Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The nicotinamide analog 6-aminonicotinamide (6AN) is undergoing preclinical evaluation as a potential agent for modulating the cytotoxicity of antineoplastic drugs. Studies performed by Berger and coworkers (1983, 1989) revealed that 6AN sensitized L1210 murine leukemia cells to bis(2-chloroethyl)nitrosourea in vitro and in a murine model. In subsequent studies, a mixture containing the modulators N-(phosphonacetyl)-L-aspartate, 6-methylmercaptopurine riboside, and 6AN has been reported to enhance the efficacy of ionizing radiation, 5-fluorouracil, doxorubicin, and paclitaxel against spontaneous autochthonous breast cancers in mice (reviewed in Martin and Schwartz, 1997). More recently, 6AN was observed to enhance the cytotoxicity of cisplatin in V79 Chinese hamster cells (Chatterjee et al., 1997) and a variety of human tumor cell lines in vitro (Budihardjo et al., 1998). Collectively, these studies raised the possibility that 6AN might enhance the effectiveness of various anticancer agents in vitro and in vivo.

Despite these promising results, studies performed to evaluate the feasibility of using 6AN to modulate the cytotoxicity of cisplatin in vivo have been disappointing (Walker et al., 1999). Optimal sensitization of A549 or K562 cells to cisplatin in vitro required pretreatment with 15 to 30 µM 6AN for at least 10 h. Pharmacokinetic studies, on the other hand, revealed that 6AN had a terminal serum half-life of only 20 to 30 min in mice (Walker et al., 1999). After bolus administration of 10 mg/kg 6AN (one-fourth the single-dose LD₅₀), serum 6AN levels peaked at ~90 µM but rapidly declined, preventing prolonged exposure to concentrations that sensitized cells to cisplatin in vitro. Conversely, a continuous infusion strategy was capable of maintaining 6AN plasma levels at 15 µM for 24 to 48 h but was uniformly lethal (Walker et al., 1999), most likely as a consequence of the well-documented neurotoxicity of 6AN (Sternberg and Philips, 1958; Herken et al., 1969). These observations indicate the need to identify a less toxic analog of 6AN, a process that would be facilitated by a better understanding of the mechanism by which 6AN modulates the action of other agents.

Our previous study (Budihardjo et al., 1998) demonstrated that 6AN enhanced cisplatin-induced apoptosis but did not affect the cytotoxicity of etoposide, topotecan, 4-hydroperoxycyclophosphamide, or chlorambucil, raising the possibility that 6AN might preferentially affect a drug-specific step in the cell death pathway. Consistent with this hypothesis, 6AN was observed to enhance cisplatin accumulation and the subsequent formation of Pt-DNA adducts (Budihardjo et al., 1998). Although the mechanisms responsible for cisplatin accumulation are not completely understood, a current model suggests that cisplatin uptake involves both passive diffusion and carrier-mediated transport (Gately and Howell, 1993). Accordingly, the observation that 6AN enhances intracellular cisplatin levels raises the possibility that 6AN alters the expression or function of polypeptide(s) involved in cisplatin transport across the plasma membrane.

A different view of 6AN action has emerged from other studies. In particular, the observation that 6AN enhances expression of the glucose-regulated protein GRP78 (Chatterjee et al., 1995, 1997) has raised the possibility that GRP78 might somehow regulate cisplatin action. GRP78, however, is a stress-inducible chaperone protein that resides in the endoplasmic reticulum, where it is hypothesized to function in the translocation of proteins from the cytosol and in the correct assembly of proteins during early protein processing (Lee, 1992; Little et al., 1994; reviewed in Brostrom et al., 1995). A mechanistic link between 6AN-induced GRP78 expression and altered cisplatin action has not been established.

In the present study, the cellular effects of 6AN were examined in greater detail. Labeling with [³⁵S]methionine, followed by two-dimensional nonequilibrium pH gradient electrophoresis (NEPHGE)/SDS-polyacrylamide gel electrophoresis (PAGE), was used to more completely define the effects of 6AN on protein expression. Protein synthesis inhibitors were employed to examine the role of 6AN-induced polypeptides on cisplatin accumulation and cytotoxicity. Assays of methionine uptake and incorporation were performed to assess the effect of 6AN on amino acid transport and protein synthesis. Finally, 18 structural analogs of 6AN were tested for their ability to inhibit protein synthesis, increase Pt-DNA adducts, and enhance cisplatin cytotoxicity. Results of these studies are consistent with a model in which 6AN-associated inhibition of protein synthesis prompts a compensatory increase in amino acid and cisplatin accumulation, with subsequent enhancement of Pt-DNA adducts and cisplatin-induced cytotoxicity.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Reagents were purchased from the following suppliers: cisplatin, L-alanine, L-serine, 6AN, 6-aminonicotinic acid, nicotinamide, methylaminoisobutyrate (MeAIB), cycloheximide, anisomycin, puromycin, DRB, actinomycin D, and digitonin from Sigma (St. Louis, MO); HindIII from Life Technologies/BRL (Gaithersburg, MD); proteinase K from Boehringer-Mannheim (Indianapolis, IN); bicinchoninic acid from Pierce (Rockford, IL); enhanced chemiluminescence reagents (ECL) or Amplify (Amersham, Arlington Heights, IL); pH 3.5 to 10 ampholytes from Pharmacia/LKB (Piscataway, NJ); mouse monoclonal anti-GRP78 antibody from StressGen (Victoria, British Columbia); and platinum standard from J.T. Baker (Phillipsburg, NJ). RNase A (RAF grade; Worthington, Freehold, NJ) was prepared as a 10 mg/ml solution in H₂O and boiled before storage at 20°C. All chemicals for synthesis were of the highest available purity.

Cell Culture. A549 nonsmall cell lung cancer cells and K562 chronic myelogenous leukemia cells from American Type Culture Collection (Manassas, VA) were cultured at 37°C in a humidified atmosphere of 95% air, 5% CO₂ in RPMI 1640 medium containing 5% heat-inactivated fetal bovine serum, 100 U/ml penicillin G, 100 µg/ml streptomycin, and 2 mM L-glutamine (medium A). To ensure logarithmic growth, cultures were maintained at densities below 1 × 10⁶ cells/ml (K562) or 70 to 80% confluence (A549). Cells were fed on the day before the start of each experiment.

Colony Forming Assays. Aliquots containing 3 to 5 × 10⁵ log phase K562 cells in 1 ml of medium A were incubated with diluent or a concentration (250 µM) of the various 6AN analogs for 18 h. A 1-µl aliquot of dimethyl sulfoxide containing the indicated final concentration of cisplatin was added for 1 h. Cells were then sedimented at 200g for 10 min, washed, diluted 1:500, and plated in 0.3% agar as described (Budihardjo et al., 1998; Walker et al., 1999). After a 10- to 14-day incubation at 37°C, colonies containing 50 cells were counted using an inverted phase-contrast microscope. Survival was expressed relative to control cells incubated with the corresponding concentration of 6AN analog in the absence of cisplatin treatment. Control plates typically contained 200 to 400 colonies.

Measurement of Whole Cell Cisplatin Accumulation and Pt-DNA Adducts. For assessment of cellular cisplatin accumulation, duplicate 100-mm plates of A549 cells grown to 70 to 80% confluence were incubated in medium A in the absence or presence of 6AN, cycloheximide, puromycin, anisomycin, DRB, or actinomycin D for 6 to 24 h as indicated. Freshly prepared cisplatin was then added to a final concentration of 40 µM from a 1000× concentrated stock. After a 2-h incubation at 37°C, cells were washed once with ice-cold PBS, briefly trypsinized, sedimented at 200g for 10 min, and washed three times with ice-cold PBS. At the last wash step, cells were resuspended in a total volume of 1 ml. A 100-µl aliquot of cells was removed, solubilized in 100 µl of 1 M NaOH at 21°C overnight, neutralized with 100 µl of 1 M HCl, and assayed for protein by the bicinchoninic acid method (Smith et al., 1985). The remainder of the cells were sedimented, reacted overnight at 21°C with 0.5 ml of concentrated nitric acid, heated to 100°C for 5 min, diluted with 0.5 ml of 30% (w/w) H₂O₂, and heated again to 100°C for 5 min. Elemental platinum in the lysates was assayed by the Metals Laboratory of the Mayo Clinic using inductively coupled plasma mass spectroscopy as recently described (Walker et al., 1999).

Methionine Labeling, Protein Electrophoresis, and Immunoblotting. A549 cells were incubated in the absence or presence of 62.5 µM 6AN in medium A for 18 to 24 h. After culture medium was removed, cells were briefly trypsinized and washed twice with RPMI 1640 medium lacking methionine. Cells were resuspended in 1 ml of methionine-free RPMI 1640 medium supplemented with 5% dialyzed fetal bovine serum and 50 µCi/ml [³⁵S]methionine in the continued absence or presence of 62.5 µM 6AN and incubated for 1 h at 37°C. After labeling, the cells were sedimented at 800g for 10 min at 4°C; washed twice with ice-cold PBS containing 5 mM unlabeled methionine; sonicated in SDS sample buffer consisting of 4 M urea, 2% (w/v) SDS, 62.5 mM Tris-HCl (pH 6.8), and 5% (v/v) -mercaptoethanol; and heated to 70°C for 30 min. Aliquots containing equal amounts of trichloroacetic acid-perceptible radiolabel (500,000 cpm) were mixed with four volumes of IEF buffer [9.5 M urea, 8% (w/v) Nonidet P-40, 2% (w/v) pH 3.5 to 10 ampholytes, and 5% (w/v) -mercaptoethanol] and subjected to NEPHGE in the first dimension followed by SDS-PAGE in the second dimension as previously described (O'Farrell et al., 1977; Kaufmann and Shaper, 1984). After the second dimension, gels were impregnated with Amplify and exposed to Kodak X-Omat AR film for 4 days at 70°C in the presence of intensifying screens as specified by the supplier.

Methionine Uptake and Incorporation. To assess methionine uptake, A549 cells in 24-well plates were treated with 0 to 1000 µM 6AN for 18 to 24 h. Cells were then treated with 1 µCi/ml [³H]methionine in medium A for 5 min, washed five times with ice-cold PBS, and lysed in 0.5 ml of 0.5% SDS. Aliquots were subjected to scintillation counting and protein measurement (Smith et al., 1985). To correct for nonspecific trapping of label, replicate aliquots were treated with [³H]methionine for <10 s and handled in an identical fashion.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Evaluation of the Effects of 6AN Analogs. Previous experiments demonstrated that the niacin derivative 6AN enhances cisplatin-induced Pt-DNA adducts and cytotoxicity in a variety of human tumor cell lines (Budihardjo et al., 1998). To search for structural analogs that might have similar effects, 18 6AN derivatives (Fig. 1) were evaluated for their ability to increase Pt-DNA adducts and cisplatin toxicity in K562 cells. Results of these experiments are summarized in Fig. 2.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Structures of 6AN analogs used in this study. 6AN is depicted as compound A.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Effects of 6AN and its analogs. A, K562 cells were incubated for 21 h with 250 µM 6AN or 250 µM analogs D, F, H, L, M, and R. Cisplatin was then added at the indicated final concentration. After a 1-h incubation with cisplatin, cells were washed and plated in drug-free agar to allow colonies to form. Data are expressed as the ratio of the number of colonies in samples treated with the 6AN analog + cisplatin to the number of colonies formed in samples treated with the 6AN analog alone. B, summary of the effects of all 18 analogs on cisplatin sensitivity in clonogenic assays performed as indicated in A. Results are expressed as the ratio of the LD90 in absence of the analog to the LD90 in the presence of the analog. Results are the means of two to six experiments with analogs B through R. C, summary of the effects of all 18 analogs on formation of Pt-DNA adducts. Data are expressed as the ratio of Pt-DNA adducts in the presence of the analog to Pt-DNA adducts in the absence of any analog. Results are the means of 17 experiments with 6AN (compound A, included as a control in each experiment) and two to four experiments with analogs B through R.

6AN Selectively Increases the Expression of Five Polypeptides. A549 cells were used for most of the subsequent studies because the effects of 6AN on cisplatin sensitivity were somewhat greater in this cell line (Budihardjo et al., 1998). All major findings presented below, however, were also replicated in K562 cells to confirm that the results were not unique to one line.

View larger version (58K): [in this window] [in a new window]   Fig. 3.   6AN treatment is accompanied by enhanced expression of multiple cellular polypeptides. Two-dimensional gels of A549 cells were treated with buffer (A) or 62.5 µM 6AN (B) for 21 h then pulsed with [35S]methionine for 1 h before lysis. NEPHGE (right to left) was followed by SDS-PAGE (top to bottom). Numbers at left are molecular weights of standard proteins (run on the edge of the gel) × 103. Numbers at top are pH values of corresponding sections of blank NEPHGE gel run in parallel. Circled areas are species that increase in 6AN-treated cells. Results are representative of three experiments.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Effect of 6AN on GRP78 expression and cisplatin cytotoxicity in A549 cells. A, A549 cells were treated with the indicated concentration of 6AN for 18 h. The indicated concentration of cisplatin was then added for 2 h. At the completion of the incubation, cells were washed and incubated for 7 days to allow colonies to form. B, from the data in A, the cisplatin LD90 was calculated at each 6AN concentration. Inset, immunoblotting for GRP78 and lamin B1. A549 cells were incubated for 18 h with the indicated concentration of 6AN, harvested for SDS-PAGE, and probed with monoclonal anti-GRP78 (top). Lamin B1 (bottom) served as a loading control. All results are representative of at least three experiments.

Protein Synthesis Inhibitors Mimic the Effects of 6AN on Formation of Pt-DNA Adducts and Cisplatin Sensitivity. To further examine the potential role of the 6AN-induced polypeptides in sensitizing cells to cisplatin, A549 cells were treated with 6AN or diluent in the presence of 107 µM cycloheximide, a concentration that inhibited protein synthesis in A549 cells by >90%, before cisplatin exposure. Treatment with 250 µM 6AN for 6 h resulted in a 4.2- ± 1.5-fold (n = 3) increase in Pt-DNA adducts (Fig. 5A). Addition of cycloheximide 30 min before 6AN did not prevent this increase. On the contrary, treatment with cycloheximide alone for 6.5 h enhanced the formation of Pt-DNA adducts 2.9- ± 0.9-fold (n = 3, Fig. 5A). Consistent with these results, pretreatment with cycloheximide also failed to diminish the effects of 6AN on cisplatin sensitivity in colony forming assays (Fig. 5B). A 6-h pretreatment with 6AN decreased the cisplatin LD₉₀ by 3.5- ± 1.5-fold (n = 6). This effect was not prevented by adding cycloheximide 30 min before 6AN. Instead, treatment with cycloheximide alone resulted in a 2.3- ± 0.2-fold decrease in the LD₉₀, and the combination of cycloheximide plus 6AN resulted in a 5.2- ± 0.4-fold decrease in the LD₉₀ (n = 3). In other words, cycloheximide mimicked the effects of 6AN when applied alone and augmented the effects of 6AN when combined with this agent. Interestingly, all the changes observed after the 6-h incubation occurred without any change in GRP78 (Fig. 5B, inset), providing additional support for the view that induction of GRP78 and effects of 6AN on cisplatin action can be dissociated.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Effect of cycloheximide on platinum-DNA adduct formation and cisplatin cytotoxicity. A, A549 cells were incubated for 6 h with 250 µM 6AN or 6.5 h with 107 µM cycloheximide. Cisplatin (40 µM) was then added, and the incubation was continued for an additional 2 h before cells were harvested for measurement of Pt-DNA adducts. Error bars, range of duplicate determinations in this experiment. The text summarizes results of multiple independent experiments. B, effect of 6AN and cycloheximide on A549 colony formation. A549 cells were incubated for 6 h with 250 µM 6AN or 6.5 h with 107 µM cycloheximide (CHX) before the addition of the indicated concentration of cisplatin. After a 2-h incubation with cisplatin, cells were washed and allowed to form colonies in drug-free medium for 7 days. Inset, in a parallel experiment, A549 cells treated with 250 µM 6AN or 107 µM cycloheximide for 6 or 6.5 h, respectively,were harvested for immunoblotting with monoclonal anti-GRP78 antibody. Both 6AN and cycloheximide enhance Pt-DNA adducts (A) and cytotoxicity of cisplatin (B) without increasing GRP78 (inset) over the time course of this exposure. C, effect of actinomycin D on cisplatin accumulation and protein synthesis. A549 cells were incubated for 22 h with the indicated concentration of actinomycin D. Cisplatin (40 µM) was then added, and the incubation was continued for an additional 2 h before cells were harvested for measurement of Pt-DNA adducts (). Alternatively, incorporation of radiolabeled methionine into TCA-precipitable material was determined (). Results in each panel present an experiment that is representative of three to six separate experiments.

Protein Synthesis Inhibitors Increase Cisplatin Accumulation. Previous observations suggested that the effect of 6AN on Pt-DNA adducts is mediated by a 6AN-induced increase in cisplatin accumulation that is partially inhibited by the nonmetabolizable amino acid MeAIB (Budihardjo et al., 1998). To determine whether protein synthesis inhibitors might be modulating the action of cisplatin by a similar mechanism, A549 cells were incubated for 24 h with 107 µM cycloheximide, 100 µM puromycin, or 38 µM anisomycin. Even though each of these agents inhibited protein synthesis by >90%, cell viability as determined by trypan blue exclusion remained 98% after 24 h. When these cells were assayed for cisplatin accumulation, treatment with each of the protein synthesis inhibitors was accompanied by increased accumulation of cisplatin (Fig. 6A). In other experiments, a 20- to 24-h exposure to actinomycin D or DRB had the same effect. Addition of 20 mM MeAIB at the same time as cisplatin partially reversed the effect of the protein synthesis inhibitors just as it partially reversed the effect of 6AN (Fig. 6B and data not shown). In contrast, serine had no effect on cisplatin accumulation (data not shown). Taken together, these data indicate that 6AN and protein synthesis inhibitors increase cisplatin accumulation, possibly through a mechanism that involves a neutral amino acid transporter.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Effect of protein synthesis inhibitors on cisplatin accumulation. A, effect of a 24-h treatment with 250 µM 6AN, 107 µM cycloheximide, 100 µM puromycin, or 38 µM anisomycin on cellular accumulation of cisplatin by A549 cells. Error bars, range of duplicate samples in this experiment. In multiple experiments, cisplatin accumulation was enhanced 2.6- ± 0.7-fold (mean ± S.D.) by cycloheximide, 3.6- ± 0.6-fold by puromycin, 2.8- ± 0.8-fold by anisomycin, and 3.8- ± 0.3-fold by actinomycin D. B, effect of MeAIB on the 6AN- and cycloheximide-induced increase in cisplatin accumulation. A549 cells were treated for 24 h with 250 µM 6AN or 107 µM cycloheximide. MeAIB was then added to a concentration of 20 mM, followed 30 min later by addition of 40 µM cisplatin. After a 2-h incubation with cisplatin, samples were harvested for determination of cell-associated platinum. Error bars, range of duplicate samples in this experiment. In three independent experiments, MeAIB diminished the effect of 6AN by 48 ± 8% and the effect of cycloheximide by 81 ± 18%.Aliquots were then pulsed for 1 h with radiolabeled methionine or treated with 40 µM cisplatin for assessment of Pt-DNA adducts. In three separate experiments, nicotinamide restored protein synthesis to 86 ± 17% of control levels and decreased Pt-DNA adducts from 560 ± 180% of control (6AN alone) to 90 ± 10% of control (6AN + nicotinamide).

6AN Inhibits Protein Synthesis at the Concentrations That Increase Cellular Cisplatin Accumulation. Because cycloheximide, puromycin, and anisomycin produced effects similar to those of 6AN, we next investigated the possibility that 6AN might be capable of inhibiting protein synthesis. When A549 cells were incubated with increasing concentrations of 6AN for 18 h, a dose-dependent decrease in incorporation of radiolabeled methionine into TCA-precipitable macromolecules was observed (Fig. 7A). Decreased methionine incorporation was evident at 8 µM 6AN and reached a maximum of 70 to 80% inhibition at 62.5 µM. To rule out the possibility that these results reflect decreased methionine uptake rather than diminished protein synthesis, uptake of radiolabeled methionine into cells was measured at time points ranging from 30 s to 5 min. In 10 separate experiments, 6AN enhanced methionine uptake by a factor of 1.8- ± 0.5-fold (e.g., inset, Fig. 7A). These results argue against the possibility that 6AN is inhibiting methionine uptake and instead suggest that 6AN is inhibiting protein synthesis. Furthermore, the effects of 6AN on protein synthesis (Fig. 7A), cellular cisplatin accumulation (Budihardjo et al., 1998), and cisplatin cytotoxicity (Fig. 4) were observed at similar concentrations (half-maximal effects at 16 µM 6AN in all cases), suggesting that these changes were related.

View larger version (26K): [in this window] [in a new window]   Fig. 7.   Effect of 6AN and 6AN analogs on protein synthesis in A549 and K562 cells. A, effect of 6AN on methionine incorporation into TCA-precipitable material in A549 cells. Cells were incubated for 18 h with increasing concentrations of 6AN, then pulsed with radiolabeled methionine for 1 h before lysis in 10% TCA. Inset, effect of 6AN on radiolabeled methionine uptake in A549 cells. Cells were treated with the indicated concentration of 6AN for 22 h, then exposed to radiolabeled methionine for 5 min. B, summary of the effects of 18 6AN analogs on methionine incorporation into TCA-precipitable material in K562 cells. Cells were incubated with the indicated 6AN analog (see Fig. 1 for key) at 250 µM for 21 h. Radiolabeled methionine was then added for 1 h. At the completion of the incubation, TCA precipitates were collected and counted. Results of a single experiment in which all 18 analogs were tested simultaneously are presented. Similar results were obtained in two to three independent experiments. C, effect of nicotinamide on 6AN-inhibited protein synthesis and formation of Pt-DNA adducts. A549 cells were incubated for 21 h with 250 µM 6AN in the absence or presence of 2.5 mM nicotinamide.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Recent studies have demonstrated that 6AN sensitizes cells to cisplatin in vitro (Chatterjee et al., 1997; Budihardjo et al., 1998). Because 6AN demonstrated neurotoxicity in animals in vivo (Sternberg and Philips, 1958; Walker et al., 1999) as well as unfavorable pharmacokinetics (Walker et al., 1999), studies were initiated to investigate the mechanism by which 6AN modulates the action of cisplatin and to search for analogs that might have similar effects but lack the neurotoxicity of 6AN. These experiments have led to a series of novel observations.

Previous studies have indicated that 6AN causes a variety of biochemical changes, including inhibition of 6-phosphogluconate dehydrogenase and depletion of products downstream from this enzyme, notably 5'-phosphoribosyl-1'-pyrophosphate and NADPH, in intact cells (Herken et al., 1969; Keller et al., 1976; Varnes, 1988; Street et al., 1996). Depletion of 5'-phosphoribosyl-1'-pyrophosphate in turn leads to the decreased synthesis of NAD⁺ observed in 6AN-treated cells (Hunting et al., 1985; Berger et al., 1987; Martin and Schwartz, 1997), whereas depletion of NADPH contributes to glutathione depletion (Varnes, 1988; Budihardjo et al., 1998). In addition, 6AN has been shown to directly inhibit poly(ADP-ribose) polymerase (Sims et al., 1982), an NAD⁺-consuming nuclear enzyme implicated in DNA repair (reviewed in de Murcia et al., 1997). Our earlier study not only confirmed that 6AN had these effects but also ruled out these biochemical changes as potential explanations for 6AN-induced enhancement of cisplatin toxicity (Budihardjo et al., 1998). In particular, 6AN enhanced cisplatin accumulation and formation of Pt-DNA adducts, whereas other agents that depleted glutathione, diminished NAD⁺ levels, or inhibited poly(ADP-ribose) polymerase did not.

Experiments performed in the present study examined the possibility that effects on formation of Pt-DNA adducts might be due to 6AN-induced changes in expression of one or more polypeptides. After 6AN treatment for 16 to 24 h, increased labeling of at least five polypeptides was observed (Fig. 3). One of these comigrated with GRP78, a glucose-regulated heat shock family member that is localized in the lumen of the endoplasmic reticulum (Lee, 1992). While this work was in progress, Berger and coworkers reported that two different GRP78 inducers, 6AN and 2-deoxyglucose, each enhanced cisplatin sensitivity (Chatterjee et al., 1997). These data prompted the suggestion that GRP78 levels might affect cisplatin-induced DNA crosslinks and cytotoxicity (Chatterjee et al., 1997). Although it is conceivable that enhanced GRP78 expression and increased cisplatin sensitivity might be mechanistically linked in some cell lines, several observations suggest that this is not the case in A549 cells. First, dose-response curves indicate that cells are sensitized to cisplatin at 6AN concentrations that have small effects on GRP78 levels (Fig. 4). Second, treatment of cells with high 6AN concentrations (e.g., 250 µM) results in increased cisplatin accumulation and cytotoxicity beginning in ~6 h (Fig. 5) even though GRP78 is not elevated at this time (Fig. 5, inset). Finally, cycloheximide prevents 6AN-induced increases in expression of GRP78 (I.I.B. and S.H.K., unpublished observations) but does not prevent the effect of 6AN on formation of Pt-DNA complexes (Fig. 5A) or cytotoxicity (Fig. 5B).

The observations with cycloheximide and other protein synthesis inhibitors are difficult to reconcile with the hypothesis that any of the 6AN-induced polypeptides (Fig. 3) play a critical role in altering the accumulation and cytotoxicity of cisplatin in A549 cells. Instead, the present results are consistent with a model (Fig. 8) in which 6AN treatment results in inhibition of protein synthesis; inhibition of protein synthesis causes enhanced uptake of a variety of compounds, including cisplatin; and enhanced accumulation of cisplatin results in increased formation of Pt-DNA adducts, which in turn results in increased cytotoxicity. This model is supported by a variety of observations.

View larger version (10K): [in this window] [in a new window]   Fig. 8.   Proposed mechanism by which 6AN enhances cisplatin cytotoxicity.

First, we observed that 6AN inhibits protein synthesis by 70 to 80% (Fig. 7A) at concentrations that sensitize cells to cisplatin (Fig. 4). In contrast, 18 inactive 6AN analogs failed to inhibit protein synthesis by more than 30% (Fig. 7B). The effects of 6AN on protein synthesis, like the effects on Pt-DNA adducts, were antagonized by nicotinamide (Fig. 7C). All of these observations provide support for a link between inhibition of protein synthesis and modulation of cisplatin action. Although the suggestion that 6AN inhibits protein synthesis (Fig. 7A) appears to conflict with the enhanced expression of a subset of cellular polypeptides (Fig. 3), there is ample precedent for this observation. Studies examining stress-induced proteins have indicated that a number of cellular stresses (e.g., heat shock) simultaneously up-regulate expression of certain genes (e.g., heat shock protein genes) and inhibit protein synthesis (Lindquist and Craig, 1988). The net effect on cellular levels of these polypeptides reflects the increased abundance of message encoding these polypeptides vis á vis the diminished overall rate of translation. 6AN appears to exert a similar effect, increasing the expression of genes such as GRP78 (Chatterjee et al., 1995) while simultaneously decreasing overall protein synthesis (Fig. 7A).

Second, our studies establish a correlation between inhibition of protein synthesis and increased Pt-DNA adducts. In addition to 6AN, four different protein synthesis inhibitors had an effect similar to that of 6AN (Figs. 5 and 6). Actinomycin D inhibits protein synthesis by inhibiting transcription; anisomycin inhibits the transpeptidation step; cycloheximide inhibits the peptidyl transferase reaction on ribosomes; and puromycin causes premature chain termination (Moldave, 1985; Pain, 1986). The demonstration that all four of these compounds exert similar effects suggests that inhibition of protein synthesis is the common feature of their action. Additional experiments (IIB, unpublished observations) have demonstrated that a 22-h treatment of A549 cells with 10 mM 2-deoxyglucose also causes decreased protein synthesis and increased formation of Pt-DNA adducts, providing a means of reconciling the results of Chatterjee et al. (1997) with the present model. The observation that other protein synthesis inhibitors, notably sparsomycin, cinnamaldehyde, anguidine, and L-histidinol, also enhance the cytotoxicity of cisplatin (Hromas and Yung, 1986; Dornish et al., 1989; Hofs et al., 1995; Warrington et al., 1996) is likewise consistent with the model presented in Fig. 8, although the effects of the latter agents on Pt-DNA adducts remain to be evaluated.

Third, as indicated in the inset to Fig. 7A, we observed that 6AN treatment resulted in increased uptake of methionine (which is imported by a neutral amino acid transporter) as well as enhanced cisplatin accumulation. The possibility that protein synthesis inhibitors can alter membrane transport processes is well established. Treatment of 3T3-L1 adipocytes with cycloheximide causes a 7-fold increase in 2-deoxyglucose transport without any alteration in the plasma membrane levels of the glucose transporter (Clancy et al., 1991; Harrison et al., 1992). Likewise, cycloheximide rapidly stimulates leucine uptake on the system L amino acid transporter in cultured vascular smooth muscle cells (Low et al., 1994). Because the polypeptide corresponding to the system L amino acid transporter has not been identified, it is not known whether levels of this polypeptide are altered by cycloheximide. Nonetheless, these examples provide precedent for alterations in transport as a consequence of treatment that inhibits protein synthesis. In both cases, the authors postulated that cycloheximide inhibited the synthesis of a short-lived polypeptide that suppressed the activity of the transporters under basal conditions (Clancy et al., 1991; Harrison et al., 1992; Low et al., 1994).

In view of previous claims that diminished accumulation is a potentially important cause of cisplatin resistance (Gately and Howell, 1993), it is tempting to speculate that the addition of protein synthesis inhibitors to platinating agents might provide one means of overcoming cisplatin resistance. Of the anticancer drugs that are currently in clinical use, actinomycin D and L-asparaginase are capable of inhibiting protein synthesis. Each of these agents, however, causes a variety of toxicities in vivo (Chabner and Loo, 1996); and neither is routinely combined with cisplatin. Accordingly, further preclinical studies are required to determine whether the mechanism of biochemical modulation outlined above can be developed into a regimen suitable for testing in the clinical setting.