Drug resistance in diploid yeast is acquired through dominant alleles, haploinsufficiency, gene duplication and aneuploidy

Abstract

Previous studies of adaptation to the glucose analog, 2-deoxyglucose, by Saccharomyces cerevisiae have utilized haploid cells. In this study, diploid cells were used in the hope of identifying the distinct genetic mechanisms used by diploid cells to acquire drug resistance. While haploid cells acquire resistance to 2-deoxyglucose primarily through recessive alleles in specific genes, diploid cells acquire resistance through dominant alleles, haploinsufficiency, gene duplication and aneuploidy. Dominant-acting, missense alleles in all three subunits of yeast AMP-activated protein kinase confer resistance to 2-deoxyglucose. Dominant-acting, nonsense alleles in the REG1 gene, which encodes a negative regulator of AMP-activated protein kinase, confer 2-deoxyglucose resistance through haploinsufficiency. Most of the resistant strains isolated in this study achieved resistance through aneuploidy. Cells with a monosomy of chromosome 4 are resistant to 2-deoxyglucose. While this genetic strategy comes with a severe fitness cost, it has the advantage of being readily reversible when 2-deoxyglucose selection is lifted. Increased expression of the two DOG phosphatase genes on chromosome 8 confers resistance and was achieved through trisomies and tetrasomies of that chromosome. Finally, resistance was also mediated by increased expression of hexose transporters, achieved by duplication of a 117 kb region of chromosome 4 that included the HXT3, HXT6 and HXT7 genes. The frequent use of aneuploidy as a genetic strategy for drug resistance in diploid yeast and human tumors may be in part due to its potential for reversibility when selection pressure shifts.

Fig 1. Aneuploidy in 2DG-resistant diploid strains.

Diploid strains were sequenced, and the read depth (y-axis) is plotted against chromosomal position (x-axis) with chromosome number shown in Roman numerals. Median read depth is indicated with a yellow line. Monosomic chromosomes are plotted in brown, trisomic chromosomes in blue and tetrasomic chromosomes in red.

Fig 2. Dominant alleles in the Snf1 kinase complex confer 2DG resistance.

A. The structure the mammalian AMPK (6c9h.pdb) bound to the activator R734 [46] is shown as a model for the Snf1 kinase heterotrimer. The alpha subunit is shown in green, the beta subunit in blue and the gamma subunit in cyan. The locations of the kinase domain active site, the allosteric drug and metabolite-binding site (ADaM) and the dominant mutations that confer 2DG resistance are shown. B-D. 2DG resistance assays performed in quadruplicate using wild type (WT) and gene deletion strains transformed with the indicated plasmids that express a wild type gene, the same gene with a single missense mutation indicated or empty plasmid vector. Mean values statistically different from cells expressing the wild type allele are indicated.

Fig 3. Haploinsufficiency of REG1 confers 2DG resistance.

A. Diploid cells with the indicated genotypes at the REG1 locus were serially diluted and spotted onto agar plates containing SC medium with 2% glucose with and without 0.1% 2DG. B. 2DG resistance assay of diploid cells with the indicated genotypes. C. 2DG resistance measured in quadruplicate using 0.1% 2DG. Diploid strains used in this experiment were as follows: REG1/REG1 (MSY1333 x MSY188); reg1Δ/REG1 (MSY1333 x MSY930); reg1-Q332Stop/REG1 (MSY188 x MSY1520); reg1Δ/reg1Δ (MSY930 x MSY841).

Fig 4. Monosomy of chromosome 4 confers 2DG resistance and reverts at high frequency.

A. Growth properties DS10 cells on agar plates. Small and large colonies of DS10 were re-streaked onto agar plates with the media indicated above. B. Growth of DS10 cells and wild type diploid cells in media containing 2DG. C. Reversion of small colony phenotype. Multiple, independent populations of small colonies from strains DS10 and DS11 were spread on plates containing SC medium. The mean percentage (±SD) of colonies that reverted to the large colony phenotype for each independent replicate is shown. D. Chromosomal ploidy analysis of DS10 and revertants. Normalized median read depth for each chromosome is plotted for wild type, DS10 and five DS10 revertants (DS10-R). Chromosomes that diverge from normal ploidy (2n) are indicated.

Fig 5. Increased copy number of DOG1/2 genes mediates 2DG resistance.

A. Normalized median read depth of chromosomes in wild type diploid strain (MSY1527), in the 2DG-resistant isolate, DS14, and in diploids generated from the haploid progeny of DS14. Chromosomes with 2n copy number are shown in black. Aneuploid chromosomes are indicated. B. 2DG resistance assay of diploid strains shown in A. C. 2DG resistance was measured in haploid strains with increasing copy number of the DOG1/2 genes. Resistance was compared to wild type (WT) haploid cells with 1n ploidy of chromosome 8 and mean values statistically different are indicated. D. 2DG resistance assays of a wild type haploid strain and haploid strains with a disomy of chromosome 8 were performed in quadruplicate. Deletions were engineered in the disomic strains to remove the DOG1/2 or HXT415 genes from one of the disomic chromosomes or the SIP2 gene on chromosome 7.

Fig 6. Duplication of HXT367 cluster on chromosome 4 confers 2DG resistance.

A. Read depth across chromosome 4 was plotted for the wild type strain (MSY1333) and the haploid strain (MSY1564) bearing the 117 kb duplicated region. Positions of Ty elements flanking the duplicated region and some of the genes within are shown. Regions present in high-copy number plasmid clones [24] covering this region are shown at top. B. 2DG resistance assay performed in quadruplicate showing mean ±SD and statistical significance for haploid strains (1n) with (MSY1564) and without (MSY1333) the 117 kb duplication. C. 2DG resistance in a wild type haploid strain transformed with high-copy number plasmids spanning the duplicated region. Statistical significance for each sample compared to vector is indicated as is a comparison of 4G6 and 4H6.

Fig 7. Effect of Aneuploidy on mRNA expression.

A. Analysis of mRNA abundance was performed using RNAseq in wild type and DS10 cells washed off agar plates and grown for 2 hours in SC media with 2% glucose with and without 2DG. The log2 ratio of mRNA abundance from aneuploid strain divided by the wild type abundance is shown. Genes on the vertical axis are clustered by chromosome. The same RNAseq analysis was also performed on wild type and DS14 cells grown to mid-log in SC media and two hours after addition of 2DG. Two biological replicates for each strain and condition are shown. B. REG1 mRNA abundance as transcripts per million mapped reads (tpm) is shown for wild type and DS10 cells with and without addition of 2DG. C. DOG1 and DOG2 mRNA abundance is shown for wild type and DS14 cells with and without addition of 2DG. D. Ribosomal protein mRNA levels for wild type and aneuploid strains with and without 2DG treatment. TPM Values for 132 mRNAs are shown in a violin plot with the median value indicated below.

Fig 8. Genetic response to 2DG in haploid and diploid cells.

A schematic representation of the yeast cell response to 2DG is shown and described in more detail in the Discussion. Genetic strategies used by haploid and diploid cells to adapt to the presence of 2DG are listed. Recessive alleles are shown in lower case and dominant alleles are shown in upper case.