Search

The phosphatidylinositol-3-kinase (PI3K)/Akt/mTOR pathway is a major target for cancer therapy. As a strategy to induce the maximal inhibition of this pathway in cancer cells, we combined allosteric mTOR inhibitors (rapamycin and RAD001) with a dual PI3K/mTOR kinase inhibitor (PI-103). Both in vitro and in vivo, the combination exhibited more activity than single agents in human ovarian and prostate cancer cells that harbor alterations in the pathway. At the molecular level, combined inhibition of mTOR prevented the rebound activation of Akt that is seen after treatment with rapamycin and its analogues and caused more sustained inhibition of Akt phosphorylation. Furthermore, the combination strongly inhibited the expression of PI3K/Akt/mTOR downstream proteins. In particular, it showed greater activity than the single agents in inhibiting the phosphorylation of 4EBP1, both in vitro and in vivo, resulting in selective inhibition of CAP-dependent translation. A proteomic approach was used to confirm the identification of c-Myc as the key regulator for the reduction in downstream proteins affected by the combined inhibition of mTOR. In conclusion, the combination of a catalytic and an allosteric inhibitor of mTOR shows greater activity, without a concomitant increase in toxicity, than either drug alone, and this may have therapeutic implications for inhibiting this pathway in the clinical setting., (©2011 AACR.)

Cell motility and invasion play an essential role in the development of metastasis. Evidence suggests that the enzyme phospholipase Cgamma1 (PLCgamma1) may be involved in tumor progression and possibly development of metastasis. In this study, we show that down-regulation of PLCgamma1 expression severely impairs activation of the small GTP-binding protein Rac and cell invasion in breast cancer cell lines and U87 in vitro. Experimental metastasis assays in nude mice show that inducible knockdown of PLCgamma1 strongly inhibits development of MDA-MB-231-derived lung metastasis and reverts metastasis formation. In addition, analysis of 60 breast cancer patients' tissues revealed an increase of PLCgamma1 expression in metastasis compared with the primary tumor in 50% of tissues analyzed. These data show a critical role of PLCgamma1 in the metastatic potential of cancer cells, and they further indicate that PLCgamma1 inhibition has a therapeutic potential in the treatment of metastasis dissemination.

The purpose of this study was to investigate the antiangiogenic and in vivo properties of the recently identified phosphatidylinositol 3-kinase (PI3K)/Akt inhibitor Inositol(1,3,4,5,6) pentakisphosphate [Ins(1,3,4,5,6)P5]. Because activation of the PI3K/Akt pathway is a crucial step in some of the events leading to angiogenesis, the effect of Ins(1,3,4,5,6)P5 on basic fibroblast growth factor (FGF-2)-induced Akt phosphorylation, cell survival, motility, and tubulogenesis in vitro was tested in human umbilical vein endothelial cells (HUVEC). The effect of Ins(1,3,4,5,6)P5 on FGF-2-induced angiogenesis in vivo was evaluated using s.c. implanted Matrigel in mice. In addition, the effect of Ins(1,3,4,5,6)P5 on growth of ovarian carcinoma SKOV-3 xenograft was tested. Here, we show that FGF-2 induces Akt phosphorylation in HUVEC resulting in antiapoptotic effect in serum-deprived cells and increase in cellular motility. Ins(1,3,4,5,6)P5 blocks FGF-2-mediated Akt phosphorylation and inhibits both survival and migration in HUVEC. Moreover, Ins(1,3,4,5,6)P5 inhibits the FGF-2-mediated capillary tube formation of HUVEC plated on Matrigel and the FGF-2-induced angiogenic reaction in BALB/c mice. Finally, Ins(1,3,4,5,6)P5 blocks the s.c. growth of SKOV-3 xenografted in nude mice to the same extent than cisplatin and it completely inhibits Akt phosphorylation in vivo. These data definitively identify the Akt inhibitor Ins(1,3,4,5,6)P5 as a specific antiangiogenic and antitumor factor. Inappropriate activation of the PI3K/Akt pathway has been linked to the development of several diseases, including cancer, making this pathway an attractive target for therapeutic strategies. In this respect, Ins(1,3,4,5,6)P5, a water-soluble, natural compound with specific proapoptotic and antiangiogenic properties, might result in successful anticancer therapeutic strategies.

Brostallicin (PNU-166196) is a synthetic alpha-bromoacrylic, second-generation DNA minor groove binder structurally related to distamycin A, presently in Phase II trials in Europe and the United States. The compound shows broad antitumor activity in preclinical models and dramatically reduced in vitro myelotoxicity in human hematopoietic progenitor cells compared with that of other minor groove binders. Brostallicin showed a 3-fold higher activity in melphalan-resistant L1210 murine leukemia cells than in the parental line (IC(50) = 0.46 and 1.45 ng/ml, respectively) under conditions in which the cytotoxicity of conventional antitumor agents was either unaffected or reduced. This melphalan-resistant cell line has increased levels of glutathione (GSH) in comparison with the parental cells. Conversely, GSH depletion by buthionine sulfoximine in a human ovarian carcinoma cell line (A2780) significantly decreased both the cytotoxic and the proapoptotic effects of brostallicin. In one experiment, human glutathione S-transferase pi (GST-pi) cDNA was transfected into A2780 cells, and four clones of A2780 with different expression levels of GST-pi were generated (i.e., two clones with high and two clones with low GST-pi expression). A 2-3-fold increase in GST-pi levels resulted in a 2-3-fold increase in cytotoxic activity of brostallicin. Similar results were obtained for GST-pi-transfected human breast carcinoma cells (MCF-7). Brostallicin showed 5.8-fold increased cytotoxicity in GST-pi-transfected versus empty vector-transfected cells with low GST-pi expression. In an in vivo experiment, A2780 clones were implanted into nude mice. The antitumor activity of brostallicin was higher in the GST-pi-overexpressing tumors without increased toxicity. Regarding the mechanism of action, brostallicin interacts reversibly with the DNA minor groove TA-rich sequences but appears unreactive in classical in vitro DNA alkylation assays. We speculated that an intracellular reactive nucleophilic species, e.g., GSH, could react with the alpha-bromoacrylamide moiety functions. Experiments on the interaction with plasmid DNA showed a change of the DNA topology from supercoiled to circular form (nicking) in the presence of GSH, whereas no change was found in its absence. In vitro incubations of brostallicin were performed with the human recombinant GST isoenzymes A1-1, M1-1, and P1-1 (alpha, mu and pi isoenzymes, respectively) in the presence of GSH. The decrease in brostallicin levels was monitored in these incubations; the rate of loss (and therefore brostallicin metabolism) was significantly higher for the M1-1 and P1-1 isoenzymes than for the A1-1 isoenzyme.

The activity of 4-demethoxy-3'-deamino-3'-aziridinyl-4'-methylsulphonyl-daunorubicin (PNU-159548), a new alkycycline with high antitumor activity against a broad range of cancer cells, was evaluated in vitro and in vivo in cells selected for resistance to different anticancer agents. Both in vitro and in vivo, PNU-159548 did retain its activity in cells expressing the multidrug resistance (MDR) phenotype, associated to MIDR-1 gene overexpression or with an alteration in the topoisomerase II gene (altered MDR), independently on the drug used for the selection of the resistant cell line. According to these data, the intracellular uptake of PNU-159548 is not influenced by the presence of MDR-1. PNU-159548 was also active, both in vitro and in vivo, against cells showing resistance to various alkylating agents iincluding cisplatin, cyclophosphamide, and melphalan) and topoisomerase I-inhibitors. Cells defective in nucleotide excision repair, which did show hypersensitivity to treatment with UV irradiation and alkylating agents, showed only a marginally increased sensitivity to PNU-159548. Similarly, the activity of the drug was not influenced by the mismatch repair system, as assessed in two different cellular systems deficient in hMLH1 expression and in which hMLH1 activity was restored by chromosome 3 transfer. The results obtained clearly indicate that the new anticancer agent PNU-159548 is able to overcome the classical mechanisms of resistance emerging after treatment with the most clinically used anticancer agents, and it could represent an alternate choice in the treatment of those tumors refractory to conventional therapy.

We examined the consequences of p73alpha overexpression on gene expression and cellular response to anticancer agents in clones from the human ovarian cancer cell line A2780. Using microarray filters, the expression of 588 genes in two clones overexpressing p73alpha (A2780/p73.4 and A2780/ p73.5) in comparison with empty vector-transfected cells was evaluated. There were clear differences in gene expression profiles. Both of the clones showed a marked increase in the expression of genes involved in DNA repair, including genes participating in nucleotide excision repair and mismatch repair. This was confirmed by reverse transcription-PCR and Northern blot analysis and was associated with an increase in the ability of p73alpha-expressing clones to repair two different DDP (cis-dichlorodiammine platinum)-damaged plasmids in a host reactivation assay. p73alpha overexpressing clones were less sensitive than parental cells to alkylating agents treatment or UV radiation but equally sensitive to the topoisomerase I inhibitor topotecan, which indicated that the increase in expression of DNA repair genes has implications for the response to DNA damaging agents.

Hemizygosity and homozygosity mapping studies show that many common sporadic cancers including lung, breast, kidney, cervical, ovarian, and head and neck cancer display deletions on the short arm of chromosome 3. For ovarian cancer, monochromosomal transfer suppression studies have identified three candidate regions for chromosome 3p ovarian cancer tumor suppressor genes (OCTSGs). To accurately map OCTSG candidate regions, we analyzed 70 ovarian tumors for loss of heterozygosity (LOH) at 20 loci on chromosome 3p that were selected to target those regions proposed to contain tumor suppressor genes for common sporadic cancers. All samples were informative for at least five markers. In 33 (52%) tumors without microsatellite instability, LOH was observed for at least one 3p marker. Analysis of 27 ovarian tumors demonstrating both loss and retention of 3p markers enabled us to define four nonoverlapping minimal deletion regions (OCLOHRs): (a) OCLOHR-1 mapped distal to D3S3591 at 3p25-26; (b) OCLOHR-2 mapped between D3S1317 and D3S1259 at 3p24-25; (c) OCLOHR-3 mapped between D3S1300 and D3S1284, an area that includes the FHIT locus at 3p14.2; and (d) OCLOHR-4 mapped between D3S1284 and D3S1274 at 3p12-13, a region known to contain overlapping homozygous deletions in lung and breast tumor cell lines. However, microsatellite markers from the chromosome 3p21.3 interval homozygously deleted in lung cancer cell lines did not identify a distinct OCLOHR. The frequency and extent of 3p LOH correlated with tumor stage such that LOH at two or more OCLOHRs was present in 53% (16 of 30) of stage III tumors but only 26% (5 of 19) of stage I/II tumors (P = 0.08). To determine the relationship between the OCLOHRs and the three candidate ovarian cancer suppression regions (OCSRs) identified previously by monochromosome transfer studies, we performed detailed genetic and physical mapping studies to define the extent of the three candidate OCSRs and to establish YAC contigs covering each region. OCSR-A at 3p25-26 and OCSR-B at 3p24 were shown to overlap with OCLOHR-1 and OCLOHR-2, respectively, providing further evidence for OCTSGs in these regions. We also show that OCSR-C overlaps with a locus at 3p21.3 previously implicated in lung and breast cancer.

We tested the ability of synthetic peptides derived from p21(WAF1), fused to the internalization peptide sequence derived from Antennapedia, to inhibit the growth of cancer cells in two human ovarian cancer cell lines expressing wild-type p53 or not. Two fused peptides corresponding to p21(WAF1) regions 17-33 and 63-77 inhibited cell growth in both cell lines while the same peptides without the internalization sequence were inactive. The fused peptides prevented growth at concentrations which inhibited cyclin-dependent kinase 2 and cdc2 activity, thus demonstrating that the peptides act by mimicking the action of p21(WAF1) on kinases. This study illustrates the potential pharmacological use of small peptides fused with the Antennapedia internalization sequence in proliferative disorders. The approach may be extended to other diseases in which cell penetration of a peptide may be of therapeutic benefit. More stable drug-like molecules with better pharmacological properties could be designed based on the results obtained with peptides.

Nine human ovarian cancer cell lines that express wild-type (wt) or mutated p53 were used to evaluate the cytotoxicity induced by paclitaxel. The IC50 calculated in the five mutated p53-expressing cell lines was not different from the four wt p53-expressing cell lines. The introduction of wt p53, by using a temperature-sensitive mutant murine p53 or the human p53 under the control of a tetracycline-dependent promoter, did not change the cytotoxicity of paclitaxel as compared to mock-transfected cells. By using for each cell line the paclitaxel IC50, we found that these concentrations were sufficient to induce an increase in p53 levels in all of the four wt p53-expressing cells, whereas in the mutated p53-expressing cells, the levels were unaffected. This increase in p53 levels led to an increase in the mRNA and protein levels of p53 downstream genes (WAF1, GADD45, and bax). In none of the cell lines examined was paclitaxel able to induce apoptosis, evaluated by terminal deoxynucleotidyl transferase-mediated nick end labeling staining and filter binding assay at concentrations closed to the IC50. By increasing the concentration of paclitaxel in the filter binding assay, we could see fragmentation of DNA in the different cell lines. We conclude that the presence of p53 is not a determinant for the cytotoxicity induced by paclitaxel in human ovarian cancer cell lines. Differences in the activation of p53 downstream genes could be observed in wt versus mutated p53-expressing cells, but this does not account either for a differential induction of apoptosis or for a change in cytotoxicity induced by paclitaxel.

N-Deformyl-N-(4-N-N,N-bis(2-chloroethylamino)benzoyl)distamy cin A (FCE 24517) is a novel cytotoxic and antitumor agent shortly to be investigated in phase I clinical trials. It was equally effective in inhibiting the growth of the murine L1210 line and of a subline (L1210/PAM) resistant to nitrogen mustards, whereas distamycin A was virtually inactive. The cellular uptake and retention of FCE 24517 and distamycin A were similar, thus excluding the possibility that this marked variation in cytotoxic activity was due to different intracellular concentrations of the two compounds. FCE 24517 did not appear to act as an inhibitor of macromolecule synthesis. As shown by radioactively labeled precursor incorporation only 24 h after drug treatment a significant inhibition of DNA synthesis was observed in L1210 or in L1210/PAM, when a marked proportion of cells was arrested in premitotic phase. FCE 24517 did not cause DNA breaks, DNA interstrand cross-links, or DNA-protein cross-links in L1210 cells exposed to active drug concentrations. A very low amount of radioactivity was found to be bound irreversibly to DNA in L1210 cells exposed for 1 h to [14C]FCE 24517. Using plasmid pBr322 DNA fragments in a modified version of the Maxam and Gilbert DNA sequencing technique we found no detectable binding of FCE 24517 to N-7-guanine (the major site of alkylation for classical alkylating agents), whereas some alkylations to adenine (presumably to N-3-adenine) were demonstrated. Thus it appears that FCE 24517 is a novel antitumor agent with a mode of action different from that of the drugs currently used in the clinic. In summary it is suggested that FCE 24517 acts by causing a few selective alkylations to adenines in the minor groove of DNA, although the precise base sequence necessary has yet to be elucidated.

The five stable metabolites [prostaglandin F2 alpha, prostaglandin D2, prostaglandin E2 (PGE2), thromboxane B2, and 6-keto-prostaglandin F1 alpha] of arachidonic acid (AA) via the cyclooxygenase pathway were measured by high-resolution gas chromatography-mass spectrometry in Lewis lung carcinoma homogenates at various times after tumor implantation (11 to 25 days). Vegetating and necrotic sections of the primary tumor and lung metastases were examined. Vegetating tumor showed a very active AA metabolism. Synthesis of PGE2, the most abundant product, markedly increased during tumor growth (up to 30 micrograms/g). A high and increasing synthetic capacity was also noted for prostaglandin D2 (up to 9 micrograms/g). Minor time differences and lower levels (up to 1.4 micrograms/g) were found for the other AA metabolites. PGE2 and prostaglandin D2 were the major products in necrotic tumor, too, but synthesis was markedly less than in vegetating tumor, and no increase was noted over time. Metastatic tissue showed a different AA metabolic profile, as compared to primary tumor and surrounding lung tissue, with PGE2 and 6-keto-prostaglandin F1 alpha being the main metabolites.

DNA damage caused by methazolastone [an analogue of 5-(3,3-dimethyl-1-triazeno)imidazole-4-carboxamide which does not require metabolic activation] was investigated in L-1210 leukemia which is sensitive to this drug and in a L-1210 subline [L-1210/N,N-bis(2-chloroethyl)-N-nitrosourea (BCNU)] which is resistant to both chloroethylnitrosoureas and methyltriazenes. Both in vitro and in vivo metazolastone caused formation of DNA alkali-labile sites (assessed by alkaline elution techniques) which were present in similar amounts and repaired at a similar rate in L-1210 and L-1210/BCNU. This suggests that these lesions are not crucial to methyltriazenes activity. DNA alkali-labile sites may be due to the removal of 7-methylguanine by 7-methylguanine-DNA glycosylase which showed the same activity in L-1210 and L-1210/BCNU. Flow cytometry studies revealed that in L-1210 but not in L-1210/BCNU methazolastone induced an arrest of cells in SL-G2-M phases. This blockade was delayed, occurring after at least two cell divisions after drug treatment and therefore appeared temporally unrelated to the presence of DNA alkali-labile sites. There was three times more O6-methylguanine-DNA methyltransferase in L-1210/BCNU than in L-1210 suggesting that methylation of O6-guanine is an important lesion for methyltriazenes activity and resistance to this drug may be linked to its repair.

N10-Propargyl-5,8-dideazafolic acid (CB 3717), a new antifolate which directly inhibits thymidylate synthase and which is now under early clinical investigation, was compared with methotrexate (MTX) for its antiproliferative activity and mode of action on M14 human melanoma cell line and NIH/3T3 murine fibroblasts transfected with human c-Ha-ras oncogene (NIH/3T3R). CB 3717 was as active as MTX on both cell lines in inhibiting colony formation, but 20-100 times less potent. After 24 h of exposure both drugs caused an accumulation of cells in the G1 phase of the cell cycle, probably because of inhibition of DNA synthesis and blockage at the G1-S boundary. In NIH/3T3R treated for 16 h with 2 microM MTX or 200 microM CB 3717, we found DNA single-strand breaks amounting to approximately 130 and 140 rad equivalents, respectively, and a considerable number of DNA double-strand breaks, far more than expected if they had been the result of the proximity of single-strand breaks on the two complementary DNA strands. No DNA-protein cross-links were detected. When cells were incubated in drug-free medium for 8 h, there was a further accumulation of single-strand breaks, possibly due to the effects of the drug retained intracellularly as polyglutamyl derivative. Simultaneous treatment with 1.77 microM cycloheximide prevented DNA damage produced by both drugs. Thymidine (10 microM), renewed in the culture medium every 24 h, also prevented DNA damage and cytotoxicity. Since after 16 h treatment with MTX or CB 3717 cells were completely viable, as assessed by [3H]thymidine release, trypan blue exclusion test, and 51Cr release, DNA damage appears to be an early event preceding cell death and may be a feature of the killing ability of the drugs. The involvement of a protein in the formation of DNA breaks is suggested by the fact that when protein synthesis was inhibited with cycloheximide DNA damage was no longer seen.

The five stable metabolites [prostaglandin F2 alpha (PGF2 alpha), prostaglandin D2 (PGD2), prostaglandin E2 (PGE2), thromboxane B2 (TXB2), and 6-ketoprostaglandin F1 alpha (6-keto-PGF1 alpha)] of arachidonic acid (AA) via the cyclooxygenase pathway were measured by high-resolution gas chromatography-mass spectrometry in M5076 ovarian reticulosarcoma (M5) homogenates at various times after tumor implantation (Days 15, 18, 21, and 24). Vegetating tumor showed an active AA overall metabolism, which significantly increased during tumor growth. Synthesis of selected products (TXB2, PGD2, and PGE2) increased markedly over time (up to 10.6, 3.5, and 0.9 micrograms/g, respectively). The overall metabolic profile was TXB2 much greater than PGD2 greater than PGF2 alpha greater than 6-keto-PGF1 alpha greater than PGE2 on Day 15 and TXB2 much greater than PGD2 much greater than PGF2 alpha greater than 6-keto-PGF1 alpha on Day 24. TXB2 was also by far the most abundant product of in vitro-cultured M5 cells. Chronic treatment of M5-bearing mice with dazmegrel (UK-38,485), a selective thromboxane synthetase inhibitor (100 mg/kg p.o. daily, from Day 7 to killing), resulted in incomplete TXB2 synthesis inhibition, AA metabolism diversion toward the other prostaglandins, and no effects of tumor growth and metastasis. More frequent dazmegrel treatment (100 mg/kg p.o. every 8 h from Day 1 to killing) resulted in complete TXB2 synthetase inhibition, AA metabolism diversion, and increased tumor growth and metastasis. These data do not support the hypothesis of thromboxane synthetase inhibitors reducing tumor growth. However, since TXB2 suppression was accompanied by the production of other products possibly interfering in tumor growth, no conclusions on the effective role of TXA2 in malignancy can be drawn.