Search

Paclitaxel (PTX) and other microtubule inhibitors cause mitotic arrest. However, low concentrations of PTX (low PTX) paradoxically cause G1 arrest (without mitotic arrest). Here, we demonstrated that unexpectedly, low PTX did not cause G1 arrest in the first cell cycle and did not prevent cells from passing through S phase and entering mitosis. Mitosis was prolonged but cells still divided, producing either two or three cells (tripolar mitosis), thus explaining a sub G1 peak caused by low PTX. Importantly, sub G1 cells were viable and non-apoptotic. Some cells fused back and then progressed to mitosis, frequently producing three cells again before becoming arrested in the next cell-cycle interphase. Thus, low PTX caused postmitotic arrest in second and even the third cell cycles. By increasing concentration of PTX, tripolar mitosis was transformed to mitotic slippage, thus eliminating a sub G1 peak. Time-lapse microscopy revealed that prolonged mitosis ensured a p53-dependent postmitotic arrest. We conclude that PTX directly affects cells only in mitosis and the duration of mitosis determines cell fate, including p53-dependent G1-like arrest.

The Gotham Prize was awarded to Alex Varshavsky for "Targeting the absence", a strategy employing negative targets of cancer therapy. This is a brilliant example of therapeutic engineering: designing a sequence of events that leads to the selective killing of one type of cell, while sparing all others. A complex molecular device (Varshavsky's Demon) examines DNA, recognizes the present target in normal cells and kills cancer cells. The strategy is limited by the delivery (transfection or infection) of DNA-based devices into each cell of our body. How can we overcome this limitation? Can therapeutic engineering be applied to small drugs? Can each small molecule reach a cell separately and, once in a cell, exert orchestrated action governed by cellular context? Here I describe how a combination of small drugs can acquire a demonic power to check, choose and selectively kill. The cytotoxicity is restricted to cells lacking (or having) one of the targets. For example, in the presence of a normal target, one drug can cancel the cytotoxic action of another drug. And by increasing a number of targets, we can increase the precision and power of such 'restrictive' combinations. Here I discuss restrictive combinations of currently available drugs that could be tested in clinical trials. Could then these combinations cure cancer today? And what does 'cure' really mean? This article suggests the answer.

Insightful articles by Kirkwood and other outstanding scientists reveal paradoxes of aging. The source of paradoxes is an assumption that aging is caused by random, cumulative molecular damage. Here I demonstrate that a concept of TOR-driven program-like aging almost automatically resolves eleven paradoxes of aging. This article discusses why the accumulation of molecular damage does not limit life span, why calorie restriction and inhibition of protein synthesis extend life span, why the non-existing 'program' for aging is nevertheless robust, why a key gene for aging cannot be found by knocking it out, why low insulin is associated with good health but low insulin response with bad health, why aging is not a disease but can be treated as a disease, why 'healthy' aging is slow aging, and how we know that calorie restriction actually slows aging in humans.

As recently suggested, the target of rapamycin (TOR) pathway, rather than molecular damage by free radicals, drives aging and diseases of aging. But may mitochondria nevertheless contribute to aging? Here, I discuss aimless program-like aging (versus altruistic program), conflict between the cell and mitochondria, cell murder (versus cell suicide) and the role of mitochondria in aging. In particular, life-long selection among mitochondria may yield "selfish" (malignant) mitochondria resistant to autophagy. And TOR may create an intra-cellular environment that is permissive for such selfish mitochondria. In theory, pharmacologic inhibitors of the TOR pathway may reverse accumulation of defective mitochondria, while also inhibiting the aging process., (Copyright (c) 2007 Wiley-Liss, Inc.)

It has become a cliché that cancer therapy fails because it does not target rare cancer stem cells (CSCs). Here we are discuss that this is not how therapy fails and not any cancer cell with stem-like properties is CSC. Paradoxically, CSCs must be resting to explain their resistance to therapy yet must be cycling to explain their persistence in cell culture. To solve contradictions, this article introduces the term cancer stemloids (or stem cell-like cells) to describe proliferating self-renewing cells. The stem cell hierarchy (stem--proliferating--terminal cells) exists exactly to separate self-renewal (immortality) from proliferation. Cancer stemloids break the stem cell hierarchy and eventually may replace other cells. While CSC is shielded from any selective pressure and therefore unable to drive tumor progression, cancer stemloids undergo clonal selection, accumulate mutations, thus determining tumor progression and therapeutic failures. Unlike CSC, cancer stemloids are a crucial target for cancer therapy, exactly because they proliferate. Furthermore, two normally mutually-exclusive properties (proliferation and stemness) provide a means to design therapy to kill cancer stemloids selectively without killing normal stem and non-stem cells. In contrast, true CSCs are not only a difficult, but also an insufficient and perhaps even an unnecessary therapeutic target, especially in advanced malignancies.

Damage-induced G1 checkpoint in mammalian cells involves upregulation of p53, which activates transcription of p21(Waf1) (CDKN1A). Inhibition of cyclin-dependent kinase (CDK)2 and CDK4/6 by p21 leads to dephosphorylation and activation of Rb. We now show that ectopic p21 expression in human HT1080 fibrosarcoma cells causes not only dephosphorylation but also depletion of Rb; this effect was p53-independent and susceptible to a proteasome inhibitor. CDK inhibitor p27 (CDKN1B) also caused Rb dephosphorylation and depletion, but another CDK inhibitor p16 (CDKN2A) induced only dephosphorylation but not depletion of Rb. Rb depletion was observed in both HT1080 and HCT116 colon carcinoma cells, where p21 was induced by DNA-damaging agents. Rb depletion after DNA damage did not occur in the absence of p21, and it was reduced when p21 induction was inhibited by p21-targeting short hairpin RNA or by a transdominant inhibitor of p53. These results indicate that p21 both activates Rb through dephosphorylation and inactivates it through degradation, suggesting negative feedback regulation of damage-induced cell-cycle checkpoint arrest.

Any drug selects for drug resistance. But super-antagonistic drug combinations can select for drug sensitivity. This has important application not only for antibacterial therapy but also for cancer therapy: to control cancer with lesser side effects and to eliminate drug-resistant cancer cells, while sparing sensitive normal cells.

Cell cycle arrest in response to DNA damage involves protein stabilization and consequent upregulation of p53, which induces transcription of cyclin-dependent kinase inhibitor p21 (CDKN1A). We now show that p21 acts as a negative regulator of the cellular levels of p53. p21 knockdown by short hairpin RNA strongly increased p53 upregulation by a DNA-damaging drug doxorubicin in HT1080 fibrosarcoma cells. A protease inhibitor N-Ac-Leu-Leu-norleucinal (ALLN) drastically increased the amount of p53 in HCT116 colon carcinoma cells, but it had no effect on the already high p53 level in a p21(-/-) derivative of this cell line. Inhibition of transcription, which increases p53 levels in different cell lines due to the degradation of p53-destabilizing proteins such as Mdm2, failed to increase but instead decreased the amount of p53 in p21(-/-) cells, despite a drastic decrease in the level of Mdm2. These results indicate that p21 acts as a negative regulator of p53 stability in different cell types. p53 regulation by p21 may provide a negative regulatory loop that limits p53 induction.

Newton did not discover that apples fall: the information was available prior to his gravitational hypothesis. Hypotheses can be tested not only by performing experiments but also by retrieving experiments from the literature (via PubMed, for example). Here I show how disconnected facts from known data, if properly connected, can generate novel predictions testable in turn by other published data. With examples from cell cycle, aging, cancer and other fields of biology and medicine, I discuss how new knowledge was and will be derived from old information. Millions of experiments have been already performed to test unrelated hypotheses and the results of those experiments are available to 'test' your hypotheses too. But most data (99% by some estimates) remain unpublished, because they were negative, seemed of low priority, or did not fit the story. Yet for other investigators those data may be valuable. The well-known story of Franklin and Watson is a case in point. By making preliminary data widely available, 'data-owners' will benefit most, receiving the credit for otherwise unused results. If posted (pre-published) on searchable databases, these data may fuel thousands of projects without the need for repetitive experiments. Enormous 'pre-published' databases coupled with Google-like search engines can change the structure of scientific research, and shrinking funding will make this inevitable.

Numerous mutations increase lifespan in diverse organisms from worms to mammals. Most genes that affect longevity encode components of the target of rapamycin (TOR) pathway, thus revealing potential targets for pharmacological intervention. I propose that one target, TOR itself, stands out, simply because its inhibitor (rapamycin) is a non-toxic, well-tolerated drug that is suitable for everyday oral administration. Preclinical and clinical data indicate that rapamycin is a promising drug for age-related diseases and seems to have anti-tumor, bone-sparing and calorie-restriction-mimicking 'side-effects'. I also discuss other potential anti-aging agents (calorie restriction, metformin, resveratrol and sirtuins) and their targets, interference with the TOR pathway and combination with antioxidants.

Here I discuss how the same mechanism--namely, inhibition of transcription during mitosis--can explain (1) apoptosis during mitotic arrest, (2) mitotic slippage, (3) G1 arrest after mitotic slippage and (4) secondary apoptosis after mitotic slippage. In fact, during mitotic arrest transcription is absent, thus depleting those short-lived proteins that have short-lived RNAs. Depletion of anti-apoptotic proteins (IAP, Mcl-1) can trigger apoptosis during prolonged mitotic arrest (apoptosis I). On the other hand, simultaneous depletion of cyclin B allows a cell to exit mitosis without division (mitotic slippage), thus preventing apoptosis I. Depletion of Mdm-2 causes accumulation of p53 during mitotic arrest. If a cell exits mitosis, transcription resumes. In turn the accumulated p53 induces p21 and Bax (and other pro-apoptotic proteins). In turn p21 can cause G1 arrest, whereas Bax can cause apoptosis II. Also, I discuss that mitotic depletion of short-lived proteins collaborates with mitotic phosphorylation of p53, Bcl-2 and BclxL. Finally this article clarifies notions of apoptosis-prone and -reluctant cells and mitotic catastrophe.

Withdrawal of mitogens (growth factors) arrests normal cells in G0 (quiescence). All other stresses and factors arrest cell cycle beyond the restriction point in G1 and G2 (non-G0 arrest), in the presence of mitogenic stimulation. Strong mitogenic stimuli by themselves cause non-G0 arrest. Unlike G0, arrest beyond restriction point is characterized by both high levels of cyclins and CDK inhibitors, activated mitogenic pathways with a secondary GF resistance, and continuous mass growth (cell hypertrophy). Prolonged hypertrophic arrest culminates in cell senescence. This review discusses that quiescence and senescence are two opposite, mutually exclusive conditions and that cell senescence can be reversed and prevented., ((c) 2006 Wiley-Liss, Inc.)

While ruling out programmed aging, evolutionary theory predicts a quasi-program for aging, a continuation of the developmental program that is not turned off, is constantly on, becoming hyper-functional and damaging, causing diseases of aging. Could it be switched off pharmacologically? This would require identification of a molecular target involved in cell senescence, organism aging and diseases of aging. Notably, cell senescence is associated with activation of the TOR (target of rapamycin) nutrient- and mitogen-sensing pathway, which promotes cell growth, even though cell cycle is blocked. Is TOR involved in organism aging? In fact, in yeast (where the cell is the organism), caloric restriction, rapamycin and mutations that inhibit TOR all slow down aging. In animals from worms to mammals caloric restrictions, life-extending agents, and numerous mutations that increase longevity all converge on the TOR pathway. And, in humans, cell hypertrophy, hyper-function and hyperplasia, typically associated with activation of TOR, contribute to diseases of aging. Theoretical and clinical considerations suggest that rapamycin may be effective against atherosclerosis, hypertension and hyper-coagulation (thus, preventing myocardial infarction and stroke), osteoporosis, cancer, autoimmune diseases and arthritis, obesity, diabetes, macula-degeneration, Alzheimer's and Parkinson's diseases. Finally, I discuss that extended life span will reveal new causes for aging (e.g., ROS, 'wear and tear', Hayflick limit, stem cell exhaustion) that play a limited role now, when quasi-programmed senescence kills us first.

Selective modulation of cell death is important for rational chemotherapy. By depleting Hsp90-client oncoproteins, geldanamycin (GA) and 17-allylamino-17-demethoxy-GA (17-AAG) (heat-shock protein-90-active drugs) render certain oncoprotein-addictive cancer cells sensitive to chemotherapy. Here we investigated effects of GA and 17-AAG in apoptosis-prone cells such as HL60 and U937. In these cells, doxorubicin (DOX) caused rapid apoptosis, whereas GA-induced heat-shock protein-70 (Hsp70) (a potent inhibitor of apoptosis) and G1 arrest without significant apoptosis. GA blocked caspase activation and apoptosis and delayed cell death caused by DOX. Inhibitors of translation and transcription and siRNA Hsp70 abrogated cytoprotective effects of GA. Also GA failed to protect HL60 cells from cytotoxicity of actinomycin D and flavopiridol (FL), inhibitors of transcription. We next compared cytoprotection by GA-induced Hsp70, caspase inhibitors (Z-VAD-fmk) and cell-cycle arrest. Whereas cell-cycle arrest protected HL60 cells from paclitaxel (PTX) but not from FL and DOX, Z-VAD-fmk prevented FL-induced apoptosis but was less effective against DOX and PTX. Thus, by inducing Hsp70, GA protected apoptosis-prone cells in unique and cell-type selective manner. Since GA does not protect apoptosis-reluctant cancer cells, we envision a therapeutic strategy to decrease side effects of chemotherapy without affecting its therapeutic efficacy.

The anti-cancer agent paclitaxel (PTX) is an effective anti-restenosis agent on drug eluting stents, primarily due to growth inhibition of coronary artery smooth muscle cells (CASMC) across a wide dose range. In this study, we compared the effects of PTX on CASMC to apoptotic-prone HL60 leukemia cells and apoptotic-reluctant A549 lung cancer cells to assess cell survival mechanisms. In comparison to HL60 and A549 cells, CASMC had a shorter mitotic arrest and a lower mitotic index. While CASMC and A549 cells did not become apoptotic and displayed a multi-nucleated phenotype, HL60 cells showed prolonged mitotic arrest followed by apoptosis. CASMC exiting mitosis were arrested in G1 as MN tetraploid cells, with decreased levels of cyclin B1 and PCNA. CASMC remained metabolically active, becoming permanently arrested as evidenced by increased levels of beta-galactosidase activity. These cells did not demonstrate elevated levels of inflammatory markers. Our findings suggest that a weak mitotic checkpoint or inhibited apoptotic cascade, or a combination of both, determine cell survival following PTX treatment. These in vitro findings suggest a mechanism for the cytostatic activity of PTX in CASMC for the inhibition of restenosis.

Degradation of p53 requires transcription, and therefore inhibitors of transcription cause p53 accumulation. When transcription resumes, the p53 that accumulated in turn induces p21. During mitosis, chromosomes are condensed, the nuclear envelope is dissolved and transcription is absent. If a cell stays in mitosis too long, (e.g. mitotic arrest caused by Taxol or nocodazole), then p53 accumulates. This explains how p53 can measure mitotic time and perhaps represents the most fundamental function of p53.

Tumor stem cells are quiescent and, therefore, resistant to therapy, yet harbor the capacity to replenish a tumor after therapy. Therefore, it is tempting to explain all therapeutic failures by the persistence of tumor stem cells. Yet, this explanation is relevant only to initial stages of stem-cell-dependent tumors (such as chronic myeloid leukemia) that, actually, are well controlled by therapy. In advanced cancers that poorly respond to therapy, quiescent tumor stem cells play a negligible role. Instead, proliferating cells determine disease progression, prognosis, therapeutic failures, and resistance to therapy. And therapy fails not because it eliminates only proliferating tumor cells, but because it does not eliminate them. With noticeable exceptions, it is the proliferating cell that should be targeted, whereas resting cancer cells including stem and dormant cells need to be targeted only when they 'wake up'. Finally, I discuss a strategy of selectively killing dominant proliferating clones, including proliferating stem-like and drug-resistant cancer cells, while sparing normal cells.

At first glance, the antiangiogenic drug Avastin (bevacizumab) must paradoxically normalize angiogenesis, in order to potentiate chemotherapy. However, this may be only a part of the story. Here I discuss that the synergy between Avastin and chemotherapy is also consistent with the classic notion that antiangiogenic therapy actually inhibits angiogenesis. It has been previously predicted that inhibition of angiogenesis (action) will induce a reactive resistance (reaction), which is mediated by the HIF-1/VEGF pathway in cancer cells, thus allowing both endothelial and cancer cells to resist therapy. Therefore, inhibitors of the reactive resistance are needed to potentiate anti-angiogenic therapy. In the combination of chemotherapy plus Avastin, it is chemotherapy that is the principal antiangiogenic agent. This role of chemotherapy requires that something should be added to block the reaction. And this is exactly what Avastin does (by blocking VEGF). While chemotherapy inhibits angiogenesis, Avastin abrogates the reactive resistance, sensitizing both endothelial and cancer cells to therapy.

Most cancers do not respond to chemotherapy. Disappointedly, even objective clinical responses to anticancer therapy often do not translate into improvements in overall survival. To explain the response-survival paradox, it has been pointed out that effective therapy is ineffective against cancer stem cells, which replenish the tumor (causing relapse). In contrast, I discuss that, according to this scenario, patient survival will be prolonged at least by the duration of remission. Furthermore, stem-cell-based relapses will be sensitive to the initial therapy, and in theory cancer could be controlled indefinitely. To explain the paradox, I discuss that effective therapy selects for resistance among proliferating cancer cells. Importantly, mechanisms of resistance can be divided into nononcogenic (e.g., drug transporters and mutation in drug-targets) and oncogenic (e.g., apoptosis avoidance and cell cycle dysregulation). Oncogenic resistance is associated with highly aggressive cancer phenotype and, therefore, there is no increase in overall survival. I further suggest that (a) therapeutic response is a prerequisite for successful therapy, (b) resistance can be exploited for therapeutic advantage, (c) each response can be translated into increased survival, and (e) in slow growing tumors, cancer can be stopped without tumor shrinkage. Therapy will control cancer if it can selectively suppress proliferating cancer cells and will improve survival as long as acquired resistance can be exploited.

Most anticancer drugs are teratogens, merely because they target vital cellular functions. Conversely, some plants produce agents that intentionally target embryonic signaling pathways, precisely to cause birth defects if pregnant animals eat such plants. Cyclopamine, a teratogen produced by a flowering plant, inhibits the Hh/Gli pathway, causing developmental defects such as cyclopia (one eye in the middle of the face). In theory, selective teratogens may suppress cancer cells that reactivate embryonic pathways, while sparing most normal cells. I discuss the potential (and limits) of teratogens in cancer therapy, linking diverse topics from morning sickness of pregnancy, embryonic pathways and poisonous plants to the mechanism of action of anticancer teratogens and their combinations with less selective cytotoxic agents.

Mutant p53 is a cancer-specific target for pharmacologic intervention. We show that histone deacetylase inhibitors such as FR901228 and trichostatin A completely depleted mutant p53 in cancer cell lines. This depletion was preceded by induction of p53-regulated transcription. In cells with mutant p53 pretreated with histone deacetylase inhibitors, DNA damage further enhanced the p53 trans-function. Furthermore, histone deacetylase inhibitors were preferentially cytotoxic to cells with mutant p53 rather than to cells lacking wild-type p53. We suggest that, by either restoring or mimicking p53 trans-functions, histone deacetylase inhibitors initiate degradation of mutant p53. Because mutant p53 is highly expressed, a sudden restoration of p53-like functions is highly cytotoxic to cells with mutant p53. In a broader perspective, this shows how selectivity may be achieved by targeting a non-cancer-specific target, such as histone deacetylases, in the presence of a cancer-specific alteration, such as mutant p53.

In the presence of oxygen and iron, hypoxia-inducible factor (HIF-1alpha) is rapidly degraded via the prolyl hydroxylases (PHD)/VHL pathways. Given striking similarities between p53 and HIF-1alpha regulation, we previously suggested that HIF-1 transcriptionally initiates its own degradation and therefore inhibitors of transcription must induce HIF-1alpha. Under normoxia, while inducing p53, inhibitors of transcription did not induce HIF-1alpha. Under hypoxia or low iron (DFX), inhibitors of transcription dramatically super-induced HIF-1alpha. Removal of inhibitors resulted in outburst of the HIF-1-dependent transcription followed by depletion of HIF-1alpha. Although hypoxia/DFX induced PHD3, we excluded the PHD/VHL pathway in the regulation of HIF-1alpha under hypoxia/DFX. The transcription-dependent degradation of HIF-1alpha under hypoxia occurs via the proteasome and is accelerated by protein acetylation. Thus, HIF-1alpha is regulated by two distinct mechanisms. Under normoxia, HIF-1alpha is degraded via the classic PHD/VHL pathway, is expressed at low levels and therefore does not activate the feedback loop. But under hypoxia, HIF-1alpha accumulates and transcriptionally activates its own degradation that is independent from the PHD/VHL pathway.

The mutation theory of cancer was always confronted by alternative (vitalistic) theories, which insist that cancer (like life itself) cannot be reduced to molecular interactions. In fact, the most fundamental feature of the somatic mutation theory of cancer is that it is a molecular theory, meaning that all the complexity of cancer on any level (e.g., tissue) can be explained on the molecular level. To emphasize the essence of mutation theory, cancer-causing mutation can be defined as any (a) molecular event that is (b) somatically inheritable and (c) selectable (e.g., provides selective advantage in restrictive/carcinogenic conditions). Here I review molecular (somatic mutation) theory and its alternatives and discuss that molecular interactions can completely explain complex tissue phenomena such as benign tumors and stroma initiated tumorigenesis. In addition, molecular theory predicts extragenetic somatic hereditary in cancer (e.g., posttranslational protein modifications that initiate and are supported by positive feedback loops) and also explains the relationship between selection for resistance, hallmarks of cancer and genetic instability. From molecules to cells to the organism, this review discusses how somatically heritable molecular alterations (genetic, epigenetic and extragenetic) alter translation of cellular signals, resulting in resistance to growth inhibition and apoptosis, that is manifested as secondary hallmarks of cancer (metastasis, angiogenesis and immortality) and, finally, as the amazing ability of some cancer cells such as canine transmissible sarcoma to 'live in a wild' like unicellular mammalian species.

Carcinogenesis and cancer therapy are two sides of the same coin, such that the same cytotoxic agent can cause cancer and be used to treat cancer. This review links carcinogenesis, chemoprevention and cancer therapy in one process driven by cytotoxic agents (carcinoagents) that select either for or against cells with oncogenic alterations. By unifying therapy and cancer promotion and by distinguishing nononcogenic and oncogenic mechanisms of resistance, I discuss anticancer- and chemopreventive agent-induced carcinogenesis and tumor progression and, vice versa, carcinogens as anticancer drugs, anticancer drugs as chemopreventive agents and exploiting oncogene-addiction and drug resistance for chemoprevention and cancer therapy.

Chemotherapy of cancer is limited by toxicity to normal cells. Drug resistance further limits the therapy. Here, we investigated selective killing of drug-resistant cancer cells by antagonistic drug combinations, which can spare (because of drug antagonism) normal cells. We used paired cell lines that are resistant to Adriamycin due to either expression of MRP1 or lack of G2 checkpoints. The goal was to selectively kill Adriamycin-resistant cancer cells with Docetaxel (Taxotere), while protecting parental (Adriamycin-sensitive) cells, using cytostatic concentrations of Adriamycin. Taxotere kills cells in mitosis. Therefore, by arresting parental cells in G2, 20 to 40 ng/mL of Adriamycin prevented cell death caused by Taxotere. Also, Adriamycin prevented the effects of Taxotere in normal human lymphocytes. In contrast, Taxotere selectively killed MRP1-expressing leukemia cells, which did not undergo G2 arrest in the presence of Adriamycin. Also, in the presence of Adriamycin, HCT116-p21-/- cancer cells with a defective G2 checkpoint entered mitosis and were selectively killed by Taxotere. Finally, 20 ng/mL of Adriamycin protected normal FDC-P1 hematopoietic cells from Taxotere. Whereas parental cells were protected by Adriamycin, the mitogen-activated protein/extracellular signal-regulated kinase inhibitor PD90598 potentiated the cytotoxic effect of Taxotere selectively in Raf-1-transformed FDC-P1 leukemia cells. We propose a therapeutic strategy to prevent normal cells from entering mitosis while increasing apoptosis selectively in mitotic cancer cells.

Remarkably, a cancer cell rarely possesses two mutant p53 proteins. Instead, mutation of one allele is usually associated with loss of the second p53 allele. Why do not two mutant p53 co-exist? We hypothesize that two different p53 may complement each other, when expressed at equal levels. By titrating trans-deficient and DNA-binding-deficient p53 in cells with mutant p53 and by co-transfecting distinct mutant p53 in p53-null cells, we demonstrated activation of p53-dependent transcription. We suggest that, due to complementation of two mutant p53, cancer cells need to delete the second p53 allele rather than mutate it.

By activating anti-apoptotic factors (e.g., Hsp70, Raf-1, Bcl-xL), Bcr-Abl blocks apoptotic pathways at multiple levels, thus rendering leukemia cells resistant to chemotherapeutic agents such as doxorubicin (DOX). In Bcr-Abl-transfected HL60 (HL/Bcr-Abl) cells, procaspase-9 was increased and partially processed. The Bcr-Abl inhibitor imatinib (Gleevec, STI-571) released the apoptotic stream. Also, HL/Bcr-Abl cells were hyper-sensitive to geldanamycin (GA), which depletes Bcr-Abl and Raf-1. Raf-1 and Bcr-Abl-transfected FDC-P1 hematopoietic cells were selectively sensitive to GA and imatinib, respectively. Remarkably, cell clones with high levels of Bcr-Abl that could not be depleted by GA were relatively resistant to both GA and imatinib. GA and flavopiridol sensitized such resistant cells to imatinib. These data suggest bi-phasic sensitivity to mechanism-based therapeutic agents. Although Bcr-Abl renders cells hyper-sensitive, an excess of Bcr-Abl results in resistance (due to the remaining activity). We discuss therapeutic approaches to overcome bi-phasic resistance to mechanisms-based agents.

As announced by Andrew von Eschenbach, the NCI has set the goal of eliminating suffering and death due to cancer by 2015. Supporting this prediction, I discuss that cancer might be controlled and even cured by combining three potential therapeutic strategies aimed at (i) cancer-specific targets, (ii) universally-vital targets with selective protection of normal cells (the selective combinations) and (iii) tissue-specific targets. Although (i) targeting cancer-specific pathways (e.g., by imatinib and gefitinib) is probable, it alone will not be sufficient to control cancer. This strategy is limited to oncogene (kinase)-dependent cancers and is further limited by therapy-induced resistance and tumor progression. Thus, targeting cancer-specific pathways needs to be complemented by two divergent therapeutic strategies: (ii) selective combinations and (iii) tissue-selective therapy. With selective protection of normal cells (based on cell cycle and apoptosis manipulation), combinations of selective and chemotherapeutic drugs can be effective in most common cancers. Alternatively, tissue-selective therapy can suppress cancer cells in a tissue-selective manner, sparing other tissues. While alone, each therapeutic strategy may cause drug resistance and even tumor progression; these obstacles can be overcome and even exploited by using all three strategies in sequence. And finally, these strategies will benefit from molecular diagnostics and can be used for chemoprevention.

During a billion years of evolution, living creatures have perfected cytotoxic agents to kill other organisms without killing themselves, thus providing us with antibiotics to kill bacteria without killing eukaryotic (e.g. human) cells. Some natural agents inhibit specifically most vital cellular structures and functions in cancer cells. However, nature was not creating antibiotics for cancer, and natural agents kill cancer cells precisely because they share targets with normal cells. To discriminate between particular cancer cells and normal cells, we can design or select artificial agents that are not necessarily lethal but are aimed either at cancer-specific targets or at dispensable and even unavailable (in cancer cells) targets. Using rational drug combinations, such selective agents can assist natural agents to eradicate cancer cells selectively.

The tumor suppressor protein p53 is emerging as a central regulator of homologous recombination (HR) processes and DNA replication. P53 may downregulate HR through multiple mechanisms including the reported associations with the Rad51 and Rad54 recombinases, and the BLM and WRN helicases. Here, we investigated whether the interaction of p53 with human replication protein A (RPA) is necessary for the regulation of HR. By employing a plasmid-based HR assay in p53-null H1299 lung carcinoma cells, we studied the HR-suppressing properties of a panel of p53 mutants, which varied in their ability to interact with RPA. Both wild-type p53 and a transactivation-deficient p53 mutant (L22Q/W23S) suppressed HR and prevented RPA binding to ssDNA in vitro and in vivo. Conversely, p53 mutations that specifically disrupt the RPA-binding domain, while not compromising p53 transactivation function (D48H/D49H and W53S/F54S), did not affect HR. Suppression of HR was also not seen with missense mutations in the p53 core domain (His175 and His273), which retained the ability to interact with RPA, suggesting that the disruption of additional binding interactions of p53, for example, with Rad51 or recombination intermediates, also impacts on HR. We hypothesize that sequestration of RPA by p53 at the sites of recombination is one means by which p53 can inhibit HR processes. Our data support and extend the previously formulated 'dual model' of p53's role as guardian of the genome.

After a decade of exciting promises, the CDK inhibitor flavopiridol has quietly failed in most clinical trials. This review discusses that flavopiridol is a potent inhibitor of global transcription. This explains not only downregulation of numerous proteins, cell cycle arrest and apoptosis but also all pleiotropic and mysterious effects of flavopiridol. Yet, flavopiridol is not just a second actinomycin D. As an inhibitor of transcription with a unique mechanism of action, flavopiridol may have tremendous clinical potentials. This article reviews the molecular and cellular effects of flavopiridol as well as mechanisms of therapeutic and side effects, suggesting its novel clinical applications as a single agent and in drug combinations.

This review discusses 26 new anticancer drugs approved by the FDA in the past decade. Based on their targets, these anticancer agents can be divided into three groups. First group contains cancer-selective or semi-selective drugs that are effective in rare kinase- addictive cancers. For other malignancies, semi-selective drugs have to be judiciously combined with nonselective agents. The second group includes analogs of classic cytotoxic agents such as DNA alkylating agents, nucleoside analogs, and anti-microtubule agents. As expected, they have a marginal advantage over the existing cytotoxic drugs, nevertheless are more effective (in common cancers) than semi-selective agents. The third is a diverse group of tissue-selective agents that essentially attack the normal tissues of tumor origin and thus exploit the tissue-specific similarities between normal and cancer cells. Our analysis predicts that monotherapy with semi-selective agents will be limited to rare cancers. In most cancers, however, two anticancer strategies may be most fruitful: (a) combinations of cytotoxic drugs with semi-selective agents aimed at matching targets and (b) tissue-selective therapy aimed at normal and tumor cells of the same tissues.

Paclitaxel (PTX), a microtubule-active drug, causes mitotic arrest leading to apoptosis in certain tumor cell lines. Here we investigated the effects of PTX on human arterial smooth muscle cell (SMC) cells. In SMC, PTX caused both (a) primary arrest in G(1) and (b) post-mitotic arrest in G(1). Post-mitotic cells were multinucleated (MN) with either 2C (near-diploid) or 4C (tetraploid) DNA content. At PTX concentrations above 12 ng/ml, MN cells had 4C DNA content consistent with the lack of cytokinesis during abortive mitosis. Treatment with 6-12 ng/ml PTX yielded MN cells with 2C DNA content. Finally, 1-6 ng/ml of PTX, the lowest concentrations that affected cell proliferation, caused G(1) arrest without multinucleation. It is important that PTX did not cause apoptosis in SMC. The absence of apoptosis could be explained by mitotic exit and G(1) arrest as well as by low constitutive levels of caspase expression and by p53 and p21 induction. Thus, following transient mitotic arrest, SMC exit mitosis to form MN cells. These post-mitotic cells were subsequently arrested in G(1) but maintained normal elongated morphology and were viable for at least 21 days. We conclude that in SMC PTX causes post-mitotic cell cycle arrest rather than cell death.

Tyrosine phosphorylation of paxillin regulates actin cytoskeleton-dependent changes in cell morphology and motility in adherent cells. In this report we investigated the involvement of paxillin tyrosine phosphorylation in the regulation of actin cytoskeleton-dependent polarization and motility of a non-adherent IL-3-dependent murine pre-B lymphocytic cell line Baf3. We also assessed the effect of phorbol myristate acetate (PMA), a phorbol ester analogous to those currently in clinical trials for the treatment of leukemia, on paxillin phosphorylation. Using tyrosine-to-phenylalanine phosphorylation mutants of paxillin and phosphospecific antibody we demonstrated that IL-3 stimulated phosphorylation of paxillin tyrosine residues 31 and 118, whereas the tyrosines 40 and 181 were constitutively phosphorylated. Phosphorylation of paxillin residues 31 and 118 was required for cell polarization and motility. In the presence of IL-3, PMA dramatically reduced the phosphorylation of residues 31 and 118, which was accompanied by inhibition of cell polarization and motility. This PMA effect was partially recapitulated by expression of exogenous tyrosine 31 and 118 mutants of paxillin. We also demonstrated that PMA inhibited the IL-3-induced and activation-dependent tyrosine phosphorylation of focal adhesion kinase. Thus, our results indicate that phosphorylation of paxillin tyrosine residues 31 and 118 regulates actin-dependent polarization and motility of pre-B Baf3 cells, both of which could be inhibited by PMA. They also suggest that inhibition of upstream signaling by PMA contributes to the decrease of paxillin phosphorylation and subsequent changes in cell morphology.

Flavopiridol (FP) inhibits gene expression and causes apoptosis, and these effects cannot be explained by inhibition of cyclin-dependent kinases that govern cell cycle. The simple and established notion that FP is an inhibitor of transcription predicts its effects. Because Mdm-2 targets p53 for degradation, FP, as predicted, dramatically induced p53 by inhibiting Mdm-2. Once p53 was induced, restoration of transcription (by removal of FP) resulted in superinduction of p21 and Mdm-2. Similarly, low concentrations of FP (50 nm) induced p21 and Mdm-2 because of their initial down-regulation. A sustained decrease of Mdm-2/p21 expression and accumulation of p53 coincided with near-maximal cytotoxicity of FP at concentrations >100 nm. Induction of p53 was a marker, not a cause, of cytotoxicity. FP caused rapid apoptosis (caspase-dependent cell death) in p53-null leukemia cells. In these cells, FP-induced apoptosis was converted to growth arrest by inhibitors of caspases. In apoptosis-reluctant A549 and PC3M cancer cells, FP inhibited cell proliferation but did not cause apoptosis. Like typical inhibitors of transcription, FP sensitized cells to apoptotic stimuli, allowing tumor necrosis factor to cause rapid and massive apoptosis in otherwise apoptosis-reluctant cells. We discuss that, as a reversible inhibitor of transcription, FP can be used clinically in novel rational drug combinations.

Activating mutations in the epidermal growth factor receptor (EGF-R) predict response to gefitinib. How does this recent discovery affect our outlook on selective (targeted) cancer therapy? It allows us to compare mutant EGF-R with Bcr-Abl as anticancer drug targets and to discuss the nature of oncogene addiction. It emphasizes molecular diagnostics to identify oncogene-addictive cancers. It also re-enforces the notion that most cancers with multiple oncogenic alterations (common cancers) will unlikely respond to selective drugs alone. In such cancers, one strategy is targeting cancer-non-specific, universal and vital structures, essential for life of all cells: microtubules, topoisomerases, histone deacetylases, the proteasome. But in order to be cancer-selective, these chemotherapeutic agents need to be combined with selective agents. Such combinations can be effective and selective in common cancers.

Although induction of apoptosis (cell death mediated by caspases) determines response to cancer therapy, this approach is limited by lack of selectivity in available apoptosis-inducing agents. Furthermore, most cancers, almost by definition, are resistant to apoptosis, growth arrest and cell senescence. Then, how can anticancer agents kill cancer cell without unacceptable toxicity to a patient? The potential therapeutic approaches range from selective inhibition of antiapoptotic pathways, antiangiogenic therapy, tissue-selective therapy (including immunotherapy) to exploitation of, for example, drug resistance, oncoprotein addiction, unrestricted cell cycles, hypermitogenic and hypoxic features of cancer cells. These overlapping and complementary approaches rely on rational drug combinations (at mechanism-based doses and sequences) aimed at matching targets. To ensure killing of cancer cells selectively, we may combine apoptosis- and senescence-inducing agents with inhibitors of apoptosis (to protect normal cells), inhibitors of signal transduction with cell cycle-dependent chemotherapy, antiangiogenic agents with hypoxia-inducible factor-1 inhibitors, tissue-selective therapy with differentiating agents and activators of death receptors with chemotherapy. In theory, consecutive use of these drug combinations may control cancer.

Angiogenesis is necessary for tumor growth (a rationale for antiangiogenic therapy), but hypoxia caused by such a therapy will, in theory, drive tumor progression and metastasis. To reconcile conflicting notions, we discuss that, first, although a shift from normoxia (21% O2) to hypoxia indeed activates cancer cells for aggressive behavior, this may not occur during therapy, because most cancers are not normoxic to start with. Second, only successful antiangiogenic therapy, which is capable of controlling cancer, will select for resistance and progression. After all, in order to occur, therapy-induced tumor progression must be preceded by tumor regression.

Instead of exploiting the differences between normal and cancer cells, seemingly unrelated anticancer modalities (from immunotherapy to hormones) exploit (a). the differences between various normal tissues and (b). tissue-specific similarities of normal and cancer cells. Although these therapies are successfully used for years to treat leukaemia and cancer, their unifying principles have never been explicitly formulated: namely, they are aimed at differentiated cells and normal tissues and target both normal and cancer cells in a tissue-specific manner. Whereas tiny differences between cancer and normal cells have yet to be successfully exploited for selective anticancer therapy, numerous tissue-specific differences (e.g. differences between melanocytes, prostate, thyroid and breast cells) provide a means to attack selectively that exact tissue that produced cancer. Despite inherent limitations, such as fostering resistance and dedifferentiation, tissue-selective therapy have enormous potentials to control cancer.

In growth-limiting conditions, cells that express telomerase and inactivate tumor suppressors have a selective advantage due to resistance to growth arrest. Accidentally such cells become immortal.

Cancer cells are intrinsically resistant to growth arrest and can further acquire multidrug resistance. Current approaches to this problem are intended to reverse, overcome or prevent the drug resistance. However, the resistance of cancer cells can be exploited to kill resistant cells selectively, while sparing sensitive normal cells. As the simplest example, multidrug-resistant cells pump out protectors, such as pharmacological inhibitors of apoptosis. A sequence of at least two agents must include an exclusive protector, which is ineffective in resistant cancer cells, and an inclusive cytotoxic drug, which kills unprotected cells. By abolishing several dose-limiting side effects of chemotherapy, this strategy provides a means to treat selectively most deranged, aggressive and resistant cancers.

A diverse range of conditions, from mitogenic stimuli to cytotoxic stress, can induce cell senescence. Here, I propose that simultaneous stimulation of mitogen-activated pathways and downstream inhibition of cyclin-dependent kinases leads, ultimately, to cell senescence. This model distinguishes between two types of growth arrest: first, exit to G0 phase, which is caused by the withdrawal of mitogens and can lead to apoptosis; and second, hypermitogenic arrest, which is stimulated by mitogens and can lead to senescence. The concept of hypermitogenic arrest defines cell senescence as a functionally active, stable and conditionally reversible state.

In search of the order, we are tempted to universally link cell death, proliferation, differentiation, and senescence. Current models (classical, conflicting signal and quantitative signal models) are restricted, precisely because they attempt to hardware a plethora of end-points of cellular responses. By defining each cellular process in molecular term, one can disconnect proliferation (CDK activation), apoptosis (caspase activation), and differentiation (tissue function genes expression), even though these responses are linked by upstream signal transduction pathways. These ambivalent pathways (e.g. mitogen-activated pathways) simultaneously transduce opposite signals (for growth arrest and cycling, for cell death and survival), which are ultimately translated in all possible combinations of cellular responses. When depicted in multidimensional axis, this universal model may also include invasiveness, senescence, metastatic and angiogenic responses and even such integral characteristics as malignant transformation.

Overexpression of Bcl-2 plays a role in the development of drug resistance in leukemia and other apoptosis-prone tumors. Raf isoforms areserine/threonine kinases that act as signal transducers in cascades initiated by many growth factors and mitogens. Raf isoform activation has been linked to drug resistance in leukemia. In this study we investigated effects of Bcl-2 and Raf-1 on doxorubicin-induced growth inhibition of MCF-7 breast cancer cells. In the absence of doxorubicin, overexpression of Bcl-2 or a constitutively active form of Raf-1 in MCF-7 cells did not affect proliferation rate. Overexpression of Bcl-2 increased resistance of MCF-7 cells to doxorubicin in 2-day, 5-day, and 8-week assays. Analysis of doxorubicin sensitivity of individual MCF/Bcl-2 clones showed that doxorubicin resistance was positively correlated with level of Bcl-2 overexpression. Overexpression of constitutively active Raf-1 also increased resistance to doxorubicin. Induction of Raf-1 activity in MCF-7 cells overexpressing Bcl-2 resulted in greater doxorubicin resistance than induction of Raf-1 activity in MCF-7 cells lacking Bcl-2 overexpression. Furthermore, levels of P-glycoprotein mRNA were increased in MCF-7 cells overexpressing a constitutively active Raf-1. MCF-7 cells overexpressing constitutively active Raf-1 were also more resistant to paclitaxel, which, like doxorubicin, is a substrate of P-glycoprotein. These observations suggest both independent and overlapping roles for Raf-1 and Bcl-2 oncogenes in the resistance to growth inhibition by doxorubicin.

Studies into the molecular mechanisms of cancer have revealed that, with a few exceptions, the disease lacks a specific drug target. Therefore, new anticancer drugs not only take many years and much money to develop but also might not outperform existing drugs. However, what are assumed to be unfavorable hallmarks of cancer, for example drug resistance, can be exploited for selective anticancer therapy and for protection of normal cells. Based on this paradigm, the drug discovery can be complemented by novel use of existing agents and even "failed" drugs. The pharmacological industry could develop low cost, effective therapeutic modalities, by "re-using" already marketed and late-stage products in cancer-selective therapeutic kits.

A phase III failure of Iressa, an inhibitor of the EGF receptor (EGFR), is viewed as a surprise. With a few exceptions, however, inhibitors of EGFR cannot be effective as a monotherapy in cancers with additional oncogenic changes (downstream of EGFR), which cause mitogen-independent proliferation. In other cases, combining these inhibitors with chemotherapy may lead to antagonism in cancer cells and/or aggravated side-effects. This review discusses why neither levels of EGFR nor doses of Iressa correlate with clinical response. We suggest a novel use of signal transduction inhibitors, including Iressa, to increase therapeutic index by modulating cycle-dependent and apoptosis-inducing chemotherapies. This approach may be most beneficial to patients who do not respond to monotherapy with kinase inhibitors. Development of molecular diagnostics will further diversify these strategies.