EGFR-mediated hyperacetylation of tubulin induced docetaxel resistance by downregulation of HDAC6 and upregulation of MCAK and PLK1 in prostate cancer cells

Abstract

Docetaxel-based chemotherapy has generally been considered as one of the effective treatments for castration-resistant prostate cancer (PCa). However, clinical treatment with docetaxel often encounters a number of undesirable effects, including drug resistance. Tubulin isoforms have been previously examined for their resistance to docetaxel in many cancers, but their real mechanisms remained unclear. In this study, a series of docetaxel-resistant PC/DX cell sublines were established by chronically exposing PC3 to progressively increased concentrations of docetaxel. Western blotting results showed significantly higher expression of acetyl-tubulin, α-tubulin, β-tubulin, γ-tubulin, and βIII-tubulin in PC/DX25 than in parental PC3 cells. PC/DX25 with greater resistance to docetaxel had higher levels of acetyl-tubulin and mitotic centromere-associated kinesin (MCAK) than PC3 cells. This study found that docetaxel induced the expression of acetyl-tubulin and MCAK in PC3 cells at a dose- and time-dependent manner. Both mRNA and protein levels of histone deacetylase 6 (HDAC6) were significantly decreased in PC/DX25 compared with PC3 cells. PC3 increased the resistance to docetaxel by HDAC6 knockdown and Tubastatin A (HDAC6 inhibitor). Conversely, PC/DX25 reversed the sensitivity to docetaxel by MCAK knockdown. Notably, flow cytometry analysis revealed that MCAK knockdown induced significantly sub G1 fraction in PC/DX cells. Overexpression of polo-like kinase-1 increased the cell survival rate and resistance to docetaxel in PC3 cells. Moreover, epidermal growth factor receptor (EGFR) activation induced the upregulation of acetyl-tubulin in docetaxel-resistant PCa cells. These findings demonstrated that the EGFR-mediated upregulated expression of acetyl-tubulin played an important role in docetaxel-resistant PCa.

INTRODUCTION

In 2022, 288,300 men in the United States were diagnosed with prostate cancer (PCa), which was the highest incidence of cancer (29%) among American male. Moreover, 34,700 died of PCa, which was the second leading cause of cancer death in American men.¹ In Taiwan, the incidence of PCa was also increasing every year.² Androgen deprivation therapy (ADT) has been used as the first-line treatment for men with advanced PCa. Although the majority of patients with PCa can benefit from ADT at the early stage, almost all patients ultimately die of castration-resistant prostate cancer (CRPC) within 3 years.³

Docetaxel is the first-line chemotherapeutic agent for improving survival in CRPC patients. The cytotoxic mechanism of docetaxel involves interaction with free tubulin, promotion of stable microtubule assembly, and simultaneous inhibition of microtubule depolymerization to disrupt the dynamic microtubule structure. Previous studies demonstrated that docetaxel induced cell cycle arrest at the G2/M phase, blocked cell division, inhibited cell proliferation, thus causing cell death.⁴ Although the use of docetaxel has improved clinical outcomes, patients often suffer many undesirable effects, including drug resistance. Therefore, it is important to identify the molecular events for the development of docetaxel resistance in PCa. Drug resistance-related mechanisms have previously been studied in many cancer cells.⁵ Upregulation of efflux transporter reduces the accumulation of drugs in the cell. ATP-binding cassette (ABC) transport protein is a transmembrane drug efflux pump that can transport drugs across the cell membrane. These transport proteins including ABCB1 (P-glycoprotein, P-gp/multidrug resistance 1, MDR1), ABCC1 (multidrug resistance protein 1, MRP1), and ABCG2 (MXR/BCRP), were highly expressed in drug-resistant cancer cells.⁶ Drug-resistant cells induced activation of survival signaling pathways to promote cell proliferation and/or inhibit apoptosis to evade drug-mediated cytotoxicity. Magadoux et al. indicated that docetaxel binding to β-tubulin mediated stress signals and further activated survival signaling pathways such as c-Jun N-terminal kinase (JNK), signal transducer and activator of transcription 1 (STAT1), STAT3, and nuclear factor kappa B (NF-κB).⁷ Previous studies showed development of drug-resistant PCa through the activation of PI3K/AKT pathway.⁸ Our recent study found that epidermal growth factor receptor (EGFR) mediated docetaxel resistance through the Akt-dependent ABCB1 expression in PCa cells. EGFR played an important role in the development of docetaxel resistance in PCa.⁹

Microtubules are assembled from heterodimers of α- and β-tubulin. There are eight α-tubulin and seven β-tubulin isotypes in the tubulin family. Microtubules are highly dynamic polymers as their ends generate rapid lengthening and shortening by respective addition and removal of tubulin heterodimers. These unique properties enable microtubules to participate in a variety of functions that are crucial to cellular processes such as cell division, differentiation, intracellular transport, and motility.¹⁰ Microtubule-associated proteins (MAPs), such as tau, MAP4, stathmin, and kinesin, regulate microtubule aggregation and disassembly, and alterations in the activity or expression of these proteins may contribute to resistance to anti-microtubule drugs.^11,12 As a member of the kinesin superfamily of molecular motors, kinesin-13s are a series of microtubule depolymerases that regulate microtubule length and play an important role in cell division. Kinesin-13s comprise KIF2A, KIF2B, and KIF2C or MCAK in mammalian cells.^13,14 In multiple species, MCAK was found at centrosomes, centromeres, astral microtubules, and the plus ends of microtubules. MCAK also regulated microtubule dynamics in mitosis. Aurora A, Aurora B, or PLK1 phosphorylated MCAK, regulated its localization to centromere and its depolymerase activity.¹⁵

Many studies demonstrated that different types of cellular stress mediated the distinct expression of tubulin isotypes, tubulin posttranslational modifications (PTMs), and MAPs in cancer.¹⁶ PTMs of tubulin include tyrosination-detyrosination, glutamylation, glycylation, acetylation, phosphorylation, and methylation. The majority of PTMs occur on the C-terminal tails of α- and β-tubulins, but some PTMs, such as acetylation, phosphorylation, polyamination, and methylation, occur on the structured core of tubulin.¹⁰ Tubulin acetylation, which is reversible and dynamic, is one of the characteristics of stabilized microtubules.¹⁷ α-Tubulin acetyltransferase (ATAT1) is the major α-tubulin acetyltransferase. Tubulin deacetylases include HDAC6 and sirtuin type 2 deacetylase (SIRT2).¹⁸ The level of acetylation can influence the affinity of specific molecules to microtubules and its transport speed. The acetylation of tubulin can promote the binding of motor molecules, dynein, and kinesin-1, binding to the microtubule, and also the brain-derived neurotrophic factor vesicular transport. Tubulin acetylation deficiency, which results in deficient development and function of nerves, is the cause of Huntington's disease.^19,20 The activity of HDAC6 is regulated by the phosphorylation of glycogen synthase kinase 3β (GSK3β). Excessive acetyl-tubulin, which blocks the mitochondrial transport, causes Parkinson's and Alzheimer's diseases.²¹ In nonneuronal cells, EGF bound to EGFR, facilitated Tyr⁵⁷⁰ of HDAC6 phosphorylation, further reduced its activity and mediated tubulin acetylation, eventually causing endocytosis of EGFR.²² Meanwhile, acetyl-tubulin enhanced binding of heat shock protein 90 (Hsp90) to microtubule and stimulated both the Akt/PKB pathway and p53 client proteins. p53 then moved into the nucleus and induced cell proliferation and survival.²³ HDAC6 inhibitors enhanced tubulin acetylation, further leading to tumor suppressor protein CYLD being transported to nucleus and bound to Bcl-3, thus inducing a delay in the G₁-to-S-phase transition.²⁴

Tubulin isoforms have previously been studied for their resistance to docetaxel in many cancers, but their real mechanisms remained unclear. This study evaluated the feasibility of employing docetaxel as a cytotoxic agent for PC3 cells and examined the role of acetyl-tubulin in docetaxel-resistant PCa.

MATERIALS AND METHODS Chemicals, reagents, and antibodies

Dimethyl sulfoxide (DMSO), 3-(4,5-dimethylthiazo-2-yl)-2,5-diphenyl tetrazolium (MTT), Tubastatin A (a potent and selective inhibitor of HDAC6), and docetaxel were purchased from Sigma Chemical Co. (MA, USA). Anti-α-tubulin (1878) was purchased from Epitomics (CA, USA). Anti-EGFR (sc-03), anti-HSP90α (sc-8262), and anti-HSP90β (sc-1057) antibodies were obtained from Santa Cruz Biotechnology (CA, USA). Anti-β-actin (GTX108370), anti-acetyl-tubulin (GTX11323), anti-GAPDH (GTX100118), anti-HDAC6 (GTX100722), anti-β-tubulin (GTX109639), anti-βIII-tubulin (TUBB3) (GTX102451), anti-γ-tubulin (GTX113286), anti-PLK1 (GTX104302), and anti-Aurora A (GTX104620) antibodies were purchased from GeneTex (CA, USA). Anti-phospho-EGFR (Tyr1068) (#2234) antibody was obtained from Cell Signaling Technology (MA, USA). Anti-HDAC1 (05-614) antibody was purchased from Upstate (MA, USA). Anti-MCAK antibody (3760-100) was purchased from Biovision (CA, USA). Anti-Aurora B (BD611082) antibody was purchased from BD Biosciences (CA, USA). EGFR siRNA and Lipofectamin 2000 were purchased from Invitrogen Inc. (CA, USA). Recombinant EGF protein was purchased from PEROTECH (NJ, USA).

Tissue specimens

Clinical PCa tissues were stored in paraffin-embedded archive blocks that were either hormone-naïve, CRPC, or docetaxel-resistant. The patient's docetaxel-resistant status was defined as a progressive increase in the PSA level after receiving docetaxel-based chemotherapy. These specimens were obtained from radical prostatectomy carried out at the National Taiwan University Hospital (NTUH). This study was approved by the Institutional Review Board of NTUH and Kaohsiung Medical University and informed consent was obtained from all participating patients (KMUH-IRB-20110394).

Cell lines and cell culture

Docetaxel-resistant PC/DX cells were established by chronically exposing PC3 to progressively increased concentrations of docetaxel (Aventis Pharma, Dagenham, England). PC/DX25 cells were cultured in 25 nM docetaxel to maintain their drug-resistant phenotypes. PC3 and PC/DX sublines were cultured in RPMI-1640 medium, supplemented with 10% FBS, 1% penicillin/streptomycin solution, and 1% L-glutamine (Gibco-Life Technologies, CA, USA). Cells were grown at 37°C in a humidified 5% CO₂ atmosphere. The PC3 cell lines were purchased from American Type Culture Collection (ATCC).

Cytotoxicity of docetaxel treatment by MTT assay

Cell viability was determined using 3-(4,5-dimethylthiazo-2-yl)-2,5-diphenyl tetrazolium (MTT) assay in vitro. In brief, cells in 100 μL culture medium were seeded into 96-well plates and incubated at 37°C for 24 h prior to drug treatment. The number of cells were titrated to keep control cells growing in the exponential phase throughout the 72-h incubation period. Cells were cultured for 72 h in the presence of different concentrations of docetaxel. MTT assay was performed as previously described.⁹ Survival rate and IC₅₀ values of docetaxel treatment were calculated using median-effect analysis and presented as mean ± standard error of the means (SEM). Cells were transfected with siRNA or expressive plasmid by lipofectamine for 8 h, followed by treatment with different concentrations of docetaxel for 48 h. Cytotoxicity of siRNA knockdown or expressive plasmid transfection was examined using MTT assay.

Quantitative real-time polymerase chain reaction

Total RNA was isolated from cells using REzol C&T reagent (Protech Technology, Taipei, Taiwan) according to the manufacturer's instructions. A total amount of 5 μg RNA was converted to cDNA using High Capacity cDNA Reverse Transcription Kits (ABI, Foster City, USA). PCR analysis used the Taq Master Mix (Protech Technology, Taipei, Taiwan). qRT-PCR analysis used the SYBR Green I Master Mix (ABI, CA, USA). Fluorescence of SYBR Green was detected by ABI 7900 Sequence Detection System. HDAC6 primer (forward, 5′-TATATCTGCCCCAGTACCTTCG-3′; reverse, 5′-AGCACCATTCAGAACCTCTCCT-3′), HDAC1 primer (forward, 5′-TCAAGCTCCACATCAGTCCTTC-3′; reverse, 5′-CAGTCGCTGTTTGATCTTCTCC-3′), KIF5B primer (forward, 5′-GCATAAAGGAAGCAGTCAGG-3′; reverse, 5′-ATTGCACTTGGGTGAGTTGG-3′), or MCAK primer (forward, 5′-AGAGCAAGTCCATTCCATCC-3′; reverse, 5′-TTCTGGGCCTTCTTCTCTTC-3′). As a control, GAPDH had expression of forward, 5′-TCTCCTCTGACTTCAACAGCGAC-3′; reverse, 5′-CCCTGTTGCTGTAGCCAAATTC-3′. The products of RT-PCR were analyzed using 2% agarose gel electrophoresis.

Western blotting analysis

Cells were resuspended in RIPA lysis buffer (0.5% sodium deoxycholate, 1% NP-40, 150 mM NaCl, 10 mM EDTA, 50 mM Tris–HCl [pH 7.5], 1 mM sodium vanadate, 0.1% sodium dodecyl sulfate [SDS] in the presence of 1× Halt™ Protease Inhibitor Cocktails, and Halt™ Phosphatase Inhibitor Cocktails [Thermo Fisher Scientific, USA]), followed by incubation on ice for 30 min and centrifugation at 14,000 rpm for 30 min at 4°C. Protein concentrations were determined using a Bio-Rad Protein assay (Bio-Rad Laboratories, CA, USA). Equal amounts of protein (30–50 μg/lane) were subjected to 10% SDS-PAGE and Western blotting was performed as previously described.⁹

Microtubule assembly assay

Separation of polymerized tubulin from tubulin dimers and analysis of the effects of docetaxel on tubulin polymerization in vitro were performed as described by Blagosklonny et al.²⁵ In brief, PC3 cells were lysed at 4°C in a hypotonic buffer containing 20 mM Tris–HCl (pH 6.8), 1 mM MgCl₂, 2 mM EGTA, protease inhibitor (1×), phosphatase inhibitor (1×), and 0.5% Nonidet P-40. Supernatants, in the form (cytosolic) of soluble tubulin heterodimers, were collected after centrifugation at 15,000 × g for 10 min at 4°C. The insoluble microtubule pellets (cytoskeleton) were dissolved in the SDS-polyacrylamide gel electrophoresis (PAGE) sampling buffer. Both soluble and insoluble forms were subjected to electrophoresis on a 10% SDS-PAGE gel, and then examined using Western blotting as previously described.⁹

Immunohistochemistry (IHC)

Immunostaining was performed on paraffin sections, using a Polymer Detection System (Bio SB, CA, USA). In brief, the sections were de-paraffinized in xylene and rehydrated through graded alcohols. Hydrogen peroxide, 0.3%, was added to block any endogenous peroxidase activity. Antigen retrieval was performed using immunoDNA Digestor (Bio SB, CA, USA) for 40 min at 95°C. Immunodetector protein block (Bio SB, CA, USA) was used for 30-min blocking at room temperature. The sections were incubated with anti-acetyl-tubulin antibody overnight and then used as a 1:100 dilution at 4°C. After washing, tissue sections were incubated at room temperature for 30 min with a Mouse/Rabbit PolyDetector secondary antibody (Bio SB). After washing, the antigen–antibody complex was applied and stained with DAB Chromogen (Bio SB, CA, USA) as a substrate. Counterstaining was performed lightly with hematoxylin (Merck, Darmstadt, Germany). Specific staining for acetyl-tubulin was detected as brown in color. Immunodetector protein block diluent was used instead of the first antibody as the negative control. All control slides yielded negative results. IHC of anti-acetyl-tubulin expression was evaluated in terms of both the proportion of positive cells and the staining intensity, as described previously.²⁶ The proportion of positive cells was scored at 0%–100% of the cells examined. Intensity was graded as follows: 0, no signal; 1, weakly positive; 2, moderately positive; and 3, strongly positive staining.

Immunofluorescence

Cells (5 × 10⁴) were cultured in 24-well plates to 70% cell confluency. Cells were washed thrice with PBS, fixed with 4% paraformaldehyde (PFA) for 20 min, permeabilized with 0.1% Triton-X for 10 min, and then blocked with PBS containing 5% normal goat serum and 1% BSA for 1 h at room temperature. Cells were incubated with primary antibody HDAC6 and acetyl-tubulin, respectively, overnight at 4°C. After being washed thrice with PBS, the cells were incubated with goat anti-rabbit IgG (Alexa Flour® 488) (Thermo Fisher Scientific, MA, USA) or goat anti-mouse (Alexa Flour® 594) (Thermo Fisher Scientific, MA, USA) secondary antibody (1:200) for 30 min at room temperature. Cells were washed twice with PBS, and the nuclei were then stained with DAPI for 10 min at room temperature. Stained cells were viewed using fluorescence microscopy.

Flow cytometry

Cells (5 × 10⁵) were plated in 6-cm culture dish for 24 h. After trypsinization, cells were washed with PBS and then recentrifuged. Then the cell pellets were fixed in 1 mL 70% cold EtOH (added 700 μL 100% cold ethanol) at 4°C overnight. Cell pellets were incubated in 0.5 mL permeabilized buffer (0.1% Triton X-100 + 0.1 mg/mL RNase A) for 1 h at 37°C, and then stained with 0.5 mL staining buffer (propidium iodide 20 μg/mL, RNase A 0.1 mg/mL, and 0.1% Triton X-100) for 30 min at 37°C in the dark room. The sample was analyzed using the Attune NxT Acoustic Focusing Cytometer (Invitrogen by Thermo Fisher Scientific).

Statistical analysis

The survival rate and levels of various targeting genes between PC3 and PC/DX25 cells were compared using t-test. All statistic tests were determined by two-sided with p < 0.05 being statistically significant.

RESULTS Upregulated expression of acetyl-tubulin was associated with docetaxel-resistant PCa cells and docetaxel-resistant CRPC tissues

A series of docetaxel-resistant sublines PC/DX were developed from PCa cell line, PC3. This study found higher expression of acetyl-tubulin, α-tubulin, β-tubulin, γ-tubulin, and βIII-tubulin in PC/DX25 than in PC3 (Figure 1A). PC/DX sublines also showed upregulated expression of acetyl-tubulin, α-tubulin, β-tubulin, and γ-tubulin at a dose-dependent manner (Figure 1B). Similarly, IHC results showed upregulated expression of acetyl-tubulin in docetaxel-resistant CRPC than in CRPC patient's tissues (Figure 1C). Furthermore, docetaxel significantly induced acetyl-tubulin expression in PC3 cells at a dose- and time-dependent manner (Figure 2A,B). Of note is that docetaxel, not cisplatin, specifically and significantly induced the expression of acetyl-tubulin in PC3 cells (Figure 2C).

FIGURE 1. Upregulated expression of acetyl-tubulin in docetaxel-resistant prostate cancer cells and docetaxel-resistant CRPC tissues. (A) Western blot analysis showed higher levels of acetyl-tubulin, α-tubulin, β-tubulin, γ-tubulin, and βIII-tubulin in PC/DX25 than in PC3 cells. (B) PC/DX25 cells exhibited upregulation of acetyl-tubulin, α-tubulin, β-tubulin, and γ-tubulin in a series of docetaxel-resistant sublines. Total proteins extracted from PC3 and PC/DX cells were examined by Western blotting. (C) Immunohistochemical staining of (a) human prostate cancer tissue of CRPC: weakly positive staining (1+); (b) DX-resistant CRPC (tumor): strongly positive staining (3+); (c) DX-resistant CRPC (normal): no signal (0); and (d) negative control (magnification 100×). Paraffin sections were stained with anti-acetyl-tubulin antibody. DX-resistant CRPC (normal) tissues were from normal-looking DX-resistant CRPC tissues. CRPC, castration-resistant prostate cancer.

FIGURE 2. Docetaxel specifically induced acetyl-tubulin protein expression in PC3 cells at a dose- and time-dependent manner. (A) PC3 cells were treated with docetaxel at concentrations of 0, 1, 3, 5, 10, and 30 nM, respectively for 5 h. (B) PC3 cells were treated with 10 nM docetaxel for 0, 0.5, 1, 2, 3, and 5 h, respectively. (C) PC3 cells were treated with 10 nM cisplatin or 10 nM docetaxel for 5 h. Total proteins were extracted and acetyl-tubulin, α-tubulin, β-tubulin, and GAPDH were examined using Western blotting.

Inhibition of HDAC6 mediated acetyl-tubulin expression and reversed sensitivity to docetaxel in PC3 cells

Previous research demonstrated that HDAC6 was a member of the HDAC family and function of specifically tubulin deacetylase.²⁷ RNA and protein levels of HDAC6 were significantly lower in PC/DX25 than in PC3 (Figure 3A–C). However, the RNA level of HDAC1, another member of the HDAC family, showed no significant difference (Figure 3B). Western blotting and immunofluorescent staining results showed knockdown of HDAC6 mediated upregulated expression of acetyl-tubulin in PC3 cells (Figure 3D,E). Knockdown of HDAC6 significantly increased IC₅₀ value of docetaxel from 1.59 to 15.87 μM (p < 0.01) in PC3 cells (Figure 3F,G). Tubastatin A (a HDAC6 inhibitor) significantly inhibited the HDAC6 expression but induced the acetyl-tubulin expression at a time-dependent manner (Figure 3H). Pretreatment with Tubastatin A enhanced docetaxel resistance (IC₅₀ = 17.28 nM) in PC3 cells as compared with that of the control (IC₅₀ = 6.34 nM) (Figure 3I). The IC₅₀ of PC3 with HDAC6 inhibitor was 2.73-fold higher than that of the control.

FIGURE 3. Downregulation of HDAC6 induced acetyl-tubulin expression and enhanced resistance to docetaxel in PC3 cells. Analysis of HDAC6 and HDAC1 gene expression in PC3 and PC/DX25 cells was performed using (A) RT-PCR and (B) real-time PCR. (C) Total proteins extracted from PC3 and PC/DX25 cells. The protein levels of HDAC6 and HDAC1 were examined using Western blotting. (D) PC3 cells were transfected with HDAC6 siRNA. After 48 h of transfection, the protein levels of HDAC6, acetyl-tubulin, α-tubulin, β-tubulin, and GAPDH were examined using Western blotting. (E) PC3 cells were transfected with HDAC6 siRNA for 8 h, double-stained with HDAC6 (green) and acetyl-tubulin (red) and then visualized with immunofluorescence. The nuclei were labeled with DAPI (blue); 400× magnification. (F) Cytotoxic sensitivity to docetaxel was examined using MTT assay. (G) The mean of triple independent IC50 values was greater in HDAC6 siRNA cells (15.87 μM) than in control cells (1.59 μM). PC3 with HDAC6 knockdown being 10-fold higher than control. (H) PC3 cells were treated with 2.5 μM of Tubastatin A for 24, 48, and 72 h. The protein levels of HDAC6, acetyl-tubulin, and GAPDH were examined by Western blotting. (I) Cytotoxic sensitivity to docetaxel was examined by MTT assay. Pretreated with Tubastatin A enhanced the resistance to docetaxel (IC50 = 17.28 nM) compared with control (IC50 = 6.34 nM) in PC3 cells. The IC50 of PC3 with HDAC6 inhibitor being 2.73-fold higher than control. HDAC6, histone deacetylase 6.

Upregulation of acetyl-tubulin expression by MCAK was associated with docetaxel resistance in PC3 cells

The gene and protein expressions of MCAK were significantly upregulated in PC/DX25 compared with PC3 cells (Figure 4A,B). Western blotting results showed higher expressions of acetyl-tubulin and MCAK protein at a dose-dependent manner in docetaxel-resistant sublines (Figure 4C). These findings revealed that acetyl-tubulin and MCAK coexisted in the cytoskeleton fraction of PC/DX25, and might with each other to bind on microtubule (Figure 4D). Moreover, docetaxel induced the upregulation of acetyl-tubulin, MCAK, Hsp90α, and Hsp90β in the cytoskeleton fraction at a dose-dependent manner (Figure 4E). Similarly, docetaxel induced the upregulation of acetyl-tubulin and MCAK in the cytoskeleton fraction of PC3 cells at a time- and dose-dependent manner (Figure 5). The survival rate of PC/DX25 was significantly reduced by MCAK knockdown compared with that of the control (Figure 6). The IC₅₀ value of MCAK knockdown and control cells were 0.34 μM and 0.62 μM, respectively, with MCAK knockdown being 0.55-fold lower than control cells. Of note is that knockdown of MCAK expression in PC/DX cells reversed sensitivity to docetaxel cytotoxicity at a dose- and time-dependent manner (Figure 6B,C). This study also found that co-treatment of MACK siRNA and docetaxel in PC/DC25 cells could significantly induce cell death than docetaxel only in subG1 fraction by flow cytometry analysis (Figure 6D).

FIGURE 4. Upregulation of microtubule-related motor protein in docetaxel-resistant PC3 cells. (A) MCAK gene expressions in PC3 and PC/DX25 cells were analyzed using real-time PCR. (B) The protein levels of MCAK and GAPDH were examined using Western blotting. (C) Higher expressions of acetyl-tubulin and MCAK proteins were examined at dose-dependent manner in docetaxel-resistant sublines by Western blotting. (D) PC/DX25 cells and (E) PC3 cells treated with docetaxel at concentrations of 0, 25, 200, and 500 nM for 6 h were collected, respectively. Total cell protein was separated into cytosolic phase; “S” and cytoskeleton phase; “P” by microtubule assembly assay. The protein levels of acetyl-tubulin, MCAK, Hsp90α, Hsp90β, α-tubulin, and β-tubulin. were examined using Western blotting. MCAK, mitotic centromere-associated kinesin.

FIGURE 5. Docetaxel induced MACK expression in PC3 cells at a dose- and time-dependent manner. (A) PC3 cells were treated with docetaxel at concentrations of 0, 1, 3, 5, 10, and 30 nM for 5 h, respectively. (B) PC3 cells were treated with 10 nM docetaxel for 0, 0.5, 1, 2, 3, and 5 h, respectively. Total cell protein was separated into cytosolic phase; “S” and cytoskeleton phase; “P” by microtubule assembly assay. The protein levels of acetyl-tubulin, MCAK, and GAPDH were examined using Western blotting. MCAK, mitotic centromere-associated kinesin.

FIGURE 6. Knockdown of MCAK expression resensitized with docetaxel in PC/DX25 cells. PC/DX25 cells were transfected with MCAK siRNA. (A) The protein levels of MCAK and GAPDH were examined using Western blotting. (B) PC/DX25 cells were transfected with MCAK siRNA for 8 h, followed by treatment with docetaxel at concentrations of 0, 5, 25, 50, 350, 500, and 750 nM for 48 h or (C) with 500 nM docetaxel for 0, 16, 24, and 48 h. Cytotoxic sensitivity to docetaxel was examined using MTT assay. (D) Cells were treated with 0 or 500 nM docetaxel for 48 h after siRNA transfection for 24 h. Cells were collected and cell cycle was analyzed using flow cytometry. MCAK, mitotic centromere-associated kinesin. *p [less than] 0.05; **p [less than] 0.01.

PLK1 was associated with docetaxel resistance in PC3 cells

Previous studies showed MCAK played a key role in chemoresistance of drug-resistant cells. As known, MACK was regulated by G2/M transition-related proteins, such as Aurora A, Aurora B, and PLK1. PC/DX sublines showed the upregulation of PLK1 and Aurora B expression at a dose-dependent manner (Figure 7A). Further overexpression of PLK1, we transfected GFP-PLK1-WT and GFP-PLK1-KD, a point-mutated kinase domain of PLK1 plasmids into PC3 cells, respectively by treatment of docetaxel for 48 h (Figure 7B). IC₅₀ values of PLK1-KD and PLK1-WT cells were 13.6 nM and 21.7 nM, respectively, with PLK1-KD knockdown being 0.62-fold lower than PLK1-WT cells (Figure 7C). Therefore, inactivation of PLK1 mediated cytotoxic sensitivity to docetaxel in PCa cells.

FIGURE 7. PLK1 was associated with docetaxel resistance in prostate cancer cells. (A) The protein levels of PLK1, Aurora A, and Aurora B in PC/DX sublines were examined using Western blotting. (B) PC3 cells were transfected with expressive GFP-PLK1-WT or PLK1-KD plasmid, respectively for 16 h, followed by treatment with docetaxel for 48 h. Protein levels of GFP-PLK1, MCAK, and GAPDH were examined using Western blotting. (C) Cytotoxic sensitivity to docetaxel was examined using MTT assay. IC50 values of 48-h docetaxel treatment was lower in PLK1-KD cells (13.6 nM) than in PLK1-WT cells (21.7 nM). MCAK, mitotic centromere-associated kinesin. *p [less than] 0.05; **p [less than] 0.01.

EGFR regulated up-expression of acetyl-tubulin

Our previous study demonstrated that the EGFR signal pathway played a key chemoresistant role in docetaxel-resistant PCa.⁹ Both EGFR and acetyl-tubulin were overexpressed in PC/DX25 (Figure 8A). Knockdown of EGFR significantly downregulated acetyl-tubulin expression in PC/DX25 (Figure 8B). Moreover, we also found EGF induced the activation of EGFR, p-EGFR, and acetyl-tubulin expression, but inhibited the HDAC6 expression at time-dependent manner (Figure 8C). The present results indicated that EGFR pathway involved in the docetaxel resistance by upregulation of acetyl-tubulin in PCa.

FIGURE 8. EGFR regulated the up-expression of acetyl-tubulin. (A) The protein levels of EGFR and acetyl-tubulin were examined in PC3 and PC/DX25 cells. (B) Knockdown of EGFR expression in PC/DX25 cells. Total proteins were extracted; and EGFR, acetyl-tubulin, α-tubulin, and β-actin were examined using Western blotting. (C) PC3 cells were treated with EGF (80 ng/mL) for 0.5, 1, 2, 3, and 5 h, respectively. Total proteins were extracted; and EGFR, p-EGFR(Y1068), HDAC6, acetyl-tubulin, α-tubulin, β-tubulin, and GAPDH were examined using Western blotting. HDAC6, histone deacetylase 6.

DISCUSSION

The present results showed significantly higher expression of acetyl-tubulin, α-tubulin, β-tubulin, γ-tubulin, βIII-tubulin, and MCAK (mitotic centromere associate kinesin) in PC/DX25 (docetaxel-resistant subline) than in PC3 (parental) cells. Conversely, PC/DX25 reversed the sensitivity to docetaxel by MCAK knockdown. Overexpression of PLK1 enhanced the cell survival rate and resistance to docetaxel in PC3. Of note is that EGFR activation induced the upregulated expression of acetyl-tubulin in PC3 cells. Upregulation of acetyl-tubulin played an important role in docetaxel-resistant PCa.

Previously, Burkhart et al. found the levels of isotype of microtubule may induce docetaxel resistance in some cancer cells.²⁸ Soucek et al. reported distinct molecular profiles of α-tubulin posttranslational modifications displayed by PCa cells.²⁹ This study found significantly higher expression of acetyl-tubulin, α-tubulin, β-tubulin, γ-tubulin, and βIII-tubulin in PC/DX25 than in PC3 cells (Figure 1). Ploussard et al. showed the isotype of βIII-tubulin played an important role in the development of docetaxel resistance in PCa cells.³⁰ The present results also confirmed upregulation of βIII-tubulin in PC/DX25 (Figure 1A), and found similarly upregulation of acetyl-tubulin, α-tubulin, β-tubulin, and γ-tubulin by increasing concentrations of docetaxel in PC/DX sublines (Figure 1B). These results explained the chemoresistant mechanism of overexpression of tubulin isotypes can escape and overcome docetaxel-induced apoptosis in docetaxel-resistant PC. Interestingly, we found docetaxel significantly induced acetyl-tubulin expression was at a dose- and time-dependent manner in PC3 cells (Figure 2). Therefore, upregulation of acetyl-tubulin may be a protective effect to resist docetaxel in PC cells. Similarly, we also found acetyl-tubulin was overexpressed in docetaxel-resistant CRPC than in CRPC patient's tissues (Figure 1C).

Previous studies showed histone deacetylase 6 (HDAC6) was involved in the reaction of tubulin de-acetylation.²⁷ We found the level of HDAC6 gene was significantly lower in docetaxel-resistant than in parental cells by cDNA microarray. Confirming data, the downregulation of HDAC6 gene and protein was examined by RT-PCR, real-time PCR, and Western blotting, respectively in PC/DX25 cells (Figure 3A–C). Conversely, PC3 cells showed the upregulation of acetyl-tubulin and significant resistance to docetaxel by knockdown of HDAC6 (Figure 3D–F). Tubastatin A, a potent and selective HDAC6 inhibitor mediated the similar phenomena of HDAC6 knockdown in PC3 cells (Figure 3G,H). Although previous data showed tubulin was a specific substrate of cytoplasmic HDAC6 enzyme,³¹ we suggested the other isoforms HDAC may involve in the deacetylation of tubulin. However, the RNA level of HDAC1, another member of the HDAC family, showed no significant difference in our study (Figure 3B). Actually, the mechanism of acetylation of tubulin mediating the docetaxel-resistance of PCa cells remained unclear.

Previous studies indicated that the motor protein kinesin can transport some specific molecules by microtubule as a track. The upregulation of microtubule acetylation promoted the binding and transport of kinesin-1, a member of the kinesin family, such as KIF5B.²⁰ Kinesin was not only with transport function, but also with the regulation of the composition and decomposition for microtubules.³² Interestingly, Ganguly et al. showed overexpression of MCAK, a member of the kinesin family, can induce the microtubule of dynamic breakdown and significantly mediate resistance to paclitaxel. They suggested these results were consistent with a hypothesis for paclitaxel resistance by dependent stability of the attachment of microtubules to their nucleating centers, and they implicate MCAK in the mechanism of microtubule detachment.³³ Based on our cDNA microarray data (data not shown), we found the upregulated expression of MACK, and confirm the gene or protein of MACK were significantly upregulated in PC/DX25, respectively (Figure 4A,B). Stathmin (STMN1), an important regulatory protein of microtubule dynamics has been well-characterized and was associated with paclitaxel resistance in breast cancer cells.³⁴ Interestingly, we also found the upregulated expression of STMN1 in PC/DX25 cells (data not shown). To further explore if the increase of tubulin acetylation would enhance the binding of MACK to the microtubules by microtubule assembly assay, as expected, we found docetaxel induced the upregulation of acetyl-tubulin, MCAK, Hsp90α, and Hsp90β in cytoskeleton fraction at dose-dependent (Figure 4D), and at time-dependent manner in PC3 cells (Figure 5), indicating that the increase coexisted in acetylated tubulin and MACK may act as a cytoprotective effect in docetaxel-resistant PC cells. Previous study demonstrated that acetyl-tubulin significantly induced Hsp90 recruitment to microtubule and stimulated the signaling pathway of the Hsp90 clients Akt/PKB and p53.²³ The dynamic interaction of acetyl-tubulin and Hsp90 may be as a carrier for the delivery of survival-related molecules to target organelle or nucleus for cellular responses. Therefore, we suggested the hyperacetylation of tubulin in microtubule may involve in the docetaxel-resistant mechanism of PC/DX25 cells. However, the real mechanism needed to be further studied in the future. Based on these results, we found the MACK played a key role in acetylation of tubulin and docetaxel resistance in PCa cells. Confirmed data, as shown in Figure 6 knockdown of MCAK expression resensitized to docetaxel cytotoxicity in PC/DX cells. Previous studies indicated MCAK was regulated by specific G2/M transition-related kinases, such as Aurora A, Aurora B, or PLK1,³⁵ phosphorylated the centromeric MCAK and regulated its localization and microtubule depolymerization activity. The Aurora A, Aurora B, or PLK1 kinases were usually upregulated in cancer cells. Similarly, our results showed the higher expression of Aurora A, Aurora B, and PLK1 in docetaxel-resistant sublines (Figure 7A). These results suggested MCAK played an important role in preventing apoptosis by evading G2/M phase arrest. Recent results demonstrated docetaxel inhibited the centrioles moved to the two poles of cell, and then led to cell apoptosis.^36–38 MCAK may play a regulation role to control the dynamic depolymerization of microtubules during cell division. Based on these findings, we suggested the chemoresistance of docetaxel induced the upregulation expression of acetyl-tubulin by suppression of HDAC6 or Sirt2 and then recruited the binding of MACK on microtubules for dynamic depolymerization or MCAK was controlled by Aurora A, Aurora B, or PLK1, and then jump out of G2/M phase to escape the G2/M arrest-mediated apoptosis.

We previously showed EGFR signal pathway was a key chemoresistant role in docetaxel-resistant PC.⁹ Interestingly, we found the hyperacetylation of tubulin and inhibition of the HDAC6 expression was through the signal pathway of EGFR activation, respectively at time-dependent manner in this study (Figure 7). Consistent finding was that treatment with EGF reduced the cytotoxicity of docetaxel in PC3 cells (data not shown). Our results showed EGFR pathway may involve the docetaxel resistance by upregulation of acetyl-tubulin in PC.

Previous studies showed that HDAC6 activity can be regulated by GSK3β²¹ or EGFR²² signal pathway, thereby to affect the acetylation of tubulin in cells. We predicted the induction of hyperacetylation tubulin was through Akt/GSK3β/HDAC signaling pathway, and the acetyl-tubulin recruited the binding and transport of Hsp90 to the microtubule; Hsp90 may act as a carrier for Akt, to mediate the Akt activation and thereby to induce cell proliferation and survival.²³ In addition, acetylation of tubulin may be regulated by the acetylated enzymes ELP3³⁹ or MEC-17 (GCN5).⁴⁰ However, these two proteins were not significantly expressed in our docetaxel-resistant PC cells (data not shown).

In this study, we found high expression of acetyl-tubulin were in docetaxel-resistant sublines and tissues, and up-expression of acetyl-tubulin to enhance the resistance to docetaxel in parental PC3 cells. Acetyl-tubulin may play a key role to regulate HDAC6 directly or indirectly through PI3K-Akt, MAPK, and STAT3 signaling pathways by EGFR activation, or acetylated enzyme GCN5 to induce the acetylation of tubulin and recruit the binding of MACK to microtubule. We supposed the acetyl-tubulin involved in the docetaxel resistance may be through the high binding of MACK and acetyl-tubulin to accelerate the dynamic disassembly of microtubules against docetaxel-mediated depolymerization of microtubules. In this study, we concluded EGFR-induced hyperacetylation of tubulin mediated docetaxel resistance by upregulation of MCAK and PLK1 in PCa cells.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.