点击显示 收起
【关键词】 Genomic
Screening the Saccharomyces cerevisiae homozygous diploid deletion library against a sublethal concentration of cisplatin revealed 76 strains sensitive to the drug. As expected, the largest category of deletions, representing 40% of the sensitive strains, was composed of strains lacking genes involved in DNA replication and damage repair. Deletions lacking function of the highly conserved vacuolar H+ translocating ATPase (V-ATPase) composed the category representing the second largest number of sensitive strains. The effect on cell death exhibited by V-ATPase mutants was found to be a general response to various DNA damaging agents as opposed to being specific to cisplatin, as evidenced by sensitivity of the mutants to hydroxyurea (a DNA-alkylating agent) and UV irradiation. Loss of V-ATPase does not affect DNA repair, because double mutants defective for V-ATPase function and DNA repair pathways were more sensitive to cisplatin than the single mutants. V-ATPase mutants are more prone to DNA damage than wild-type cells, indicated by enhanced activation of the DNA damage checkpoint. Vacuole function per se is not cisplatin-sensitive, because vacuolar morphology and vacuolar acidification were unaffected by cisplatin in wild-type cells. V-ATPase also controls cytoplasmic pH, so the enhanced sensitivity to DNA damage may be associated with the drop in pHi associated with V-ATPase mutants. The increased loss in cell viability induced by cisplatin at lower pH in V-ATPase mutants supports this hypothesis. The loss in viability seen in wild-type cells under the same conditions was far less dramatic.
Cisplatin [cis-diamminedichloroplatinum(II)] is one of the most widely used anticancer drugs. Platinum-based chemotherapy cures most cases of advanced testicular cancer and has high efficacy in the treatment of other solid tumors, such as ovarian and small-cell lung cancers. The principal cytotoxic mechanism associated with cisplatin is the generation of platinum-DNA adducts, the most significant DNA lesions being 1,2-intrastrand cross-links that form across adjacent guanines (Wang and Lippard, 2005).
Unfortunately, acquired resistance to cisplatin can limit therapeutic potential (Perez, 1998). There are several resistance mechanisms, including decreased intracellular drug accumulation, enhanced cellular detoxification by glutathione and metallothionein, altered DNA repair, and inhibition of apoptosis (Perez, 1998; Huang et al., 2005). However, these mechanisms do not completely account for the observed in vivo unresponsiveness of certain tumors to cisplatin (Niedner et al., 2001; Schenk et al., 2003). Therefore, novel pathways mediating cisplatin resistance exist.
Use of model organisms, such as the yeast Saccharomyces cerevisiae, has been instrumental in revealing the molecular basis of cisplatin toxicity. Complex systems driving signal transduction, DNA repair, and the cell cycle are all highly conserved throughout the eukaryotic lineage. The range of mechanisms that can be probed using S. cerevisiae are those relating to maintenance of viability at the unicellular as opposed to the multicellular level. However, conclusions drawn from studies using these organisms are unambiguous because of the ability to disrupt expression of individual genes. Not surprisingly, these studies confirm the importance of DNA repair pathways including NER, RER, and PR (Grossmann et al., 2001; Beljanski et al., 2004; Wu et al., 2004). A significant advantage conferred by the use of S. cerevisiae genome-wide resources has been the identification of new genes not associated with DNA repair that mediate response to cisplatin. For example, elevated levels of phosphodiesterase 2 and the transcription factor Cin5 confer resistance to cisplatin (Burger et al., 2000; Furuchi et al., 2001). Cells lacking the serine/threonine kinase Sky1, the copper transporter Ctr1, and the nitrogen permease regulator Npr2, are also resistant to cisplatin (Ishida et al., 2002; Schenk et al., 2003, 2004). These strategies, exploiting gene overexpression from genomic libraries or transposon-mediated gene disruption, do not provide comprehensive coverage of the entire genome, because overexpression libraries rarely include all genes, and transposon insertion libraries do not disrupt genes in a random manner. Therefore, we screened the entire set of 4728 homozygous deletion strains, representing deletion of all nonessential open reading frames in S. cerevisiae, to identify genes that lead to sensitivity to cisplatin. A similar screen found 130 deletion strains that were sensitive to cisplatin (Wu et al., 2004). This screen, however, involved mixing all deletions followed by exposure to cisplatin and extraction of genomic DNA. Molecular barcodes identifying each deletion were amplified, and hybridized to an oligonucleotide array, enabling abundance of each deletion strain to be determined (Wu et al., 2004). Cisplatin itself is a potent DNA damaging agent. As a result, a DNA barcode associated with a strain that is sensitive to cisplatin might not be amplified, because it has sustained excessive damage. This would be the case in barcodes bearing consecutive guanines, given the nature of cisplatin-induced DNA damage. Lack of amplification is more likely in strains most sensitive to the drug. To avoid this problem, we individually assessed each deletion, as described previously in screens for strains sensitive to the DNA-alkylating agent methyl methanesulfonate and tirapazamine, a topoisomerase II inhibitor (Chang et al., 2002; Hellauer et al., 2005). Our approach was vindicated because we present identification of 49 cisplatin-sensitive deletions not identified by the screen involving amplification of molecular barcodes. Among the deletions novel to our screen were eight strains that each lacked a component of the highly conserved vacuolar H+ translocating ATPase (V-ATPase). Work in mammalian systems correlates V-ATPase activity with the response to cisplatin. Genes encoding subunits of the V-ATPase are induced when human cell lines are treated with cisplatin and are up-regulated in cisplatin-resistant cell lines (Murakami et al., 2001; Torigoe et al., 2002). Many genes are up-regulated in tumors, although not all of the corresponding proteins play a role in tumor progression. Our data suggest that correlation between V-ATPase function and cisplatin sensitivity underpins a key relationship between the enzyme and sensitivity to the drug. Furthermore, we showed that reduction in V-ATPase activity increased susceptibility to DNA damage per se, explaining why V-ATPase inhibitors render tumors more sensitive to DNA-damaging agents in general.
Yeast Strains and Media. The homozygous diploid deletion strains and individual haploid deletion strains were made by the Saccharomyces Gene Deletion Project (Winzeler et al., 1999). They were obtained from the European Saccharomyces cerevisiae Archive for Functional Analysis (Frankfurt, Germany). The parental diploid strain BY4743 was used as control in the screening of the deletion library for sensitivity to cisplatin. Strain BY4742 was used as the control for work using haploid strains. Genes encoding Vma6 or Vma8 subunits of V-ATPase were deleted in the BY4742 background using a Hygromycin B resistance gene as selectable marker, as described by Goldstein and McCusker (1999). Combinations of mutants in the same strain were constructed by standard procedures (Rose et al., 1990). The genotypes of strains used in this study are listed in Table 1. Yeast was grown in YPD (1% yeast extract, 2% peptone and 2% dextrose) prepared as described in Rose et al., 1990. Cisplatin, HU, and MMS were obtained from Sigma (St. Louis, Missouri). Cisplatin stock solutions were prepared in YPD or phosphate-buffered saline (PBS) and stored as aliquots at -20°C.
TABLE 1 S. cerevisiae strains used in this study.
Screen for Cisplatin-Sensitive Yeast Strains. We determined that the sublethal concentration of cisplatin required to result in visibly slower growth of the parental yeast diploid strain (BY4743) was 250 µg/ml. Deletion strains were maintained by growth as arrays of 384 colonies on solid YPD. All replications were automated and were carried out using a 384-pin replicator operated by a Biomek FX Laboratory Automation Workstation (Beckman Coulter, Fullerton, CA). Colonies were picked and resuspended in 50 µl of YPD in 384-well plates and incubated for 48 h at 30°C (to stationary phase). Each culture (5 µl) was transferred to fresh 50 µl of YPD in 384-well plates and grown to log-phase (14 h at 30°C with intermittent agitation). Replicas of these plates were made on solid YPD media with and without cisplatin (250 µg/ml), followed by incubation at 30°C. During incubation for 3 days, growth was scored by colony size compared with that of the wild-type strain BY4743 (as described in Hellauer et al., 2005). Mutants showing a significant growth defect or absence of growth after 1 day in the presence of 250 µg/ml cisplatin were scored as "x" or "xx." Mutants showing a significant growth defect or absence of growth after 3 days in the presence of 250 µg/ml cisplatin were scored as "xxx" or "xxxx."
Assessing Sensitivity or Viability of Individual Strains to Cisplatin. Cultures were grown in liquid YPD at 30°C to exponential phase and diluted to equal cell density. Six-fold serial dilutions were spotted across YPD, or YPD containing the concentration of cisplatin indicated in figure legends, followed by incubation at 30°C for 3 days. For viability assays, equal cell densities of exponential phase cultures were incubated in PBS, pH 7.4, containing the concentration of cisplatin indicated in figure legends; aliquots were removed after 1 h, diluted, and spread over YPD plates. Percentage viability was determined from the number of colonies that appeared after incubation for 3 days at 30°C, as a percentage of the number of colonies that appeared for each strain without cisplatin treatment.
Sensitivity to DNA Damaging Agents Other than Cisplatin. Cultures were grown at 30°C to exponential phase and diluted to equal cell density (1 x 107 cells/ml). Four successive 6-fold serial dilutions were spotted across YPD or YPD plus 20 mM hydroxyurea (HU) or 0.01% methyl methanesulfonate (MMS). For assessing sensitivity to UV irradiation, dilutions spotted across YPD were exposed to 40 J/m2 UV using a UV cross-linker (Syngene, Frederick, MD). All plates were subsequently incubated for 3 days at 30°C.
Vacuole Staining. FM4-64 was obtained from Invitrogen (Carlsbad, CA) and quinacrine from Sigma (St. Louis, MO). Staining with FM4-64 was performed as described by Conibear and Stevens (2002). FM4-64 was loaded into cells at 40 µM for 15 min followed with one wash to remove free dye and a chase period for 60 min at 30°C in YPD, or YPD with 100 µg/ml cisplatin. Cells were examined by confocal laser microscopy using Texas Red filters. Staining with 200 µM quinacrine (in the presence or absence of 100 µg/ml cisplatin) was performed as described by Roberts et al. (1991). Once stained, cells were visualized within 10 min by confocal laser microscopy using fluorescein filters.
Western Blot Analysis of Rad53. Exponential phase cultures grown at 30°C were incubated for 1 h in the presence of cisplatin at concentrations indicated in the figure legends. Yeast protein extracts were prepared from TCA-treated cells, and Rad53 was detected with a rabbit polyclonal antibody (Tercero et al., 2003).
Effect of pH on Cisplatin Sensitivity. Equal cell densities of exponential phase cells were incubated for 1 h at 30°C in PBS (pH 7.4 or 5.8) containing cisplatin, at the concentrations of drug indicated in figure legends, followed by dilution and inoculation over YPD plates. Percentage viability was determined from the number of colonies that appeared after incubation for 3 days at 30°C and expressed as a percentage of the number of colonies that appeared for each strain without cisplatin treatment.
Screen for Cisplatin-Sensitive Yeast Strains. Hypersensitive strains revealed by screening with a sublethal concentration of cisplatin would reveal the identities of genes that normally function in pathways sensitive to this drug. We performed a high-throughput robot-aided screen for sensitivity to 250 µg/ml cisplatin, using a collection of 4728 homozygous diploid yeast deletion mutants corresponding to nonessential yeast genes. Sensitive mutants were scored by comparison with colony size of the isogenic wild-type strain, after incubation for 1 day. The same colonies were scored again after a further 2-day incubation. Overall, 76 strains were sensitive to cisplatin (Table 2). The sensitivity of these mutants was confirmed by individually pinning serial dilutions of deletion strains on 250 µg/ml cisplatin and control plates.
TABLE 2 Cisplatin-sensitive deletion strains
Mutants showing a significant growth defect or absence of growth after 1 day in the presence of 250 µg/ml cisplatin scored as "x" or "x x." Mutants showing a significant growth defect or absence of growth after 3 days in the presence of 250 µg/ml cisplatin scored as "x x x" or "x x x x."
Four of the seventy-six strains, however, correspond to deleted open reading frames that overlap with other genes. These are the YLR235c, YGL167c, YOR331c, and YKL118w deletions, which overlap with TOP3, HUR1, VMA4, and VPH2, respectively. Two of the overlapped genes (TOP3 and HUR1) are involved in DNA repair. The other two are involved in V-ATPase structure and function. In each case, it is likely that the resistance is due to deletion of the overlapping gene (TOP3, HUR1, VMA4, and VPH2), given 1) the well established importance of DNA repair mechanisms in the processing of cisplatin-DNA adducts, and 2) the importance of V-ATPase activity in resistance to cisplatin, demonstrated in the work presented here and inferred through correlative studies by others (Murakami et al., 2001; Torigoe et al., 2002). The genes are presented in categories based on known or inferred function (Table 2).
Exposure to cisplatin leads to DNA damage, so it was not surprising to find that the largest category, representing 42% of the sensitive strains, was composed of 31 strains deleted for genes involved in DNA replication and damage repair. These fell mainly into the following categories of DNA repair: NER (six genes, including RAD1 and RAD2), RER (eight genes, including RAD51 and RAD52); replication-dependent repair (RR; eight genes, including MUS81 and POL32), and PR (three genes, including RAD18). Deletions of two DNA helicases (HPR5 and SGS1) and one topoisomerase (TOP3) were also sensitive. Twenty-three (of the 31 strains) were reported previously to be cisplatin-sensitive (Birrell et al., 2002; Wu et al., 2004); and seven were reported to be sensitive to tirapazamine, another anticancer drug that causes DNA damage (Hellauer et al., 2005). Deletion of genes in the following functional categories also gave rise to cisplatin sensitivity: cell cycle (10 genes: Table 2), cell stress, and signal transduction (YDJ1, PPH3, SOD1, ZUO1, SEP1, and HAL5), protein synthesis (EAP1, MRF1, RPL13B, RPL7A, and SRO9), transcription (RPB9 and HCM1), and two transporters (TPK1 and GUP1).
The category represented by the second largest number of cisplatin-sensitive deletions was composed of strains lacking function of the vacuolar H+ translocating ATPase (V-ATPase). The disruption of V-ATPase function is lethal in all eukaryotic organisms except S. cerevisiae (Graham et al., 2003). The lethality in higher systems is connected to aspects of V-ATPase function that are more significant in such organisms than they are in a unicellular organism like yeast, such as receptor-mediated endocytosis. Yeast has one hormone receptor, the action of which is nonessential; this is in contrast to various essential receptors in membranes of complex organisms (Graham et al., 2003). This made budding yeast an ideal system to further investigate the role of V-ATPase in cisplatin resistance. We present data to explain why such mutants show more sensitivity to cisplatin as well as other DNA-damaging agents.
V-ATPase Mutants Are Sensitive to Cisplatin. The V-ATPase is composed of thirteen subunits (Graham et al., 2003). Loss of any one of eight V-ATPase subunits led to cisplatin sensitivity (Table 2). Four strains lacked a component of the catalytic (V1) domain responsible for ATP hydrolysis, and the other four lacked a component of the proton translocating (Vo) domain. In addition, deletion of VPS33, a gene required for assembly of V-ATPase (though itself not a part of the enzyme) also led to cisplatin sensitivity. To confirm these results, we generated two strains, one lacking a component of the Vo domain (vma6, Table 1), the other lacking a component of the V1 domain (vma8, Table 1). As expected, both vma6 and vma8 strains exhibited severely impaired growth in the presence of cisplatin compared with growth on media without the drug (Fig. 1A). In a cell survival assay, rapid loss of viability was exhibited by these mutants in the presence of cisplatin (Fig. 1B).
Fig. 1. Sensitivity of the V-ATPase mutants to cisplatin. A, wild-type and V-ATPase mutants were grown at 30°C to exponential phase and diluted to equal cell density. Six-fold serial dilutions were spotted across YPD or YPD containing 150 µg/ml cisplatin; plates were incubated at 30°C for 3 days. B, exponential phase wild-type and V-ATPase mutants diluted to equal cell density were incubated in PBS, pH 7.4, containing cisplatin at the concentration indicated; aliquots were removed after 1 h, diluted, and spread over YPD plates. Viability was determined from the number of colonies that appeared after incubation at 30°C for 3 days-as a percentage of the number of colonies that appeared for each strain without cisplatin treatment.
Vacuolar Morphology and Acidity Are Not Disrupted by Cisplatin. The vacuole of budding yeast is required for protein turnover, nutrient recycling, osmoregulation, storage of amino acids and inorganic phosphate, and maintenance of cytoplasmic pH (Graham et al., 2003). The vacuole (or an aspect of vacuolar function) may be a target of cisplatin, in that loss of V-ATPase function rendered cells hypersensitive to the drug. To investigate this possibility, we assessed the effect of cisplatin on vacuole morphology and vacuole acidification. In general, one to five vacuoles can be visualized in wild-type cells when stained with the fluorescent dye FM4-64 (Conibear and Stevens, 2002). Many mutants defective for vacuole function exhibit aberrant morphology of the vacuole itself (Raymond et al., 1992). Cisplatin at 100 µg/ml had no effect on the viability of wild-type cells but clearly led to loss of viability in the vma6 strain (Fig. 1B). However, under these conditions, vacuole morphology was unaltered in either strain, appearing normal in size, shape, and number (Fig. 2A). Even though cisplatin does not affect vacuolar morphology, it could disrupt vacuole acidification. The lumen of the vacuole is more acidic than the surrounding cytoplasm, so the resulting luminal pH drives numerous vacuole-associated processes (Graham et al., 2003). Cells were stained with quinacrine, a fluorescent weak base that accumulates only in the vacuolar lumen upon acidification (Roberts et al., 1991). Intensity of fluorescence is directly proportional to the degree of vacuolar acidification. Wild-type cells treated with up to 200 µg/ml cisplatin accumulate the same amount of quinacrine as cells incubated without the drug (Fig. 2B). As expected, vma6 cells could not be stained with quinacrine because of the failure of V-ATPase-mediated vacuolar acidity (Fig. 2B). These data indicate that vacuoles in wild-type cells are acidified and morphologically normal in cells treated with cisplatin, suggesting that the vacuole per se is not the target of cisplatin.
Fig. 2. Vacuolar morphology and acidity are not disrupted by cisplatin. Wild-type and V-ATPase mutants were grown at 30°C to exponential phase and diluted to equal cell density, followed by incubation with cisplatin at the concentration indicated. Cells were stained with FM4-64 for visualization of vacuolar morphology (A) or quinacrine for assessing the acidity of the vacuole (B).
Fig. 3. Sensitivity of V-ATPase mutants to various DNA damaging agents. Wild-type and V-ATPase mutants were grown at 30°C to exponential phase and diluted to equal cell density. Six-fold serial dilutions were spotted across YPD, YPD containing 20 mM HU or 0.01% MMS, or YPD followed by exposure to 40 J/m2 UV; plates were incubated at 30°C for 3 days.
V-ATPase Mutants Are Hypersensitive to Other DNA-Damaging Agents. Cisplatin hypersensitivity in V-ATPase mutants could be due to increased levels of DNA damage, or may diminish the effectiveness of DNA repair mechanisms. We assessed the response of V-ATPase mutants to various DNA-damaging agents. MMS is a DNA-alkylating agent, and HU is a DNA replication inhibitor, giving rise to stalled replication forks that are sensed by the cell as abnormal DNA structures (Tercero et al., 2003). vma6 and vma8 strains are both hypersensitive to each agent (Fig. 3). The vacuole itself can act as a detoxification mechanism by accumulation of small molecules via the endocytic machinery. The vma6 and vma8 strains, however, are both also hypersensitive to UV irradiation compared with the wild type (Fig. 3). Therefore, it is likely that the increased sensitivity to cisplatin, HU, and MMS exhibited by V-ATPase mutants is not due to a defect in a mechanism that involves vacuolar sequestration of these cytotoxic agents. Instead, DNA could be more prone to damage in these cells or repair mechanisms may be less efficient.
Assessing Sensitivity of V-ATPase/DNA Repair Double Mutants to Cisplatin. The cisplatin-induced DNA cross-linked adducts in S. cerevisiae are repaired mainly by three DNA repair mechanisms: the NER, RER, and PR pathways. To determine the possibility that loss of V-ATPase would diminish the effectiveness of these repair pathways, we performed epistasis analysis using V-ATPase/DNA repair double mutants. RAD1, encoding a single-stranded DNA endonuclease, is a classic NER gene. REV3 encodes a subunit of DNA polymerase zeta, which is involved in PR. RAD52 encodes a protein that stimulates strand exchange during RER. Double mutants were constructed that were defective for V-ATPase function and one DNA repair pathway. All three double mutants (vma6rad1, vma6rev3, and vma6rad52) were more sensitive to cisplatin than the single mutants (Fig. 4). This suggests that involvement of V-ATPase with DNA damage sensitivity is independent of DNA repair pathways. Instead, V-ATPase mutants could be more sensitive to cisplatin simply because more damage is caused by a given concentration of drug.
Fig. 4. Sensitivity of [V-ATPase/DNA repair] double mutants to cisplatin. Exponential phase wild type and single and double mutants were diluted to equal cell density and incubated in PBS, pH 7.4, containing cisplatin at the concentration indicated; aliquots were removed after 1 h, diluted, and spread over YPD plates. Viability was determined from the number of colonies that appeared after incubation at 30°C for 3 days and expressed as a percentage of the number of colonies that appeared for each strain without cisplatin treatment.
Activation of the DNA Damage Checkpoint in V-ATPase Mutants. When yeast cells are treated with cisplatin, the DNA damage checkpoint is activated, leading to cell cycle arrest at G2/M. Mutants in components of the checkpoint display increased sensitivity to cisplatin (Grossmann et al., 1999). The sensitivity to cisplatin exhibited by strains lacking the S-phase checkpoint protein MRC1 is in agreement with this (Table 2). Rad53, the effector protein kinase in this pathway, is activated via hyperphosphorylation in response to DNA damage (Tercero et al., 2003). Phosphorylated Rad53 is detected as a smear of slowly migrating forms of the protein on Western blots immunoprobed with anti-Rad53 antisera. We incubated cells in the presence of various concentrations of cisplatin for 1 h. The incubation of both wild-type and vma6 cells with 200 µg/ml cisplatin led to the appearance of hyperphosphorylated forms of Rad53, suggesting that lack of V-ATPase activity did not impair the DNA damage checkpoint pathway (Fig. 5). A single band that indicated the nonphosphorylated form of Rad53 was detected in the absence of the drug, in both wild-type and vma6 cells (Fig. 5). In vma6 cells, however, hyperphosphorylated Rad53 was detected on incubation with 50 µg/ml cisplatin. In contrast, hyperphosphorylated Rad53 was barely detectable in wild-type cells treated in the same way, suggesting that V-ATPase mutants suffered more DNA damage than the wild-type cells.
Fig. 5. Phosphorylation of the Rad53 checkpoint effector kinase. Wild-type or V-ATPase mutant cells were grown to early log phase at 30°C and then incubated for 1 h with cisplatin at the concentration indicated. Rad53 was visualized by probing Western blots with anti-Rad53 antibody. The positions of unphosphorylated Rad53 and phosphorylated Rad53 (Rad53-p) are shown on the right.
Effect of pH on Cisplatin Sensitivity. One of the major functions of V-ATPase is maintenance of intracellular pH by translocating protons from the cytosol into the lumen of the vacuole (Graham et al., 2003). In both unicellular and multicellular organisms, cytoplasmic pH is reduced when function of the V-ATPase is compromised (Moreno et al., 1998; Murakami et al., 2001). The increased DNA damage sensitivity in the V-ATPase mutants could be due to the acidified cytoplasm. To investigate the effect of pH on cisplatin sensitivity, we followed viability of cells after incubation with cisplatin in neutral (pH 7.4) or acidic (pH 5.8) conditions. Wild-type yeast cells are efficient at maintaining a constant intracellular pH, even when they are incubated in media buffered to varying pH. This is due principally to the action of a plasma membrane H+-translocating ATPase. This enzyme pumps H+ out of the cell, consuming up to 40% of total cellular ATP. To remove the possibility that action of this enzyme would compensate for the effect of incubating cells at varying pH, we incubated cells in the absence of a carbon source so that cells could not make ATP to drive H+ extrusion by PM ATPase. Intracellular pH of S. cerevisiae incubated under these conditions closely approaches the pH of the extracellular medium, with intracellular pH changes being due to H+ leakage across the plasma membrane (Brett et al., 2005).
Both wild-type and vma6 cells were more sensitive to cisplatin when incubated at lower pH. The effect of low pH, however, was more dramatic in vma6 cells (Fig. 6). In the presence of 20 µg/ml cisplatin, the viability of vma6 cells at pH 5.8 was decreased by 76% (by 24.% at neutral pH), whereas the viability of the wild-type cells at pH 5.8 was decreased by 25% (by 10% at neutral pH). This implies that lower cytoplasmic pH increases sensitivity to DNA damage in V-ATPase mutants.
Fig. 6. Effect of pH on cisplatin sensitivity. Exponential phase cultures diluted to equal cell density were incubated in PBS, pH 7.4 or 5.8, with cisplatin at the concentration indicated; aliquots were removed after 1 h, diluted, and spread over YPD plates. Percentage viability was determined from the number of colonies that appeared after incubation at 25°C for 3 days, expressed as a percentage of the number of colonies that appeared for each strain without cisplatin treatment.
Pathways that modulate cisplatin sensitivity could be induced in tumor cells, and development of agents to inhibit these pathways can overcome the resistance that emerges frequently during treatment. Development of new platinum drugs by substitution of ligands or chloride, leaving groups of cisplatin has not been successful in terms of overcoming drug resistance. New approaches can be developed based on discovery of mechanisms that mediate toxicity of this drug.
The budding yeast S. cerevisiae has been used as a powerful tool to identify and investigate pathways targeted by drugs. In the present study, we have applied a systematic approach to search for the nonessential genes in yeast that play a role in the response to cisplatin, leading to the identification of 76 deletion strains sensitive to this drug.
More than 40% of the cisplatin-sensitive strains lacked various components involved in DNA damage repair. The deleted genes were mainly involved in three DNA repair pathways: RR and PR, RER, and NER. However, none of the components involved in base excision repair or mismatch repair were identified in this screen, suggesting that PR, RR, and NER are the main mechanisms by which cells repair cisplatin-induced DNA damage, whereas base excision repair and mismatch repair are not involved. This is in agreement with previous studies (Grossmann et al., 2001; Beljanski et al., 2004; Wu et al., 2004).
A novel insight from our screen is that sister chromatid cohesion plays a role in the response to cisplatin. Dcc1, Ctf8, and Ctf18 are components of a complex required for establishment of sister chromatid cohesion. Cells lacking the genes encoding any of these three proteins are sensitive to cisplatin. Cells lacking another gene involved in cohesion (Ctf4) are affected in the same way. Recruitment of the cohesin complex to sites of DNA damage is necessary for recombination-mediated repair of double-strand breaks (Strom et al., 2004). Therefore, it is likely that the Dcc1/Ctf8/Ctf18 complex and Ctf4 facilitate RER of cisplatin-induced DNA damage. In addition, two mutants (asf1 and mrc1), known to impair the Rad53-dependent DNA damage checkpoint pathway, were sensitive to cisplatin. This is in agreement with work showing that cisplatin causes a checkpoint-dependent G2/M arrest (Grossmann et al., 1999). To support this notion, we showed that cisplatin resulted in hyperphosphorylation of Rad53, indicating the activation of this checkpoint pathway (Fig. 5).
Several mutants that compromised cell stress tolerance and signal transduction were sensitive to cisplatin. In mammalian cells, the oxidative and osmolar stress responses protect cells from cisplatin-induced nephrotoxicity (Hanigan et al., 2005). In addition, genes involved with ribosomal function and protein synthesis were identified as cisplatin-resistant genes. In agreement with this are reports describing cisplatin-induced disruption of the translation initiation complex and overexpression of a ribosomal protein conferring resistance to cisplatin (Rosenberg and Sato, 1993; Shen et al., 2006).
It is noteworthy that a group of nine genes encoding V-ATPase subunits and an assembly factor for this enzyme were identified as cisplatin-hypersensitive strains. It is not surprising that so many genes involved in V-ATPase function were identified, given that loss of any V-ATPase subunit or assembly factor is known to result in loss of V-ATPase activity (Graham et al., 2003). In yeast, this enzyme is localized to the membrane of the vacuole, with a smaller population of V-ATPase complexes localized to the endosomal network. In mammalian cells, the enzyme is similarly localized to the lysosome (equivalent to the yeast vacuole) and endocytic compartments. In addition, the enzyme is localized to the membranes of specialized cells, notably the brush-border membranes of renal proximal tubules (Stevens and Forgac, 1997). In mammalian cells, one of the V-ATPase subunits is induced by cisplatin (Torigoe et al., 2002). Furthermore, several genes encoding V-ATPase subunits are up-regulated in drug-resistant tumor cell lines (Martinez-Zaguilan et al., 1999; Murakami et al., 2001). This increase in levels of V-ATPase in cisplatin-resistant cells is correlative, and could be a consequence of drug treatment rather than a cause of resistance. The data we present, however, points toward V-ATPase activity directly contributing to drug tolerance.
V-ATPase translocates H+ from the cytoplasm to the vacuole. Two consequences of this are regulation of cytoplasmic pH and acidification of the vacuole; the latter is crucial for maintenance of processes associated with the vacuole lumen. We showed that cisplatin did not affect the morphology of vacuoles in wild-type cells or a V-ATPase mutant. Furthermore, acidification of vacuoles in wild-type cells was also unaffected. This implied that sensitivity to cisplatin exhibited by V-ATPase mutants was associated with an effect on processes outside the vacuole, such as DNA repair, or the extent to which cisplatin damages DNA in the first place. V-ATPase mutants were also sensitive to the DNA-alkylating agent MMS, the DNA replication inhibitor HU, and UV irradiation. This suggested that the activity of the V-ATPase was required for limiting the effects of DNA damaging agents in general. This concept is supported by a recent report describing the sensitivity of yeast V-ATPase mutants to tirapazamine, an anticancer drug that targets topoisomerase II (Hellauer et al., 2005).
V-ATPase mutants may be defective in DNA damage repair mechanisms or may lead to enhanced DNA damage. PR, RR, and NER are the mechanisms that repair damage induced by cisplatin. Epistasis analysis ruled out the possibility that a defect in V-ATPase diminished the effect of repair pathways, because [V-ATPase/repair pathway] double mutants were far more sensitive to cisplatin than the single mutants (Fig. 4). Activation of the DNA damage checkpoint by low concentrations of cisplatin was enhanced in V-ATPase mutants. This suggested that loss of the V-ATPase function facilitates the DNA damage caused by cisplatin.
V-ATPase is a regulator of cytoplasmic pH. Consequently, loss of V-ATPase activity leads to intracellular acidification, which may lead to greater levels of cisplatin-mediated DNA damage. At lower pH, a greater proportion of hydrolyzed cisplatin has an aqua ligand, rather than a hydroxoligand. This enhances the reactivity of cisplatin because the aquated form is more labile. Chemical activity of cisplatin in vitro is greater at lower pH, promoting DNA platination (Murakami et al., 2001). The increased sensitivity to cisplatin in a V-ATPase mutant incubated at lower pH supports this hypothesis. At 50 µg/ml cisplatin, it is notable that cell viability of the wild type in pH 5.8 buffer, is similar to cell viability of vma6 at pH 7.4. This may well reflect a similar intracellular pH in these cells.
The role played by lower pH, however, must be more complex than straightforward increases in chemical reactivity of drugs, because V-ATPase mutants were also hypersensitive to UV irradiation. Low pH can change DNA conformation (Robinson et al., 1992). Therefore, sensitivity to DNA damaging drugs or UV irradiation in V-ATPase mutants may be associated with altered DNA conformation at low cytosolic pH, rendering DNA more prone to damage. This may explain why use of V-ATPase inhibitors in human cell lines renders them more sensitive to cisplatin (Laurencot et al., 1995; Murakami et al., 2001; Luciani et al., 2004).
In mammalian cells, cellular acidosis is an early event in apoptosis. Limiting the drop in cytoplasmic pH represses apoptosis, this being frequently associated with an up-regulation of V-ATPase subunits in tumors (Torigoe et al., 2002; Izumi et al., 2003). Accordingly, enhanced cell death in yeast V-ATPase mutants incubated with cisplatin could be due, in part, to activation of apoptosis caused by lowering of intracellular pH. Some controversy does surround the concept of yeast apoptosis. However, yeast demonstrates several markers typical of apoptosis including DNA fragmentation, phosphatidylserine externalization, chromatin condensation, and histone H2B phosphorylation (Madeo et al., 2004; Ahn et al., 2006). Furthermore, a growing list of genes that regulate apoptosis in mammalian cells have been identified in yeast, including the Yca1 caspase and the apoptosis-inducing factor Aif1 (Madeo et al., 2004).
We also demonstrated that loss of four genes associated with ribosome function and protein synthesis rendered cells sensitive to cisplatin. Accordingly, we predict that enhancing protein synthesis would confer drug resistance. This is in agreement with recent work involving use of a human epidermoid carcinoma cell line that showed cisplatin resistance is induced by overexpression of the ribosomal protein gene RPL36 (Shen et al., 2006). Another study revealed that inactivation of the Sky1 kinase in S. cerevisiae leads to cisplatin resistance. Monitoring levels of the orthologous protein in testicular tumors revealed that expression of the kinase in cisplatin resistant cells are lower than in tumors from patients who were responding to platinum drug-based therapy, so levels of the protein could predict the response to the drug (Schenk et al., 2004). Such work clearly demonstrates that conclusions drawn from use of the S. cerevisiae model system are of direct relevance to mammalian systems. This indicates that other genes identified in our screen are worthy of further investigation, because they may be important for predicting the responsiveness of tumors to cisplatin.
Acknowledgements
We thank Drs. Mehdi Mollapour and Nicholas L. Harris (University of Sheffield, Sheffield, UK) for assistance with rearraying the genomic library from 96 to 384 colony format, and we thank Professor Peter W. Piper (University of Sheffield) for allowing access to robotic facilities used for the re-arraying procedure. We are grateful to Dr. John F. X. Diffley (Cancer Research UK, Clare Hall, UK) for providing antisera against Rad53. We also thank Dr. Katherine E. Bowers (University College London, London, UK) for advice pertaining to work with V-ATPase mutants.
ABBREVIATIONS: NER, nucleotide excision repair; RER, recombination dependent repair; PR, postreplication repair; RR, replication-dependent repair; V-ATPase, vacuolar H+-transporting ATPase; YPD, yeast extract/peptone/dextrose; PBS, phosphate-buffered saline; HU, hydroxyurea; MMS, methyl methanesulfonate.
【参考文献】
Ahn S, Diaz RL, Grunstein M, and Allis CD (2006) Histone H2B deacetylation at lysine 11 is required for yeast apoptosis induced by phosphorylation of H2B at serine 10. Mol Cell 24: 211-220.
Beljanski V, Marzilli LG, and Doetsch PW (2004) DNA damage-processing pathways involved in the eukaryotic cellular response to anticancer DNA cross-linking drugs. Mol Pharmacol 65: 1496-1506.[Abstract/Free Full Text]
Birrell GW, Brown JA, Wu HI, Giaever G, Chu AM, Davis RW, and Brown JM (2002) Transcriptional response of Saccharomyces cerevisiae to DNA-damaging agents does not identify the genes that protect against these agents. Proc Natl Acad Sci USA 99: 8778-8783.[Abstract/Free Full Text]
Brett CL, Tukaye DN, Mukherjee S, and Rao R. The yeast endosomal Na+(K+)/H+ exchanger Nhx1 regulates cellular pH to control vesicle trafficking. Mol Biol Cell 16: 1396-1405, 2005.[Abstract/Free Full Text]
Burger H, Capello A, Schenk PW, Stoter G, Brouwer J, and Nooter K (2000) A genome-wide screening in Saccharomyces cerevisiae for genes that confer resistance to the anticancer agent cisplatin. Biochem Biophs Res Commun 269: 767-774.
Chang M, Bellaoui M, Boone C, and Brown GW (2002) A genome-wide screen for methyl methanesulfonate-sensitive mutants reveals genes required for S phase progression in the presence of DNA damage. Proc Natl Acad Sci USA 99: 16934-16939.[Abstract/Free Full Text]
Conibear E and Stevens TH (2002) Guide to yeast genetics and molecular and cell biology part C. Methods Enzymol 351: 408-432.
Furuchi T, Ishikawa H, Miura N, Ishizuka M, Kajiya K, Kuge S, and Naganuma A (2001) Two nuclear proteins, Cin5 and Ydr259c, confer resistance to cisplatin in Saccharomyces cerevisiae. Mol Pharmacol 59: 470-474.[Abstract/Free Full Text]
Goldstein AL and McCusker JH (1999) Three new dominant drug resistance cassettes for gene disruption in Saccharomyces cerevisiae. Yeast 15: 1541-1543.
Graham LA, Flannery AR, and Stevens TH (2003) Structure and assembly of the yeast ATPase. J Bioenerg Biomembr 35: 301-312.
Grossmann KF, Brown JC, and Moses RE (1999) Cisplatin DNA cross-links do not inhibit S-phase and cause only a G2/M arrest in Saccharomyces cerevisiae. Mutat Res 434: 29-39.
Grossmann KF, Ward AM, Matkovic ME, Folias AE, and Moses RE (2001) S. cerevisiae has three pathways for DNA interstrand crosslink repair. Mutat Res 487: 73-83.
Hanigan MH, Deng M, Zhang L, Taylor PT Jr, and Lapus MG (2005) Stress response inhibits the nephrotoxicity of cisplatin. Am J Physiol 288: F125-F132.
Hellauer K, Lesage G, Sdicu AM, and Turcotte B (2005) Large-scale analysis of genes that alter sensitivity to the anticancer drug tirapazamine in Saccharomyces cerevisiae. Mol Pharmacol 68: 1365-1375.[Abstract/Free Full Text]
Huang RY, Eddy M, Vujcic M, and Kowalski D (2005) Genome-wide screen identifies genes whose inactivation confer resistance to cisplatin in Saccharomyces cerevisiae. Cancer Res 65: 5890-5897.[Abstract/Free Full Text]
Ishida S, Lee J, Thiele DJ, and Herskowitz I (2002) Uptake of the anticancer drug cisplatin mediated by the copper transporter Ctr1 in yeast and mammals. Proc Natl Acad Sci USA 99 22: 14298-14302.
Izumi H, Torigoe T, Ishiguchi H, Uramoto H, Yoshida Y, Tanabe M, Ise T, Murakami T, Yoshida T, Nomoto M, et al. (2003) Cellular pH regulators: potentially promising molecular targets for cancer chemotherapy. Cancer Treat Rev 29: 541-549.
Laurencot CM, Andrews PA, and Kennedy KA (1995) Inhibitors of intracellular pH regulation induce cisplatin resistance in EMT6 mouse mammary tumor cells. Oncol Res 7: 363-639.
Luciani F, Spada M, De Milito A, Molinari A, Rivoltini L, Montinaro A, Marra M, Lugini L, Logozzi M, Lozupone F, et al. (2004) Effect of proton pump inhibitor pretreatment on resistance of solid tumors to cytotoxic drugs. J Natl Cancer Inst. 96: 1702-1713.[Abstract/Free Full Text]
Madeo F, Herker E, Wissing S, Jungwirth H, Eisenberg T, and Frohlich KU (2004) Apoptosis in yeast. Curr Opin Microbiol 7: 655-680.
Martinez-Zaguilan R, Raghunand N, Lynch RM, Bellamy W, Martinez GM, Rojas B, Smith D, Dalton WS, and Gillies RJ (1999) pH and drug resistance. I. Functional expression of plasmalemmal V-type H+-ATPase in drug-resistant human breast carcinoma cell lines. Biochem Pharmacol 57: 1037-1046.
Moreno SNJ, Zhong L, Lu HG, de Souza W, and Benchimol M (1998) Vacuolar-type H+-ATPase regulates cytoplasmic pH in Toxoplasma gondii tachyzoites. Biochem J 380: 853-860.
Murakami T, Shibuya I, Ise T, Chen ZS, Akiyama S, Nakagawa M, Izumi H, Nakamura T, Matsuo K, Yamada Y, et al. (2001) Elevated expression of vacuolar proton pump genes and cellular pH in cisplatin resistance. Int J Cancer 93: 869-874.
Niedner H, Christen R, Lin X, Kondo A, and Howell SB (2001) Identification of genes that mediate sensitivity to cisplatin. Mol Pharmacol 60: 1153-1160.[Abstract/Free Full Text]
Perez RP (1998) Cellular and molecular determinants of cisplatin resistance. Eur J Cancer 34: 1535-1542.
Raymond CK, Howald-Stevenson I, Vater CA, and Stevens TH (1992) Morphological classification of the yeast vacuolar protein sorting mutants: evidence for a prevacuolar compartment in class E vps mutants. Mol Biol Cell 3: 1389-1402.
Roberts CJ, Raymond CK, Yamashiro CT, and Stevens TH (1991) Guide to yeast genetics and molecular and cell biology part A. Methods Enzymol 194: 644-661.
Robinson H, van der Marel GA, van Boom JH, and Wang AH (1992) Unusual DNA conformation at low pH revealed by NMR: parallel-stranded DNA duplex with homo base pairs. Biochem 31: 10510-10517.
Rose MD, Winston F, and Hieter P (1990) Methods in Yeast Genetics: A Cold Spring Harbor Laboratory Course Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Rosenberg JM and Sato PH (1993) Cisplatin inhibits in vitro translation by preventing the formation of complete initiation complex. Mol Pharmacol 43: 491-497.
Schenk PW, Brok M, Boersma AWM, Brandsma JA, Dulk HD, Brok M, Burger H, Stoter G, Brouwer J, and Nooter K (2003) Anticancer drug resistance induced by disruption of the Saccharomyces cerevisiae NPR2 gene: a novel component involved in cisplatin- and doxorubicin-provoked cell kill. Mol Pharmacol 64: 259-268.[Abstract/Free Full Text]
Schenk PW, Stoop H, Bokemeyer C, Mayer F, Stoter G, Oosterhuis JW, Wiemer E, Looijenga LHJ, and Nooter K (2004) Resistance to platinum-containing chemotherapy in testicular germ cell tumors is associated with down-regulation of the protein kinase SRPK1. Neoplasia 6: 297-301.
Shen D-W, Liang X-J, Suzuki T, and Gottesman MM (2006) Identification by functional cloning from a retroviral cDNA library of cDNAs for ribosomal protein L36 and the 10-kDa heat shock protein that confer cisplatin resistance. Mol Pharmacol 69: 1383-1388.[Abstract/Free Full Text]
Stevens TH and Forgac M (1997) Structure, function and regulation of the vacuolar (H+)-ATPase. Annu Rev Cell Dev Biol. 13: 779-808.
Strom L, Lindroos HB, Shirahige K, and Sjogren C (2004) Postreplicative recruitment of cohesin to double-strand breaks is required for DNA repair. Mol Cell 16: 1003-1015.
Tercero JA, Longhese MP, and Diffley JFX (2003) A central role for DNA replication forks in checkpoint activation and response. Mol Cell 11: 1323-1336.
Torigoe T, Izumi H, Ishiguchi H, Uramoto H, Murakami T, Ise T, Yoshida Y, Tanabe M, Nomoto M, Itoh H, et al. (2002) Enhanced expression of the human vacuolar H+-ATPase c subunit gene (ATP6L) in response to anticancer agents. J Biol Chem 277: 36534-36543.[Abstract/Free Full Text]
Wang D and Lippard SJ (2005) Cellular processing of platinum, anticancer drugs. Nat Rev Drug Discovery 4: 307-320.
Winzeler EA, Shoemaker DD, Astromoff A, Liang H, Anderson K, Andre B, Bangham R, Benito R, Boeke JD, Bussey H, et al. (1999) Functional characterization of the S. cerevisiae genome by gene deletion and parallel analysis. Science (Wash DC) 285: 901-906.[Abstract/Free Full Text]
Wu HI, Brown JA, Dorie MJ, Lazzeroni L, and Brown JM (2004) Genome-wide identification of genes conferring resistance to the anticancer agents cisplatin, oxaliplatin, and mitomycin C. Cancer Res 64: 3940-3948.[Abstract/Free Full Text]
作者单位：Pharmaceutical Science Research Division (C.L., B.H., B.P.) and Genomics Facility (M.J.A.), School of Biomedical and Health Sciences, King's College London, London, United Kingdom