h. The resin was then treated with a small molecule compound, rotated end to overend at room temperature for 4 h, and then washed away of any excess compound. The resin was then washed with three rounds of high pH buffer (0.1 M Tris–HCl, 0.5 M NaCl, pH = 8)/low pH buffer (0.1 M AcOH/NaAcOH, 0.5 M NaCl, pH = 4). Samples were desalted using a Vivapure C18 spin column (Sartorius) before LC-MS analysis [
18].
For the drug-loaded affinity column, after the resin was swelled, washed, and added into coupling buffer, FW-04-806 was dissolved in dimethyl sulfoxide (DMSO) and mixed into the resin (up to 10 μmoles per mL of medium). The mixture was rotated end to overend for 4 h at room temperature, and then washed away of the excess ligand with coupling buffer. Any remaining active groups were blocked with the capping solution for 2 h at room temperature, and the column was equilibrated with coupling buffer. The test proteins were added into the resin, the mixture was rotated overnight at 4°C, and then washed away of any excess proteins. The resin was added to loading buffer, boiled for 10 min, separated with 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis, and then assayed by western blotting.
LC-MS detection
Samples were analyzed on Agilent 6410B Triple Quadrupole LC/MS system. Peptides were separated on a BioBasic Picofrit C18 capillary column (New Objective). Elution was performed with an acetonitrile gradient from 0 to 100% over 1 h with an overall flow rate of 1 mL/min.
ATP-Sepharose binding assay
ATP-Sepharose binding assay was modified base on previous protocol [
19]. Different concentrations of FW-04-806 or 17AAG were added into recombinant NBD Hsp90 protein (10 μg), and then mixtures were incubated with 25 µL preequilibrated γ-phosphate-linked ATP-Sepharose (Jena Bioscience GmbH) in 200 µL incubation buffer (10 mM Tris–HCl, 50 mM KCl, 5 mM MgCl
2, 20 mM Na
2MoO
4 , 0.01% NP-40, pH 7.5) for 4 h at 4°C. The protein bound to Sepharose beads was separated with 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and assayed with protein immunoblotting.
Colorimetric determination of ATPase activity
Malachite green reagent [
20,
21] was prepared on the day of use and contained malachite green (0.0812%, w/v), polyvinyl alcohol (2.32%, w/v, dissolves with difficulty and requires heating), ammonium molybdate (5.72%, w/v, in 6 M HCl), and argon water mixed in a ratio of 2:1:1:2 to a golden yellow solution. The assay buffer consisted of 100 mM Tris–HCl, 20 mM KCl, and 6 mM MgCl
2, with a pH of 7.4. The experiments were performed in 100 μL of test solution containing 80 μL of malachite green reagent. The test solution contained 0.5 μM Hsp90 protein, 1 mM ATP, and 25, 50, 100, or 200 μM FW-04-806 or vehicle (DMSO).
Immunoprecipitation
Samples (500 μg of total protein) were incubated overnight with 2 μg of primary antibody at 4°C, after which 20 μL of protein A Mag Sepharose™ (GE Healthcare, UK) was added to the mixture, which was then incubated for 2 h at 4°C. The immunoprecipitated protein complexes were washed once with lysis buffer and twice with ice-cold PBS. After the supernatant was discarded, the antibody/protein complexes were resuspended in 30 μL of loading buffer and boiled for 5 min. The entire sample was separated with 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and assayed with protein immunoblotting.
Small interfering RNA (siRNA) gene knockdown
SKBR3 cells were seeded in antibiotic-free normal growth medium supplemented with fetal bovine serum. Single siRNA oligonucleotides targeting human Hsp90α/β (sc-35608, Santa Cruz Biotechnology) and control siRNA (sc-37007) were diluted in siRNA Transfection Medium (sc-36868) and mixed with siRNA Transfection Reagent (sc-29528) according to the manufacturer’s protocol. SKBR3 cells were incubated with the transfection complexes for 6 h and in the normal growth medium for 48 h. Cells then were treated with DMSO or FW-04-806 for 24 h before cell lysates were prepared and analyzed with western blot.
Quantitative real-time PCR
Total RNA extraction was performed with TRIzol reagent (Life Technologies Corporation), and first strand cDNA was synthesized using 1 μg of total RNA (concentrations measured by NANODROP 2000, Thermo Scientific) treated with avian myeloblastosis virus (AMV) reverse transcriptase (Promega) according to the manufacturer’s instructions. Quantitative real-time reverse transcription polymerase chain reaction (RT-PCR) analysis was performed in triplicate with FastStart Essential DNA Green Master (Roche) using LightCycler 96 (Roche). The ΔΔCT method was used to calculate relative expression. Primer sequences used in RT-PCR for human Akt (forward 5′-TTGAGAGAAGCCACGCTGT-3′ and reverse 5′-CGGAGAACAAACTGGATGAA-3′), HER2 (forward 5′-TGCTGTCCTGTTCACCACTC-3′ and reverse 5′-TGCTTTGCCACCATTCATTA-3′), Raf-1 (forward 5′-CACCTCCAGTCCCTCATCTG-3′ and reverse 5′-CTCAATCATCCTGCTGCTCA-3′), Hsp90 (forward 5′-GGGCAACACCTCTACAAGGA-3′ and reverse 5′-ATCAACTGGGCAATTTCTGC-3′), GAPDH (forward 5′-AGAAGGCTGGGGCTCATTTG-3′ and reverse 5′-AGGGGCCATCCACAGTCTTC-3′).
MTS assay
Cells (5 × 103/well) were seeded into 96-well plates and treated with 10, 20, or 40 μM of FW-04-806 or vehicle (DMSO) for 48 h. At the end of the incubation period, cell viability was assessed by MTS assay(Promega)according to the manufacturer’s instruction. The number of living cells is proportional to the absorbance at 490 nm. The results are presented as means ± standard deviation from three independent experiments. Inhibition graphs used mean values obtained from each concentration relative to control values, and the half maximal inhibitory concentration (IC50) were calculated by SPSS.
Cell cycle analysis
Cells were seeded in 6-well plates and treated with various doses of FW-04-806 or vehicle (DMSO) for 24 h. The cells were harvested, washed with phosphate-buffered saline (PBS), and fixed with 70% ethanol at −20°C overnight. After an additional washing, cells were incubated with RNase A (20 μg/mL) at 37°C for 30 min, stained with propidium iodide (100 μg/mL; Sigma Aldrich) for 10 min, and analyzed with flow cytometry (BD FACSC autoTM II).
Apoptosis assay
Apoptosis was determined with the Annexin-V: FITC Apoptosis Detection Kit I (BD Biosciences) according to the manufacturer’s protocol. Briefly, the vehicle control (DMSO) and FW-04-806-treated cells were collected via centrifugation and washed once with PBS. The cells were subsequently stained with fluorescein and propidium iodide for 15 min at room temperature and analyzed with flow cytometry.
Western blot analysis
After treatment, cancer cells were collected, washed with PBS, lysed with NP-40 lysis buffer (50 mmol/L Tris pH 8.0, 150 mmol/L NaCl, and 1% NP-40) supplemented with phenylmethanesulfonyl fluoride (Sigma Aldrich) and PhosSTOP (Roche Diagnostics) for 30 min at 4°C, and centrifuged at 12,000 × g for 10 min. The supernatant was collected as the total protein extract. Protein concentration was estimated using a Pierce BCA Protein Assay Kit (Thermo Scientific, USA) according to the manufacturer’s protocol. Equal amounts of protein were analyzed with sodium dodecyl sulfate polyacrylamide gel electrophoresis. Thereafter, proteins were transferred to polyvinylidene fluoride membranes and blotted with specific primary antibodies. Proteins were detected via incubation with horseradish peroxidase-conjugated secondary antibodies and visualized with SuperSignal WestPico (Thermo Scientific, USA). All the western blot detections were repeated three or more times.
Animals, tumor xenografts, and test agents for in vivo studies and efficacy
BALB/c (nu/nu) athymic mice were purchased from Shanghai SLAC Laboratory Animal Co. LTD. For SKBR3 and MCF-7 xenografts, 6-mm3 tumor fragments were implanted into the subcutaneous tissue of the axillary region using a trocar needle, and the animals were randomly divided into groups (n = 6) when the bearing tumor reached approximately 20 mm3. FW-04-806 was suspended at the desired concentration for each dose group in an aqueous vehicle containing 10% ethanol, 10% polyethylene glycol 400, and 10% Tween 80. The control group was given 0.4 mL/mouse vehicle solution i.g.; mice in other groups were given 50, 100, or 200 mg/kg of FW-04-806. Doxorubicin hydrochloride (ADM, Shenzhen Main Luck Pharmaceuticals Inc, China) was purchased as 10 mg injections and diluted with saline as necessary to achieve the prescribed concentration.
Tumor volumes were calculated using the following ellipsoid formula: [D × (d2)]/2, in which D is the large diameter of the tumor, and d is the small diameter. Tumor growth inhibition was determined using the following formula: 100 % × [(WC–WT)/WC], in which WC represents mean tumor weight of a vehicle group, and WT represents mean tumor weight of a treated group. All animal experiments were approved by animal care and use committee, Fujian Medical University, China.
Immunohistochemistry
Immunohistochemistry was performed on tumors harvested from each xenograft group treated with FW-04-806 or vehicle. Tumors were cut into 3- to 5-mm pieces, fixed in 4% paraformaldehyde for 6 h, dehydrated, paraffin-imbedded, sectioned, and placed on slides (Zhongshan Biotechnology Company, China). Antigen retrieval was performed in 0.1 M citrate buffer, pH 6, at 100°C for 2 min. After incubation with 3% hydrogen peroxide for 10 min and three washes with PBS buffer, primary antibody Neu (rabbit poly-clonal, Santa Cruz Biotechnology, sc-284) was used at a dilution of 1:200 for 2 h. The slides were sequentially incubated with biotin-conjugated secondary antibodies (1:200) followed by horseradish peroxidase-conjugated streptavidin (1:100). The first step lasted 20 min and the second step lasted 30 min at 37°C, with 5 min PBS washes three times for each step. The reactions were revealed using diaminobenzidine substrate, and the slides were then washed under running tap water. Contrast was applied with hematoxylin, and the slides were mounted in Canadian balsam and observed with a light microscope.
Statistical analysis
Analysis of variance was used for comparisons across multiple groups. The data are reported as means ± standard deviation. Statistical analysis was conducted using PASWstatistics 18 (SPSS, Inc); p < 0.05 was considered statistically significant.
Discussion
Most Hsp90 inhibitors have been developed to inhibit Hsp90 chaperone function by binding to Hsp90 at the N-terminus and blocking the ATP/ADP pocket [
22]. The antibiotics benzoquinone ansamycins, such as geldanamycin (GA) and its derivative 17-allyamino-geldanamycin (17AAG), were the first identified Hsp90 inhibitors [
23]. The binding of GA in the N-terminal ATP pocket arrests the catalytic cycle of Hsp90 in the ADP-bound conformation, inactivating chaperone activity, which results in the ubiquitination and proteasomal degradation of client proteins [
24‐
26]. Although GA and its derivatives have exhibited potent anticancer effects, severe hepatotoxicity has prevented clinical development [
27]. This study showed that a natural product, FW-04-806, a novel Hsp90 inhibitor, inhibits Hsp90 function through binding the N-terminus of Hsp90 and blocking formation of the Hsp90-Cdc37 complex (Figures
1 and
2) in an ATP-binding independent manner, therefore the mechanism of action is clearly different from those classic Hsp90 inhibitors.
The wide-ranging functions of Hsp90 require a series of co-chaperones to drive the chaperone cycle to completion [
22]. Therefore, affecting co-chaperone function by specifically targeting various co-chaperone/Hsp90 interactions may offer an alternative way to achieve the outcomes of direct Hsp90 inhibition [
28,
29]. Cdc37 is an essential co-chaperone and functions as an adaptor in the recruitment of client proteins, predominantly kinases such as HER2, epidermal growth factor receptor, non-receptor tyrosine kinases (Src), lymphocyte-specific protein tyrosine kinase, Raf-1, and CDK4, to Hsp90 [
30‐
34]. The targeting of the Hsp90/Cdc37 interaction is a potential alternative to direct Hsp90 inhibition that may offer greater specificity and an improved side effect profile owing to the elevated expression of Cdc37 in cancer [
5]. To date, only a few medicines were discovered to target Hsp90/Cdc37 interaction. Celastrol is a quinine methide triterpene extracted from
Tripterygium wilfordii. It has recently been found to disrupt Hsp90/Cdc37 association, which results in the degradation of AKT and CDK4 and the induction of apoptosis in the pancreatic cell line Panc-1 [
28]. But recent nuclear magnetic resonance (NMR) studies have suggested that celastrol binds to Cdc37 instead of the Hsp90 N-terminus domain [
35,
36]. Withaferin A (WA), a steroidal lactone extracted from
Withania somnifera, disrupts the Hsp90/Cdc37 complex by binding to the C-terminus domain of Hsp90 and changing Hsp90 conformation to prevent Cdc37 binding [
37,
38]. Sulforaphane, a dietary component from broccoli sprouts, blocks Hsp90-Cdc37 complex by interacting with Ile amino acids residues of the N-terminal and middle domain of Hsp90 [
19]. Our work here found a new medicine targeting Hsp90/Cdc37 interaction with new mechanism which is quite different with the medicines above.
We have showed that FW-04-806 is a Hsp90 inhibitor that directly binds to the N-terminus of Hsp90 and attenuates Hsp90/Cdc37 chaperone/co-chaperone interactions, leading to the degradation of multiple Hsp90 client proteins via the proteasome pathway, which may be the primary mechanism mediating the anticancer activities of FW-04-806. The antagonistic efficacy of FW-04-806 against human breast cancer lines has been investigated at both the molecular and cellular levels. It has been demonstrated that FW-04-806 inhibits the HER2-overexpressed and HER2-underexpressed breast cancer cell lines SKBR3 and MCF-7 in a dose and time-dependent manner with IC50 values of 12.11 and 39.44 μM, respectively. Moreover, it was shown that FW-04-806 arrests cell cycle progression and induces programmed cell death.
It has further been shown that FW-04-806 displays antitumor effects in an
in vivo animal model as well as in the
in vitro settings previously described. Studies were conducted to investigate the effect of FW-04-806 on tumors derived from cancer cell lines SKBR3 and MCF-7. High dose administration of FW-04-806 displayed an inhibitory effect on SKBR3-derived tumors was more preferable in both the antitumor activity and mouse body weight than that of ADM, one of the most widely used chemotherapy drugs. Importantly, we found that FW-04-806 displays a better antitumor effect in SKBR3 tumor xenograft model than in MCF-7. The result is consistent with cell proliferation assay and
in vitro apoptosis assay applied for SKBR3 and MCF-7. As these cell lines are HER2-overexpressed and HER2-underexpressed respectively, and HER2 is among the most sensitive Hsp90 clients [
39], we assume that FW-04-806 has a preferential inhibitory effect on HER2-overexpressed cancer cells. This assumption is now being tested on other cancer cell lines. Moreover, mice survived at the dose of 900 mg/kg in the acute toxicity test, and all xenografts mice had no appreciable adverse effects during the treatment. No histological abnormalities was found in lung, liver, heart, and kidneys of mice (Data not shown), suggested that FW-04-806 had a favorable toxicity profile.
Competing interest
The authors declare that they have no competing interests.
Authors’ contribution
WH and MY carried out the mechanism studies, participated equally in the experiments and drafted the manuscript. MY also conceived of the project, and participated in its design and coordination. LZ carried out the computational docking. QW and MZ participated in the western blot assay. JX and WZ conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.