PHA-848125

The cyclin-dependent kinase inhibitor PHA-848125 suppresses the in vitro growth of human melanomas sensitive or resistant to temozolomide, and shows synergistic effects in combination with this triazene compound
Simona Caporalia, Ester Alvinob, Giuseppe Staraceb,c, Marina Ciomeid, Maria Gabriella Brascad, Lauretta Levatia, Alberto Garbina, Daniele Castigliae, Claudia Covaciue,
Enzo Bonmassarb,f, Stefania D’Atria,∗
a Laboratory of Molecular Oncology, Istituto Dermopatico dell’Immacolata-IRCCS, Via dei Monti di Creta 104, 00167 Rome, Italy
b Institute of Neurobiology and Molecular Medicine, National Council of Research, Via del Fosso del Cavaliere 100, 00133 Rome, Italy
c Regina Elena Cancer Institute, Via delle Messi d’Oro 156, 00158 Rome, Italy
d Nerviano Medical Sciences Srl, Business Unit Oncology, Viale Pasteur 10, 20014 Nerviano (MI), Italy
e Laboratory of Molecular and Cell Biology, Istituto Dermopatico dell’Immacolata-IRCCS, Via dei Monti di Creta 104, 00167 Rome, Italy
f Department of Neuroscience, School of Medicine, University of Rome “Tor Vergata”, Via Montpellier 1, 00133 Rome, Italy

a r t i c l e i n f o a b s t r a c t

Article history:
Received 3 September 2009 Received in revised form 14 December 2009
Accepted 14 December 2009

Keywords:
Melanoma
Cyclin-dependent kinase inhibitor Temozolomide
Cell growth Cell cycle
PHA-848125 is a novel cyclin-dependent kinase inhibitor under Phase I/II clinical investigation. In this study, we describe, for the first time, the effect of PHA-848125 on human melanoma cells in vitro. Seven melanoma cell lines with different sensitivity to temozolomide (TMZ) were exposed to PHA-848125 for 5 days and then assayed for cell growth. In all cases, including TMZ-resistant cells, PHA-848125 IC50 values were significantly below the maximum plasma concentrations achievable in the clinic. In the most PHA-848125-sensitive cell line, the drug caused a concentration-dependent G1 arrest. PHA-848125 also impaired phosphorylation of the retinoblastoma protein at CDK2 and CDK4 specific sites, decreased retinoblastoma protein and cyclin A levels, and increased p21Cip1 , p27Kip1 and p53 expression. Combined treatment with fixed ratios of TMZ plus PHA-848125 was studied in three melanoma cell lines. PHA- 848125 was added to the cells 48 h after TMZ and cell growth was evaluated after 3 additional days of culture. Parallel experiments were performed in the presence of O6 -benzylguanine (BG), to prevent repair of methyl adducts at O6 -guanine induced by TMZ. Drug combination of TMZ plus BG and PHA- 848125 produced additive or synergistic effects on cell growth, depending on the cell line. In the absence of BG, the combination was still more active than the single agents in the cell line moderately sensitive to TMZ, but comparable to PHA-848125 alone in the two TMZ-resistant cell lines. When TMZ plus BG were used in combination with PHA-848125 against cultured normal melanocytes, neither synergistic nor additive antiproliferative effects were observed. Our results indicate that PHA-848125 can have a therapeutic potential in melanoma patients, alone or combined with TMZ. Moreover this agent appears to be particularly attractive on the bases of its effectiveness against TMZ-resistant melanoma cells.
© 2009 Elsevier Ltd. All rights reserved.

⦁ Introduction

Abbreviations: CDK, cyclin-dependent kinase; RB, retinoblastoma protein; TMZ, temozolomide, i.e. 8-carbamoyl-3-methyl-imidazo[5,1-d]-1,2,3,5-tetrazin- 4(3H)-one; MMR, mismatch repair; CM, complete medium; MGM, melanocyte growth medium; BG, O6 -benzylguanine; MGMT, O6 -methylguanine-DNA methyl- transferase; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; DMSO, dimethyl sulfoxide; PBS, phosphate buffered saline; CI, combination index; ECL, enhanced chemiluminescence; O6 -G, O6 -guanine; O6 -MeG, O6 -methylguanine; GI, growth inhibition.
∗ Corresponding author. Tel.: +39 06 66464735; fax: +39 06 6642430.
E-mail address: [email protected] (S. D’Atri).
Cell cycle progression is tightly regulated by a family of ser- ine/threonine kinases that form heterodimeric complexes with cyclins and operate in distinct phases of the cell cycle (reviewed in [1–4]). Cyclin-dependent kinase (CDK) 4 and CDK6, in associa- tion with type D cyclins, and CDK2, in association with cyclin E, sequentially phosphorylate the retinoblastoma protein (RB), lead- ing to the G1 → S transition. Cyclin A/CDK2 is required for S phase progression, whereas cyclin B/CDK1 controls the G2 M transi- tion [1–4]. Regulation of CDK activity occurs at multiple levels, including phosphorylation on inhibitory sites, activating phos- phorylation, association with endogenous inhibitors belonging to


1043-6618/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phrs.2009.12.009

Cip/Kip (p21Cip1, p27Kip1 and p57Kip2) or INK4 (p15INK4b, p16INK4a, p18INK4c and p19INK4d) family, cyclin synthesis and degradation [1–4].
A fundamental aspect of cancer is represented by dys- regulated cell cycle control. In tumor cells, genetic/epigenetic events resulting in overexpression of cyclins or CDKs, or down- regulation/loss of CDK inhibitors, or RB function, provide a selective growth advantage [2,4]. Chemical inhibitors of cell cycle may have therefore potential application in suppressing malignant growth.
Because of their central role in the control of cell cycle and their participation in the regulation of transcription, CDKs have been tar- geted for drug discovery and numerous CDK inhibitors have been developed and have entered clinical trials (reviewed in [5–7]).
In preclinical studies, CDK inhibitors have shown the ability to inhibit cancer cell proliferation, to induce apoptosis and to synergize with conventional chemotherapy [5–7]. In the clinic, flavopiridol and roscovitine, the two most studied CDK inhibitors, showed modest antitumor activity when used alone. However, encouraging results were obtained when flavopiridol was com- bined with classical antitumor agents [6,7]. Presently, several novel CDK inhibitors more active than flavopiridol and roscovitine are being tested in clinical trials [8].
Metastatic melanoma is highly refractory to chemotherapy and has a very poor prognosis, with a median survival rate of 6 months and a 5-year survival rate not exceeding 5% (reviewed in [9]). The methylating agent dacarbazine, which is still considered to be the reference single agent for advanced disease, has objective response rates in the range of 10–15% [9]. Temozolomide (TMZ), a triazene compound that spontaneously hydrolyzes to the active metabo- lite of dacarbazine (reviewed in [10,11]), shows antitumor activity comparable to dacarbazine. However, TMZ has the advantage to penetrate the central nervous system, and might be therefore active against brain metastases [9,11]. A number of combination regimens containing multiple chemotherapeutic agents, multiple biological agents or both have shown an improved overall response rate with respect to single-agent dacarbazine. However, no combina- tion therapy has been found to increase the overall survival rate in comparison to dacarbazine alone [9]. Therefore, new treatment strategies are urgently needed.
Germline mutations in the CDKN2A locus, which encodes p16INK4a and p14ARF, have been detected in 25–40% of melanoma- prone families (reviewed in [12,13]). Inactivation of p16INK4a also occurs in a high percentage of sporadic melanomas, mainly as a result of deletions or promoter methylation [13]. Germline and somatic mutations have been detected in the CDK4 gene, all of which interfere with CDK4 protein binding to p16INK4a. Amplifica- tion of cyclin D1 and CDK4, as well as overexpression of CDK6 have also been reported in melanoma [12,13]. All these molecular alter- ations lead to the dysregulation of the cyclin D/CDK4/6-p16INK4a-RB pathway, and can also increase the activity of cyclin E/CDK2 and A/CDK2, as a result of Cip/Kip protein sequestration in cyclin D/CDK4/6 complexes [3,6]. Therefore, drugs targeting CDKs might have a therapeutic potential in melanoma.

PHA-848125 is a pyrazolo[4,3-h]quinazoline identified as a highly effective cyclin A/CDK2 inhibitor [14] currently undergoing Phase I and II clinical trials. In the present study we determined for the first time the growth inhibitory effects of the drug on a panel of human melanoma cell lines endowed with different sensitivity to TMZ, including those markedly resistant to this drug. Moreover, we investigated the effects of PHA-848125 on cell cycle progres- sion and also on the expression of a number of effector molecules involved in the G1 S transition. Finally, this report highlights the additive/synergistic inhibitory effects of a combined treatment with TMZ and PHA-848125 on human melanoma cell proliferation in vitro.
⦁ Materials and methods

⦁ Cell lines

Seven human melanoma cell lines were used in this study. M10 was kindly provided by Dr. G. Zupi (Regina Elena Cancer Insti- tute, Rome, Italy); GL-MEL was a gift of Dr. F. Guadagni (IRCCS San Raffaele Pisana, Rome, Italy); SK-Mel-28 and WM-115 were obtained from the American Type Culture Collection (Manassas, VA); CN-Mel, DR-Mel and PR-Mel were established in our labo- ratory. WM-115 was derived from a primary melanoma, while the remaining cell lines were originated from metastatic lesions. PR-Mel cell line is mismatch repair (MMR)-deficient due to bial- lelic somatic inactivation of MLH1 [15]. CN-Mel, GL-Mel, M10, SK-Mel-28 cell lines are MMR-proficient [16]. DR-Mel and WM- 115 were analyzed for the expression of the MMR proteins MSH2, MSH6, MLH1 and PMS2 in the present study. The cells were cul- tured at 37 ◦C in 5% CO2 humidified atmosphere and maintained in RPMI-1640 (GIBCOTM, Invitrogen Corporation, Paisley, UK) sup- plemented with 10% fetal calf serum (GIBCOTM), 2 mM l-glutamine, and 50 µg/ml gentamycin (GIBCOTM) (hereafter referred to as com- plete medium, CM).
The human colon cancer cell line LoVo was obtained from Amer- ican Type Culture Collection and cultured in DMEM (GIBCOTM) supplemented with 10% fetal calf serum, 2 mM l-glutamine and 50 µg/ml gentamycin. LoVo cells are homozygous for a partially deleted MSH2 gene [17].
Human melanocytes were isolated from normal skin biopsies of two different donors and cultured in melanocyte growth medium (MGM), as previously described [18].
All biological material was obtained with the patient’s informed consent, and the study was conducted according to the Declaration of Helsinki Principles.

⦁ Drugs and reagents

}
{

PHA-848125, i.e. N,1,4,4-tetramethyl-8- [4-(4-methylpipera- zin-1-yl)phenyl]amino -4,5-dihydro-1H-pyrazolo[4,3-h]quinazo- line-3-carboxamide, was synthesized at Nerviano Medical Sci- ences [14]. TMZ was kindly provided by Schering-Plough Research Institute (Kenilworth, NJ). O6-Benzylguanine (BG), a competi- tive inhibitor of the DNA repair protein O6-methylguanine-DNA methyltransferase (MGMT), (E.C. 2.1.1.63), (reviewed in [19,20]), and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro- mide (MTT) were purchased from Sigma (St. Louis, MO). TMZ was always prepared freshly in CM, as the drug readily decomposes in aqueous solution. BG was dissolved in ethanol (2.4 mg/ml), and PHA-848125 was dissolved in dimethyl sulfoxide (DMSO) (5.7 mg/ml). The two agents were stored as stock solutions at 80 ◦C, and diluted in culture medium just before use. MTT was prepared at a concentration of 5 mg/ml in phosphate buffered
saline (PBS), and stored at 4 ◦C.
Reagents for SDS-polyacrylamide gel electrophoresis were all purchased from Bio-Rad (Hercules, CA).

⦁ Evaluation of cell sensitivity to PHA-848125 or TMZ

Melanoma cells were suspended in CM at a concentration of
×
2 104 cells/ml, dispensed in 50 µl aliquots into flat-bottom 96- well plates (Falcon, Becton and Dickinson Labware, Franklin Lakes, NJ) and allowed to adhere overnight at 37 ◦C. Graded amounts of PHA-848125 or TMZ were then added to the wells (4 wells per point) in 50 µl of CM and the plates were incubated at 37 ◦C ina 5% CO2 humidified atmosphere for 5 days. The cytotoxic effects of TMZ were also evaluated in combination with the MGMT inhibitor BG. To this end, 10 µM BG was added to the plates 2 h before TMZ and left

in culture for the entire period of cell exposure to the drug. Control groups were represented by untreated cells, and cells treated with BG or DMSO alone. The growth of the cells treated with BG or DMSO alone did not differ from that of untreated cells (data not shown). MGMT activity of BG-treated cells was undetectable 2 h after the addition of the inhibitor and remained essentially undetectable up to the end of the assay (data not shown).
×
Normal melanocytes were suspended in MGM at the con- centration of 1.6 105 cells/ml, plated (50 µl/well) and exposed to TMZ + BG or to PHA-848125 as described for melanoma cells.
At the end of the incubation period, cell growth was evaluated by the MTT assay as previously described [16]. Briefly, 0.1 mg of MTT (in 20 µl of PBS) was added to each well and cells were incubated
at 37 ◦C for 4 h. Cells were then lysed with a buffer (0.1 ml/well)
containing 20% SDS and 50% N,N-dimethylformamide, pH 4.7. After overnight incubation, the absorbance was read at 595 nm using a 3550-UV microplate reader (Bio-Rad).
Cell sensitivity to drug treatment was expressed in terms of IC50 (drug concentration producing 50% inhibition of cell growth, calcu- lated on the regression line in which absorbance values at 595 nm were plotted against the logarithm of drug concentration).

⦁ Evaluation of cell sensitivity to combined treatment with PHA-848125 and TMZ

To study the effects of combined treatment with TMZ and PHA- 848125 on melanoma cell or melanocyte proliferation, we used the median effect method described by Chou and Talalay [21]. Accord- ing to this method, a parameter called combination index (CI) is calculated based on the following equation:
= + +
CI DA DB ˛DADB
(Dx)A (Dx)B (Dx)A(Dx)B
where DA and DB are the concentrations of drug A and drug B that have x effect when used in combination at a fixed ratio, (Dx)A and (Dx)B are the concentrations of drug A and drug B that have the same effect when used individually, and ˛ = 1 or 0 depending on whether the two drugs are assumed to be mutually nonexclusive (i.e. they have different modes of action or act independently) or mutually exclusive (i.e. they act on the same targets). In this method, for each level of effect exerted by the drug combination, synergy is indicated by CI < 1, additivity by CI = 1, and antagonism by CI > 1.
Briefly, melanoma cells were set up for MTT assay as described above, and then exposed to graded amounts of PHA-848125 or TMZ alone or to both drugs combined at a fixed ratio, according to the following schedule: the cultures were ini- tially incubated with CM alone, with 10 µM BG alone, with graded concentrations of TMZ alone, or with 10 µM BG (2 h before adding TMZ) + graded concentrations of TMZ. After 48 h, the cells were further treated as follows: (a) the cultures incu- bated with CM alone were divided into three subgroups, i.e. one left untreated, one treated with DMSO and one treated with graded concentrations of PHA-848125 alone; (b) the cul- tures treated with BG alone were divided into two subgroups one not subjected to further treatment and the other exposed to DMSO; (c) the cultures treated with TMZ or TMZ + BG were divided into two subgroups each, one not subjected to further treatment and the other exposed to graded concentrations of PHA-848125. After additional 72 h, all groups were subjected to the MTT assay. Growth inhibition of cells treated with TMZ, TMZ + BG, PHA-848125, TMZ + PHA-848125 or TMZ + BG + PHA-
848125 was evaluated with respect to the following controls: untreated cells (TMZ), BG-treated cells (TMZ + BG), DMSO-treated cells (PHA-848125, TMZ + PHA-848125), BG/DMSO-treated cells (TMZ + BG + PHA-848125).
Melanocyte sensitivity to TMZ + BG combined with PHA-848125 was evaluated as described for melanoma cells. In this case,
× ×
8 103 cells/well instead of 1 103 cells/well were plated for the
MTT assays.
The CalcuSyn software (Cambridge Biosoft, Cambridge, UK) was used to calculate the CI values from the experimentally determined concentration–response curves for TMZ, PHA-848125 or the com- binations of both drugs. The CI was calculated according to the more stringent statistical assumption that TMZ and PHA-848125 are mutually nonexclusive drugs. When BG was included in the assay, TMZ + BG was considered as a single drug.

⦁ Western blot analysis

Rabbit polyclonal antibodies against CDK2 (M2), CDK4 (C-22), cyclin A (H-432), cyclin D1 (M-20), p16INK4a (N-20) and p27Kip1 (C-19), as well as mouse monoclonal antibody (mAb) against cyclin E (HE12) were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA); mouse mAb against p21Cip1 (clone 70) and MSH6 (clone 44) were purchased from BD Transduction Labora- tories (Heidelberg, Germany); rabbit polyclonal antibody against p53 was from Cell Signaling Technology Inc. (Danvers, MA); rabbit polyclonal antibody against phospho-RB-Thr821 (44–582G), and rabbit polyclonal antibody against phospho-RB-Thr826 (44–576) were purchased from Invitrogen Corporation (Camarillo, CA) and Biosource International Inc. (Camarillo, CA), respectively; mouse mAb against MSH2 (clone GB12) and PMS2 (clone 9) were pur- chased from Oncogene Research Products (Boston, MA); mouse mAb against MLH1 (clone G168-15), and mouse mAb against RB (clone G3-245) were obtained from BD PharMingen (Heidelberg, Germany); mouse mAb against actin (clone AC-40) was obtained from Sigma.
Total cellular extracts were prepared by incubating cells on
ice in lysis buffer (50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 1 mM EGTA, 1% NP-40, 0.25% sodium deoxycholate, 1 mM NaF, 1 mM Na3VO4, 1 mM AEBSF) supplemented with 1x of a protease inhibitor cocktail (Complete Mini EDTA-free, Roche Diagnostic, Mannhein, Germany) and 1x of a phosphatase inhibitor cocktail (PhosSTOP, Roche Diagnostic), for 10 min. The cell lysates were then clari- fied by centrifugation, diluted in 5x Lämmli sample buffer, and boiled for 5 min. Twenty-five micrograms protein per sample were run on SDS-polyacrilamide gels, transferred to nitrocellulose mem- branes (Amersham Biosciences, Little Chalfont, UK) and blocked with 5% non-fat milk in Tris-buffered saline supplemented with 0.1% Tween 20 for 1 h at room temperature. The membranes were then incubated in the same solution overnight at 4 ◦C with primary antibodies at the following dilutions: anti-CDK2, anti- CDK4, anti-cyclin A, anti-cyclin D1, anti-cyclin E, anti-p16INK4a, and anti-p27Kip1 1:200; anti-p21Cip1, anti-p53, and anti-phospho-RB- Thr821 1:500; anti-phospho-RB-Thr826, anti-RB; and anti-actin 1:1000. The latter antibody was used as an internal standard for loading. Immunodetection was carried out using appropriate horseradish peroxidase-linked secondary antibodies and enhanced chemiluminescence (ECL) detection reagents.

⦁ Cell cycle and apoptosis analysis


Adherent and floating cells were collected from control and PHA-848125-treated cultures, fixed with 70% ethanol at 20 ◦C and stained with propidium iodide as previously described [22]. The propidium iodide fluorescence was measured on a linear scale using a FACS Vantage SE/DiVa flow cytometer (Becton and Dickinson, San Jose, CA). The fraction of cells in the cycle phases was evaluated by a mathematical model of the cell cycle [23]. Apoptotic cells were determined by their sub-G1 peak.

⦁ MGMT assay


Melanoma cells were removed from continuous culture washed twice with PBS and stored as pellets at 80 ◦C until used. MGMT activity in cell extracts was determined by measuring the transfer of 3H-methyl groups from a DNA substrate to the MGMT protein as previously described [16]. MGMT activity was expressed in terms of fmoles of 3H-methyl groups transferred per mg of protein in cell extract.

⦁ Mutational analysis of TP53 and CDKN2A

Genomic DNA was extracted from the melanoma cell lines using the DNeasy Tissue kit (Qiagen, Hilden, Germany). All coding exons and flanking intronic sequences of TP53 were amplified using pub- lished primer pairs [24]. PCRs were performed in 50 µl reaction volumes using the AmpliTaq Gold polymerase (1.5 U) and standard reagents (Applied Biosystems, Foster City, CA). Each PCR amplicon was subjected to direct sequencing in both orientations using an ABI Prism 377 semiautomated sequencing system (Applied Biosys- tem, Foster City, CA). Sequencing analysis was initially carried out on exons 5, 6, 7, and 8; if a mutation was found the screening was stopped, otherwise the analysis was extended to the remaining exons.
Mutational screening of exon 1α, 2 and 3 of the CDKN2A locus was performed using primers and PCR conditions previously described [25].

⦁ Statistical analysis

Statistical significance among different IC50 values was assessed using two-tailed Student’s t-test analysis.
Statistical significance of PHA-848125-induced alterations of cell cycle distribution was assessed according to two-tailed paired Student’s t-test analysis.

⦁ Results

⦁ Growth inhibitory activity of PHA-848125

Previous studies demonstrated that PHA-848125 is a potent ATP-competitive inhibitor of cyclin A/CDK2, showing an IC50 value of 0.045 µM in biochemical assays [14]. PHA-848125 also inhibits, although with lower potency, the activity of cyclin H/CDK7 (IC50:
0.150 µM), cyclin D1/CDK4 (IC50: 0.160 µM), p35/CDK5 (IC50:
0.265 µM), cyclin E/CDK2 (IC50: 0.363 µM), and cyclin B/CDK1 (IC50: 0.398 µM) [14]. Among 37 additional kinases tested, only Thropomyosin Receptor Kinase A was inhibited by PHA-848125 in the same nanomolar range of CDKs [14]. The drug also shows in vitro and in vivo antitumor activity against A2780 human ovarian carcinoma cells [14].
As a first approach to the evaluation of the therapeutic potential of PHA-848125 in melanoma, we decided to compare the antitumor effects of TMZ and the CDK inhibitor in a panel of melanoma cell lines.
At concentrations achievable in the clinic, cytotoxicity of TMZ, like that of dacarbazine, is primarily due to the methylation of the O6 position of guanine (O6-G) in DNA [10,19,20]. Therefore, high levels of MGMT, which removes small alkyl groups from O6- G in a stoichiometric and autoinactivating reaction, antagonize the antitumor activity of TMZ [10,19,20]. Correspondingly, chem- ical depletion of MGMT activity (e.g. by BG) increases tumor cell sensitivity to TMZ both in vitro and in vivo [10,19,20]. The cyto- toxic effects of unrepaired O6-methylguanine (O6-MeG) rely on the formation of O6-MeG:T mispairs in the course of the first DNA dupli- cation after drug treatment, which in turn are recognized by the
MMR system (reviewed in [26,27]). The DNA damage produced by unsuccessful processing of O6-MeG:T mispairs by the MMR system activates a signaling cascade resulting in cell cycle arrest at the G2 phase of the second cell doubling event, followed by either apopto- sis, mitotic catastrophe, or a senescence-like state [10,19,20]. Cells with defective MMR are highly resistant to TMZ regardless of their MGMT activity, and, as expected, BG fails to influence substantially this type of resistance [10,19,20].
Based on the mechanism of action of TMZ, we selected five melanoma cell lines (i.e. CN-Mel, GL-Mel, M10, PR-Mel and SK- Mel-28) previously characterized for MGMT and MMR activity [16], and two additional cell lines (i.e. DR-Mel and WM-115) that we tested in the present study for MGMT activity and MMR protein expression. The cells were incubated with graded concentrations of PHA-848125 or TMZ or TMZ + BG (10 µM), for 5 days and then assayed for cell growth by the MTT assay.
As illustrated in Fig. 1, the previously characterized MMR- proficient cell lines (i.e. CN-Mel, GL-Mel, M10, SK-Mel-28) as well as DR-Mel cells expressed all the MMR proteins. The MMR-deficient cell line PR-Mel was confirmed to be devoid of the MLH1 compo- nent of MMR system, and also to express barely detectable levels of the PMS2 protein, which is unstable in the absence of MLH1. The expression of MSH6 was found to be extremely low in WM-115 cells as compared to that detectable in the other melanoma cell lines, suggesting a possible impairment of MMR activity.
Table 1 illustrates the MGMT activity of the cell lines and their sensitivity to TMZ, alone or combined with BG, or to PHA-848125. In agreement with our previous studies [16], CN-Mel, M10, and SK-Mel-28 cells displayed high MGMT activity, while GL-Mel and PR-Mel showed low or undetectable MGMT activity, respectively. MGMT activity was also high in DR-Mel cells, while it was unde- tectable in WM-115 cells. In the absence of BG, all the cell lines, with the exception of GL-Mel, can be considered resistant to TMZ, although to a different extent. In fact, they showed IC50 values exceeding the peak plasma concentrations (i.e. 200–300 µM) that can be reached in patients receiving the triazene compound as a single dose of 1000 mg/m2 [28]. In the previously character- ized MMR-proficient cell lines and in DR-Mel cell line, the growth inhibitory effect of TMZ was markedly enhanced by BG. However, CN-Mel cells maintained high levels of drug resistance even in the

Fig. 1. MMR protein expression in human melanoma cell lines. Twenty-five micrograms of whole cell extracts were subjected to electrophoresis on a 7% SDS- polyacrylamide gel. The proteins were transferred to a nitrocellulose membrane and incubated with antibodies against MSH2, MSH6, MLH1 or PSM2. Incubation with anti-actin mAb was performed as a loading control. The immune complexes were visualized using ECL. LoVo cells, expressing MLH1 and PSM2 but not MSH2 and MSH6 proteins, were included in the analysis as a control. The molecular weights of the analyzed proteins are as follows: MSH2, 100 kDa; MSH6, 160 kDa; MLH1, 85 kDa; PMS2, 96 kDa; actin, 42 kDa.

Table 1
MGMT activity and sensitivity to TMZ, TMZ + BG and PHA-848125 of human melanoma cell lines.

Cell linea MGMTb IC50 (µM)c
TMZ
TMZ+ BGd
pe

PHA-848125
GL-Mel 64 ± 7 205 ± 7 58 ± 3 <0.01 0.123 ± 0.010
M10 513 ± 20 362 ± 17 23 ± 2 <0.01 0.526 ± 0.047
DR-Mel 444 ± 13 403 ± 8 138 ± 12 <0.01 0.680 ± 0.074
SK-Mel-28 633 ± 35 740 ± 32 266 ± 28 <0.01 0.464 ± 0.044
CN-Mel 813 ± 18 806 ± 16 413 ± 23 <0.01 0.422 ± 0.071
PR-Mel NDf >1000 >1000 – 0.179 ± 0.020
WM-115 ND >1000 >1000 – 0.300 ± 0.048
a Cells were incubated with graded concentrations of the indicated drugs for 5 days and then analyzed for cell growth by the MTT assay.
b MGMT activity is expressed in terms of fmoles of 3 H-methyl groups transferred per mg of protein in cell extract. Each value represents the mean ± standard error of the mean of at least three independent experiments.
±
c Drug concentration required to inhibit cell growth by 50%. Each value represents the mean standard error of the mean of at least three independent experiments.
d Exposure to BG was performed by incubating the cells with 10 µM BG for 2 h before treatment with TMZ and maintaining the inhibitor in culture up to the end of the assay.
e Probability calculated according to Student’s t-test comparing for each cell line the IC50 values of TMZ obtained in the presence of BG with those obtained without the MGMT inhibitor.
f ND, not detectable.

presence of BG, probably as a consequence of additional mecha- nisms of resistance not dependent on MMR or MGMT. As expected, BG failed to sensitize to TMZ the MMR-deficient PR-Mel and the MGMT-deficient WM-115 cells. Noteworthy, this latter cell line displayed a high level of resistance to TMZ despite undetectable MGMT activity, most probably as a result of its barely detectable expression of MSH6 (Fig. 1).
The melanoma cell lines under investigation showed differ- ent levels of sensitivity to PHA-848125, with IC50 values ranging between 0.123 and 0.680 µM. Noteworthy, WM-115 and PR-Mel cell lines, which were found to be highly resistant to TMZ appear to be, together with GL-Mel, the most susceptible cell lines to the growth inhibitory effects of PHA-848125.

⦁ Effects of PHA-848125 on cell cycle progression and apoptosis

The effects of PHA-848125 on cell cycle progression and target cell apoptosis were investigated in asynchronously growing GL- Mel and M10 cell lines, selected for showing a 5-fold difference in drug-induced cell growth inhibition. The cells were cultured in the presence of 0.156 or 0.625 µM PHA-848125, and the percentage of cells in each phase of the cell cycle or in apoptosis was determined by flow cytometric analysis of DNA content after 24, 48, 72 and 96 h of drug exposure.
In both cell lines, drug-induced perturbations of the cell cycle were already evident after 24 h of treatment, and remained detectable up to 96 h of culture (Table 2, Fig. 2 and data not shown). Exposure of GL-Mel cells to 0.156 µM PHA-848125 induced a moderate increase in the G1 fraction with no significant changes concerning the S and the G2/M fractions. Treatment with 0.625 µM PHA-848125 caused instead a marked accumulation of the cells in the G1 phase, accompanied by a concomitant strong reduc- tion of the percentage of cells in the S and G2/M phases (Fig. 2 and Table 2). M10 cells resulted less susceptible than GL-Mel cells to PHA-848125-induced alterations of cell cycle. A moderate, although significant, variation of the percentage of cells in the G1 and G2/M phases was only observed upon treatment with 0.625 µM PHA-848125, while the percentage of cells in S phase showed no changes at both drug concentrations tested (Fig. 2 and Table 2).
No apoptosis induction was observed in both cell lines at the concentrations and time points tested. Apoptotic cell death was also not triggered in GL-Mel cells exposed to 1.25 µM PHA-848125 (a concentration about 10-fold higher than that corresponding to the IC50) for up to 96 h (data not shown).
To investigate the reversibility of cell cycle perturbations induced by PHA-848125, GL-Mel cells were treated with 0.156
or 0.625 µM of the drug for 24 h, washed and analyzed for cell cycle distribution after additional 24, 48 and 72 h of culture in CM alone. As illustrated in Fig. 3, the cells treated with 0.156 µM PHA-848125 showed a cell cycle distribution comparable to that of control as early as 24 h after drug removal. On the other hand, the cells exposed to 0.625 µM PHA-848125 resumed a control-like cell cycle distribution after 48 h of culture in drug-free medium.

⦁ Mutational analysis of the TP53 gene in GL-Mel and M10 cells and effect of PHA-848125 on p53 and p21Cip1 protein expression

Previous studies demonstrated that several CDK inhibitors can stabilize and activate the tumor-suppressor protein p53, thereby enhancing their suppressive effects on cancer cell growth (reviewed in [29]). Moreover, Mohapatra et al. [30] recently showed that the CDK2 inhibitor roscovitine induces apoptosis only in melanoma cell lines expressing wild type p53. We therefore decided to investigate the functional status of p53 in GL-Mel and M10 cell lines and the effect of PHA-848125 treatment on p53 expression. To this end, the TP53 gene was subjected to mutational screening by PCR amplification of genomic DNA followed by direct sequencing of the resulting PCR products. The analysis covered all the coding regions in GL-Mel cell line, which we found to harbor a wild type TP53 gene. In M10 cells, the mutational analysis covered only exons 5 through 8. In this cell line indeed, a C-to-T nucleotide transition at nucleotide position 637 (c.637C > T, exon 6) was iden- tified. Chromatogram inspection revealed a single peak for the T, which indicated a homozygous/hemizygous status for the muta- tion. The c.637C > T change results in the formation of the nonsense codon p.R213X, which leads to a premature translational stop. The p.R213X mutation, which is considered a loss-of-function muta- tion, has been previously detected in a number of tumor specimens and cell lines [31,32,33], including one melanoma cell line (COSMIC database; http://www.sanger.ac.uk).
We next evaluated the expression of p53 and its transcriptional target p21Cip1 in GL-Mel and M10 cells, either untreated or exposed
to 0.156 or 0.625 µM PHA-848125 for 24 h. The p53 protein was clearly expressed in GL-Mel cells, but barely detectable in M10 (Fig. 4). Under basal conditions, GL-Mel cells also expressed higher level of p21Cip1 with respect to M10 cells (Fig. 4). At the tested concentrations, PHA-848125 slightly up-regulated the expression of p53 in GL-Mel cells, while it induced an increase in the lev- els of p21Cip1 in both GL-Mel and M10 cells (Fig. 4). However, in the latter cell line, drug-induced up-regulation of p21Cip1 was less pronounced.

Table 2
Cell cycle perturbations induced by PHA-848125 in GL-Mel and M10 cell lines.

PHA-848125a Time Percentage GL-Mel
G1 of cellsb

S

G2 /M
M10 G1

S

G2 /M
– 54.1 ± 1.7 34.4 ± 1.7 11.5 ± 1.6 42.9 ± 2.4 38.0 ± 1.6 19.1 ± 1.4
0.156 µM 24 h 62.3 ± 2.7* 30.3 ± 2.7 7.9 ± 1.2 47.7 ± 4.4 37.0 ± 2.9 15.3 ± 2.6
0.625 µM 87.1 ± 2.2* 11.0 ± 2.4* 1.7 ± 0.3* 52.1 ± 3.4* 33.8 ± 2.6 14.1 ± 1.7*
– 56.4 ± 2.1 31.3 ± 2.3 12.2 ± 1.8 48.0 ± 2.4 36.5 ± 1.2 15.5 ± 1.4
0.156 µM 96 h 62.1 ± 3.6 27.9 ± 4.7 9.9 ± 1.3 52.2 ± 3.2 35.9 ± 1.3 11.9 ± 2.4
0.625 µM 81.1 ± 0.4* 12.7 ± 1.2* 6.2 ± 1.6* 56.5 ± 3.4* 33.2 ± 2.5 10.3 ± 1.4*
a Melanoma cells were cultured in the presence of the indicated concentrations of PHA-848125 for 24 or 96 h and then processed for cell cycle analysis. Control groups were treated with DMSO alone.
b The percentage of cells in each phase of cell cycle was evaluated by flow cytometric analysis of DNA content. Each value represents the mean ± standard error of the
mean of at least three independent experiments.
* p < 0.05, according to paired Student’s t-test, comparing each treated group with the corresponding control.

⦁ Effects of PHA-848125 on the expression of cell cycle regulators

To get further insight into the mechanism of action of PHA- 848125, and to identify possible cellular determinants affecting the response to the drug, expression of CDK2, CDK4, total RB, RB phos- phorylated at either Thr821 (CDK2 specific site) or Thr826 (CDK4 specific site), cyclin A, D1, and E, p16INK4a, and p27Kip1 was evalu- ated in GL-Mel and M10 cell lines, either untreated or exposed to 0.156 or 0.625 µM PHA-848125 for 24 h.
The results illustrated in Fig. 5 show that under basal condi- tions, the amount of total RB, CDK2, and RB phosphorylated on
Thr826 was higher in GL-Mel than in M10 cell line, while CDK4, RB phosphorylated on Thr821, cyclin A, and cyclin E, were more expressed in M10 cells. The p16INK4a protein was absent in GL-Mel cells, while it was expressed in M10 and confirmed to be wild type by the mutational analysis of the corresponding coding gene.
In both cell lines, treatment with PHA-848125 did not affect the expression of CDK2, and CDK4 (Fig. 5). In GL-Mel cells, PHA-848125 down-regulated the level of total RB, and RB phosphorylated on Thr821 and on Thr826, and produced a reduction in the expression of cyclin A. In contrast, cyclin D1, cyclin E, and p27Kip1 were up- regulated by drug treatment (Fig. 5). In M10 cells, treatment with PHA-848125 was also followed by a reduction of RB phosphorylated

Fig. 2. Cell cycle perturbations induced by PHA-848125 in GL-Mel and M10 cells. Melanoma cells were cultured with the indicated concentrations of PHA-848125 for 24 h and then the percentage of cells in each phase of the cell cycle was evaluated by flow cytometric analysis of DNA content. Control groups were treated with DMSO alone. Data collection was gated utilizing forward light scatter and side light scatter to exclude cell debris and cell aggregates. The PI fluorescence was measured on a linear scale. The percentage of the cells in the different phases of the cell cycle is reported on each histogram.

Fig. 3. Reversibility of cell cycle perturbations induced by PHA-848125 in GL-Mel cells. The cells were exposed to the indicated concentrations of PHA-848125 for 24 h, washed and then cultured in CM alone for additional 72 h. The percentage of cells in each phase of the cell cycle was evaluated by flow cytometric analysis of DNA content at the end of treatment and every 24 h. Control groups were treated with DMSO alone. Data collection was gated utilizing forward light scatter and side light scatter to exclude cell debris and cell aggregates. The PI fluorescence was measured on a linear scale. The percentage of the cells in the different phases of the cell cycle is reported on each histogram. The results are representative of two independent experiments.

on Thr821. However, drug-induced changes in the expression of this protein were less pronounced with respect to that found in GL-Mel cells. In contrast, no changes were observed in the levels of total RB, RB phosphorylated on Thr826, cyclin A, cyclin D1, and cyclin E (Fig. 5). No changes were also observed in the expression of p16INK4a.

⦁ Antiproliferative effects of combined treatment with TMZ and PHA-848125

Previous studies demonstrated that CDK inhibitors can syn- ergize with several conventional cytotoxic agents, usually in a sequence-dependent manner. In particular, for the majority of

Fig. 4. Effect of PHA-848125 on p53 and p21 expression in GL-Mel and M10 cells. The cells were treated with the indicated concentrations of PHA-848125 for 24 h. Whole cell extracts were prepared and resolved on 10% (p53) or 15% (p21Cip1 ) SDS- polyacrylamide gels. Proteins were transferred to nitrocellulose membranes and probed with antibodies against p53 and p21Cip1 . Incubation with the anti-actin mAb was performed as a loading control. The immune complexes were visualized using ECL. The results are representative of three independent experiments. The molecular weights of p53 and p21Cip1 proteins are 53 and 21 kDa, respectively.
the drugs analyzed so far, optimal treatment schedules require that CDK inhibitors are administered after the antineoplastic agent [5–7].
To assess whether TMZ and PHA-848125 could act synergisti- cally to inhibit melanoma growth, we incubated tumor cells with TMZ for 48 h and then PHA-848125 was added to the cultures for additional 72 h. This treatment schedule was designed to allow the cells to complete a first round of DNA duplication after TMZ treatment, and to generate O6-MeG:T mispairs required for the engagement of the MMR system able to trigger the cytotoxic cas- cade.
Three cell lines (i.e. GL-Mel, M10 and SK-Mel-28) showing dif- ferent sensitivity to PHA-848125, TMZ, and TMZ combined with 10 µM BG (hereafter referred to as TMZ/BG), were selected for these studies. For each cell line, concentration–response curves were determined for PHA-848125, TMZ, TMZ/BG, and for the associa- tion of TMZ or TMZ/BG with PHA-848125. Concentration–response curves for the combinations were set up maintaining a constant ratio between TMZ and PHA-848125, and using up to four differ- ent ratios to generate different concentration–response curves. In particular, TMZ/BG was combined with PHA-848125 at two differ- ent ratios, i.e. the ratio corresponding approximately to that of the IC50 values of TMZ/BG and of PHA-848125, and the half of this ratio (Supplementary Table S1). TMZ was combined with PHA-848125 at four different ratios, i.e. the ratio of the IC50 values of the two drugs given alone, the half of this ratio, and the ratios tested for the TMZ/BG + PHA-848125 combination (Supplementary Table S1). The CI of the various combinations at a level of growth inhibition of 75% and 95% was calculated according to Chou and Talalay [21].
In all cell lines treated with PHA-848125 48 h after exposure to TMZ/BG, the tested combinations were more effective than the sin- gle agents, as indicated by the shift of the concentration–response

Fig. 5. Effect of PHA-848125 on the expression of cell cycle regulators in GL-Mel and M10 cells. The cells were treated with the indicated concentrations of PHA- 848125 for 24 h. Whole cell extracts were prepared and resolved on 8% (total RB, phospho-RB-Thr821, phospho-RB-Thr826), 10% (cyclin A and cyclin E), 12% (cyclin D1) or 15% (CDK2, CDK4, p16INK4a, p27Kip1 ) SDS-polyacrylamide gels. Proteins were transferred to nitrocellulose membranes and probed with antibodies against the indicated proteins. Incubation with the anti-actin mAb was performed as a loading control. The immune complexes were visualized using ECL. The results are repre- sentative of three independent experiments. The molecular weights of the analyzed proteins are as follows: RB, 110 kDa; cyclin A, 60 kDa; cyclin D1, 37 kDa; cyclin E, 50 kDa; p16INK4a, 16 kDa; p27Kip1 , 27 kDa; CDK2, 33 kDa; CDK4, 34 kDa.

curves of the combinations to the left of those corresponding to TMZ/BG or PHA-848125 alone (Fig. 6 and data not shown). The CI values reported in Table 3 indicate that in M10 cells, the effects of TMZ/BG and PHA-848125 were nearly additive. In GL-Mel and SK-

Table 3
CI values for the association of TMZ/BG and PHA-848125.

Cell line TMZ/BG + PHA-848125 combinationa

TMZ:PHA-848125 ratio CIb

GI 75% GI 95%
GL-Mel 500:1 0.61 ± 0.02 0.87 ± 0.05
250:1 0.71 ± 0.05 0.76 ± 0.04
M10 50:1 0.88 ± 0.11 1.06 ± 0.04
25:1 1.02 ± 0.22 0.92 ± 0.17
SK-Mel-28 500:1 0.61 ± 0.03 0.44 ± 0.04
250:1 0.56 ± 0.02 0.31 ± 0.04
a Melanoma cells were incubated with graded amounts of TMZ in the pres- ence of 10 µM BG (TMZ/BG) for 48 h. Graded amounts of PHA-848125 were then added to the cultures to obtain the indicated TMZ:PHA-848125 ratios. Cell growth was evaluated by the MTT assay 72 h after the addition of PHA-848125. Concentration–response curves were also set up for TMZ/BG and PHA-848125 alone. b For each TMZ/BG and PHA-848125 ratio, the combination index (CI) value was calculated at drug-induced growth inhibition (GI) of 75% and 95%. Each value rep- resents the mean ± standard error of the mean of three independent experiments.
Synergistic effects are indicated by CI < 1, additive effects by CI = 1.
Mel-28 cells, synergistic effects between TMZ/BG and PHA-848125 were observed at both drug ratios tested (Table 3).
When the cells were exposed to TMZ in the absence of BG, and the concentration–response curves of the association of TMZ and PHA-848125 were set up using the same drug ratios selected for the TMZ/BG and PHA-848125 combination, synergy between TMZ and the CDK inhibitor was still detected in GL-Mel cells (Fig. 7 and data not shown), which are moderately sensitive to TMZ alone. In M10 and SK-Mel-28 cells, TMZ concentrations lower than 100 µM (M10) or 400 µM (SK-Mel-28) did not alter cell proliferation, and the growth inhibitory effect of the tested combinations strictly overlapped that of PHA-848125 alone (Fig. 7 and data not shown). Additive or synergistic effects were observed in M10 and GL- Mel cells when TMZ and PHA-848125 were combined using higher TMZ concentrations (Table 4). In SK-Mel-28 cells, the antiprolifera- tive effect of the combinations set up at TMZ:PHA-848125 ratios of 1000:1 or 2000:1 was not higher than that obtained with the single
agents alone (data not shown).
± ±
±
Further experiments were carried out with two primary cul- tures of normal melanocytes (i.e. NM-1 and NM-2 cells) obtained from two different donors. On the basis of 5-day MTT assays the mean IC50 values (±standard error) relative to TMZ/BG were 811 ± 53 µM and 239 ± 25 µM for NM-1 and NM-2 respectively. The mean IC50 values ( standard error) relative to PHA-848125 were 0.583 0.168 µM and 0.123 0.017 µM for NM-1 and NM-2, respectively. In the experiments performed to test the effect of the combined treatment with the two agents, the cells were exposed to TMZ/BG, to PHA-848125 or to both drugs at TMZ:PHA-848125 ratio of 500:1. The range of drug concentrations utilized in this case was similar to that selected for SK-Mel-28 cell line. This experimental design was adopted with the intent to treat the normal counterpart of melanoma cells with the highest levels of antineoplastic agents utilized in the present study, optimizing the antitumor activity of TMZ through MGMT suppression. The results, illustrated in Fig. 8, confirmed that NM-1 cells were more resistant than NM-2 cells to either TMZ/BG or PHA-848125, and showed that a similar sensi- tivity pattern was detectable for both drugs used in combination. Differently from melanoma cells exposed to TMZ/BG + PHA-848125 combination, no synergistic or additive but rather antagonistic effects were detected in both melanocyte cultures.

⦁ Discussion

The present study illustrates novel aspects of the antitumor activity of PHA-848125, a new and potent cyclin A/CDK2 inhibitor, also active against D1/CDK4, E/CDK2, B/CDK1 and H/CDK7 [14],

Table 4
CI values for the association of TMZ and PHA-848125.

Cell line TMZ + PHA-848125 combinationa

TMZ:PHA-848125 ratio CIb

GI 75% GI 95%
GL-Mel 2000:1 0.96 ± 0.20 0.72 ± 0.08
1000:1 0.99 ± 0.22 0.92 ± 0.11
M10 800:1 1.10 ± 0.03 0.95 ± 0.04
400:1 0.76 ± 0.01 0.67 ± 0.08
a Melanoma cells were exposed to graded amounts of TMZ for 48 h. Graded amounts of PHA-848125 were then added to the cultures to obtain the indicated TMZ:PHA-848125 ratios. Cell growth was evaluated by the MTT assay 72 h after the addition of PHA-848125. Concentration–response curves were also set up for TMZ and PHA-848125 alone.
b For each TMZ and PHA-848125 ratio, the combination index (CI) value was
calculated at drug-induced growth inhibition (GI) of 75% and 95%. Each value rep- resents the mean ± standard error of the mean of three independent experiments. Synergistic effects are indicated by CI < 1, additive effects by CI = 1.

Fig. 6. Effect of combined treatment with TMZ/BG and PHA-848125 on melanoma cell growth. GL-Mel, M10 and SK-Mel-28 cell lines were cultured in the presence of increasing concentrations of TMZ plus 10 µM BG (TMZ/BG) for 48 h. Graded amounts of PHA-848125 were then added to the cultures to obtain, for each point of the concentration–response curve of the drug combination, a constant ratio between TMZ and PHA-848125 (i.e. 500:1 for GL-Mel cells; 50:1 for M10 cells and 500:1 for SK-Mel- 28 cells). Cell growth was evaluated by the MTT assay 72 h after the addition of PHA-848125. Concentration–response curves for TMZ/BG or PHA-848125 alone were also set up (see Section 2). For each cell line, growth inhibition induced by the drug combination is plotted as a function of TMZ concentration (A, C, E) or PHA-848125 concentration (B, D, F). Each value represents the mean of three independent experiments, with bars indicating standard error of the mean.

showing that the drug is markedly active against melanoma cells at clinical concentrations.
±
In all melanoma cell lines tested, PHA-848125 produced cell growth inhibition with IC50 values ranging from 0.123 to 0.680 µM. In Phase I clinical studies, the drug reached a peak plasma con- centration of 1.47 0.51 µM when given for 7 consecutive days every 2 weeks, at the dose of 150 mg/day (i.e. the recommended Phase II dose with this administration modality) [34]. Therefore,
the observed IC50 values of PHA-848125, far below this peak plasma concentration, suggest that the drug can have a therapeutic efficacy in melanoma patients. Most important for its clinical application is the finding that PHA-848125 is effective against melanoma cells markedly resistant to TMZ. Actually, only 10–15% of melanoma patients respond to triazene compounds, and rapid onset of resis- tance to these agents occurs both in the clinic and in in vitro models [9,35]. Moreover, up to now no treatment regimen has been shown

Fig. 7. Effect of combined treatment with TMZ and PHA-848125 on melanoma cell growth. GL-Mel, M10 and SK-Mel-28 cell lines were cultured in the presence of increasing concentrations of TMZ for 48 h. Graded amounts of PHA-848125 were then added to the cultures to obtain, for each point of the concentration–response curve of the drug combination, a constant ratio between TMZ and PHA-848125 (i.e. 500:1 for GL-Mel cells; 50:1 for M10 cells and 500:1 for SK-Mel-28 cells). Cell growth was evaluated by the MTT assay 72 h after the addition of PHA-848125. Concentration–response curves for TMZ or PHA-848125 alone were also set up (see Section 2). Each value represents the mean of three independent experiments, with bars indicating standard error of the mean. For GL-Mel cells, combination index values at drug-induced growth inhibition of 75% and 95% were as follows: (a) 0.85 ± 0.14; 0.67 ± 0.08, TMZ:PHA-848125 ratio of 500:1; (b) 0.59 ± 0.09; 0.29 ± 0.04, TMZ:PHA-848125 ratio of 250:1.

Fig. 8. Effect of combined treatment with TMZ/BG and PHA-848125 on the growth of normal melanocytes. NM-1 and NM-2 cells were cultured in the presence of increasing concentrations of TMZ plus 10 µM BG (TMZ/BG) for 48 h. Graded amounts of PHA-848125 were then added to the cultures to obtain, for each point of the concentration–response curve of the drug combination, a constant ratio between TMZ and PHA-848125 (i.e. 500:1). Cell growth was evaluated by the MTT assay 72 h after the addition of PHA-848125. Concentration–response curves for TMZ/BG or PHA-848125 alone were also set up (see Section 2). Each value represents the mean of three independent experiments, with bars indicating standard error of the mean. For NM-1 and NM-2, combination index values at drug-induced growth inhibition of 75% were
6.88 ± 0.95 and 2.24 ± 0.38, respectively.

to improve patient survival with respect to that employing TMZ or dacarbazine alone.
The antitumor effect of PHA-848125 appears to be mainly dependent on impairment of cell cycle progression. Indeed, the highly sensitive GL-Mel cells showed a marked G1 arrest after expo- sure to 0.625 µM of the inhibitor. At the same drug concentration, the less sensitive M10 cells displayed only a moderate, although significant, increase of the G1 fraction and decrease of the G2/M fraction. On the other hand, neither GL-Mel cells nor M10 cells underwent apoptosis when exposed to 0.156 or 0.625 µM PHA- 848125 for up to 96 h. Moreover, apoptosis did not occur in GL-Mel cells treated with the drug at a concentration 10-fold higher than that corresponding to the IC50 value. Finally, GL-Mel cells rapidly recovered from cell cycle perturbations induced by a 24 h-exposure to PHA-848125. Taken together, our data indicate that, at least in GL-Mel and M10 cells, PHA-848125 acts mainly as a cytostatic agent.
Sequential phosphorylation of RB by cyclin D/CDK4-6 and cyclin E/CDK2 impairs RB-mediated suppression of the activity of mem- bers of the E2F family of transcription factors. This is followed by full and coordinated expression of genes critical for S-phase entry, including cyclin A, cyclin E, PCNA, and thymidine kinase (reviewed in [36,37]). In addition to RB, cyclin E/CDK2 phospshory- lates other proteins that regulate cell division, including p27Kip1 that, after phosphorylation, is ubiquitinated and degraded by the proteasome, facilitating S-phase entry [1]. In GL-Mel cells PHA- 848125 down-regulated RB phosphorylation at both CDK4 and CDK2 specific sites, reduced the levels of total RB and cyclin A, and increased the amount of p27Kip1. These molecular events are consistent with PHA-848125 ability to inhibit cyclin D/CDK4 and cyclin A-E/CDK2 activity in biochemical assays and can be involved in the strong G1 arrest in GL-Mel cells. Indeed, in M10 cells, which are less sensitive to the drug than GL-Mel cells, PHA-848125 did not affect the level of total RB, RB phosphorylated on Thr826, cyclin A and p27Kip1, and only moderately reduced the expres- sion of RB phosphorylated on Thr821. Noteworthy, drug-induced impairment of p27Kip1 degradation found in GL-Mel might create a positive feedback loop in which cyclin E/CDK2 inhibition gen- erates high p27Kip1 levels that in turn reinforce the repression of cyclin E/CDK2 and of other CDKs. A molecular characteris- tic that could also underlie the higher PHA-848125 sensitivity of GL-Mel as compared to that of M10, is the lack of p16INK4a expres-
sion in GL-Mel cells. Indeed, inactivation of p16INK4a was found to be associated with increased sensitivity of melanoma cells to the antiproliferative activity of the pan-CDK inhibitor flavopiridol [38].
Up-regulation of p53, although unable to trigger apoptosis, can contribute to the strong G1 arrest induced by PHA-848125 in GL- Mel cells, and loss of p53 function can, a least in part, explain the lower sensitivity of M10 cells to PHA-848125. A key regulator of G1 arrest downstream p53 induction is p21Cip1 (reviewed in [39]). Consistently, we found a strong up-regulation of p21Cip1 in PHA- 848125-treated GL-Mel cells. However, p21Cip1 levels increased, although to a lower extent, also in drug-treated M10 cells, indi- cating that in response to PHA-848125 a positive modulation of p21Cip1 can occur in a p53-independent manner.

Unexpectedly, cyclin D1, and cyclin E, which are positive reg- ulators of the G1 S transition, were found to be moderately up-regulated in PHA-848125-treated GL-Mel cells. Presently, the molecular mechanisms underlying this effect remain to be exper- imentally elucidated. In any case, up-regulation of cyclin D1 and cyclin E does not prevent the strong G1 arrest caused by PHA- 848125 in GL-Mel cells.
In this study we addressed the question of whether combina- tion of TMZ and PHA-848125 could result in additive or synergistic effects on melanoma cell growth. To this end, malignant cells were exposed to TMZ for 48 h and then to PHA-848125 for additional 72 h. This treatment schedule that was tested either in the absence or in the presence of BG, was designed taking into consideration the mechanism of action of TMZ and PHA-848125. Indeed, we consid- ered that because of its ability to induce a G1 arrest, PHA-848125 treatment before or concomitantly with TMZ would have reduced the percentage of cells entering the S-phase required to generate O6-MeG:T mispairs, thus compromising the cytotoxic activity of the triazene compound.
Our results demonstrate that when the activity of TMZ was optimized through MGMT impairment, the TMZ and PHA-848125 combination produced additive effects in the most TMZ-sensitive M10 cell line, while the effects were synergistic in GL-Mel and SK-Mel-28 cell lines. Notably, under similar experimental con- ditions, no synergistic or additive but rather antagonistic effects were observed with normal melanocytes. Different cell cycle kinet- ics between normal and malignant melanocytic cells can offer a reasonable explanation for this finding. Actually, the prolifera-

tion rate of primary cultures of melanocytes is by far lower than that of melanoma cells. Therefore, in TMZ/BG-treated melanocytes, DNA duplication-dependent O6-MeG:T mismatches are less effi- ciently generated before exposure to PHA-848125. It follows that the antiproliferative activity of the triazene compound could be partially compromised by the subsequent addition of the CDK inhibitor.
When melanoma cells were exposed to TMZ and PHA-848125 combination in the absence of BG, and the cells were treated with the same range of TMZ concentrations adopted in the presence of BG, synergy occurred only in the GL-Mel cells. This can be explained considering that the high MGMT activity displayed by M10 and SK-Mel-28 cells prevents TMZ-induced toxicity. Therefore, cell growth inhibition afforded by TMZ + PHA-848125 combination simply reflects the activity of the CDK inhibitor. Although addi- tive or even synergistic effects were obtained in M10 cells at the TMZ:PHA-848125 ratios of 800:1 and 400:1, in the highly TMZ-resistant SK-Mel-28 cell line, the TMZ and PHA-848125 com- binations set up at the highest TMZ:PHA-848125 ratios (i.e. 2000:1 and 1000:1) did not result superior to the single agents alone. This is in line with the observation that, in the absence of BG, synergis- tic or additive effects between the two drugs are detectable when melanoma cells are not endowed with exceedingly high resistance to TMZ. In any case, it is reasonable to speculate that in most of the two-drug combination experiments described here, the growth of cells surviving the in vitro exposure to the DNA methylating agent could be suppressed by additional treatment with the CDK inhibitor.
In conclusion, the present study demonstrates, for the first time,
that PHA-848125 is highly effective in inhibiting melanoma cell growth and cell cycle progression at concentrations achievable in the clinic. Most importantly, the drug shows remarkable growth suppressive activity against melanoma cells highly resistant to TMZ. Moreover, our data show that the combined treatment with TMZ and PHA-848125 produces additive and even synergistic effect on melanoma cell growth, when target cells are at least moderately susceptible to TMZ. This can occur when malignant cells are MMR- proficient and express low levels of MGMT, either spontaneously or following exposure to MGMT inhibitors. Therefore, the present findings support the hypothesis that PHA-848125 can have a ther- apeutic potential in melanoma patients, either alone or combined with TMZ.

Acknowledgement

This study was supported by Istituto Superiore di Sanità, grant number ACC12. The authors wish to thank Massimo Teson for skil- ful assistance.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.phrs.2009.12.009.

References

⦁ Sherr CJ, Roberts JM. Living with or without cyclins and cyclin-dependent kinases. Genes Dev 2004;18:2699–711.
⦁ Deshpande A, Sicinski P, Hinds PW. Cyclins and CDKs in development and cancer: a perspective. Oncogene 2005;24:2909–15.
⦁ Neganova I, Lako M. G1 to S phase cell cycle transition in somatic and embryonic stem cells. J Anat 2008;213:30–44.
⦁ Malumbres M, Barbacid M. Cell cycle, CDKs and cancer: a changing paradigm. Nat Rev Cancer 2009;9:153–66.
⦁ Grant S, Roberts JD. The use of cyclin-dependent kinase inhibitors alone or in combination with established cytotoxic drugs in cancer chemotherapy. Drug Resist Updat 2003;6:15–26.
⦁ Shapiro GI. Cyclin-dependent kinase pathways as targets for cancer treatment. J Clin Oncol 2006;24:1770–83.
⦁ Deep G, Agarwal R. New combination therapies with cell-cycle agents. Curr Opin Investig Drugs 2008;9:591–604.
⦁ Malumbres M, Pevarello P, Barbacid M, Bischoff JR. CDK inhibitors in cancer therapy: what is next? Trends Pharmacol Sci 2008;29:16–21.
⦁ Tawbi HA, Kirkwood JM. Management of metastatic melanoma. Semin Oncol 2007;34:532–45.
⦁ Payne MJ, Pratap SE, Middleton MR. Temozolomide in the treatment of solid tumours: current results and rationale for dosing/scheduling. Crit Rev Oncol Hematol 2005;53(3):241–52.
⦁ Quirt I, Verma S, Petrella T, Bak K, Charette M. Temozolomide for the treat- ment of metastatic melanoma: a systematic review. Oncologist 2007;12: 1114–23.
⦁ Chin L, Garraway LA, Fisher DE. Malignant melanoma: genetics and therapeu- tics in the genomic era. Genes Dev 2006;20:2149–82.
⦁ Fecher LA, Cummings SD, Keefe MJ, Alani RM. Toward a molecular classification of melanoma. J Clin Oncol 2007;25:1606–20.
⦁ { }
⦁ Brasca MG, Amboldi N, Ballinari D, Cameron A, Casale E, Cervi G, et al. Identifi- cation of N,1,4,4-Tetramethyl-8- [4-(4-methylpiperazin-1-yl)phenyl]amino - 4,5-dihydro-1H-pyrazolo[4,3-h]quinazoline-3-carboxamide (PHA-848125), a potent, orally available cyclin dependent kinase inhibitor. J Med Chem 2009.
⦁ Castiglia D, Pagani E, Alvino E, Vernole P, Marra G, Cannavo E, et al. Biallelic somatic inactivation of the mismatch repair gene MLH1 in a primary skin melanoma. Genes Chromosomes Cancer 2003;37:165–75.
⦁ Pepponi R, Marra G, Fuggetta MP, Falcinelli S, Pagani E, Bonmassar E, et al. The effect of O6 -alkylguanine-DNA alkyltransferase and mismatch repair activities on the sensitivity of human melanoma cells to temozolo- mide, 1,3-bis(2-chloroethyl)1-nitrosourea, and cisplatin. J Pharmacol Exp Ther
2003;304:661–8.
⦁ Umar A, Boyer JC, Thomas DC, Nguyen DC, Risinger JI, Boyd J, et al. Defective mismatch repair in extracts of colorectal and endometrial can- cer cell lines exhibiting microsatellite instability. J Biol Chem 1994;269: 14367–70.
⦁ Bondanza S, Maurelli R, Paterna P, Migliore E, Giacomo FD, Primavera G, et al. Keratinocyte cultures from involved skin in vitiligo patients show an impaired in vitro behaviour. Pigment Cell Res 2007;20:288–300.
⦁ Kaina B, Christmann M, Naumann S, Roos WP. MGMT: key node in the bat- tle against genotoxicity, carcinogenicity and apoptosis induced by alkylating agents. DNA Repair (Amst) 2007;6:1079–99.
⦁ Verbeek B, Southgate TD, Gilham DE, Margison GP. O6 -Methylguanine-
DNA methyltransferase inactivation and chemotherapy. Br Med Bull 2008;85:17–33.
⦁ Chou TC, Talalay P. Quantitative analysis of dose–effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 1984;22:27–55.
⦁ D’Atri S, Tentori L, Lacal PM, Graziani G, Pagani E, Benincasa E, et al. Involve- ment of the mismatch repair system in temozolomide-induced apoptosis. Mol Pharmacol 1998;54:334–41.
⦁ Bertuzzi A, Gandolfi A, Germani A, Vitelli R. Estimation of cell DNA synthesis rate from flow-cytometric histograms. Cell Biophys 1983;5:223–36.
⦁ Soto JL, Cabrera CM, Serrano S, Lopez-Nevot MA. Mutation analysis of genes that control the G1 /S cell cycle in melanoma: TP53, CDKN1A, CDKN2A, and CDKN2B. BMC Cancer 2005;5:36.
⦁ Soufir N, Avril MF, Chompret A, Demenais F, Bombled J, Spatz A, et al. Preva- lence of p16 and CDK4 germline mutations in 48 melanoma-prone families in France. The French Familial Melanoma Study Group. Hum Mol Genet 1998;7: 209–16.
⦁ Jiricny J. The multifaceted mismatch-repair system. Nat Rev Mol Cell Biol 2006;7:335–46.
⦁ Casorelli I, Russo MT, Bignami M. Role of mismatch repair and MGMT in response to anticancer therapies. Anticancer Agents Med Chem 2008;8:368–80.
⦁ Rudek MA, Donehower RC, Statkevich P, Batra VK, Cutler DL, Baker SD. Temo- zolomide in patients with advanced cancer: phase I and pharmacokinetic study. Pharmacotherapy 2004;24:16–25.
⦁ Wesierska-Gadek J, Schmid G. Dual action of the inhibitors of cyclin-dependent kinases: targeting of the cell-cycle progression and activation of wild-type p53 protein. Expert Opin Investig Drugs 2006;15:23–38.
⦁ Mohapatra S, Coppola D, Riker AI, Pledger WJ. Roscovitine inhibits differ- entiation and invasion in a three-dimensional skin reconstruction model of metastatic melanoma. Mol Cancer Res 2007;5:145–51.
⦁ Milner BJ, Allan LA, Eccles DM, Kitchener HC, Leonard RC, Kelly KF, et al. p53 mutation is a common genetic event in ovarian carcinoma. Cancer Res 1993;53:2128–32.
⦁ O’Connor PM, Jackman J, Bae I, Myers TG, Fan S, Mutoh M, et al. Characteriza- tion of the p53 tumor suppressor pathway in cell lines of the National Cancer Institute anticancer drug screen and correlations with the growth-inhibitory potency of 123 anticancer agents. Cancer Res 1997;57:4285–300.
⦁ Holstege H, Joosse SA, van Oostrom CT, Nederlof PM, de Vries A, Jonkers J. High incidence of protein-truncating TP53 mutations in BRCA1-related breast cancer. Cancer Res 2009;69:3625–33.
⦁ Tibes R, Jimeno A, Von Hoff DD, Walker R, Pacciarini MA, Scaburri A, et al.ASCO Annual Meeting Proceedings (Post-Meeting Edition) vol. 26. 2008 (no. 15S May 20 Supplement).
⦁ Alvino E, Castiglia D, Caporali S, Pepponi R, Caporaso P, Lacal PM, et al. A single cycle of treatment with temozolomide, alone or combined with O6 – benzylguanine, induces strong chemoresistance in melanoma cell clones in

vitro: role of O6 -methylguanine-DNA methyltransferase and the mismatch repair system. Int J Oncol 2006;29:785–97.
⦁ Giacinti C, Giordano A. RB and cell cycle progression. Oncogene 2006;25:5220–7.
⦁ Du W, Pogoriler J. Retinoblastoma family genes. Oncogene 2006;25:5190–200.
⦁ Robinson WA, Miller TL, Harrold EA, Bemis LT, Brady BM, Nelson RP. The effect of flavopiridol on the growth of p16+ and p16− melanoma cell lines. Melanoma Res 2003;13:231–8.
⦁ Gartel AL. p21(WAF1/CIP1) and cancer: a shifting paradigm? Biofactors 2009;35:161–4.

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>