SEARCH

SEARCH BY CITATION

Keywords:

  • chromatin modulation;
  • targeted therapy;
  • histone deacetylase;
  • colon cancer;
  • prostate cancer

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Inhibition of histone deacetylase (HDAC) activity as stand-alone or combination therapy represents a promising therapeutic approach in oncology. The pan- or class I HDAC inhibitors (HDACi) currently approved or in clinical studies for oncology give rise to dose-limiting toxicities, presumably because of the inhibition of several HDACs. This could potentially be overcome by selective blockade of single HDAC family members. Here we report that HDAC11, the most recently identified zinc-dependent HDAC, is overexpressed in several carcinomas as compared to corresponding healthy tissues. HDAC11 depletion is sufficient to cause cell death and to inhibit metabolic activity in HCT-116 colon, PC-3 prostate, MCF-7 breast and SK-OV-3 ovarian cancer cell lines. The antitumoral effect induced can be mimicked by enforced expression of a catalytically impaired HDAC11 variant, suggesting that inhibition of the enzymatic activity of HDAC11 by small molecules could trigger the desired phenotypic changes. HDAC11 depletion in normal cells causes no changes in metabolic activity and viability, strongly suggesting that tumor-selective effects can be achieved. Altogether, our data show that HDAC11 plays a critical role in cancer cell survival and may represent a novel drug target in oncology.

Inhibition of histone deacetylase (HDAC) activity by small molecules has been shown to provide a therapeutic benefit to patients with diverse diseases including malignancies.1 The HDAC family is divided into classes I, IIa/b and IV, which together comprise 11 zinc-dependent enzymes, as well as into class III, which encompasses the 7 NAD+-dependent sirtuins.2 HDACs deacetylate both nuclear histone and nonhistone proteins such as transcription factors, and thereby play important roles in the regulation of numerous cellular processes.2 Analysis of knockout mice lacking single HDACs has revealed highly specific functions for the different family members during both, development and adulthood.3 The specific roles of individual HDACs in cancer are not yet fully clear,4 but deciphering these specificities has the power to support development of truly targeted therapeutics for diverse tumor entities.5, 6

Small molecule HDAC-inhibitors (HDACi) tested in patients in the context of clinical trials or approved as drugs to treat cutaneous and peripheral T cell lymphomas are currently all pan-HDACi or target the most abundantly expressed class I HDACs (HDAC1, 2, 3 and 8).7 Apart from the success achieved in treating T cell lymphomas, efficacies observed against other tumor types, especially solid tumors, have been disappointing.5, 8 The dose-limiting toxicities observed may account for the low efficacy.9 For both, “broad spectrum”-HDACi (vorinostat) and the newer class I HDACi (depsipeptide, MGCD0103 and MS-275/entinostat), the major toxicities reported include severe fatigue, gastrointestinal toxicity and hematological side effects, including reversible thrombocytopenia and neutropenia.5, 10 These observations are not surprising if one considers the central role of HDACs as key regulators of chromatin structure and post-translational modifiers of numerous proteins, which at least in the case of the class I HDACs, are ubiquitously expressed in many cell types and tissues. Given that individual HDAC family members control fundamentally different genetic programs,2 selective inhibitors may lead to both, improved efficacy and drug safety.

We recently unraveled an HDAC11-controlled molecular mechanism promoting malignancy of childhood neuroblastoma through epigenetic suppression of the BMP4 gene.11 Analysis of the HDAC11 mRNA expression in a variety of solid cancers and respective healthy tissues using the web-based ONCOMINE platform12 revealed that HDAC11 was among the top 1% differentially overexpressed genes in mixed lobular and ductal breast carcinoma versus normal breast tissue (p-value: 1.76 E−6), the top 2% differentially overexpressed genes in hepatocellular carcinoma versus normal liver tissue (p-value: 1.02 E−15) and among the top 4% differentially overexpressed genes in renal pelvis urothelial carcinoma versus normal kidney tissue (p-value: 2.18 E −8). Here we investigate the effects of depleting HDAC11, the most recently identified and sole class IV HDAC family member,13 on metabolic activity and apoptosis in different human carcinoma cell lines as compared to normal cells, thereby assess the broader application of novel drugs targeting HDAC11 beyond treatment of neuroblastoma.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Cell Culture

The HCT-116, MCF-7 and PC-3 cancer cell lines were obtained from the German Collection of Microorganisms and Cell Cultures (DSMZ, Braunschweig, Germany), and the SK-OV-3 cancer cell line from the American Type Culture Collection (ATCC, LGC Standards GmBH, Wesel, Germany). Primary human fibroblasts (PHFs) were isolated from foreskin. HCT-116 and PC-3 cell lines as well as PHFs were cultured in DMEM/HAM's F12 (Invitrogen, Darmstadt, Germany) supplemented with 10% FCS (Sigma Aldrich, Hamburg, Germany) and 1% nonessential amino acids (Invitrogen) at 37°C, 5% CO2. SK-OV-3 cells were grown in DMEM supplemented with 10% FCS. MCF-7 cells were cultured in RPMI 1640 (Biochrom, Berlin, Germany) supplemented with 10% FCS. The primary human mammary epithelial cells (HMECs) were from Lonza (Basel, Switzerland). Human nontumorigenic hTERT-immortalized mammary epithelial cells with reduced p53 levels (HMEC-T53) were generated at Bayer Pharma AG, Berlin, Germany. HMEC-T53 cells were cultured in HuMEC basal serum-free medium (Invitrogen) with HuMEC Supplement (Invitrogen) and bovine pituitary extract (Invitrogen), likewise at 37°C, 5% CO2. Cells were regularly monitored for mycoplasma, Acholeplasma laidlawii and squirrel monkey retrovirus (SMRV) infection by the in-house high-throughput multiplex cell contamination test (McCT) service.14

Small Interfering RNA, Short Hairpin RNA and Plasmid DNA

For short-term HDAC11 silencing, cells were transiently transfected with 25 nM HDAC11-specific small interfering RNA (siRNA; Qiagen, Hildesheim, Germany) or nonsilencing negative control siRNAs (AllStars Negative Control siRNA, Qiagen; siGenome RISC-free control siRNA, Thermo Fisher Scientific, Schwerte, Germany) using HiPerFect (Qiagen). For comparison, cells were transiently transfected with 25 nM HDAC1-, HDAC2-, HDAC3- or HDAC8-specific siRNAs (Qiagen) or treated with 0.5–2 μM of the pan-HDAC-inhibitor vorinostat (SAHA; Selleck Biochemicals, Munich, Germany) dissolved in dimethyl sulfoxide (DMSO). For the generation of cell clones with stable silencing of HDAC11, preselected short hairpin RNAs (shRNAs) were first cloned into a lentiviral vector derived from BLOCK-iT pLenti6 (Invitrogen). The shRNA sequences of the forward oligonucleotides were as follows: negative control shRNA: 5′-TTCTCCGAACGTGTCACGTTTCAAGAGAACGTGACACGTTCGGAGAA-3′; positive control shRNA (shCOPA-4044): 5′-GGATCAGTCCTCTGCAGTTTCTTCAAGAGAGAAACTGCAGAGGACTGATCC-3′; HDAC11 shRNA #1: 5′-CACCGCACACGAGGCGCTATCTTAATTCAAGAGATTAAGATAGCGCCTCGTGTGC-3′; HDAC11 shRNA #2: 5′-CACCGGAGAGGAACATCAAGAAATCTTCAAGAGAGATTTCTTGATGTTCCTCTCC-3′; HDAC11 shRNA #3: 5′-CACCGCATCATTGCTGACTCCATACTTCAAGAGAGTATGGAGTCAGCAATGATGC-3′. Lentiviruses were produced by cotransfecting the respective pGT396-Puro construct with lentiviral Packaging Mix (Invitrogen) into HEK-293FT cells (Invitrogen). Viral supernatants were collected and concentrated by ultracentrifugation (2 hr, 50,000g). Viral titers were determined using the HIV p24 ELISA (Perkin Elmer). Transductions were carried out using p24 concentrations of 1 μg/ml for 6 hr at 37°C. Selection was performed in presence of the appropriate antibiotic.

For plasmid transfection, 4–8 μg DNA were used together with Lipofectamine (Invitrogen). pExp3.2-HDAC11wt was constructed by cloning the full length human HDAC11 sequence with the Gateway System into pcDNA3.2/V5 (Invitrogen) (pExp3.2-HDAC11 = pcDNA3.2/V5-DEST × pENTR221-HDAC11). The catalytic domain of HDACs consists of a set of conserved amino acids, which participate in the Asp-His charge relay system mediating the removal of acetyl groups.15 To generate a mutant pExp3.2-HDAC11 construct, the corresponding residues His-142 and His-143 (http://us.expasy.org/sprot/) were mutated to alanine by introducing two point mutations with the QuikChange® II Site-Directed Mutagenesis Kit (Stratagene) into the HDAC11 sequence (H142A/H143A). For HDAC8, the mutant H142A/H143A form was shown to have lost more than 80% of its catalytic activity relative to wild-type.16 For the expression of Myc-tagged HDAC11, both wild-type and mutant forms of HDAC11 were transferred via an LR-reaction into a Gateway-compatible version of pcDNA3.1 directing the expression of the Myc-tag (MEQKLISEEDL) amino-terminal to the inserted open reading frame. The correct sequences of all inserts were verified by DNA sequencing (GATC, Germany).

RNA Extraction, Reverse Transcription, Gene Expression Profiling and Quantitative RT-PCR

Total RNA was isolated from cell cultures with Qiagen's RNeasy MiniKit. Gene expression profiles of HCT-116 cells 48 and 96 hr after HDAC11 knockdown were generated with one negative control siRNA and one HDAC11-specific siRNA in three biological replicates using the whole-genome Centrix-8 bead-array platform (Illumina). M-MLV reverse transcriptase (Invitrogen) was used to transcribe cDNAs using random hexamers (Invitrogen) as primers. Gene expression data were validated by qRT-PCR with SYBR Green Dye (Eurogentec, Cologne, Germany) on an ABI Prism 7700 thermal cycler and normalized to the averaged ATCB and GAPDH expression. These genes were expressed at constant levels in all experimental conditions. All primers were obtained from Qiagen.

Trypan Blue Assay

Cells were seeded, treated as indicated and evaluated for number and viability using the VI-CELL Cell Viability Analyzer (Beckman Coulter, Krefeld, Germany).17 Trypan blue is a diazo dye, which selectively enters cells with disrupted membranes and not viable cells with intact cell membranes.

WST-1 Metabolic activity Assay

Cells were seeded and treated as indicated. The WST-1 reagent (Roche, Mannheim, Germany) was added, and cells were reincubated for another 0.5–4 hr. Thereafter, the formazan formed was quantitated at 450 nm on an ELISA plate reader.

Flow Cytometry

For the characterization and quantification of potentially apoptotic cells, HCT-116 cells were labeled with Annexin V as described by the manufacturer (Becton Dickinson, Heidelberg, Germany) or labeled with Nicoletti stain as described.18 Samples were analyzed with a FACS Calibur (Becton Dickinson) using the Cell Quest Pro software (Becton Dickinson). Measurements were acquired in Fl-2 in logarithmic mode.

Caspase 3-Like Activity Assay

Cells were seeded, treated as indicated and lysed in cell lysis buffer (Biovision, Mountain View, CA) for 10 min on ice. Thereafter, the reaction buffer containing the AFC-labeled caspase 3-specific peptide, DEVD (Biovision), was added. Caspase 3-like activity was measured at 37°C in black 96-well plates using a fluorescence plate reader with a 380 nm excitation filter and a 530 nm emission filter.18

Data Analysis and Statistics

At least three independent biological replicates were conducted for all cell culture experiments. Results were compared using an unpaired t-test, and p-values <0.05 were considered significant. Genetic pathway analysis of the datasets generated on a whole-genome Centrix-8 bead-array platform was performed using the MetaCore (GeneGo) Pathway Analysis Software, Version 6.4. Gene IDs of these datasets were mapped onto the gene IDs represented by pathway maps and networks in MetaCore and terms enriched were thereby identified. HDAC11 mRNA expression in solid cancers versus respective healthy tissues was studied using the ONCOMINE 4.4 Research Edition. ONCOMINE is a microarray database and integrated data mining platform,12 based on high-throughput cancer profiling data so that target expression across a large volume of cancer types can be assessed online (http://www.oncomine.org).

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

HDAC11 Depletion Reduces Metabolic Activity and Viability of Cancer but not Normal Cells

As a basis for functional analyses, HDAC11 mRNA expression was assessed in 305 cell lines by reanalysis of in-house expression data gathered on Affymetrix HGU133Plus2.0 arrays (Supporting Information Table 1).19 Among others, HDAC11 was strongly expressed in colon, prostate, ovarian and breast cancer cell lines. The established cell lines HCT-116, PC-3, SK-OV-3 and MCF-7 representing high incidence tumor entities were chosen as model systems for further in vitro functional studies. We tested whether HDAC11 depletion in these cancer cell lines altered metabolic activity and viable cell number. HDAC11 was depleted using knockdown via small interfering RNAs (siRNAs). Cells grown in 6-well plates or 100 mm dishes were transiently transfected with three different siRNAs targeting HDAC11, a scrambled control or a Cy3-labeled negative control siRNA. Transfection efficiency was assessed microscopically using the Cy3-labeled negative control siRNA, and knockdown efficiency was determined by qRT-PCR. Transfection efficiency ranged between 85 and 95% at 24 hr (Supporting Information Fig. 1a), and mRNA knockdown efficiency between 70 and 95% at 36 hr (Supporting Information Fig. 1b). All HDAC11 siRNAs used had previously been investigated for specificity and potential compensatory effects on other HDACs.11 At 96 hr, clear phenotypic differences indicative of impaired viability were microscopically observed in all investigated cancer cell lines upon HDAC11 depletion (Fig. 1a). Metabolic activity was reduced by ∼65% in the colon cancer cell line HCT-116, 59% in the prostate cancer cell line PC-3, 62% in the ovarian cancer cell line SK-OV-3 and 48% in the breast cancer cell line MCF-7 in the WST-1 assay at 96 hr (Fig. 1b). For comparison, the most responsive cancer cell line HCT-116 was treated for 72 hr with 0.5–2 μM of the pan-HDAC-inhibitor vorinostat20 or transfected with siRNAs targeting the most abundantly expressed class I HDACs 1, 2 and 3 as well as the fourth class I HDAC family member, HDAC8.21 These analyses revealed a comparable effect of HDAC11 depletion and 1 μM vorinostat treatment (Supporting Information Fig. 2a). Further, HDAC11 depletion caused a similar decrease in metabolic activity as the depletion of HDAC1 and HDAC2 (Supporting Information Fig. 2b), whereas the depletion of HDACs 3 and 8 induced a weaker reduction in metabolic activity (Supporting Information Fig. 2b). Having shown a comparable or stronger effect upon HDAC11 depletion as compared to pan-HDAC-inhibitor treatment or class I HDAC depletion, we turned back to the panel of cancer cell lines and analyzed the influence of HDAC11 depletion upon cancer cell viability. The number of viable cells was reduced 2.9-fold in HCT-116, 1.8-fold in PC-3 and 1.4-fold in SK-OV-3 and MCF-7 cells (Fig. 1c).

thumbnail image

Figure 1. HDAC11 depletion in human carcinoma cell lines reduces metabolic activity and viability. (a) Light microscopy of HCT-116, PC-3, SK-OV-3 and MCF-7 cells 96 hr after transfection with nonsilencing AllStars negative control siRNA negative control siRNA #1 (Neg Co siRNA #1; upper panel) or HDAC11 siRNA #1 (lower panel). Magnification = 200×. (b) Metabolic activity of HCT-116, PC-3, SK-OV-3 and MCF-7 cells 96 hr after transfection with HDAC11 siRNAs or negative control siRNAs. Numbers are expressed relative to untreated cells (ut). Error bars indicate SD. (c) Viable cell numbers were assessed by automated trypan blue stained cell counting for HCT-116, PC-3, SK-OV-3 and MCF-7 cells 96 hr after transfection. Numbers are expressed relative to untreated cells (ut). Error bars indicate SD.

Download figure to PowerPoint

PHFs and human hTERT-immortalized mammary epithelial cells with reduced levels of p53, HMEC-T53, were used to investigate the role of HDAC11 in nontumorigenic, normal cells. Both cell lines expressed HDAC11 detectably but to a lower extent than the cancer cell lines (Supporting Information Table 2). Transfection efficiency with siRNAs ranged between 90 and 95% (Supporting Information Fig. 3a) and mRNA knockdown efficiency between 70 and 94% (Supporting Information Fig. 3b). No morphological differences were seen in PHFs and HMEC-T53 cells upon HDAC11 depletion (Fig. 2a). The metabolic activity was 99% in PHFs and 98% in HMEC-T53 cells as compared to controls at 96 hr, respectively (Fig. 2b). The number of viable PHFs and HMEC-T53 cells remained unchanged as compared to controls (Fig. 2c). Together, HDAC11 depletion clearly impaired both viability and metabolic activity of all cancer cell lines examined, regardless of the entity from which they were derived, while normal cells were unaffected.

thumbnail image

Figure 2. HDAC11 depletion in normal human cells has no influence on metabolic activity and viability. (a) Light microscopy of primary human fibroblasts (PHFs) and human mammary epithelial cells (HMEC-T53) 96 h after transfection with negative control siRNA #1 (Neg Co siRNA #1; upper panel) or HDAC11 siRNA #1 (lower panel). Magnification = 200×. (b) Metabolic activity of PHFs and HMEC-T53 cells 96 hr after transfection with HDAC11 siRNAs or negative control siRNAs. Numbers are expressed relative to untreated cells (ut). Error bars represent SD. n.s., not significant. (c) Viable cell numbers assessed by automated trypan blue stained cell counting for PHFs and HMEC-T53 cells 96 hr after transfection. Numbers are expressed relative to untreated cells (ut). Error bars represent SD. n.s., not significant.

Download figure to PowerPoint

HDAC11 Depletion Induces Apoptosis in Cancer but not Normal Cells

We next analyzed the number of dead cells after siRNA-mediated HDAC11 depletion in the most responsive cancer cell lines HCT-116 and PC-3 and compared them with those in nontumorigenic PHFs and HMEC-T53 cells using automated trypan blue staining. When compared to cells transfected with nonsilencing negative control siRNAs, HDAC11 depletion increased the number of trypan blue positive, dead cells 6.8-fold in HCT-116 and 4.5-fold in PC-3 cells at 72 hr, respectively (Figs. 3a and 3b). In PHFs and HMEC-T53 cells, no changes in the number of dead cells were observed upon HDAC11 depletion (Figs. 3c and 3d). We investigated whether HDAC11 depletion had caused death by apoptosis in HCT-116 cells by performing caspase 3-like activity assays. Caspase 3-like activity was induced, with up to 4.5- and 9.6-fold compared to the controls at 72 and 96 hr, respectively (Fig. 4a). In line with the trypan blue assay, no induction of caspase 3-like activity was detected after HDAC11 depletion in PHFs (Fig. 4b). Next, we performed flow cytometric analyses after Annexin V staining of HCT-116 cells to detect early apoptotic cells and Nicoletti staining to detect HCT-116 cells with fragmented and released DNA. We found an ∼2.4-fold increase in Annexin V positive HCT-116 cells (Fig. 4c) and a 3.5-fold increase in cells with fragmented DNA in subG1 (Fig. 4d), thereby confirming the induction of apoptosis by HDAC11 depletion in HCT-116 cells. We next tested whether sustained HDAC11 depletion affected the viability of normal cells in longer-term assays. Our data obtained in the short-term assays with siRNA-mediated knockdown of HDAC11 in cancer and normal cell lines suggested that HDAC11 depletion might mediate tumor-selective effects. To use HDAC11 inhibition therapeutically, longer time-courses would be required. Thus, we tested the effects of sustained HDAC11 depletion with shRNAs delivered by lentiviruses into PHFs and, as a positive control, into HCT-116 cells. All shRNAs used yielded a knockdown efficiency of at least 75% (Supporting Information Figs. 1c and 3c). This analysis revealed no effects on viability of PHFs (Fig. 5a), but strong and sustained antitumoral effects in HCT-116 cells (Fig. 5b). Taken together, the results show HDAC11 depletion to trigger apoptotic programs in cancer cells but not in normal cells. To decipher the pathways triggering the HDAC11 depletion-phenotype in cancer cells, whole-genome expression profiling of HCT-116 cells 48 and 96 hr after HDAC11 knockdown was performed. All genes up- and down-regulated ≥2-fold 48 hr after siRNA transfection were selected. Together, 418 Illumina transcripts were down-regulated and 310 up-regulated at 48 hr. The fold regulation of all 728 Illumina transcripts was also visualized at 96 hr of transfection, and the pathway analysis performed by the MetaCore (GeneGo) Pathway Analysis Software, Version 6.4. This analysis revealed that HDAC11 depletion in HCT-116 causes a differential expression of genes involved in cytoskeleton remodeling, chromatin assembly and transcription (Supporting Information Table 3). Of note, WNT and PPAR-signaling pathways were regulated by HDAC11 targeting. Both pathways are strongly linked to colon cancer tumorigenesis.22, 23 On single gene level, several genes associated with survival in patients with colon carcinoma (HMOX1),24 growth inhibition (GCNT3),25 apoptosis (AK2, TFAP2A)26, 27 and differentiation of colorectal cancer stem cells (BMP4)28 were found to be favorably regulated. The differential expression of these genes was validated by qRT-PCR (data not shown). Taken together, gene expression profiling of HCT-116 cells following HDAC11 depletion identified the differential expression of genes linked to signaling pathways causally involved in cytoskeleton remodeling, chromatin assembly and transcription.

thumbnail image

Figure 3. HDAC11 depletion induces cell death in human carcinoma cell lines but not in normal cells. Numbers of trypan blue stained, dead cells from automated trypan blue stained cell counting for (a) HCT-116, (b) PC-3, (c) PHFs and (d) HMEC-T53 72 hr after transfection with HDAC11 siRNAs or nonsilencing negative control siRNAs. Numbers are expressed relative to untreated cells (ut). Error bars represent SD. n.s., not significant.

Download figure to PowerPoint

thumbnail image

Figure 4. HDAC11 depletion induces apoptosis in HCT-116 cells but not in PHFs. Caspase 3-like protease activity in (a) HCT-116 cells and (b) PHFs 72 and 96 hr after transfection with HDAC11 siRNAs or negative control siRNAs. Shown are fold changes in caspase 3-like activity over that measured in untreated (ut) cells. HCT-116 cells 72 hr after transfection with HDAC11 siRNAs or negative control siRNAs were further analyzed by (c) Annexin V staining and (d) Nicoletti staining. Error bars represent SD. n.s., not significant.

Download figure to PowerPoint

thumbnail image

Figure 5. HDAC11 depletion in PHFs has no long-term effects on proliferation. Metabolic activity measured by WST-1 assay of (a) PHFs and (b) HCT-116 cells 7, 14 and 21 days following sustained depletion of HDAC11 with three different lentivirally transduced shRNAs. Numbers are expressed relative to negative control (Neg Co) transduced cells. Error bars represent SD. Pos Co shRNA, positive control shRNA.

Download figure to PowerPoint

Enforced Expression of a Catalytically Impaired HDAC11 Variant (H142A, H143A) in HCT-116 Cells Mimics the Phenotype Caused by RNAi

In a complementary approach to the RNAi studies, we transiently overexpressed either wild-type HDAC11 or the catalytically impaired H142A/H143A HDAC11 variant in HCT-116 cells (Supporting Information Fig. 1d). The enforced expression of wild-type HDAC11 promoted proliferation of HCT-116 cells (Figs. 6a and 6b). Enforced expression of the catalytically impaired variant, however, recapitulated the HDAC11 loss-of-function phenotype observed after short- or long-term HDAC11 knockdown, resulting in decreased cell number as compared to controls (Figs. 6a and 6b). Thus, targeting the enzymatic activity of HDAC11 by small molecule inhibitors could trigger the desired antitumoral phenotypic changes. Finally, we cotransfected HCT-116 cells with HDAC11 siRNA and the wild-type HDAC11 construct (Supporting Information Fig. 1d) and analyzed by trypan blue staining the number of viable and dead HCT-116 cells. This analysis revealed that the enforced expression of wild-type HDAC11 protein significantly counteracted the HDAC11 knockdown-induced reduction in viable cells by ∼30% and the increase in dead cells by ∼10% (Figs. 6c and 6d). Taken together, the data obtained showed clear phenotypic changes upon HDAC11 depletion in models representing tumor entities with a high incidence and evidence for tumor selectivity as no major changes in morphology, metabolic activity and number of viable and dead cells were observed in normal cells.

thumbnail image

Figure 6. Enforced expression of a catalytically impaired HDAC11 variant mimics the siRNA-mediated phenotype. (a) Light microscopy (200x magnification) of HCT-116 cells transfected with wild-type (WT) Myc-HDAC11 or the catalytically impaired H142A/H143A variant (mut Myc-HDAC11). (b) Evaluation of viable cell numbers using automated trypan blue stained cell counting 72 hr after transfection of HCT-116 cells with wild-type (WT) Myc-HDAC11 or the catalytically impaired H142A/H143A variant (mut Myc-HDAC11). Numbers are expressed relative to empty vector transfected cells. *p < 0.05, **p < 0.01. (c) Plasmid-mediated enforced expression of wild-type (WT) HDAC11 protein counteracted the reduction in viable HCT-116 cells caused by HDAC11 knockdown. HCT-116 viable cell numbers measured by trypan blue staining were evaluated 96 hr after siRNA transfection and 72 hr after plasmid transfection. **p < 0.001. (d) Plasmid-mediated enforced expression of wild-type (WT) HDAC11 protein counteracted the increase in dead HCT-116 cells caused by HDAC11 knockdown. HCT-116 dead cell numbers measured by trypan blue staining were evaluated 96 hr after siRNA transfection and 72 hr after plasmid transfection. **p < 0.001.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

We show here that both HDAC11 depletion and enforced expression of a catalytically impaired HDAC11 variant induced apoptosis and inhibited metabolic activity of colon, breast, prostate and ovarian cancer cell lines. In contrast to the strong antiproliferative and proapoptotic effects observed in these carcinoma cell lines, HDAC11 depletion had no detectable effect on two different types of normal cells. These results suggest tumor selectivity and a relatively broad therapeutic window, which is important for the future development of small molecules selectively inhibiting HDAC11.

HDAC11 is the most recently identified zinc-dependent HDAC family member and has little homology to the other classical HDACs belonging to classes I (HDAC1, HDAC2, HDAC3 and HDAC8), IIa (HDAC4, HDAC5, HDAC7 and HDAC9) and IIb (HDAC6 and HDAC10).13 Phylogenetic analyses revealed that it is most closely related to both HDAC3 and HDAC8,15 but its overall similarity is so low that it was classified as a novel group IV.2 HDAC11 is highly conserved among different species, and in humans, high transcript levels are detected in different regions of the brain, the skeletal muscle, kidney and testis.13 Enforced expression of HDAC11 in HEK-293 cells followed by coimmunoprecipitation indicates that HDAC11 and HDAC6 interact directly or indirectly.13 Moreover, HDAC11 was recently shown to interact with the replication licensing factor, Cdt1.29 To date, however, no reports have linked HDAC11 to the multiprotein HDAC complexes containing HDAC1, 2 or 3.2 HDAC11 was shown to suppress the expression of interleukin 10 (IL10) in antigen-presenting cells of the rat.30 Otherwise, little is currently known about the function of HDAC11 during development, in healthy tissues or during disease.

Most recently, we showed that HDAC11 represses BMP4 transcription in highly malignant neural crest-derived neuroblastoma, an embryonic childhood tumor with often fatal outcome in the high-risk constellation.11 Reverting the epigenetic silencing of BMP4 by HDAC11 depletion or inhibition of the enzymatic activity of HDAC11 triggers BMP4 signaling in neuroblastoma cells, which in turn induces a tumor cell transcriptome predictive of favorable patient outcome when expressed in the primary neuroblastomas. Moreover, stimulated BMP4 signaling overcomes the malignant potential of neuroblastoma in all types of preclinical neuroblastoma models including cell lines, primary cells and xenografts in mice.11 Future studies aim at elucidating the pathways controlled by HDAC11 in other tumor entities including those investigated in this study. For example, BMP4 signaling was shown to contribute to the malignant potential of neuroblastoma and to the biology of colon cancer initiating cells.28 Upon exposure to recombinant BMP4 protein, these cells forfeit their tumorigenic potential.28 In summary, the data presented here show that HDAC11 plays an important role in the control of proliferation and survival pathways of several carcinoma cell lines, and thereby, represents a potential novel drug target in oncology.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The authors wish to thank Jasmin Wünschel for excellent technical assistance, and Kathy Astrahantseff for comments on and editing of the manuscript. They are also indebted to Florian Prinz for generation of lentiviral shRNAs.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IJC_27876_sm_SuppFig1.tif23777KSupporting Information Figure 1
IJC_27876_sm_SuppFig2.tif22301KSupporting Information Figure 2
IJC_27876_sm_SuppFig3.tif23258KSupporting Information Figure 3
IJC_27876_sm_SuppTab1.doc327KSupporting Information Table 1
IJC_27876_sm_SuppTab2.doc35KSupporting Information Table 2
IJC_27876_sm_SuppTab3.doc39KSupporting Information Table 3

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.