Search

Apoptosis occurs by distinct pathways that involve the cell surface, mitochondria or the endoplasmic reticulum. Previous studies had shown that deoxyadenosine-resistant L1210 cells (Y8) proceeded to apoptosis under conditions in which the parental L1210 cell line (WT) did not undergo an apoptotic response. Combinations of drugs, acting at different molecular targets, markedly potentiated the apoptotic response in the Y8 cells without inducing apoptosis in the WT cells. In the present study, induction of apoptosis by parthenolide and BAY 11-7085, drugs that targeted nuclear factor kappa B activation, was blocked by the presence of N-acetylcysteine (NAC). On the other hand, the levels of apoptosis induced by parthenolide or BAY 11-7085 were increased by pre-treatment of the cells with glutathione lowering L-buthionine-(S,R)-sulfoximine (BSO). Western blot analyses showed that the levels of the stress proteins, Grp 78 and Gadd 153 were reduced in the parthenolide-treated Y8 cells, but not in those co-treated with NAC. Protection of the cells from apoptosis induced by parthenolide or BAY 11-7085 by NAC was relatively specific as the induction of apoptosis in the Y8 cells by MG-132, flavopiridol, Gemcitabine or PRIMA-1 was not decreased by NAC. These data suggest that multiple pathways, one of which is ER-stress induced, may ultimately be involved and interactive in the induction of apoptosis in specific cell lines.

It has become increasingly clear that the induction of apoptosis in tumor cells can occur by at least three different pathways involving the cell surface receptors, the mitochondria and the endoplasmic reticulum. Specific drugs and conditions can trigger the apoptotic response via one of the three known pathways. What is less clear is how these three pathways can interact synergistically or antagonistically to influence a common convergence step leading to the programmed cell death. In this report we present data to show that fenretinide (a synthetic retinoid) potentiates the apoptotic effects of parthenolide (a drug that inhibits the activation of NF-kappa B) and BAY 11-7085 (an inhibitor of I-kappa B-alpha kinase). This potentiation of apoptosis by fenretinide is seen in the p53-deficient, deoxyadenosine-resistant L1210 cells, but not in the parental L1210 cells that express a mutant p53. These effects are seen at a concentration of fenretinide that have little effect by itself. These data strongly suggest that fenretinide activates or inhibits some step or pathway that interacts with the inhibition of NF-kappa B activation required for the apoptotic response. Since parthenolide, BAY 11-7085 and fenretinide are well known drugs in clinical trials, an understanding of the nature of the interactions between or among the apoptotic pathways could lead to the design of better clinical protocols using these drugs that will promote apoptosis in tumor cells.

Bis(2-hydroxybenzylidene)acetone is a potent inducer of the phase 2 response through the Keap1-Nrf2-ARE pathway. This double Michael reaction acceptor reacts directly with Keap1, the sensor protein for inducers, leading to enhanced transcription of phase 2 genes and protection against oxidant and electrophile toxicities. In our efforts to identify potent chemoprotective agents, we found that in rapidly growing murine leukemia cells (L1210) low concentrations (in the submicromolar range) of bis(2-hydroxybenzylidene)acetone markedly increased the activities of NAD(P)H:quinone acceptor oxidoreductase 1 (NQO1) and glutathione reductase, and the levels of total glutathione, three markers of the phase 2 response. In contrast, at high concentrations (in the micromolar range) the same compound caused G2/M cell cycle arrest and apoptosis. Importantly, a mutant L1210 cell line (Y8), selected for resistance to deoxyadenosine and lacking expression of p53 protein, was considerably more sensitive to the apoptotic effects of bis(2-hydroxybenzylidene)acetone. When caspase activities were evaluated in cell-free extracts prepared from treated wild type or mutant L1210 cells, the activities of caspase-3, the terminal caspase in the cascade leading to apoptosis, and caspase-10 were found to be markedly elevated. The activities of other caspases measured, caspase-1, -6 and -8, were not appreciably affected. Thus, both induction of the phase 2 response and p53-independent, caspase-3-mediated apoptosis could act cooperatively in chemoprotection. The concentration-dependent differential effects on these two pathways should be carefully considered in mechanistic explanations and strategic designs.

Asparaginase is a key component of the chemotherapy protocols used in the treatment of acute lymphoblastic leukemia (ALL). The current treatment protocols are remarkable in that childhood ALL cure rates are approaching 85%. As the name implies, asparaginase catalyzes the deamination of asparagine to aspartic acid. What is not generally realized is that asparaginase also catalyzes, essentially to the same extent, the removal of the amide nitrogen from glutamine to form glutamic acid. Glutamine is a required substrate for three enzymes involved in the de novo synthesis of purine nucleotides and two enzymes involved in the de novo synthesis of pyrimidine nucleotides. In this review, the specific roles of glutamine in the de novo synthesis of nucleotides are defined and an appropriate explanation for the cell cycle arrest and cytotoxicity induced in proliferating malignant lymphoblasts by asparaginase treatment is provided.

The effects of PRIMA-1 on wild-type (WT) mouse leukemia L1210 cells and drug-resistant L1210 cells (Y8) were studied with respect to the induction of apoptosis and necrosis in these cell lines. The WT L1210 cells express mutant p53 while the Y8 L1210 cells do not express p53 mRNA or protein, but do express WAF1/p21 and Gadd 45 mRNA's and proteins. It was found that, in response to treatment with PRIMA-1, the WT L1210 cells became necrotic with little apoptosis while the Y8 L1210 cells showed a much higher level of apoptosis than necrosis. Flavopiridol in combination with PRIMA-1 caused a synergistic increase in necrosis in the WT L1210 cells while LY 294002 in combination with PRIMA-1 caused a synergistic increase in apoptosis in the Y8 L1210 cells. These studies showed that PRIMA-1 had an effect not only on cells expressing mutant p53, but also on cells that do not express p53, suggesting that PRIMA-1 and PRIMA-1-like molecules have multiple sites of action independent of restoring p53 function and that these can interact with other signaling pathways involving CDK's and PI3 kinases.

A mouse leukemia L1210 cell line (Y8), selected for resistance to deoxyadenosine, has a markedly altered phenotypic expression that includes loss of sensitivity to dATP as an allosteric inhibitor of ribonucleotide reductase, increased expression of c-myc, c-fos and WAF1/p21, but decreased expression of p53. In addition, the Y8 cells have a Very strong apoptotic response to a variety of agents under conditions in which the parental wild-type cells do not apoptose. In these studies, we show that flavopiridol (a cdk inhibitor) causes the Y8 cells to undergo apoptosis via a caspase-3 activation process. The apoptotic response to flavopiridol is markedly enhanced by LY294002. Data also show that the apoptotic response of the Y8 cells to roscovitine (a cdk inhibitor) is enhanced by UCN-01 (a PKC inhibitor). These data show that simultaneous blockage of specific pathways leads to increased apoptosis in the Y8 cells with essentially no effects on the parental wild-type L1210 cells.

The effects of a series of phenolic compounds were compared to the effects of sodium salicylate (2-hydroxybenzoate) on the growth, cell cycle and apoptotic effects in wild-type (WT) and deoxyadenosine-resistant (Y8) L1210 leukemia cells. These compounds included: salicylaldehyde, salicylaldoxime, salicylhydroxamic acid, salicylamide, 5-aminosalicylate and 5-sulfosalicylate. The IC50 values for inhibition of tumor cell growth ranged from 40 microM for salicylaldhyde to greater than 4 mM for 5-sulfosalicylate. There appeared to be an excellent correlation between the IC50 value for a compound and the ratio of octanol/aqueous distribution. Salicylamide caused a G2/M block in both the WT and Y8 L1210 cells, while salicylalehyde caused a G0/G1 block in both the WT and Y8 cells. Salicylamide and salicylaldoxime caused a much greater apoptotic effect in the Y8 cells than in the parental WT L210 cells. These data suggest that salicylaldehyde and salicylaldoxime, the most active compounds in this series, may provide the lead chemicals from which other more active drugs can be synthesized.

The effects of Bay 11-7085, an inhibitor of I-kappa-B-alpha kinase, were compared in the wild-type and deoxyadenosine-resistant mouse leukemia cell lines. At higher concentrations, Bay 11-7085 caused apoptosis and necrosis in the two cell lines. However, at much lower concentrations of Bay 11-7085, the deoxyadenosine-resistant cells became much more apoptotic than the parental wild-type L1210 cells. Under these conditions (low drug concentrations), the level of apoptotic cells far exceeded the fraction of necrotic cells. The apoptotic effects of Bay 11-7085 on the deoxyadenosine-resistant cells was both time- and concentration-dependent. Caspase-3 activity was activated in the Bay 11-7085-treated cells. The molecular basis for the difference in the apoptotic response between the wild-type and deoxyadenosine-resistant L1210 cells is not defined at this time, but these cell lines may provide a comparative model system in which differences in the cells that lead to apoptotic or necrotic responses to drugs can be defined and used in development of appropriate drugs for clinical use.

An L1210 cell line (Y8) selected for resistance to deoxyadenosine does not express p53 mRNA or protein but expresses WAF1/p21 even under basal conditions. The Y8 cell line had been previously shown to have an increased apoptotic response to a variety of agents that included DNA damaging agents, kinase inhibitors and drugs directed at NFkappa B activation. In this study we show that lactacystin (LC, an inhibitor of proteasome activity) in combination with parthenolide (PA) caused a synergistic increase in the apoptotic fraction of the Y8 cells. LC (2.5 microM) alone and PA (5.0 microM) caused less than 20% of the Y8 cells to undergo apoptosis. However, the combination of LC (2.5 microM) plus PA (5.0 microM) caused 60% of the Y8 cells to undergo apoptosis. The combination of drugs had no effects on the parental wild-type L1210 cells. Pretreatment of the intact Y8 cells with the caspase-3 inhibitor, Ac-DEVD-CHO, resulted in a marked decrease in the apoptosis caused by the LC plus PA combination. Cell-free extracts prepared from the LC plus PA combination-treated cells had activated caspase activities in the caspase cascade: caspase-3 >> caspase-8 > caspase-6 and caspase-10. These results suggest that there are interacting pathways involving aspects of NFkappa B activation and proteasome activity that could be exploited in therapy directed at p53-deficient tumor cells that would lead to caspase-3 activation and apoptosis bypassing the p53-dependent chemotherapy insensitivity.

Mouse leukemia L1210 cells selected for resistance to drugs targeted specifically at each of the protein subunits of ribonucleotide reductase were studied for their ability to grow in vivo. The life-span of the mice injected with hydroxyurea-resistant L1210 cells, which have elevated levels of the mRNA and protein for the non-heme iron (NHI, R2) subunit of ribonucleotide reductase, was approximately twice that of the mice injected with equal numbers of the parental wild-type L1210 leukemia cells. The life-span of mice injected with the L1210 cells that had alterations in the effector-binding subunit (EB, R1) was considerably shorter than the mice injected with the parental wild-type L1210 cells. These results provide direct evidence that tumor cells with alterations in the properties of ribonucleotide reductase grow differently in vivo, with defined effects on the host mouse that cause either an increased survival time or a decreased survival time compared to the effects of wild-type L1210 leukemia cells on tumor-bearing mice.

Two new lipophilic tetraazamacrocycles were prepared and, in contrast to non-lipophilic analogs, found to be potent inhibitors of tumor cell growth in vitro with IC50 values below 10 micromolar.

A mouse leukemia L1210 cell line (Y8) selected for resistance to deoxyadenosine was found to be deficient in the expression of p53 mRNA and protein while maintaining the expression of WAF1/p21 mRNA and protein even under basal conditions. The Y8 cells were shown to be more sensitive to apoptosis induced by a variety of agents when compared to the parental wild-type (WT) L1210 cells. Roscovitine, an inhibitor of cdk 2 and cdk5, was one of the agents that caused increased apoptosis in the Y8 cells through a pathway that ultimately involved the activation of caspase-3 activity. In these studies, the effects of leflunomide and parthenolide (drugs reported to alter the activation of NFkappaB in a variety of cell types) were studied for their cell cycle and apoptotic effects in WT and Y8 cells as single agents and in combination with roscovitine. Leflunomide at IC50 concentrations had little effect on the cell cycle distribution of either the WT or Y8 cells while at higher concentrations caused a G0/G1 block in Y8 cells. Parthenolide, at IC50 concentrations, caused a G0/G1 cell cycle block in the WT and Y8 cells but at higher concentrations caused a G2/M block in the Y8 cells. The combinations of leflunomide and roscovitine or parthenolide and roscovitine did not alter, in a significant way the cell cycle distribution of the Y8 cells. However, in the presence of the combinations of leflunomide and roscovitine or parthenolide and roscovitine there were large increases in the fraction of Y8 cells undergoing early apoptosis without a corresponding increase in the necrotic fraction of cells. These data show that combinations of agents directed at different pathways or different steps of pathways involved in apoptosis can cause the cells to reach an apoptotic threshold that results in synergistic apoptosis.

Mouse leukemia L1210 cells selected for resistance to deoxyadenosine contain ribonucleotide reductase that is not feedback inhibited by dATP. These deoxyadenosine-resistant cells (Y8) also do not express p53 protein but do have WAF1 and Gadd45 mRNA and protein. The Y8 cells show increased sensitivity to DNA damaging agents and kinase inhibitors. In these studies we show that in the presence of sodium salicylate (NaSal), the parental wild-type (WT) cells block in G2/M phase of the cell cycle while the Y8 cells show a marked increased in the G0/G1 population of cells. The Y8 cells are more sensitive to apoptosis induced by NaSal than the WT cells. NaSal treatment causes the induction of caspase-3-like activity in Y8 cells but no induction of caspase-3 activity in the WT cells. The caspase inhibitor, Ac-DEVD-CHO, decreased the percentage of Y8 cells in the early apoptotic fraction, but this decrease was reflected by an increase in the percent of cells in the late apoptotic/necrotic fraction. SB20358, a p38-MAP kinase inhibitor did not protect the Y8 cells from NaSal-induced apoptosis indicating that the p38-MAP kinase pathway was not involved in the NaSal-induced apoptotic pathway in the p53-independent Y8 cells.

An L1210 cell line (Y8) selected for resistance to deoxyadenosine contains ribonucleotide reductase that is not subject to inhibition by dATP. In addition, the Y8 cells have other phenotypic expressions that include increased sensitivity to apoptosis induced by various agents such as radiation, doxorubicin, anisomycin and roscovitine. The Y8 cells were found to be more sensitive to apoptosis induced by methotrexate (MTX), tiazofurin (TZ), deoxyguanosine (dGuo) and N-(phosphonoacetyl)-L-aspartate (PALA). Deoxyguanosine, at concentrations that did not cause apoptosis in the Y8 cells, prevented the apoptotic response of the Y8 cells to MTX and TZ. Deoxycytidine had no effect. Since caspase-3 activation is involved in apoptotic pathways, the effects of the caspase-3 inhibitor, Ac-DEVD-CHO, were studied on the dGuo-, MTX- or TZ-induced apoptosis in the Y8 cells. Ac-DEVD-CHO caused a marked decrease in the fraction of cells in the early phase of apoptosis. However, there was a corresponding increase in the fraction of cells in the late apoptotic/necrotic stages of cell death. This is in marked contrast to the dGuo-induced decrease in apoptosis seen in the MTX- and TZ-treated Y8 cells in which there were no increases in the late apoptotic/necrotic fraction of cells. These data show that alterations of nucleotide pools in the Y8 cells cause marked increases in the apoptotic response which may indicate that the Y8 cells are much more susceptible to the effects of misincorporation of nucleotides into DNA than are the parental WT L1210 cells.

The MCF-7/Adr cells overexpress MDR-1 which contributes to the drug-resistant phenotype. Our studies show: 1. The retention of daunomycin in the MCF-7/Adr cells relates to a temperature-dependent and energy-dependent process. 2. The MCF-7/Adr cells retain less rhodamine-123 than the parental MCF-7 cells. 3. The MCF-7/Adr cells retain less daunomycin than the parental MCF-7 cells as measured by mean fluorescence or radioactive daunomycin. 4. Cyclosporin A and verapamil effectively block the effluxes of rhodamine-123 and daunomycin from the MCF-7/Adr cells. 5. On short-term incubation, 2-deoxyglucose lowers the NTP levels to a greater extent than sodium azide, showing the importance of glycolysis in the MCF-7 cell lines. 6. Although the MCF-7/Adr cells show cross-resistance to VP-16, actinomycin D and vinblastine, these drugs do not compete with daunomycin for the efflux pump. 7. These data suggest that either there must be multiple MDR-1 pumps that differ in substrate specificity or that there are distinct substrate sites on MDR-1.

The enzyme RR catalyzes the conversion of ribonucleoside diphosphates to their deoxyribonucleotide counterparts. RR is critical for the generation of the cytosine, adenine, and guanine deoxyribonucleotide 5'-triphosphate building blocks of DNA, which are present in cells as exceedingly small intracellular pools. Therefore, interference with the function of RR might well result in an agent with significant antineoplastic activity, particularly against rapidly proliferating tumor cells. HUr is the only inhibitor of RR in clinical usage; this agent, however, is a relatively poor inhibitor of the enzyme and has a short serum half-life. Consequently, HUr is a relatively weak anticancer agent. In an effort to develop a more potent inhibitor of RR with utility as an anticancer agent, we have synthesized 3-AP and demonstrated (a) potent inhibition of L1210 leukemia cells in vitro, (b) curative capacity for mice bearing the L1210 leukemia, (c) marked inhibition of RR, and (d) sensitivity of HUr-resistant cells to 3-AP. These findings collectively demonstrate the clinical potential of 3-AP as an antineoplastic agent.

Recent studies have implicated nucleotides in diverse and unexpected functions related to p53 levels, p53-dependent G0/G1 cell cycle arrest, and the role of dATP in the activation of the caspase-induced apoptosis. Using deoxyadenosine-resistant L1210 cells (ED2 and Y8) that had ribonucleotide reductase that was not sensitive to inhibition by dATP and also exhibited other metabolic alterations, the properties of these cells with respect to the role(s) of nucleotides in these functions were explored. In the ED2 and Y8 cells that did not express p53 protein, the pools of UTP, CTP, ATP, and GTP were markedly decreased. The decreased cellular levels of UTP and CTP did not result in these cells being more sensitive to either PALA or acivicin. The ED2 and Y8 cells did not block in G0/G1 in response to PALA treatment even though the basal cellular concentrations of UTP and CTP were reduced 50 to 80%. While it has been shown that dATP in combination with cytochrome c is involved in the apoptotic pathway, the concentration of exogenous deoxyadenosine required to induce apoptosis in the parental L1210 cells was far in excess of the concentration required to inhibit cell growth. Deoxyadenosine did not cause an increase in apoptosis in the deoxyadenosine-resistant Y8 cells. These data suggest that the new roles ascribed to nucleotides may be specific for the particular cell type under very specific conditions.

L1210 MQ-580 is a murine leukemia cell line resistant to the cytotoxic activity of the alpha-(N)-heterocyclic carboxaldehyde thiosemicarbazone class of inhibitors of ribonucleotide reductase. The line is cross-resistant to etoposide, daunomycin, and vinblastine. L1210 MQ-580 cells expressed 8-fold resistance to 3-aminopyridine-2-carboxaldehyde thiosemicarbazone (3-AP), a relatively newly developed inhibitor of ribonucleotide reductase. The accumulation of [14C]3-AP by L1210 MQ-580 cells was 5- to 6-fold less than by parental L1210 cells. An increased rate of efflux of 3-AP was responsible for the lower steady-state concentration of 3-AP in resistant cells. In reverse transcription-polymerase chain reaction assays, L1210 MQ-580 cells were found to overexpress the multidrug resistance genes mdr1, mdr3, and mrp, but not the mdr2 gene, compared with parental L1210 cells. Measurement of the steady-state concentration of doxorubicin, a potential substrate for both the mdr and mrp gene products, demonstrated that L1210 MQ-580 cells accumulated 4-fold less anthracycline than parental cells. These findings indicate that drug efflux is a major determinant of the pattern of cross-resistance of L1210 MQ-580 cells. To extrapolate these observations to the human homologues of the mdr1, mdr3, and mrp murine genes, the effects of 3-AP were measured in L1210/VMDRC0.06 and NIH3T3 36-8-32 cells transfected with human MDR1 and MRP cDNAs, respectively. The transfectants were 2- to 3-fold resistant to the cytotoxic effects of 3-AP and accumulated less [14C]3-AP than their parental mock-transfected counterparts. Moreover, the cytotoxic activity of 3-AP was significantly greater in two double mrp gene knockout cell lines than in parental W 9.5 embryonic stem cells. Thus, the results suggest that 3-AP is a substrate for both the P-glycoprotein and MRP and that baseline MRP expression has the capacity to exert a protective role against the toxicity of this agent.

A mouse leukemia L1210 cell line, denoted MQ-580, that was selected for resistance to the ribonucleotide reductase inhibitor, 4-methyl-5-amino-1-formylisoquinoline thiosemicarbazone (MAIQ), in addition to having altered properties at the ribonucleotide reductase site had other alterations that contributed to its resistant phenotype; these included the expression of p-glycoprotein and the multi-drug resistance associated protein (MRP). The efflux of rhodamine 123 (Rh-123) or daunomycin (Dau) was greatly increased in MQ-580 cells compared to parental wild-type (WT) cells. The effluxes of Rh-123 and Dau were ATP- and temperature-dependent. The p-glycoprotein inhibitors, verapamil, cyclosporin A and reserpine blocked the efflux of both Rh-123 and Dau. In contrast, the inhibitors of MRP, MK571, BSO-treatment, arsenite and genistein did not block the efflux of either Rh-123 or Dau from MQ-580 cells. These findings suggest that the p-glycoprotein is the major transporter involved in effluxing Rh-123 and Dau from MQ-580 cells.

Mouse leukemia L1210 cells were generated for resistance to deoxyguanosine by two different methods. In one case the L1210 cells were subjected to gradual increases in deoxyguanosine (dGuo-R); in the second approach, the cells were subjected to deoxyguanosine at a concentration ten times the IC50 value and plated out on soft agar (D-92). The dGuo-R and D-92 cell lines had different phenotypic expressions. The dGuo-R cells showed a higher degree of resistance to dGuo than the D-92 cells. The levels of resistance to other cytotoxic drugs such as araC or 2-chloro-2'-deoxyadenosine did not necessarily correlate with the degree of resistance to dGuo. Deoxycytidine kinase activity was decreased in both of the cell lines, although there was a larger decrease in the dGuo-R cell line. The levels of kinase activities toward the other substrates were not all coordinately decreased in these cell lines. The degree of resistance of these cell lines to dGuo cannot be ascribed solely to an alteration at the deoxycytidine kinase site.

L1210 cell lines selected for resistance to deoxyadenosine exhibit altered steady-state levels of the mRNA for the early response genes and p53. In the deoxyadenosine-resistant cell lines (Y8 and ED2), the levels of the mRNAs for p53 and c-jun were markedly decreased while the steady-state levels for mRNAs for c-myc, c-fos and jun B were elevated in the Y-8 and ED2 cell lines. The levels of the mRNAs for PCNA and c-myb were the same in the wild type and mutant cell lines. The levels of the mRNAs for krox-24 were extremely low in the wild type and mutant cell lines. Cycloheximide (CHX) treatment of the cells resulted in the increase in the mRNA levels for c-jun, jun B, krox 24 and p53 in the Y-8 and ED2 cell lines. The time courses and the extents of the increases in the mRNA levels following CHX treatment were not the same for all of these mRNAs. The level of p53 RNA increased with no lag following CHX treatment while the levels of the mRNAs for c-myc, c-jun and krox-24 increased after a one-hour lag period. The level of the mRNA for p53 and c-myc increased 20- and 7-fold, respectively while the mRNA level for knox-24 increased 80-fold following CHX treatment. The Y8 and ED2 cell lines that lack steady-state levels of p53 show decreased sensitivity to cisplatin and increased frequency of gene amplification as measured by PALA resistance in a manner similar to other cell lines lacking p53. On the other hand, the ED2 and Y8 cell lines do not show a G1-block in response to PALA treatment. The cell lines appear to offer an experimental system in which to study the interactions between/among these early response genes and the p53-dependent and independent pathways.

A series of 3- and 5-alkylamino derivatives, as well as other structurally modified analogues of pyridine-2-carboxaldehyde thiosemicarbazone, have been synthesized and evaluated as inhibitors of CDP reductase activity and for their cytotoxicity in vitro and antineoplastic activity in vivo against the L1210 leukemia. Alkylation of 3- and 5-amino-2-(1,3-dioxolan-2-yl)pyridines (1, 2) resulted in corresponding 3-methylamino, 5-methylamino, 3-allylamino, 5-ethylamino, 5-allylamino, 5-propylamino, and 5-butylamino derivatives (5, 6, and 11-15), which were then condensed with thiosemicarbazide to yield the respective thiosemicarbazones (7, 8, and 16-20). Oxidation of 3,5-dinitro-2-methylpyridine (21) with selenium dioxide, followed by treatment with ethylene glycol and p-toluenesulfonic acid, produced the cyclic ethylene acetal, 23. Oxidation of 2-(1,3-dioxolan-2-yl)-4-methyl-5-nitropyridine (26) with selenium dioxide, followed by sequential treatment with sodium borohydride, methanesulfonyl chloride, and morpholine afforded the morpholinomethyl derivative 30. Catalytic hydrogenation of 23 and 30 with Pd/C yielded the corresponding amino derivatives 24 and 31. Catalytic hydrogenation of 5-cyano-2-methylpyridine (33) with Raney nickel, followed by treatment with acetic anhydride, gave the amide derivative 35. N-Oxidation of 35, followed by rearrangement with acetic anhydride, produced the acetate derivative, 5-[(acetylamino)methyl]-2-(acetoxymethyl)pyridine (37). Repetition of the N-oxidation and rearrangement procedures with compound 37 yielded the diacetate derivative 39. Condensation of compounds 24, 31, and 39 with thiosemicarbazide afforded the respective 3,5-diaminopyridine-, 4-(4-morpholinylmethyl)-5-aminopyridine-, and 5-(aminomethyl)pyridine-2-carboxaldehyde thiosemicarbazones (25, 32, and 40). The most biologically active compounds synthesized were the 5-(methylamino)-, 5-(ethylamino)-, and 5-(allylamino)pyridine-2-carboxaldehyde thiosemicarbazones (8, 17, and 18), which were potent inhibitors of ribonucleotide reductase activity with corresponding IC50 values of 1.3, 1.0, and 1.4 microM and which produced significant prolongation of the survival time of L1210 leukemia-bearing mice, with corresponding optimum % T/C values of 223, 204, and 215 being obtained when administered twice daily for six consecutive days at dosages of 60, 80, and 80 mg/kg, respectively.

Mouse leukemia L1210 cells were generated for resistance to 4-methyl-5-amino-1-formylisoquinoline thiosemicarbazone (MAIQ), a potent inhibitor of ribonucleotide reductase that is directed in the nonheme iron subunit (NHI) of the enzyme. The resistant cells, MQ-580, showed an 8-fold increase in IC50 toward MAIQ, a 4-fold increase in IC50 toward hydroxyurea, and also showed resistance to other ribonucleotide reductase inhibitors. In addition, the MQ-580 cell line was resistant to nonribonucleotide reductase inhibitors such as etoposide, daunomycin and vinblastine, but not to cisplatin. The mRNA for the NHI subunit was increased 7-fold in the MQ-580 cells with essentially no change in the mRNA level for the effector-binding subunit. The ribonucleotide reductase activity in the cell-free extracts prepared from the MQ-580 cells was only slightly elevated (30%). However, passage of the cell-free extract from the MQ-580 cells over Sephadex G-25 resulted in a 4.8-fold increase in specific activity over that of the wild-type cells. While the reductase activity in the cell-free extract from the MQ-580 cells did not show altered sensitivity to MAIQ, the reductase activity in the cell-free extract from the MQ-580 cells was much more sensitive to the effects of the iron-chelating agents Desferal and EDTA. The cell pellets from the MQ-580 cells were much darker in color than the pellets from the wild-type cells or hydroxyurea-resistant cells. The supernatant fraction from the MQ-580 cells after-SDS-PAGE showed the appearance of a strong Coomassie blue-staining band at 50 kDA that was not apparent in either the wild-type or hydroxyurea-resistant cells. This new resistant cell line offers an opportunity to explore differences in resistance mechanisms of drugs (e.g. MAIQ and hydroxyurea) that are directed at the same subunit of ribonucleotide reductase.

These studies, using three different reagents, show that the substrate properties of ribonucleotide reductase are specific but can be variable depending upon the nature of the interaction of the reagent with the holoenzyme or the individual subunit. Etheno-CDP, which acts as a competitive inhibitor with respect to CDP, interacts with the active site of the holoenzyme. This interaction was the result of rather tight structural requirements as epsilon-ADP did not result in a similar level of inhibition of either CDP or ADP reductase activities. The YL 1/2 antibody which binds very tightly to the NHI subunit has a much greater effect on CDP reductase activity than ADP reductase activity. The nonapeptide that corresponds to the C-terminus amino acid sequence of the NHI subunit and which binds to the EB subunit and aborts the formation of the NHI-EB active complex has a greater effect on ADP reductase activity than on CDP reductase activity. The use of reagents such as these can be helpful in dissecting the subtle but important differences in the substrate properties of mammalian ribonucleotide reductase.

The synthesis of a new series of pyridine-2-carboxaldehyde thiosemicarbazones (HCTs) that have amino groups in the 3- and 5-positions has allowed the comparison of the structure/function relationships with regard to inhibition of ribonucleotide reductase activity, L1210 cell growth in culture and L1210 leukemia in vivo. 3-Aminopyridine-2-carboxaldehyde thiosemicarbazones are more active than the corresponding 3-hydroxy-derivatives. The 3-amino-2-pyridine carboxaldehyde thiosemicarbazones were also more active then the 5-amino-2-carboxaldehyde thiosemicarbazones in inhibiting ribonucleotide reductase activity and L1210 cell growth in culture and in vivo. N-Acetylation of the 3-amino derivative resulted in a compound that was much less active both in vitro and in vivo; N-acetylation of the 5-amino derivative did not alter the in vitro inhibitory properties, but did eliminate the antitumor properties in vivo. When the most active HCTs were studied in more detail, it was found that the incorporation of [3H]thymidine into DNA was inhibited completely without the inhibition of [3H]uridine incorporation into RNA. Further, the conversion of [14C]cytidine to deoxycytidine nucleotides and incorporation into DNA was inhibited by the HCTs without an effect on the incorporation of cytidine into RNA. These data support the conclusion that ribonucleotide reductase is the major site of action of these HCTs. The 3-aminopyridine-2-carboxaldehyde thiosemicarbazones emerge as strong candidates for development for clinical trials in cancer patients.

A new series of alpha-(N)-heterocyclic carboxaldehyde thiosemicarbazones (HCTs) was studied for their effects on L1210 cell growth in culture, cell cycle transit, nucleic acid biosynthesis and ribonucleotide reductase activity. 3-Aminopyridine-2-carboxaldehyde thiosemicarbazone (3-AP) and 3-amino-4-methylpyridine-2-carboxaldehyde thiosemicarbazone (3-AMP) were the most active compounds tested with respect to inhibition of cell growth and ribonucleotide reductase activity. 5-Aminopyridine-2-carboxaldehyde thiosemicarbazone (5-AP) and 4-methyl-5-aminopyridine-2-carboxaldehyde thiosemicarbazone (5-AMP) were slightly less active. 3-AP, 3-AMP, 5-AP and 5-AMP inhibited the incorporation of [3H]thymidine into DNA without affecting the rate of incorporation of [3H]uridine into RNA. The uptake and incorporation of [14C]cytidine into cellular ribonucleotides and RNA, respectively, were not decreased by 3-AP or 3-AMP; however, the incorporation of cytidine into DNA via ribonucleotide reductase was inhibited markedly. Thus, a pronounced decrease in the formation of [14C]deoxyribonucleotides from radioactive cytidine occurred in the acid-soluble fraction of 3-AP- and 3-AMP-treated L1210 cells. Consistent with an inhibition of DNA replication that occurred at relatively low concentrations of 3-AP and 3-AMP, cells gradually accumulated in the S-phase of the cell cycle; at higher concentrations of 3-AP and 3-AMP, a more rapid accumulation of cells in the G0/G1 phase of the cell cycle occurred, with the loss of the S-phase population, implying that a second less sensitive metabolic lesion was created by the HCTs. N-Acetylation of 3-AMP resulted in a compound that was 10-fold less active as an inhibitor of ribonucleotide reductase activity and 8-fold less active as an inhibitor of L1210 cell growth. N-Acetylation of either 5-AP or 5-AMP did not alter the inhibitory properties of these compounds. The results obtained provide an experimental rationale for the further development of the HCTs, particularly 3-AP and 3-AMP, as potential drugs for clinical use in the treatment of cancer.

A series of substituted 2-acylpyridine-alpha-(N)-hetarylhydrazones was prepared and studied for their effects on mammalian ribonucleotide reductase activity using a highly purified enzyme preparation from Ehrlich tumor cells and on mouse leukemia L1210 cell growth in culture. Pyridine-2-aldehyde-2-pyridylhydrazone (PH 22), ethyl-2-pyridylketone-I-phthalazinylhydrazone (PH 22-25) and pyridine-2-aldehyde-2'-quinolylhydrazone (PQ 22) inhibited purified ribonucleotide reductase activity and inhibited L1210 cell growth in culture. PH 22-25 inhibited [3H]thymidine incorporation into DNA and inhibited ribonucleotide reductase activity in situ (as measured bvy [14C]cytidine metabolism and as a result inhibited DNA synthesis. There was no effect on RNA synthesis. These data indicate that these substituted hydrazones are potent inhibitors of tumor cell growth through the inhibition of ribonucleotide reductase.

The 2'-deoxy-2'-methylene derivatives of adenosine (MdAdo), guanosine (MdGuo), tubercidin (MdTu), cytidine (MdCyd) and uridine (MdUrd) were synthesized as mechanism-based inhibitors directed at ribonucleotide reductase. It was shown that MdCyd 5'-diphosphate irreversibly inactivated ribonucleotide reductase from Escherichia coli (Baker et al., J Med Chem 34: 1879-1884, 1991). In studies reported here, MdAdo/EHNA, MdGuo and MdCyd inhibited L1210 cell growth with IC50 values of 3.4, 10.6 and 1.4 microM, respectively. Since MdAdo is a substrate for adenosine deaminase, the presence of EHNA was required to give maximal growth inhibition. 8-Aminoguanosine was not required to maximize the cytotoxic effects of MdGuo. The 2'-deoxy-2'-methylene derivatives of tubercidin and uridine did not inhibit L1210 cell growth at concentrations as high as 50 microM (MdTu) or 100 microM (MdUrd). L1210 cell lines resistant to hydroxyurea (directed at the non-heme iron subunit of ribonucleotide reductase) or deoxyadenosine (directed at the effector binding subunit of ribonucleotide reductase) were not resistant to MdCyd. An L1210 cell line that was highly resistant to dGuo due to the loss of a relatively specific deoxyribonucleoside kinase (Cory et al., J Biol Chem 268: 405-409, 1993) had a 6.6-fold increase in the IC50 value toward MdCyd, but showed only a 2-fold increase in resistance to MdGuo. Another L1210 cell line that was markedly deficient in adenosine kinase activity was highly resistant to MdAdo. Analysis by flow cytometry showed that MdCyd showed the transit of the cells through the G2/M phase of the cell cycle resulting in the buildup of the G2/M population. MdAdo, MdGuo and MdCyd inhibited the incorporation of [14C]cytidine into DNA without an effect on RNA synthesis or total cellular uptake of [14C]cytidine. The conversion of [14C]cytidine to deoxycytidine nucleotides was partially inhibited by MdGuo, but not by MdAdo or MdCyd. These data show that the 2'-deoxy-2'-methylene derivatives of adenosine, guanosine and cytidine are activated via specific nucleoside kinases and that the modes of action of these compounds are not identical.

Mouse leukemia L1210 cell lines that were selected for resistance to deoxyguanosine (dGuo-R) or lacked adenosine kinase activity (ED2) were used to evaluate the nature of the nucleoside kinase that was required to phosphorylate nucleoside analogs to their respective active nucleotide form. The dGuo-R cells had reduced levels of kinase activity toward araC, dGuo and 2-CldAdo as substrates with essentially no loss of activity toward dCyd. This cell line showed resistance to dGuo, araC, araG, FdAdo, and Fara A but not to dAdo or araA. The ED2 cell line was resistant to pyrazofurin and 6-methylmercaptopurine riboside and to araA/EHNA but not to 2-CldAdo or 2-Cl-2'-FaraA. The study of the effects of newer nucleoside analogs such as dFdCyd, MdAdo, MdCyd and MdGuo in these cell lines showed that some of these agents are primarily phosphorylated by deoxyribonucleoside kinase (dFdCyd) or by adenosine kinase (MdAdo) or in some instances by multiple kinases (FaraA). These cell lines will be useful in defining the nature of the kinase(s) responsible for activating new nucleoside analogs and defining cross-resistance patterns.

A mouse leukemia L1210 cell line was selected for resistance to deoxyguanosine. The deoxyguanosine-resistant cells (dGuo-R) were 126-fold less sensitive to deoxyguanosine than the wild-type cells. The IC50 values for araC and araG were increased, but only 10-12-fold in the dGuo-R cells when compared with the wild-type cells. The dGuo-R cell line showed an increased level of resistance to 2-fluoro-2'-deoxyadenosine and 2-fluoroadenine arabinoside (11-14-fold), but essentially no increase in resistance to deoxyadenosine or adenine arabinoside. Deoxyribonucleoside kinase activity was decreased only slightly (19%) when deoxycytidine was utilized as substrate; when cytosine arabinoside or deoxyguanosine was used as the substrate, the kinase activity in the extracts from the dGuo-R cells was only 10% of the enzyme activity in the extracts from the wild-type cells. The determination of the kinetic parameters, Km and Vmax, indicated that there were marked decreases in the Vmax values for deoxyguanosine and cytosine arabinoside as substrates, but not for deoxycytidine as substrate; the Km values for deoxycytidine and cytosine arabinoside were increased in the extracts from the dGuo-R cells. By use of high-performance liquid chromatography, the kinase activities in the extracts from the wild-type and resistant cells could be resolved. There was the specific loss of kinase activity toward cytosine arabinoside and deoxyguanosine as substrates. These data indicate that the dGuo-R cells have decreased levels of a specific deoxyribonucleoside kinase activity.

2',2'-Difluorodeoxycytidine (Gemcitabine, dFdCyd) is a cytotoxic agent which is active toward a variety of tumor cells. It has been shown that there are multiple intracellular sites of action which include ribonucleotide reductase and DNA polymerase. In these studies, the effects of dFdCyd on wild-type mouse leukemia L1210 cells and variant L1210 cell lines which had alterations at the ribonucleotide reductase site or at the deoxyribonucleoside kinase site were studied. For cell growth, the IC50 value for dFdCyd in wild-type L1210 cells was 3.1 nM. In the variant cell lines, the IC50 values were: hydroxyurea-resistant (HU), 3.3 nM; deoxyadenosine-resistant (Y8), 1.8 nM; pyrazoloimidazole/deoxyadenosine-resistant (ED2), 1.9 nM; and deoxyguanosine-resistant (dGuo-R), 44.7 nM. The dGuo-R cell line had a relatively specific loss of the deoxyribonucleoside kinase responsible for phosphorylating deoxyguanosine and cytosine arabinoside with little loss of the deoxycytidine kinase activity. DFdCyd had no effect on the total uptake of [14C]cytidine into the cells or incorporation into RNA. DFdCyd inhibited the conversion of [14C]cytidine to deoxycytidine nucleotides and incorporation into DNA. However, the incorporation of cytidine into DNA was inhibited to a greater extent than was the inhibition of in situ ribonucleotide reductase activity. Ribonucleotide reductase activity in cell-free extracts prepared from L1210 cells treated with dFdCyd (20 nM) overnight was reduced by 50%. These results show that cell lines which have increased levels of ribonucleotide reductase activity (HU and ED2) or loss of feedback inhibition by dATP (ED2 and Y8) are still sensitive to dFdCyd. The findings indicate that ribonucleotide reductase is not the primary site of inhibition by dFdCyd.(ABSTRACT TRUNCATED AT 250 WORDS)

Previous studies have shown that N-hydroxy-N'-aminoguanidine (HAG) derivatives [RCH = NNHC(= NH)NHOH-tosylate] inhibit ribonucleotide reductase activity and block the growth of leukemia L1210 cells and human colon carcinoma, HT-29, cells in culture. In the current studies, the role of the side chains and the location of the bond of the side chain moiety to HAG were investigated using a new series of HAG derivatives which contained as the R-group--cyclohexyl, phenyl-, pyridyl- or napthyl moieties. The effects of these compounds as inhibitors of L1210 cell growth and ribonucleotide reductase activity were compared with the parent compound. N-hydroxy-N'-aminoguanidine was less inhibitory to ribonucleotide reductase activity and L1210 cell growth than hydroxyurea. The phenyl-HAG compounds which included 1-benzyloxybenzylidene- and 4-cyclohexylmethoxybenzylidene-HAG inhibited CDP reductase with IC50s which ranged from 50-110 microM. 1-Naphthylmethylene-HAG was more inhibitory than 2-naphthylmethylene-HAG and more inhibitory than the phenyl-HAG compounds. 2-Pyridylmethylene-HAG was more inhibitory than 3-pyridylmethylene- or 4-pyridylmethylene-HAG. While HAG inhibited CDP and ADP reductase activities essentially to the same extent, the HAG-derivatives inhibited ADP reductase activity to a greater extent than CDP reductase activity. Cyclohexylmethylene-HAG did not inhibit either L1210 cell growth or ribonucleotide reductase activity. There was good correlation between the inhibition of ribonucleotide reductase activity and L1210 cell growth by these HAG-derivatives. These data indicate that not only is the nature of the side chain substitution important, but also the location of the HAG-moiety on the ring position.

L1210 cell lines, selected for resistance to deoxyadenosine due to the loss of allosteric inhibition of ribonucleotide reductase by dATP, had altered steady-state levels of the mRNAs for c-myc, fos, and p53. Wild-type L1210 cells had constitutive steady-state levels of c-myc and p53 with little or no fos mRNA. Two different deoxyadenosine-resistant cell lines (Y8 and ED2) had elevated steady-state levels of c-myc and fos but essentially no p53 mRNA. Hydroxyurea-resistant L1210 cells had the same levels of c-myc, fos, and p53 as the wild-type cells. There was no amplification of the gene for c-myc in the Y8 or ED2 cell lines. The half-life for c-myc mRNA was essentially the same in the wild-type and the Y8 and ED2 cells. Nuclear runoff experiments showed that the rates of transcription for c-myc in the Y8 and ED2 cells were elevated and could account for the increased steady-state levels of c-myc in these two cell lines. The transcription rate for p53 mRNA was not decreased in the Y8 and ED2 cells and therefore did not account for the loss of the steady-state levels of p53 in the cells. Cycloheximide treatment of the Y8 and ED2 cells resulted in a marked increase in the steady-state p53 mRNA level, indicating that a protein which was rapidly turned over was responsible for the extremely short half-life of p53 mRNA in these two cell lines.

A new tetrazolium analog of 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) was evaluated as a substitute for MTT in the microculture screening assay for in vitro cell growth. This new tetrazolium compound, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2- (4-sulfophenyl)-2H-tetrazolium, inner salt (MTS), in the presence of phenazine methosulfate (PMS), gave a water-soluble formazan product that had an absorbance maximum at 490-500 nm in phosphate-buffered saline. The amount of colored product formed was proportional to the number of cells and the time of incubation of the cells with MTS/PMS. MTS/PMS was reactive in all the cell lines tested which included mouse leukemia L1210 cells, mouse Ehrlich tumor cells, mouse 3T3 fibroblasts, and human colon tumor cells (HT-29). HT-29 and 3T3 fibroblasts reduced MTS/PMS more efficiently than they reduced MTT. Comparable to the amount of product formed from MTT, MTS/PMS gave excellent product formation. The IC50 value for pyrazoloimidazole obtained using MTS/PMS was 200 microM; for 5-fluoro-2'-deoxyuridine, the IC50 value was 0.9 nM. These values compared very favorably with the IC50 values obtained by direct cell counts. Further, the same IC50 values were obtained when the absorbances of the formazan product in the 96-well plates were determined after different times of incubation.

Studies were carried out to determine the effects of preincubation of 4-methyl-5-amino-1-formylisoquinoline thiosemicarbazone (MAIQ) with hepatic microsomes on the ability of MAIQ to inhibit CDP reductase activity in vitro. An aliquot from the 100,000 x g supernatant fraction from this incubation was used in the CDP reductase assay. MAIQ incubated in the absence of microsomes inhibited CDP reductase activity in a dose-dependent manner. At high MAIQ concentration (5 microM) CDP reductase activity was inhibited 95%. When MAIQ (5 microM) was first incubated in the presence of hepatic microsomes and NADPH, CDP reductase activity was inhibited only 30%. This attenuation of MAIQ inhibition was dependent on time of incubation and microsomal protein concentration and showed an obligatory requirement for NADPH or NADH. Significant attenuation was observed at pyridine nucleotide concentrations as low as 0.1 mM. Heat denaturation of microsomal proteins inactivated their ability to attenuate the MAIQ inhibition. Microsomes prepared from Ehrlich tumor cells were ineffective as inactivators of MAIQ. Results of our studies show that hepatic microsomes contain an enzyme(s) which can inactive MAIQ as an inhibitor of CDP reductase.