Histone deacetylase inhibitors and hydroxyurea modulate the cell cycle and cooperatively induce apoptosis


Therapy resistance represents a major problem for disease management in oncology. Histone deacetylase inhibitors (HDACi) have been shown to modulate the cell cycle, to induce apoptosis and to sensitize cancer cells for other chemotherapeutics. Our study shows that the HDACi valproic acid (VPA) and the ribonucleotide reductase inhibitor hydroxyurea (HU) potentiate the pro-apoptotic effects of each other towards several cancer cell lines. This correlates with the HU-induced degradation of the cyclin-dependent kinase inhibitors (CDKI) p21 and p27, mediated by the proteasome or caspase-3. Moreover, we found that caspase-3 activation is required for VPA-induced apoptosis. Remarkably, p21 and p27 can confer resistance against VPA and HU. Both CDKI interact with caspase-3 and compete with other caspase-3 substrates. Hence, p21 and p27 may contribute to chemotherapy resistance as apoptosis inhibitors. Since the biological effects of VPA and HU could be achieved at concentrations used in current treatment protocols, the combined application of these compounds might be considered as a potential strategy for cancer treatment.


Although disease management in cancer has improved significantly, therapy resistance leading to tumour recurrence still hampers improvement of long-term survival (La Porta, 2007). Consequently, understanding the molecular basis of therapy resistance is of utmost importance for the development of novel therapeutic strategies (Koon and Atkins, 2007).

In cancer cells, genes regulating the cell cycle and apoptosis are commonly mutated or aberrantly expressed (Hanahan and Weinberg, 2000). The expression of such genes is determined by chromatin remodelling and histone modifications such as acetylation. Enzymes reciprocally controlling histone acetylation are histone acetyltransferases and histone deacetylases (HDACs; Kouzarides, 2000). Genes inducing differentiation, cell cycle arrest and apoptosis are induced by HDAC inhibitors (HDACi; Krämer et al., 2001), which are currently tested in clinical studies (Boyle et al., 2005; Bug et al., 2005; Kuendgen et al., 2006).

Several reports describe that apoptosis induction by HDACi depends on deregulated cell cycle checkpoints (Finzer et al., 2001; Shin et al., 2003). Interestingly, a reduction of the cyclin-dependent kinase inhibitor (CDKI) p21WAF/CIP1 (p21), which arrests cells in the G1 phase of the cell cycle, enhances HDACi-induced cell death (Rosato et al., 2003; Nguyen et al., 2004). This can be explained by the fact that p21 attenuates caspase-3 (Suzuki et al., 1999, 2000), which is crucial for apoptosis induction by several chemotherapeutics. The p21-related CDKI p27Kip1 (p27) also regulates the cell cycle and apoptosis (Blagosklonny, 2002), but its contribution to HDACi-triggered processes remains to be fully uncovered (Klisovic et al., 2003; Maggio et al., 2004).

Besides HDACi, numerous other agents induce cell cycle arrest. An example is the ribonucleotide reductase inhibitor hydroxyurea (HU), which stalls cells in S phase (Szekeres et al., 1997), induces apoptosis (Schrell et al., 1997a) and sensitizes tumours to chemotherapeutics (Van den Berg and Von Hoff, 1995). Since HU is a potent, pro-apoptotic anticancer drug with low toxicity in vivo, it is used to treat solid and haematological tumours (Schrell et al., 1997b; Montefusco et al., 2001; Bug et al., 2005; Kuendgen et al., 2006).

Although HDACi and HU are promising anticancer drugs, tumour cells resistant to each of them exist (Harris, 1985; Krämer et al., 2006). We investigated how the HDACi valproic acid (VPA; Göttlicher et al., 2001) and HU affect various cancer cell lines and observed increased sensitivity against the combination of these drugs. Our results suggest that the CDKI p21 and p27, and caspase-3 critically determine apoptosis induction by VPA and HU.


HDACi and HU synergistically trigger apoptosis in melanoma cells

In our initial experiments, we used the VPA-sensitive melanoma cell line SK-Mel-37 as a model. Treatment with the HDACi VPA and trichostatin A (TSA) as well as with HU resulted in reduced viability of SK-Mel-37 cells as measured by MTT test (Figure 1a). Interestingly, we observed significantly reduced cell viability following coadministration of HDACi and HU. These effects were observed at clinically achievable concentrations of VPA and HU (Bug et al., 2005; Kuendgen et al., 2006).

Figure 1

Hydroxyurea (HU) enhances histone deacetylase inhibitor (HDACi)-induced apoptosis. (a) The viability of SK-Mel-37 cells was determined by MTT test after exposure to valproic acid (VPA; V, 1.5 mM) or trichostatin A (TSA; T, 10 nM) and HU (1.5 mM) alone or combined for 24–72 h (Ctl, untreated cells). *P<0.05; two-sided t-test. (b) Assessment of caspase-3 activation by western blot and measurement of caspase-3 activity by colorimetric assay. SK-Mel-37 cells were treated for 24 h (Ctl, untreated; V, 1.5 mM VPA; HU, 1.5 mM HU; fl, full-length; cf, cleaved forms p12/p17/p19). Ac-DEVD-pNA is cleaved by caspase-3 into pNA, which is measured at OD405 nm after a 16 h incubation time of the assay. (c) PARP-cleavage was analysed by western blot of lysates from MZ-Mel-19 cells treated with 1.5 mM VPA (V) or 100 nM TSA (T) for 24–48 h or combined with 1.5 mM HU (HU) for 24 h (Ctl, untreated).

The reduced viability of SK-Mel-37 cells observed upon drug treatment appeared to be caused by apoptosis. Indeed, we detected enhanced cleavage of caspase-3, -8 and -9 as well as an increased caspase-3 activity in lysates from cells treated with VPA and HU, compared to control cells or cells treated with only one compound (Figure 1b and data not shown). These results were confirmed by independent apoptosis assays, such as detection of cleavage of the caspase-3 substrate PARP (Figure 1c), counting of apoptotic nuclei after Hoechst-staining and the caspase-induced inactivation of EGFP (Supplementary Figures A and B).

Notably, multiple cancer cell lines, representing various tumour entities, are sensitive to the combined VPA/HU-treatment (Supplementary Figure C and Supplementary Table).

HDACi and HU modulate the expression of p21 and p27

Apoptosis induction by HDACi involves deregulated cell cycle checkpoints. We therefore examined whether treatment with VPA and HU caused cell cycle alterations. Fluorescence-activated cell sorter (FACS) analysis revealed that VPA arrested SK-Mel-37 cells in G1 and reduced the S and G2/M populations (Figure 2a). HU mainly ablated the G0/G1 population and increased the number of cells in S phase. Compared to untreated cells, the combination of both agents decreased the number of cells in G1 and G2/M and, potently, induced apoptotic DNA cleavage (sub-G1 fraction). The pan-caspase inhibitor Z-VAD-FMK blocked DNA fragmentation, which shows that VPA combined with HU triggered caspase-dependent apoptosis.

Figure 2

Hydroxyurea (HU) and histone deacetylase inhibitors (HDACi) alter the cell cycle and decrease cyclin-dependent kinase inhibitor (CDKI) expression. (a) Apoptotic chromatin fragmentation and cell cycle alterations induced by valproic acid (VPA; V, 1.5 mM) and HU (1.5 mM) added alone or combined for 24 h were detected by propidium iodide–FACS analysis (Ctl, untreated). Z-VAD-FMK (Z-VAD, 20 μ M) was added if stated. (b) Expression of p21, p27 and caspase-3 (cf, cleaved forms) were investigated by western blot of SK-Mel-37 lysates. Cells were untreated (Ctl) or incubated with 1.5 mM VPA (V), 100 nM trichostatin A (TSA; T) and/or 1.5 mM HU for 24 h. (c) Effect of 1.5 mM VPA (V) and/or 1.5 mM HU added for 24 h on the p21 and p27 mRNA levels in SK-Mel-37 cells was analysed by reverse transcription (RT)–PCR. (d) CDKI expression in SK-Mel-37 and NW-Mel-450 melanoma cells was investigated by western blot (Ctl, untreated; V, 1.5 mM VPA; T, 100 nM TSA; 24 h). (e) Effects of VPA (V, 1.5 mM) and HU (1.5 mM) on p21 and p27 expression in NW-Mel-450 cells were analysed by western blot (24 h treatment; Ctl, untreated). (f) Influence of the proteasome on the HU-mediated decrease of p21 and p27 was analysed by western blot. NW-Mel-450 cells were incubated for 24 h with HU (1.5 mM), MG-132 (1 μ M) or left untreated (Ctl). (g) Effect of Z-VAD-FMK on the HU-mediated decrease of p27 was analysed by western blot. NW-Mel-450 cells were incubated for 24 h with HU (1.5 mM) alone or with Z-VAD-FMK (Z-VAD, 20 μ M; Ctl, untreated).

As CDKI are critical cell cycle regulators, we tested whether the expression of p21 and p27 is affected by VPA/HU treatment. Immunoblot analysis showed that HDACi induced the accumulation of both proteins, whereas their expression was reduced by HU (Figure 2b). To discriminate between transcriptional and post-transcriptional effects, we first analysed p21 and p27 mRNA levels by reverse transcription (RT)–PCR (Figure 2c). Whereas p21 mRNA levels were induced by VPA or VPA/HU treatment, they were not affected by HU alone. In contrast, p27 mRNA expression remained unaffected by VPA, but was induced by VPA/HU.

These data imply post-transcriptional mechanisms counteracting the translation of elevated p21/p27 mRNA levels into increased protein expression. To dissect further this mechanism, we used NW-Mel-450 melanoma cells, which express high endogenous levels of both CDKI (Figure 2d). This allowed us to analyse the effects of HU in cell lines of the same tumour type with different endogenous p21 and p27 levels. Similar to SK-Mel-37 cells, p21 and p27 protein levels were reduced by HU in NW-Mel-450 cells (Figure 2e), which could be due to degradation by caspases or the proteasome. Interestingly, the proteasomal inhibitor MG-132 prevented the HU-induced degradation of p21, whereas this compound decreased p27 levels and increased apoptosis (Figure 2f and data not shown). In contrast, Z-VAD-FMK counteracted the HU-induced p27 degradation (Figure 2g) and increased p27 induction by TSA (Supplementary Figure D). These findings are in agreement with the described cleavage of p27 by caspase-3 (Levkau et al., 1998), which appears different from the proteasomal p27 degradation at the G1–S transition (Pagano et al., 1995).

Our data show that HU counteracts the induction of p21 and p27 by HDACi. This correlates with increased apoptosis, indicating that p21 and p27 levels might be critical for VPA/HU-induced apoptosis.

HU can sensitize HDACi-resistant melanoma cells to HDACi-induced apoptosis

Our observation that HU increases the sensitivity of HDACi-sensitive SK-Mel-37 to VPA encouraged us to treat HDACi-resistant NW-Mel-450 cells (Krämer et al., 2006) with a combination of VPA and HU. FACS analysis showed that VPA/HU treatment dose-dependently induced apoptosis of these cells (Figure 3a). Z-VAD-FMK reduced the apoptotic DNA degradation and PARP cleavage (Figure 3b), indicating that these are caspase-mediated processes in VPA/HU-treated NW-Mel-450 cells. Thus, HU sensitizes HDACi-resistant cells to apoptosis and this correlates with decreased p21 and p27 levels (Figure 2e).

Figure 3

Hydroxyurea (HU) sensitizes histone deacetylase inhibitor (HDACi)-resistant cell lines to valproic acid (VPA)-induced apoptosis and decreases cyclin-dependent kinase inhibitor (CDKI)-expression. (a) Sensitivity of NW-Mel-450 cells treated with VPA (V, 0.75–1.5 mM) and HU (0.5–1.5 mM) for 24 h was determined by propidium iodide–FACS analysis. The occurrence of the apoptotic sub-G1 fraction is shown (Ctl, untreated). (b) Z-VAD-FMK (Z, 20 μ M) blocks apoptosis which was measured as in (a) and by western blot against PARP. NW-Mel-450 cells were treated with VPA (1.5 mM) and HU (1.5 mM) for 24 h (Ctl, untreated). (c) NW-Mel-450 cells were treated 24 h with 1.5 mM VPA and/or 1.5 mM HU (Ctl, untreated). TRADD, caspase-8 (fl, full-length; cf, cleaved form), HDAC2 and tetraacetylated histone H4 (AcH4) were detected by western blot.

Moreover, the pro-apoptotic protein TRADD, which triggers caspase-8 activation, was increased by VPA and HU (Figure 3c), while HDAC2, a factor overexpressed in tumour cells (Krämer et al., 2003; Zhu et al., 2004), was reduced. These effects were not dependent on histone hyperacetylation, since HU even decreased histone acetylation levels (Figure 3c).

We conclude that HU dose-dependently sensitizes HDACi-resistant cells to VPA and that this process involves several pro-apoptotic pathways.

p21 and p27 can bind and inhibit caspase-3

Previous reports and our data indicate that p21 and p27 can be cleaved by caspase-3 and that p21 inhibits pro-caspase-3 activation (Levkau et al., 1998; Suzuki et al., 1999, 2000). Such processes require physical interaction of these proteins. Superose-6 fractionation of NW-Mel-450 cell lysates showed that caspase-3 partially comigrates with p21 and p27 (Figure 4a). Colorectal RKO cells harbouring an inducible p21 or p27 expression system (RKOp21i or RKOp27i; Schmidt et al., 2000) permit to analyse protein interactions in the presence of varying amounts of p21 or p27. Immunoprecipitation experiments revealed a strong increase in CDKI–caspase-3 association upon p21/p27 induction, likely due to increased CDKI levels (Figure 4b). Interaction of p21/p27 with caspase-3 was also detectable in NW-Mel-450 cells (data not shown). We next investigated the localization of p21, p27 and caspase-3 in NW-Mel-450 cells by indirect immunofluorescence using confocal microscopy. In agreement with our biochemical data, we observed a significant colocalization of p21 and p27 with caspase-3 (Supplementary Figure E).

Figure 4

Caspase-3 and cyclin-dependent kinase inhibitors (CDKI) interact physically and functionally. (a) Western blot analysis of Superose-6 column fractions (7–24, 7: void) to investigate the presence of complexes containing endogenous caspase-3, p21 and p27 in untreated NW-Mel-450 cells. Fraction numbers (#) and corresponding molecular weights (MW; kDa) are indicated. (b) RKO cells containing muristerone A-inducible p21 or p27 expression constructs were untreated or incubated with muristerone A (3 μ M) for 24 h. Endogenous caspase-3 was immunoprecipitated with a rabbit polyclonal antibody and western blot was performed to detect p21 and p27 with mouse monoclonal antibodies. Reprobing confirmed equal caspase-3 precipitation (pre, normal rabbit pre-immune serum; Input: 12% of material used for immunoprecipitation). (c) Recombinant caspase-3 (p17-p12)2-activity against Ac-DEVD-pNA in the presence of in vitro translated p21, p27, Gal or Z-VAD-FMK (Z-VAD, 50 μ M) was analysed by colorimetric assay at the time points indicated (8/12/24 h; Gal protein, negative control; Z-VAD-FMK (Z-VAD, 50 μ M), positive control for caspase-3 inhibition). (d) Cleavage of recombinant in vitro translated p21 and p27 (TNT) by caspase-3 (p17-p12)2 after 24 h was assessed by western blot.

Subsequently, we analysed whether CDKI also affect the enzymatic activity of caspase-3. In the presence of in vitro translated p21 or p27, we noticed reduced cleavage of the substrate Ac-DEVD-pNA by recombinant caspase-3 (Figure 4c). This inhibition could be overcome by prolonged incubation, probably due to caspase-3-mediated degradation of p21 and p27 (Figure 4d).

Hence, p21 and p27 are not only able to bind caspase-3 in vitro and in vivo but equally inhibit its activity in trans.

Roles of caspase-3, p21 and p27 in the interplay of VPA and HU

We used RKOp21i and RKOp27i cells to analyse the impact of p21 and p27 on VPA/HU-mediated apoptosis. In RKO control cells, VPA did not trigger apoptosis (Figure 5a) but induced growth arrest (data not shown). FACS analysis revealed that, similar to NW-Mel-450 cells, HU treatment of RKO, RKOp21i and RKOp27i cells increased the apoptotic sub-G1 fraction, which was further enhanced by VPA. In contrast, apoptosis was dramatically reduced upon induction of either p21 or p27, emphasizing their functional significance for the regulation of programmed cell death by VPA/HU (Figure 5a). In vitro caspase-3 assays confirmed that endogenous caspase-3 was activated in RKO, but not in induced RKOp21i/RKOp27i cells (Figure 5b).

Figure 5

Expression of p21 and p27 inversely correlates with histone deacetylase inhibitor (HDCAi)/hydroxyurea (HU)-induced apoptosis and caspase-3 induction. (a) Apoptosis induction in RKO, RKO p21- and RKO p27-inducible cells (RKO; p21i; p27i) due to valproic acid (VPA) and HU treatment for 48 h (Ctl, untreated; V, 1.5 mM VPA; HU, 0.5 mM) was analysed by propidium iodide –FACS analysis. If indicated, muristerone A (3 μ M) was administered immediately after VPA/HU. Muristerone A was added to RKO cells to control for unspecific effects. (b) Analysis of caspase-3 activity by Ac-DEVD-pNA proteolysis after 16 h was performed with lysates from cells treated as described in (a). (c) Cyclin-dependent kinase inhibitors (CDKI) expression in cells treated as described in (a) was investigated by western blot (p27i/p21i, muristerone A-induced p27/p21 expression). (d) MCF-7 cells stably transfected with empty vector (MCF-7pc3.1) or with full-length caspase-3 (MCF-7Casp3) were treated with 1.5 mM VPA, 1.5 mM HU or both substances for 48 h. Apoptosis induction was analysed by propidium iodide fluorescence-activated cell sorter (PI–FACS) analysis (sub-G1, apoptotic cells; Ctl, untreated). (e) MCF-7 cells were transfected with p21-EGFP. Cells were treated with 1.5 mM VPA, 1.5 mM HU or both for 24 h. Tagged and endogenous p21, p27 and acetylated histones H3/H4 (acetylated histone H3 (AcH3)/acetylated histone H4 (AcH4)) were detected by western blot (Ctl, untreated). (f) MCF-7pc3.1 or MCF-7Casp3 were treated with 1.5 mM VPA for 48 h (Ctl, untreated). Western blot was used to detect p21, p27, caspase-3 (fl, full-length; cf, cleaved form) and AcH4. (g) MCF-7Casp3 cells were transfected with small interfering RNAs (siRNAs) against both, p21 and p27 (40 pmol) on 2 consecutive days. After 48 h, cells were exposed to 1.5 mM VPA for 48 h. Western blot was employed to detect p21, p27, caspase-3 and PARP (fl, full-length; cf, cleaved form). (h) Diagram of proteins affected by VPA and HU.

Since HU reduces p21 and p27 (Figures 2b–g), we investigated their levels in VPA/HU-treated, induced RKOp21i and RKOp27i cells. In contrast to endogenous p21/p27, both proteins were unaffected by HU and VPA, probably reflecting overloading and inhibition of the proteasome and caspase system (Figure 5c).

To study directly the functional involvement of caspase-3 in apoptosis mediated by VPA/HU, we used caspase-3-negative MCF-7 breast cancer cells (MCF-7pc3.1) (Jänicke et al., 1998) and MCF-7 cells reconstituted with caspase-3 (MCF-7Casp3). Whereas MCF-7pc3.1 cells underwent differentiation in response to VPA (data not shown), cell death was triggered in MCF-7Casp3 cells treated with VPA or VPA/HU (Figure 5d). These results demonstrate that pro-apoptotic VPA effects require caspase-3 and illustrate HDACi-mediated sensitization of cells to HU.

In accordance with Figures 2b–g, endogenous p21 and ectopically expressed p21-EGFP (under control of an HDACi-sensitive CMV-promoter) were induced by VPA and this was prevented by coincubation with HU (Figure 5e). In contrast, p27 remained stable under such treatment conditions unless caspase-3 was reintroduced (Figure 5f). This result underscores the crucial role of caspase-3 for the degradation of p27.

To assess whether p21 and p27 are the major targets of the apoptotic interaction of VPA and HU, both CDKI were knocked down by RNAi in MCF-7Casp3 cells. This approach showed that a reduction of p21 and p27 promoted apoptosis in response to VPA as indicated by the enhanced cleavage of caspase-3 and PARP (Figure 5g), increased in vitro caspase-3 activity and decreased cell viability (Supplementary Figure F).

We conclude that p21 and p27 are key regulators of apoptotic pathways. The levels of these CDKI determine whether cells treated with VPA and HU undergo apoptosis.


Cancer results from multiple defects in key control elements regulating cell proliferation and apoptosis. Thus, the combination of drugs acting on different cellular targets has been recognized as a promising therapeutic strategy. We analysed the interplay between VPA and HU and demonstrate here for the first time that their combined application additively or synergistically induced apoptosis in various cancer cells. Interestingly, cells resistant to one compound even became sensitive by addition of the other. Other analyses also showed increased cytotoxicity of HDACi and S-phase toxins in combination (St Croix et al., 1996; Maggio et al., 2004). These data support our results and indicate different modes of action and diverse targets of these drugs.

We found that HDACi and HU severely perturbed the cell cycle and the expression of cell cycle and apoptosis regulators (Figure 5h). The fact that HU reduced p21 and p27 levels may explain why cell lines strongly varying in basal expression of these CDKI similarly underwent S-phase arrest and apoptosis upon incubation with HU. Moreover, the HDACi-induced expression of p21 and p27 was clearly too low to prevent apoptosis (Figures 2a–d and 3). Only massive expression of p21/p27 prevented caspase-3 activation and apoptosis induction by VPA/HU (Figures 5a–c). This could be due to cell cycle arrest. However, CDKI can protect cells from various pro-apoptotic stimuli, for example, paclitaxel (Schmidt et al., 2000), etoposide (Eymin et al., 1999) and proteasomal inhibitors (Drexler and Pebler, 2003). Hence, anti-apoptotic effects of CDKI are not limited to drugs targeting S phase. Likewise and in contrast to HDACi-sensitive MCF-7Casp3 cells, the reduction of CDKI alone could not render HDACi-resistant NW-Mel-450 cells sensitive to VPA (Figure 5g, Supplementary Figure F and data not shown). This indicates that additional HU effects, such as initiation of the caspase cascade are required to induce apoptosis in NW-Mel-450 cells (Figure 3c).

The CDKI p21 and p27 are dynamic proteins, which execute different biological functions in the nucleus and the cytoplasm (Blagosklonny, 2002). In homeostasis, balanced CDKI levels restrict cell growth and may prevent an unwanted execution of the apoptotic programme. However, CDKI levels equally correlate with the chemoresistance of tumours and poor disease prognosis (Pantazis et al., 1999; Mouriaux et al., 2000; Klisovic et al., 2003). This raises the question why p21 and p27 attenuate pro-apoptotic stimuli. We analysed the mechanism of VPA- and HU-induced cell death by overexpression of CDKI, small interfering RNA (siRNA) approaches, interaction assays and functional studies. On the basis of these data, we propose that p21 and p27 associate with caspase-3 dose-dependently and decrease apoptosis execution. In turn, downregulation of p21 and p27 releases caspase-3 from inhibition by these CDKI, which leads to enhanced apoptosis induction. Furthermore, p21 and p27 may well disable apoptosis by alternative pathways, for example by acting as caspase-3 substrates and, thereby, channelling caspase-3 away from other ‘death substrates’ (Figures 4 and 5).

Clinical studies already reported that VPA partially ameliorated cancer progression and suggested enhanced effectiveness in combination with HU compared to single administration (Bug et al., 2005; Kuendgen et al., 2006). Although the combination of drugs in a clinical setting requires caution and tight monitoring, the combined use of VPA and HU may have therapeutic value and should therefore be explored further.

Materials and methods

Cell lines and transfections

Cells were grown as described (Krämer et al., 2006). SiRNAs against human p21 and p27 were from Santa Cruz Biotechnology (Heidelberg, Germany) and were transfected with Lipofectamine (Invitrogen, Karlsruhe, Germany) or FuGene (Roche, Mannheim, Germany) according to the manufacturers on two consecutive days.

Western blot and immunofluorescence imaging

Floating and adherent cells were collected. Lysate preparations, Superose-6 fractionation, coimmunoprecipitations and western blots were carried out as in Krämer et al. (2006). Antibodies were from Santa Cruz Biotechnology: caspase-3, Tradd, p21, p27, HDAC2 and normal rabbit serum (pre); Sigma (Munich, Germany): actin; NEB (Frankfurt, Germany): caspase-8 and -9; Pharmingen (Heidelberg, Germany): PARP; DAKO (Glostrup, Denmark): pan-cytokeratin, acetylated histone H3 (AcH3) and acetylated histone H4 (AcH4) (Göttlicher et al., 2001). Actin serves as loading control for all western blots. Immunofluorescence, quantitation, image analysis and presentation were performed as described in Knauer et al. (2006). DNA was visualized with Hoechst 33258 or TO-PRO-3 (Invitrogen).

Measurement of apoptosis, cell cycle and viability

MTT-assays, propidium iodide (PI)-FACS analyses and caspase-3 assays were performed as described in Krämer et al. (2006). Caspase-3 activity was measured by colorimetric assay (Ac-DEVD-pNA is cleaved by caspase-3 into pNA, which was measured at OD405 nm) and by in vitro cleavage of p21 and p27. The reactions were performed in a total volume of 20 μl in caspase-3 cleavage buffer. Where stated, 5 μl of wheat germ lysate containing TNT-translated p21/p27/Gal (Promega, Mannheim, Germany) were added to recombinant caspase-3 subunits (p17-p12)2 produced in E. coli (provided by B Dälken and W Wels (Frankfurt, Germany)).

Plasmids and RT–PCR

NW-Mel-450 or SK-Mel-37 cDNAs served as templates. Primers: p21-forward: IndexTermATGTATCCATATGATGTTCCAGATTATGCTATGTCAGAACCGGCTGGGG; p21-reverse: IndexTermCCCAAGCTTGGGCTTCCTCTTGGAGAAG; p27-forward: IndexTermATGTATCCATATGATGTTCCAGATTATGCTATGTCAAACGTGCGAGTGTC; p27-reverse: IndexTermCCCAAGCTTCGTTTGACGTCTTCTGAGGCC. PCR products were cloned into pc3.1-TOPO and sequenced with T7-/BGH-primers (Invitrogen). Sequencing confirmed that p21 and p27 were wild-type in NW-Mel-450 and SK-Mel-37 cells. RNA was isolated with RNeasy (Qiagen, Hilden, Germany). cDNA was produced as recommended by Amersham (Munich, Germany). RT–PCRs were carried out with the Taq-polymerase kit (Qiagen). RT–PCR linearity was controlled with β-actin: forward: IndexTermATGATATCGCCGCGCTCGTCGTC; reverse: IndexTermTTCTCGCGGTTGGCCTTGGGGTTCAG and was optimal at 20–25 cycles.


  1. Blagosklonny MV . (2002). Are p27 and p21 cytoplasmic oncoproteins? Cell Cycle 1: 391–393.

  2. Boyle GM, Martyn AC, Parsons PG . (2005). Histone deacetylase inhibitors and malignant melanoma. Pigment Cell Res 18: 160–166.

  3. Bug G, Ritter M, Wassmann B, Schoch C, Heinzel T, Schwarz K et al. (2005). Clinical trial of valproic acid and all-trans retinoic acid in patients with poor-risk acute myeloid leukemia. Cancer 104: 2717–2725.

  4. Drexler HC, Pebler S . (2003). Inducible p27(Kip1) expression inhibits proliferation of K562 cells and protects against apoptosis induction by proteasome inhibitors. Cell Death Differ 10: 290–301.

  5. Eymin B, Haugg M, Droin N, Sordet O, Dimanche-Boitrel MT, Solary E . (1999). p27Kip1 induces drug resistance by preventing apoptosis upstream of cytochrome c release and procaspase-3 activation in leukemic cells. Oncogene 18: 1411–1418.

  6. Finzer P, Kuntzen C, Soto U, zur Hausen H, Rosl F . (2001). Inhibitors of histone deacetylase arrest cell cycle and induce apoptosis in cervical carcinoma cells circumventing human papillomavirus oncogene expression. Oncogene 20: 4768–4776.

  7. Göttlicher M, Minucci S, Zhu P, Krämer OH, Schimpf A, Giavara S et al. (2001). Valproic acid defines a novel class of HDAC inhibitors inducing differentiation of transformed cells. EMBO J 20: 6969–6978.

  8. Hanahan D, Weinberg RA . (2000). The hallmarks of cancer. Cell 100: 57–70.

  9. Harris AL . (1985). DNA repair: relationship to drug and radiation resistance, metastasis and growth factors. Int J Radiat Biol Relat Stud Phys Chem Med 48: 675–690.

  10. Jänicke RU, Sprengart ML, Wati MR, Porter AG . (1998). Caspase-3 is required for DNA fragmentation and morphological changes associated with apoptosis. J Biol Chem 273: 9357–9360.

  11. Klisovic DD, Katz SE, Effron D, Klisovic MI, Wickham J, Parthun MR et al. (2003). Depsipeptide (FR901228) inhibits proliferation and induces apoptosis in primary and metastatic human uveal melanoma cell lines. Invest Ophthalmol Vis Sci 44: 2390–2398.

  12. Knauer SK, Bier C, Habtemichael N, Stauber RH . (2006). The Survivin-Crm1 interaction is essential for chromosomal passenger complex localization and function. EMBO R 7: 1259–1265.

  13. Koon HB, Atkins MB . (2007). Update on therapy for melanoma: opportunities for patient selection and overcoming tumor resistance. Expert Rev Anticancer Ther 7: 79–88.

  14. Kouzarides T . (2000). Acetylation: a regulatory modification to rival phosphorylation? EMBO J 19: 1176–1179.

  15. Krämer OH, Baus D, Knauer SK, Stein S, Jager E, Stauber RH et al. (2006). Acetylation of Stat1 modulates NF-kappaB activity. Genes Dev 20: 473–485.

  16. Krämer OH, Göttlicher M, Heinzel T . (2001). Histone deacetylase as a therapeutic target. Trends Endocrinol Metab 12: 294–300.

  17. Krämer OH, Zhu P, Ostendorff HP, Golebiewski M, Tiefenbach J, Peters MA et al. (2003). The histone deacetylase inhibitor valproic acid selectively induces proteasomal degradation of HDAC2. EMBO J 22: 3411–3420.

  18. Kuendgen A, Schmid M, Schlenk R, Knipp S, Hildebrandt B, Steidl C et al. (2006). The histone deacetylase (HDAC) inhibitor valproic acid as monotherapy or in combination with all-trans retinoic acid in patients with acute myeloid leukemia. Cancer 106: 112–119.

  19. La Porta CA . (2007). Drug resistance in melanoma: new perspectives. Curr Med Chem 14: 387–391.

  20. Levkau B, Koyama H, Raines EW, Clurman BE, Herren B, Orth K et al. (1998). Cleavage of p21Cip1/Waf1 and p27Kip1 mediates apoptosis in endothelial cells through activation of Cdk2: role of a caspase cascade. Mol Cell 1: 553–563.

  21. Maggio SC, Rosato RR, Kramer LB, Dai Y, Rahmani M, Paik DS et al. (2004). The histone deacetylase inhibitor MS-275 interacts synergistically with fludarabine to induce apoptosis in human leukemia cells. Cancer Res 64: 2590–2600.

  22. Montefusco E, Fazi F, Cordone I, Ariola C, Nanni M, Spadea A et al. (2001). Molecular remission following high-dose hydroxyurea and fludarabine plus cytarabine in a patient with simultaneous acute myeloid leukemia and low-grade lymphoma. Leuk Lymphoma 40: 671–674.

  23. Mouriaux F, Maurage CA, Labalette P, Sablonniere B, Malecaze F, Darbon JM . (2000). Cyclin-dependent kinase inhibitory protein expression in human choroidal melanoma tumors. Invest Ophthalmol Vis Sci 41: 2837–2843.

  24. Nguyen DM, Schrump WD, Chen GA, Tsai W, Nguyen P, Trepel JB et al. (2004). Abrogation of p21 expression by flavopiridol enhances depsipeptide-mediated apoptosis in malignant pleural mesothelioma cells. Clin Cancer Res 10: 1813–1825.

  25. Pagano M, Tam SW, Theodoras AM, Beer-Romero P, Del Sal G, Chau V et al. (1995). Role of the ubiquitin-proteasome pathway in regulating abundance of the cyclin-dependent kinase inhibitor p27. Science 269: 682–685.

  26. Pantazis P, Chatterjee D, Han Z, Wyche J . (1999). Differentiation of human malignant melanoma cells that escape apoptosis after treatment with 9-nitrocamptothecin in vitro. Neoplasia 1: 231–240.

  27. Rosato RR, Almenara JA, Grant S . (2003). The histone deacetylase inhibitor MS-275 promotes differentiation or apoptosis in human leukemia cells through a process regulated by generation of reactive oxygen species and induction of p21CIP1/WAF1 1. Cancer Res 63: 3637–3645.

  28. Schmidt M, Lu Y, Liu B, Fang M, Mendelsohn J, Fan Z . (2000). Differential modulation of paclitaxel-mediated apoptosis by p21Waf1 and p27Kip1. Oncogene 19: 2423–2429.

  29. Schrell UM, Rittig MG, Anders M, Kiesewetter F, Marschalek R, Koch UH et al. (1997a). Hydroxyurea for treatment of unresectable and recurrent meningiomas. I. Inhibition of primary human meningioma cells in culture and in meningioma transplants by induction of the apoptotic pathway. J Neurosurg 86: 845–852.

  30. Schrell UM, Rittig MG, Anders M, Koch UH, Marschalek R, Kiesewetter F et al. (1997b). Hydroxyurea for treatment of unresectable and recurrent meningiomas. II. Decrease in the size of meningiomas in patients treated with hydroxyurea. J Neurosurg 86: 840–844.

  31. Shin HJ, Baek KH, Jeon AH, Kim SJ, Jang KL, Sung YC et al. (2003). Inhibition of histone deacetylase activity increases chromosomal instability by the aberrant regulation of mitotic checkpoint activation. Oncogene 22: 3853–3858.

  32. St Croix B, Florenes VA, Rak JW, Flanagan M, Bhattacharya N, Slingerland JM et al. (1996). Impact of the cyclin-dependent kinase inhibitor p27Kip1 on resistance of tumor cells to anticancer agents. Nat Med 2: 1204–1210.

  33. Suzuki A, Kawano H, Hayashida M, Hayasaki Y, Tsutomi Y, Akahane K . (2000). Procaspase 3/p21 complex formation to resist fas-mediated cell death is initiated as a result of the phosphorylation of p21 by protein kinase A. Cell Death Differ 7: 721–728.

  34. Suzuki A, Tsutomi Y, Miura M, Akahane K . (1999). Caspase 3 inactivation to suppress Fas-mediated apoptosis: identification of binding domain with p21 and ILP and inactivation machinery by p21. Oncogene 18: 1239–1244.

  35. Szekeres T, Fritzer-Szekeres M, Elford HL . (1997). The enzyme ribonucleotide reductase: target for antitumor and anti-HIV therapy. Crit Rev Clin Lab Sci 34: 503–528.

  36. Van den Berg C, Von Hoff DD . (1995). Use of hydroxyurea to alter drug resistance of human tumor cells. Cancer Treat Res 78: 95–114.

  37. Zhu P, Martin E, Mengwasser J, Schlag P, Janssen KP, Göttlicher M . (2004). Induction of HDAC2 expression upon loss of APC in colorectal tumorigenesis. Cancer Cell 5: 455–463.

Download references


We thank S Reichardt, G Greiner and A Schimpf for excellent technical assistance. Expression constructs, antibodies and cells were generously provided by M Zörnig, W Wels, B Dälken, E Jäger and E Jänicke. Grant support: NGFN to TH (N1KR-S31T30) and Deutsche Krebshilfe to RHS (FKZ:102362).

Author information



Corresponding authors

Correspondence to O H Krämer or T Heinzel.

Additional information

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc).

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Krämer, O., Knauer, S., Zimmermann, D. et al. Histone deacetylase inhibitors and hydroxyurea modulate the cell cycle and cooperatively induce apoptosis. Oncogene 27, 732–740 (2008). https://doi.org/10.1038/sj.onc.1210677

Download citation


  • VPA
  • HDAC inhibitor
  • hydroxyurea
  • caspase-3
  • CDKI p21/p27
  • chemotherapy resistance

Further reading