Conceived and designed the experiments: TJW JRM CBB. Performed the experiments: TJW SMW AYG ALS TSW CBB. Analyzed the data: CBB. Contributed reagents/materials/analysis tools: ALG KLB JAOJ. Wrote the paper: JRM CBB.
The authors have declared that no competing interests exist.
The chemotherapeutic doxorubicin (DOX) induces DNA double-strand break (DSB) damage. In order to identify conserved genes that mediate DOX resistance, we screened the
Doxorubicin (DOX) is a highly effective anthracycline chemotherapeutic agent for many solid tumors including those of the breast however, dosage has to be carefully monitored to avoid the potentially life threatening complications associated with cardiotoxicity. Furthermore, in some cases tumors can acquire resistance to DOX greatly reducing its efficacy. In some cases these two factors can severely limit the clinical usage of this class of drugs. The mechanism of cardiotoxicity is unclear but it has been suggested that multiple processes are involved
The ability of tumors to simultaneously develop resistance to many drugs has been termed multidrug resistance (MDR) and frequently occurs following DOX treatment. Potential mechanisms for this acquired resistance include upregulation of transporters that promote drug efflux
To identify highly conserved targets that mediate resistance to DOX, many studies have successfully utilized the genetic accessibility of the model organism
To further elucidate the mechanism of DNA damage resistance in
To test this, we directly compared the relative sensitivity of diploid
Deletions of individual non-essential radiation resistance genes (or ORFs) were made in
A doxorubicin (DOX) stock solution (10 mg/ml in sterile H20) was used to prepare DOX YPD agar plates at two concentrations (25 or 50 µg/ml). DOX was added to cooled YPD agar at the time of pouring and plates were allowed to solidify at room temperature and used immediately for screening the diploid deletion collection. Strains from the frozen deletion collection individually arrayed in 96 well dishes were thawed and aliquots (∼2 µl) were transferred using a multi (48) pin “pronging” device to YPD and YPD DOX. Concomitant with the DOX screen, zymocin screening of the deletion collection was also performed by replica pronging directly from the thawed 96 well dishes onto YPD plates containing 0, 33% or 66% crude zymocin. Zymocin containing YPD plates were made as previously described
Selected strains identified as DOX or zymocin sensitive or zymocin resistant in the primary screen were subsequently confirmed by growing individual isolates (and WT) in 200 µl of YPD in 96 well dishes for two days. These cells were serially diluted (5-fold) in liquid YPD and ∼2 µl of each dilution was replica plated by pronging to either YPD, YPD containing DOX or YPD containing zymocin. Resistance to zymocin was scored following 2 days incubation. Zymocin sensitivity was scored following 3 days incubation at 30°C.
Strains were screened for hydroxyurea (HU) and methyl methane sulfonate (MMS) sensitivity using a similar dilution plating procedure as previously described
Deletion strains were exposed to zymocin on plates either directly from the diploid deletion collection arrayed in 96 well dishes or using the dilution pronging technique described above. Alternatively, selected deletion strains and WT were grown for two days in liquid YPD (filter sterilized) in 96 well plates and serial 5 fold dilutions were made in water. Cells (∼2 µl of each dilution) were replica transferred to YPD and YPD+zymocin plates. YPD plates containing zymocin were made by growing
WT and selected DOX sensitive deletion strains were examined for cell cycle progression following exposure to DOX as previously described
The PCR mediated gene conversion assay utilizing the
WT and various diploid deletion strains were patched from single colony isolates or single colonies themselves grown on YPD plates were mated on fresh YPD plates to mating type tester strains 147 (
We have described a large interactive network of ionizing radiation resistance genes in which the CCR4-NOT complex plays a key role
(A) Isogenic ionizing radiation (IR) sensitive diploid deletion strains were grown at 30°C for two days in liquid YPD medium in 96 well plates. Serial 5-fold dilutions were made in sterile water and 2 µl aliquots were replica plated to YPD solid medium with and without the indicated doses of doxorubicin. Plates were subsequently incubated for 3 days at 30°C. Arrows indicate the direction of decreasing cell concentration. When compared to WT (row 1), defects in genes within the CCR4-NOT complex (rows 2–5) confer checkpoint adaptation functions and show intermediate sensitivity to doxorubicin. Defects in members of the RAD52 recombination repair group (rows 6–8) are required for double strand break repair and are hypersensitive to doxorubicin. (B) Diploid WT,
Deletion of genes within the CCR4 damage response network results in cell cycle checkpoint adaptation defects during the G1 to S phase transition following DSBs or replication stress
Interaction of DOX within mitochondria has been proposed to result in the generation of reactive oxygen species (ROS) that contributes to cellular lethality. To examine whether mitochondrial processing of DOX contributes significantly to lethality and/or checkpoint arrest in diploid yeast we “cured” the WT,
In order to determine if the loss of mitochondrial functions affected cell cycle responses to DOX we examined cell cycle progression of logarithmic diploid WT,
The diploid deletion collection has been useful for identifying ionizing radiation (IR) repair associated genes that function specifically in G1
A total of 376 diploid deletion strains demonstrated either hypersensitivity (n = 209) or reduced growth rate (n = 167) when exposed to DOX. This represents ∼8% of the non-essential genes represented within the diploid deletion collection. Remarkably, this collection of DOX sensitive gene deletions is significantly larger (>5 fold) than that found in a similar screen using the isogenic haploid deletion collection from which only 71 deletion strains were identified (
Zymo |
IR |
G1 size |
Oxid |
Yeast DOX resistance gene | Conserved human ortholog |
S | S | S | S | ASF1, CCR4, DBF2, HFI1, MMS22, POP2, RAD50, RTT109, YDJ1 | ASF1A, CNOT6, STK38L, none, ANKRD12, CNOT8, RAD50, RTT109, HSP40 |
R | ADK1, AKR1, ARP5, BEM1, MDM20, RPB9 | AK2, ZDHHC17, ACTR5, SH3PXD2B, C12orf30, POLR2I | |||
R | S | TOP3, TPS1, YAF9, YEL033W | TOP3A, none, YEATS4, none | ||
R | CLC1, CTF4, DHH1, DOC1, GRR1, GUP1, NOT5, OCH1, RAD51, RAD52, RAD54, TSR2, TUP1, VMA7, XRS2, YLR235C | CLTA, WDHD1, DDX6, ANAPC10, FBXL20, HHATL, CNOT3, none, RAD51, RAD52, RAD54L, TSR2, WDR5, ATP6V1F, NBS1, none | |||
R | S | S | GAL11, IFM1, IMP2', MSE1, MSM1, PEP3, PHO85, RNR4, ROX3, RTS1, SNF5, SUV3, SWI6 | MLL2, MTIF2, SFRS12, EARS2, MARS2, VPS18, CDK2, RRM2, none, PPP2R5D, MLL2, SUPV3L1, AKAP9 | |
R | ADH1, ANP1, BEM4, BUD25, IES6, MIP1, MNN9, MNN10, MSD1, PIN4, RNR1, SHP1, SPT7, TCO89, VPS34 | ADH1B, TNRC6A, none, none, C18orf37, POLG, none, none, DARS2, MLL5, RRM1, NSFL1C, BAZ1A, DSPP, PIK3C3 | |||
R | S | BUD23, ERG4, LST4, PFK26, PGD1, PHO2, PKR1, PTC1, REG1, SNF2, SNF6, SOD1, SWI3, TAT1, VAN1, VMA2, VMA4, YJL175W | WBSCR22, LBR, LOC100133790, PFKFB3, MUC7, PITX1, none, PPM1B, DSPP, SMARCA2, none, SOD1, SMARCC2, SLC7A14, none, ATP6V1B2, ATP6V1E1, none | ||
R | ACO1, BUD16, CCW12, CUP5, DOA4, ERG6, GAS1, HEX3, HOM2, HOM3, HTZ1, KHA1, MSY1, PER1, RRN10, SAC7, SER2, SLM4, NAB6, VPS64, VMA5, YOL050C, YOR331C, YPL205C | ACO2, PDXK, LOC100132635, ATP6VOC, USP8, TGS1, MUC21, HRNR, none, none, H2AFV, TMCO3, YARS2, PERLD1, none, ARHGAP6, PSPH, none, none, SLMAP, ATP6V1C1, none, none, none | |||
R | S | S | S | PAT1, SLX8, YJL188C | PATL1, RNF10, none, |
R | BCK1, FUN12, HPR1, LGE1, NPL3, PLC1, THO2 | MAP3K3, EIF5B, THOC2, FLG, HNRNPR, PLCD4, THOC2 | |||
R | S | RSA1 | AKAP9 | ||
R | ADE12, GON7, LSM7, MMS4, NUP133, RAD55, RAD57, RAD59, VPH2, YDL041W, YDR433W, YKL118W, YML009C-A | ADSSL1, none, LSM7, none, none, RAD51L3, RAD51L1, RAD52, none, none, none, none, none | |||
R | S | S | DBP7, ECM33, MSN5, RPL35A, RPL43A, SAC1, SAC3, SIN3, SSZ1, UAF30 | DDX31, MUC21, XPO5, RPL35, RPL37A, SAC1L, MCM3AP, SIN3A, HSPA8, SMARCD1, | |
R | ASC1, BUD22, CTK3, FYV5, HIT1, KRE6, MET7, OPI11, PRO1, RPL39, RPS10A, | GNB2L1, LOC100133599, none, none, none, DSPP, FPGS, none, ALDH18A1, LOC100133222, RPS10, | |||
R | S | CBC2, GCR2, HAL5, KCS1, LSM1, NSR1, PDR1, RPL27A, RPS4A, RPS11B, SAT4, SIN4, VMA13, YAR1 | NCBP2, MUC21, PRKAA1, IHPK3, LSM1, NCL, none, RPL27, RPS4X, RPS11, CHEK1, none, ATP6V1H, FEM1C | ||
R | AKL1, CKB1, CKB2, CTI6, YPL182C, CTK1, EDC3, EGD1, ERV41, GET1, HEM14, HHF1, MDM35, MMS1, MTQ2, NEW1, NFI1, PSK2, PUS1, PUS7, RDS2, RIS1, RPA49, RPL12A, RPL13B, RPL20B, RPP1A, RTG1, SER1, SPT20, TAF14, TCM62, TFP3, THP1, TRK1, VMA6, YCL007C, YDR049W, YGR160W, YNL140C, YOL046C, YOR152C, YPL260W, YPL261C | AAK1, CSNK2B, CSNK2B, CYLC1, POU2F1, CRKRS, ATP6V1D, BTF3L4, ERGIC2, none, PPOX, HIST1H4A, TRIAP1, none, N6AMT1, GCN20, PIAS4, PASK, PUS1, PUS7, FAM135A, HLTF, POLR1E, RPL12, RPL13, RPL18A, RPLP1, none, PSAT1, none, MLLT3, HSPD1, ATP6VOD1, PCID2, DSPP, ATP6VOD1, none, ANKZF1, LOC645490, none, none, ANKRD26, none, none |
Resistance to the G1 specific toxin zymocin was determined in a screen that was performed in parallel to that for the identification of DOX resistance mutants. A total of 806 diploid deletion strains (16.6% of nonessential genes) were found to be hypersensitive to zymocin. A total of 106 DOXS deletion mutants (50.7%) were found to be cross sensitive to the lethal effects of zymocin. This is 3 fold greater than that expected by chance alone.
A total of 204 ionizing radiation resistance genes (4% of nonessential genes) were identified in the diploid deletion collection as previously described
Approximately 500 gene deletions (∼10% of nonessential genes) in the haploid deletion collection were found to significantly affect cell size control that is determined in G1 and regulated by the checkpoint at “START”
A total of 456 deletion mutants in the haploid deletion collection (9.4% of nonessential genes) were identified that demonstrated enhanced sensitivity to oxidative DNA damaging agents
DOX is a well-characterized chemotherapeutic that induces DNA damage by multiple mechanisms including the production of ROS by interaction with the mitochondria, direct inhibition of topoisomerase II or direct DNA interactions (by intercalation, alkylation and/or crosslinking). All of these processes are known to induce DSB damage. Therefore, it is not surprising that a subset of DOX sensitive genes significantly overlap with those that show sensitivity to IR and oxidative damage induced by H2O2 and other chemicals that act in G1 (
Among the diploid DOX resistance mutants, many (24.3%) overlap with those that were found to affect cell size control (
Zymocin is a toxin secreted by the yeast
Among the zymocin sensitive diploid deletion strains identified in the primary screen, 103 were found to overlap with our previously described set of IR sensitive diploid deletion strains
The overlapping sensitivity of our diploid DOX sensitive deletion strains to the G1 specific toxin zymocin as well as with mutants sensitive to oxidative damage and those that regulate cell size control in G1 suggests that a significant fraction of the lethal activity of DOX occurs in the G1 phase of the cell cycle. Furthermore, as compared to haploids, diploid yeast are capable of repairing DSB via recombination in G1 due to the availability of a chromosome homolog. This suggests that among the mutants identified exclusively in the diploid screen and absent in the haploid screen, some may exert repair activity specifically in G1. Alternatively, since we utilized a dose higher than that used in the haploid screen (50 and 25 µg/ml as compared to ∼11 µg/ml), this may have allowed the identification of more DOX resistance genes. To test this directly, we compared (relative to WT) the haploid and diploid DOX sensitivities for 26 mutants detected in the diploid DOX screen but not found in the haploid DOX screen as well as 4 diploid sensitive mutants (
(A) Haploid (1n
Yeast deletion |
DOX |
HU |
MMS |
|||
2n | 1n | 2n | 1n | 2n | 1n | |
>SSS | >SSS | >SSS | >SSS | S | SS | |
>SSS | >SSS | >SSS | >SSS | >SSS | SSS | |
SS | >SSS | SS | S | >SSS | >SSS | |
>SSS | >SSS | >SSS | >SSS | >SSS | SSS | |
SSS | SSS | >SSS | >SSS | >SSS | >SSS | |
SSS | SSS | >SSS | >SSS | >SSS | >SSS | |
SSS | >SSS | SSS | SSS | >SSS | >SSS | |
SSS | SSS | S | S | S | SSS | |
SSS | SSS | >SSS | >SSS | SSS | SSS | |
SS | SSS | S | S | SSS | SSS | |
SSS | SSS | - | - | >SSS | >SSS | |
>SSS | >SSS | SSS | SSS | >SSS | >SSS | |
>SSS | >SSS | S | S | S | S | |
SS | SS | S | S | SSS | SSS | |
>SSS | >SSS | >SSS | >SSS | S | S | |
>SSS | >SSS | SSS | >SSS | SS | SSS | |
>SSS | >SSS | >SSS | >SSS | >SSS | >SSS | |
>SSS | >SSS | S | S | >SSS | >SSS | |
>SSS | SS | >SSS | >SSS | SS | >SSS | |
>SSS | >SSS | SS | SSS | SSS | SSS | |
>SSS | >SSS | SSS | SSS | SS | - | |
>SSS | >SSS | - | - | SSS | >SSS | |
>SSS | >SSS | SS | SS | >SSS | >SSS | |
SS | SS | - | - | SSS | SSS | |
n = 30 | 27 | 24 | 26 | |||
These deletion strains were detected in the haploid DOX screen
Yeast deletions identified in the diploid deletion DOX screen were cross sensitive to ionizing radiation (see
Relative sensitivity of the diploid (2n) versus haploid (1n) deletion strains to DOX was determined at a concentration of 50 ug/ml. Cells were grown in liquid YPD for two days and serial 5 fold dilutions made in sterile water. Two ul aliquots were then spotted to YPD and the DOX plates and allowed to grow for 3 days. >SSS denotes an enhanced sensitivity for a given deletion mutant that was greater than 125 fold over that observed for the isogenic WT of the same ploidy; SSS denotes a 125 fold enhanced sensitivity of the mutant when compared to WT; SS denotes a 25 fold enhanced sensitivity of the mutant when compared to WT; S denotes a five fold enhanced sensitivity of the mutant when compared to WT; “ - “ denotes no enhanced sensitivity of the mutant when compared to WT.
Relative sensitivity to hydroxyurea (HU) was determined at 200 uM.
Relative sensitivity to methyl methanesulfonate (MMS) was determined at 2 uM.
For six of the deletion strains examined (
Both
Since mating type transcription regulation alters relative expression levels and useage of DSB repair pathways (homologous recombination
It has been established that in diploid cells which exhibit altered
The identification of DOX sensitive gene deletions that are diploid-specific suggests that these genes may mediate repair functions prior to the completion of DNA replication. Functionally, these genes may impact recombinational repair of DOX-induced lesions or alternatively, they may affect cell cycle progression (checkpoint) in G1 or early S phase. For those mutants that have defects affecting DNA damage checkpoint response, they may fail to elicit checkpoint arrest and continue to progress rapidly in the presence of damage (similar to the
(A) WT and mutant diploid deletion strains were grown to logarithmic phase in liquid YPD. Single unbudded (G1) cells were arrayed into 5×4 cell grids on YPD with and without doxorubicin (50 µg/ml). Representative photomicrographs of mutant cells arrested in G1 or at G1/S following exposure to doxorubicin have been shown following 15 or 30 hr growth at 30°C. Only WT diploid cells were capable of forming viable microcolonies when exposed to doxorubicin. Most unbudded cells (>70%) from the WT and mutant diploid strains demonstrated rapid cell cycle progression and microcolony formation in the absence of DOX (data not shown). The mean gene conversion frequency of the
Strikingly, the diploid
To confirm that the severe cell cycle defects and diploid-specific hypersensitivity to DOX was due to the observed diploid mutations, we complemented the
Similar to the effects of deleting members of the RAD52 recombinational repair genes, defects in recombination pathways may result in hypersensitivity to DOX. Furthermore, we have detected mating-type expression defects among mutants that show diploid-specific hypersensitivity to DOX suggesting these mutants may be decreased in their ability to undergo recombinational repair. We therefore examined the diploid-specific DOX sensitive mutants for spontaneous PCR-mediated gene conversion of the endogenous
Loss of
Using previously identified genetic and physical interactions compiled at SGD, we determined the interaction network for the 9 genes (
The resulting union of the genetic and physical interaction maps produced a combined map that was defined by 500 nodes with 1157 interconnected direct interactions (
(A) Using the 9 diploid-specific DOX resistance genes identified in this study (
Within the combined genetic/proteomic interaction map are 377 gene nodes that interconnect with the major diploid-specific DOX resistance gene node hubs yet were not detected in the DOX screen. Since many of these genes display genetic interconnectivity with multiple major DOX resistance gene nodes in a pattern similar to that for other DOX resistance genes identified in the screen, we examined 14 of these genetically predicted and multiply interconnected diploid deletion strains for sensitivity to DOX, HU and/or MMS. Of the 14 mutants examined that interconnect to multiple major DOX resistant gene nodes (
In a similar manner we utilized the proteomic interaction map within
Large sets of novel genes that mediate resistance to a variety of DNA damaging agents have been identified using the isogenic yeast deletion strain collections (reviewed in
Our diploid screen identified 376 gene deletions sensitive to the DNA damaging agent doxorubicin (DOX), many of which overlap with those that function in recombinational repair and were also identified in our previous diploid IR screens (
Nine diploid specific DOX resistance genes were identified and appear to be functionally interrelated as numerous genetic and proteomic interactions have been documented for these genes (
Detailed examination of the cell cycle progression of single, unbudded G1 cells for six of these mutants clearly demonstrated that all have cell cycle progression defects associated with G1 or G1/S phase transition. However, since defects in recombination repair could promote the persistence of DSB damage and cause extended cell cycle delays, we examined these six mutants for the ability to undergo PCR mediated gene conversion of the
The magnitude of the hyper-recombination phenotype for the
Although IR induced DSBs were predominantly described as arresting cells at G2/M, recent reports describing the genetic checkpoint controls associated with DNA damage occurring in G1 are accumulating. Evidence for recombinational repair of DSBs specifically in G1 is sparse due to the continued preference for using haploid yeast in checkpoint and DNA damage related studies. Furthermore, although it was originally thought that DSB resection, which is required for homologous recombination, did not occur during G1 in haploids
Damage induced arrest in G1 or at G1/S transition has not been as well characterized in
We have previously identified the CCR4 damage response network and the checkpoint associated roles of
Since homozygosity at the mating-type locus in diploid yeast can decrease the resistance to DNA damage and especially DSBs, we investigated whether deletions of our diploid-specific DOX resistance genes could affect mating type expression in diploids. In some, (
Repression of
Using previously published genetic and proteomic interactions, we successfully predicted and identified new IR resistance genes based on interactions with members of the CCR4 damage response network
Our yeast screen identified 376 DOX resistance genes and the majority (76%) are conserved suggesting they may have clinical relevance. DOX is a highly effective anthracycline chemotherapeutic agent that targets solid tumors of the breast and other cancers; however, dosage has to be carefully regulated and monitored to avoid the potentially life threatening complications associated with cardiotoxicity. DOX is a DNA damaging agent that produces DNA DSBs in part through the production of reactive oxygen species (ROS). The site of ROS production appears to be the mitochondrion and yeast mutants that lack functional mitochondria are indeed more resistant to DOX (
DOX resistance in tumors can occur which decreases the efficacy of this chemotherapeutic agent. In some cases this can severely limit the clinical usage of this otherwise effective class of drugs. We propose that tumor hypersensitivity and/or resistance to DOX is genetically determined and that the orthologs identified in this study offer many new potential genes that could be targeted for inactivation to increase tumor sensitivity to DOX chemotherapy. Validating these human orthologs as genes which confer resistance to DOX could allow strategies to be designed that sensitize DOX resistant cancers that would be normally refractory to treatment with this drug. From our extensive list of highly conserved DOX resistance genes identified in yeast, we utilized BLAST analysis to identify five DOX resistance targets that show high homology to proteins previously identified to be mutated in breast cancer (
Yeast diploid deletion mutants hypersensitive to the lethal effects of doxorubicin with associated sensitivity to the toxin zymocin. Doxorubicin hypersensitivity in the diploid deletion strains was scored from 1–3 (complete description in Results section of text) with 1 being the least sensitive and 3 the most sensitive. Sensitivity to zymocin in the diploid deletion strains was scored 1–3 with 1 being the least sensitive and 3 being the most sensitive (see complete description in Results section of text). Diploid deletions that are cross sensitive to ionizing radiation (IR) have been indicated in bold. References that describe haploid deletion strains that are sensitive to doxorubicin, are defective in G1 cell size control and cross sensitive to oxidative damaging agents are described in the text (see
(0.11 MB XLS)
Yeast diploid deletion mutants that show reduced (slow) growth in response to doxorubicin. Table listing diploid yeast gene deletions that have a slow growth rate when exposed to doxorubicin. Deletion strains were scored 1–2 with 1 being the least inhibited and 2 being more inhibited when exposed to doxorubicin (see text Results section for complete description). Zymocin sensitivity of diploid deletion strains has been described in
(0.10 MB XLS)
Highly conserved mitochondrial gene targets that mediate doxorubicin resistance in diploid yeast. Table listing doxorubicin sensitive yeast diploid gene deletions with products implicated in mitochondrial function (see text Discussion section for complete description).
(0.06 MB DOC)
Yeast doxorubicin resistance genes whose protein products are orthologs of human proteins encoded by genes mutated in breast cancer. Table listing yeast doxorubicin resistance proteins that are orthologs of human proteins encoded by genes found to be mutated in breast cancer (see text Discussion section for complete description).
(0.08 MB DOC)
We would like to thank Drs. D. Lew, E. Perkins and C. Tucker for the gift of plasmids and strains used in this study.