BZ was purchased from Selleck Chemicals (Houston, TX, USA). CAM was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan), SAHA was from Cayman Chemical Co. (Ann Arbor, MI, USA), and tubacin was from Sigma-Aldrich (St. Louis, MO, USA). BZ, SAHA and tubacin were dissolved in dimethyl sulfoxide to make stock solutions at concentrations of 1, 10 and 1 mM, respectively. CAM was dissolved in ethanol to prepare stock solutions of 5 mg/ml.

For this study, the MM cell lines IM-9 and RPMI-8226, and the lung carcinoma cell line H226 were obtained from the American Type Culture Collection (ATCC) (Manassas, VA, USA). The human MM cell line KMS-12-PE was obtained from the Japanese Collection of Research Bioresources (JCRB) (Osaka, Japan). A CHOP^−/− murine embryonic fibroblast (MEF) cell line (CHOP-KO-DR) established from a 13.5-day-old CHOP^−/− mouse embryo by SV-40 immortalization and a CHOP^+/+ MEF cell line (DR-wild-type) established by SV-40 immortalization as a control cell line for CHOP-KO-DR were also obtained from the ATCC. These authorized cell lines were expanded and frozen in aliquots within 1 month after obtaining from the cell banks. Each aliquot was thawed and the cells were used for the experiments within 2 months after thawing. IM-9, H226, RPMI-8226, and KMS-12-PE cells were cultured in RPMI-1640 medium (Sigma-Aldrich) supplemented with 10% fetal bovine serum (FBS) (Biowest SAS, Nuaillé, France), 2 mM L-glutamine, penicillin (100 U/ml), and streptomycin (100 μg/ml) (Wako Pure Chemicals Industries, Tokyo, Japan). CHOP-KO-DR and DR-wild-type cells were maintained in Dulbecco’s modified Eagle’s medium (Sigma-Aldrich) supplemented with 10% FBS, 2 mM L-glutamine, penicillin (100 U/ml), and streptomycin (100 μg/ml). All cell lines were cultured in a humidified incubator containing 5% CO₂ and 95% air at 37°C.

The number of viable cells was assessed using CellTiter-Blue Cell Viability Assay kit (Promega Corp., Madison, WI, USA) according to the manufacturer’s instructions as previously described in detail (16).

Immunoblotting was performed as previously described (15,16). In brief, cells were lysed with RIPA lysis buffer supplemented with a protease and phosphatase inhibitor cocktail (both from Nacalai Tesque, Kyoto, Japan). Cellular proteins were quantified using a DC Protein Assay kit (Bio-Rad, Hercules, CA, USA). Equal amounts of proteins were loaded onto the gels, separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), and transferred onto Immobilon-P membrane (Millipore, Billerica, MA, USA). The membranes were probed with primary antibodies (Abs) such as anti-acetylated-α-tubulin (6–11B-1) mAb, anti-α-tubulin (B-7) mAb, anti-GAPDH (6C5) mAb, anti-HDAC6 (H-300) Ab, and anti-Ub (P4D1) mAb, which were all purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA) and anti-PARP Ab (Cell Signaling Technology, Inc., Danvers, MA, USA). Immunoreactive proteins were detected with horseradish peroxidase-conjugated secondary Abs (Cell Signaling Technology, Inc.) and an enhanced chemiluminescence reagent (Millipore). Densitometry was performed using a Molecular Imager ChemiDoc XRS System (Bio-Rad).

For the gene silencing of HDAC6 in RPMI-8226 and H226 cells, HDAC6 siRNA and control siRNA were purchased from Life Technologies (Grand Island, NY, USA) and whose sequences are described as follows: HDAC6 sense, CCAGCACAGUCUUAUGGAUGCUAU and antisense, AUAGCCAUCCAUAAGACUGUGCUGG. siRNAs were diluted to a final concentration of 33 nM in Opti-MEM I (Life Technologies). Transfection was performed with the cells at 40% confluency using Lipofectamine RNAiMAX transfection reagent (Life Technologies) according to the manufacturer’s instructions. Knockdown efficiency was assessed by immunoblotting.

Total RNA was isolated from cell pellets using Isogen (Wako Pure Chemicals Industries) and genomic DNA was removed using RQ1 RNase-Free DNase (Promega Corp.) at 37°C for 30 min, followed by extraction with phenol chloroform and ethanol precipitation. Reverse-transcription using a PrimeScript RT Master Mix (Takara Bio, Inc., Shiga, Japan) was performed according to the manufacturer’s instructions. Real-time polymerase chain reaction (PCR) was performed on 3 ng of cDNA using validated SYBR-Green gene expression assays for human ER stress-related genes (CHOP, BAX, BIM, DR5, GADD34 and GRP78) in combination with SYBR Premix Ex Taq II (Takara Bio, Inc.). The sequences of primers and reaction conditions were previously described (16). Quantitative real-time PCR was performed in duplicates in a Thermal Cycler Dice Real-Time System TP800 (Takara Bio, Inc.). The data were analyzed using Thermal Cycler Dice Real-Time System Software version 5.00 (Takara Bio, Inc.), and the comparative Ct method (2^−ΔΔCt) was used for the relative quantification of gene expression. The data of real-time PCR products were standardized to GAPDH as an internal control. To confirm the specific amplification of target genes, each gene product was further separated by using 1.5% agarose gel after real-time PCR to detect a single band at the theoretical product size, as well as by analysis of the dissociation curve for detecting a single peak.

Cells were spread on slide glasses using Cytospin 4 Centrifuge (Thermo Fisher Scientific, Inc., Rockford, IL, USA) to make slide glass preparations. Cells were fixed for 20 min in ice-cold methanol and permeabilized with 0.1% Triton X-100 for 20 min, followed by blocking with 2% bovine serum albumin in TBST (25 mM Tris, 137 mM NaCl, 2.7 mM KCl, 0.05% Tween-20, pH 7.4) for 1 h. Cells were immunostained with primary Abs such as mouse anti-vimentin (V9) mAb, mouse anti-Ub mAb, and lysosomal-associated membrane protein-1 (LAMP-1) (H4A3) mAbs (all from Santa Cruz Biotechnology, Inc.). The secondary Abs used for fluorescence detection were Alexa Fluor^® 488 F(ab′)2 fragment of goat anti-mouse IgG (H+L) Ab (Life Technologies). Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) (Wako Pure Chemicals Industries). Slides were mounted with SlowFade Gold antifade reagent (Life Technologies). Analysis by confocal microscopy was performed using the confocal laser scanning fluorescence microscope FV10i-DOC.(Olympus Corp., Tokyo, Japan).

Cells were lysed with Triton X-100 lysis buffer (10 mM Tris-HCl, 150 mM NaCl, 2% Triton X-100, pH 7.8) supplemented with a protease inhibitor cocktail (Nacalai Tesque). The lysates were centrifuged at 12,000 g for 30 min at 4°C. The supernatant was then collected as the soluble fraction. The pellets (which contain the insoluble protein) were then resuspended in sodium dodecyl sulfate (SDS) lysis buffer (10 mM Tris-HCl, 150 mM NaCl, 2% SDS, pH 7.8) and sonicated for 30 sec with a tip sonicator VP-5S (Taitec, Saitama, Japan) to prepare the insoluble fraction. Equal volumes of each pellet and supernatants were boiled for 5 min in SDS-PAGE sample buffer (125 mM Tris-HCl, 4% SDS, 20% glycerol, 0.002% BPB, pH 6.8) and analyzed by SDS-PAGE.

Cells were fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) for 1 h. The samples were further fixed in 1% osmium tetroxide for 1 h, dehydrated in graded ethanol (30–100%), and embedded in Quetol 812 epoxy resin (Nisshin EM Co., Ltd., Tokyo, Japan). Ultrathin sections were cut with an Ultracut J microtome (Reichert-Jung, Vienna, Austria). The sections were stained with lead nitrate and uranium acetate, and subjected to electron microscopic analysis using the scanning electron microscope JEM-1200 EXII (JEOL, Tokyo, Japan).

All data are expressed as mean ± SD. Statistical analysis was performed using Mann-Whitney U test (two-tailed).

Treatment with SAHA for 48 h resulted in a dose-dependent inhibition of cellular growth in all MM cell lines (Fig. 1A). The IC₅₀ was 1.2 μM in KMS-12-PE, 1.5 μM in RPMI-8226, and 2.1 μM in IM-9 cells. SAHA-treated MM cells exhibited apoptotic morphologic features such as fragmentation of the nucleus and formation of an apoptotic body along with cleavage of PARP and caspase-3 (data not shown). The regulation of the stability and function of a microtubule has been associated with α-tubulin reversible acetylation, and HDAC6 functions as α-tubulin deacetylase (27). Immunoblotting using a specific Ab for the acetylated α-tubulin revealed that, in response to SAHA (at 0.5 μM for KMS-12-PE and RPMI-8226, at 1 μM for IM-9), the acetylation of α-tubulin was detectable within 16 h and persisted for at least 48 h in all three cell lines tested (Fig. 1B). In our previous studies, autophagy flux was shown to be blocked by CAM, and that the BZ and CAM combination treatment resulted in ER-stress overloading followed by enhanced induction of apoptosis in MM and breast cancer cells (16,17). Since HDAC6 has been implicated in aggresome formation that results in sequestering overabundant intracellular unfolded proteins (18), we examined whether the SAHA plus BZ and/or CAM combination treatment further increased cytotoxicity via ER stress loading. As shown in Fig. 2A, the combined treatment with SAHA/BZ, but not with SAHA/CAM, potentiated cell growth inhibition. Notably, although CAM treatment alone at 50 μg/ml did not inhibit the growth of cells, the combined SAHA/BZ/CAM treatment showed a clearly pronounced cytotoxicity compared with the SAHA/BZ or BZ/CAM treatment in all MM cell lines. This enhanced cytotoxicity was mediated through apoptosis induction since the pronounced expression of the cleaved PARP was observed in response to the suboptimal concentrations of the SAHA/BZ/CAM combination in IM-9 cells (Fig. 2B).

To confirm that this enhanced cytotoxicity is mediated through HDAC6 inhibition, we next attempted to knock down HDAC6 with siRNA. Pre-treatment with HDAC6 siRNA effectively suppressed the expression level of HDAC6 in RPMI-8226 cells. Under this condition, the cell growth inhibition by the BZ/CAM treatment was pronounced compared with the cells pre-treated with control siRNA. In addition, tubacin, a specific HDAC6 inhibitor, reproduced the pronounced cytotoxicity when combined with BZ and BZ/CAM (Fig. 3A and B) (28). These results suggest that HDAC6 inhibition appears to be involved in the enhancement of cytotoxicity induced by the addition of SAHA to BZ/CAM-containing cell culture medium.

To further confirm that the pronounced cytotoxicity by SAHA shown in Fig. 2A was due to the inhibition of HDAC6 activity involved in aggresome formation, we assessed aggresome formation in the presence or absence of SAHA. In RPMI-8226 cells, immunocytochemistry using anti-vimentin mAb showed that the combined treatment with BZ/CAM induced a dense deposit of vimentin in the perinuclear region, which is a well-known marker for aggresome (29). This vimentin deposit was clearly suppressed in the presence of SAHA (Fig. 4A). However, since the MM cell lines including RPMI-8226 were very sensitive to the SAHA/BZ/CAM treatment, in which apoptosis was induced within 18 h of exposure to these reagents (data not shown), precise analysis of aggresome formation was practically difficult using MM cells. We therefore used the lung squamous carcinoma cell line H226 which is less sensitive, but shows pronounced cytotoxicity in response to the SAHA/BZ/CAM combination similarly to MM cell lines (Fig. 4B). As shown in Fig. 4C, the prominent vimentin deposit in the perinuclear region was observed after combined treatment with BZ/CAM for 24 h. The vimentin deposit with high fluorescent intensity was evidently suppressed in the presence of SAHA, indicating that SAHA inhibits BZ/CAM-induced aggresome formation (Fig. 4C). Similarly to SAHA, treatment with HDAC6 siRNA suppressed the vimentin condensation in the perinuclear region in H226 cells (Fig. 4D–F).

In addition, immunocytochemistry using anti-Ub Ab exhibited the clustering of Ub-positive dots in the perinuclear region by BZ/CAM, whereas they were dispersed in the presence of SAHA in H226 cells (data not shown). As an alternative assessment of aggresome, we treated KMS-12-PE with BZ, CAM, or BZ/CAM in the presence or absence of SAHA. The cells were then treated with lysis buffer containing 2% Triton X-100, and cellular proteins were subsequently fractionated into the detergent-soluble and the -insoluble fraction. The detergent-insoluble fraction (cell pellets) was sonicated for 30 sec in the presence of 2% SDS and boiled for 5 min. Thereafter, the samples were loaded on the gels, separated by 11.25% SDS-PAGE, and immunoblotted with anti-Ub Ab. The samples derived from the detergent-soluble fractions were processed similarly. As shown in Fig. 5, the increased intracellular polyubiquitinated proteins were detectable after treatment with BZ and BZ/CAM, but not with CAM or SAHA in the detergent-soluble fraction (upper panel). In the detergent-insoluble fraction (lower panel) at higher molecular weights, prominent polyubiquitinated proteins were detected as aggresome contents in the cells treated with BZ/CAM. Treatment with BZ alone also induced accumulation of the detergent-insoluble polyubiquitinated proteins but to a lesser extent than that induced by treatment with BZ/CAM. It was noteworthy that the increased ubiquitinated proteins in the detergent-insoluble fraction were suppressed in the presence of SAHA. This also indicates the inhibition of aggresome in response to SAHA.

Immunocytochemistry using anti-LAMP-1 mAb showed lysosome accumulation in the perinuclear region along with aggresome formation after BZ/CAM treatment. As is the case of vimentin-positive aggresome formation, the clustering of LAMP-1-positive dots was suppressed in the presence of SAHA (Fig. 6A). We therefore performed electron microscopy for precise analysis of the cytosolic area strongly stained with anti-vimentin Ab. However, contrary to what we had expected, electron microscopy exhibited no typical features of the aggresome structure such as massive accumulation of closely packed electron-dense particles surrounded by a cage of bundles of intermediated filaments in the region of the centrosome, which was first described by Johnston et al in 1998 (30). Instead, there was the clustering of a number of lysosomes, autophagosomes, autolysosomes, and mitochondria with increased filaments in response to the BZ/CAM treatment (Fig. 6B). A few numbers of unstructured particles with a rather high density, which might be the aggresome precursor, were detectable in this area (Fig. 6C). It was noteworthy that many lysosomal fusions or lysosomal aggregates in various sizes were observed. These might represent the phenomenon termed ‘lysophagy’ which has recently been reported by others, that is, the lysosome engulfed into an autophagosome or autolysosome (Fig. 6D) (31). In the presence of SAHA, the clustering of these organelles in the perinuclear region was suppressed to some extent along with the decreased filamentous structures. However, an increased number of lysosomes including autolysosomes was still observed compared with the cells treated with either CAM or BZ alone (Fig. 6A).

Since aggresome is reported to form as an adaption process against overabundant intracellular unfolded or misfolded proteins, we hypothesized that HDAC6 inhibition with SAHA further enhances BZ/CAM-induced ER stress loading to the cells. As we expected, the expression levels of ER stress-related genes including CHOP were markedly pronounced in the presence of SAHA; the expression ratios of CHOP to untreated control cells increased to 5-fold by BZ/CAM treatment, and further increased to 25-fold by BZ/CAM/SAHA treatment in IM-9 cells (Fig. 7A). In H226 cells, similarly to SAHA, HDAC6 siRNA treatment enhanced BZ/CAM-induced CHOP expression compared with the cells treated with control siRNA (Fig. 7B). CHOP is an ER stress-related pro-apoptotic transcription factor which upregulates pro-apoptotic genes (e.g., BIM, BAX, DR5) and one of the critical molecules playing a role in the induction of apoptosis in response to ER stress (9,10). In addition, enhanced cytotoxicity was shown by a wild-type MEF cell line along with CHOP upregulation by the combination of the three reagents, whereas the CHOP^−/− MEF cell line derived from a CHOP-deficient mouse almost completely canceled this pronounced cytotoxicity (Fig. 7C). Therefore, the pronounced cytotoxicity and apoptosis in response to the SAHA/BZ/CAM combination treatment shown in Fig. 2A is strongly suggested to be due to the upregulation of CHOP in response to ER stress loading.