Global Response of Saccharomyces Cerevisiae to an Alkylating Agent
Total Page：16
File Type：pdf, Size：1020Kb
Proc. Natl. Acad. Sci. USA Vol. 96, pp. 1486–1491, February 1999 Genetics Global response of Saccharomyces cerevisiae to an alkylating agent SCOTT A. JELINSKY AND LEONA D. SAMSON* Department of Cancer Cell Biology, Division of Toxicology, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115 Edited by David Botstein, Stanford University School of Medicine, Stanford, CA, and approved December 7, 1998 (received for review September 16, 1998) ABSTRACT DNA chip technology enables simultaneous CUPr) was used in this study and was grown in 1% yeast examination of how '6,200 Saccharomyces cerevisiae gene tran- extracty2% peptoney2% glucose at 30°C. Cells were grown to a script levels, representing the entire genome, respond to envi- density of 5 3 106 cells per ml as measured by counting duplicated ronmental change. By using chips bearing oligonucleotide ar- dilutions. Cultures were split into two; MMS (0.1%) was added rays, we show that, after exposure to the alkylating agent methyl directly to one culture, and both cultures were incubated at 30°C ' methanesulfonate, 325 gene transcript levels are increased and for 1 h. Cells were pelleted and washed once in distilled H2O and '76 are decreased. Of the 21 genes that already were known to once in AE buffer (50 mM NaOAc, pH 5.2y10 mM EDTA) be induced by a DNA-damaging agent, 18 can be scored as immediately before RNA extraction. inducible in this data set, and surprisingly, most of the newly RNA Extraction. Total RNA was isolated by using a hot-phenol identified inducible genes are induced even more strongly than method (15). Poly(A)1 RNA was purified from total RNA with these 18. We examined 42 responsive and 8 nonresponsive ORFs Oligotex oligo(dT) selection step (Qiagen, Chatsworth, CA). by conventional Northern blotting, and 48 of these 50 ORFs Poly(A)1 RNA was amplified and biotin-labeled as follows. responded as they did by DNA chip analysis, with magnitudes Poly(A)1 RNA (1 mg) was converted into double-stranded cDNA displaying a correlation coefficient of 0.79. Responsive genes fall by using a modified oligo(dT) primer with a T7 RNA polymerase into several expected and many unexpected categories. Evidence promoter sequence at the 59 end and the Superscript Choice for the induction of a program to eliminate and replace alkylated system for cDNA synthesis (GIBCO). Double-stranded cDNA proteins is presented. was purified by phenolychloroform extractions, precipitated with ethanol, and resuspended at a concentration of 0.5 mgymlin 3 Exposure to DNA-damaging agents can increase DNA repair diethyl pyrocarbonate-treated H2O. Phase-Lock Gel (5 Prime capacity and activate cell-cycle checkpoints. Such exposures may 3 Prime) was used for all organic extractions to increase recovery. also induce enzymes that metabolize toxicants to facilitate their In vitro transcription was performed with T7 RNA polymerase elimination from the organism or may activate programmed cell (T7 Megascript kit, Ambion, Austin, TX) and with 0.5–1.0 mgof death (apoptosis) to eliminate highly damaged cells. Thus, it cDNA, 7.5 mM unlabeled ATP and GTP, 5.3 mM unlabeled has long been known that cells induce the expression of a UTP and CTP, and 1.9 mM biotin-labeled CTP and UTP variety of genes after toxic exposure, and gene regulation in (biotin-11-CTP, biotin-16-UTP, Enzo Diagnostics). Reactions response to DNA-damaging agents has been well studied in many were carried out for6hat37°C, and cRNA was purified by RNA organisms (1–5). affinity resin (RNeasy spin columns, Qiagen). A sample was Random lacZ gene fusions and differential hybridization pre- separated on a 1% agarose gel to check the size range, and then viously have identified 21 Saccharomyces cerevisiae genes whose 10 mg of cRNA was fragmented randomly to an average size of transcript levels are increased in response to DNA-damaging 50 bases by heating at 94°C for 35 min in 40 mM Triszacetate, pH agents (1, 6–8). Both approaches produced catalogs of genes of 8.1y100 mM KOAcy30 mM MgOAc. known and unknown function, but the lack of redundancy with GeneChip Hybridizations. A set of four oligonucleotide arrays which they were identified indicates that the search for such (GeneChip Ye6100 arrays, Affymetrix, Santa Clara, CA) con- inducible genes is far from complete (1, 8). taining probes for 6,218 yeast ORFs were used for hybridizations. We previously studied the inducible transcription of an S. Hybridizations were done in 200 ml of AFFY buffer (Affymetrix) cerevisiae DNA repair gene (MAG1, encoding a 3-methyladenine at 40°C for 16 h with constant mixing. After hybridization, arrays DNA glycosylase) in response to simple alkylating agents such as were rinsed three times with 200 mlof63 sodium chloridey MAG1 methyl methanesulfonate (MMS; refs. 9–13). Upstream sodium phosphateyEDTAyTriton (SSPE-T; 13 0.15 M NaCly15 regulatory elements were identified, and similar elements are mM phosphate, pH 7.6y1 mM EDTAy0.005% Triton) and then found upstream of numerous DNA repair and metabolism genes, washed with 200 mlof63 SSPE-T (pH 7.6) for 1 h at 50°C. The suggesting common transcriptional regulatory mechanisms (12– arrays were rinsed twice with 0.53 SSPE-T (pH 7.6) and washed 14). We therefore decided to identify all the genes that are with 0.53 SSPE-T (pH 7.6) at 50°C for 15 min. Staining was done regulated coordinately with MAG1. Here, we report that DNA with 2 mgyml streptavidin-phycoerythrin (Molecular Probes) and chip analysis has expanded by more than 15-fold the catalog of 1mgyml acetylated BSA (Sigma) in 63 SSPE-T (pH 7.6). The genes that are inducible by a DNA-damaging agent. In addition, m DNA chip analysis has identified a class of genes whose tran- arrays were read at 7.5 m with a confocal scanner (Molecular scripts are repressed. Global responses to a DNA-damaging agent Dynamics) and analyzed with GENECHIP software, version 3.0. have now come into focus, and we present evidence that exposure The samples were normalized by using the total average differ- to an alkylating agent elicits a program to eliminate and replace ence between the perfectly matched probe and the mismatched alkylated proteins from S. cerevisiae cells. probe (16). Northern-Blot Analysis. RNA was isolated from log-phase MATERIALS AND METHODS cells exposed to 0.1% MMS for 0, 30, 60, or 120 min by using a m Strains, Media, and Growth Conditions. S. cerevisiae strain hot-phenol extraction method (15). Total RNA (25 g) was DBY747 (MATa his3-D1 leu2-3,112 ura3-52 trp1-289a galS can1 fractionated in a 1% denaturing agarose gel, blotted, and probed The publication costs of this article were defrayed in part by page charge This paper was submitted directly (Track II) to the Proceedings office. Abbreviations: MMS, methyl methanesulfonate; SSPE-T, sodium payment. This article must therefore be hereby marked ‘‘advertisement’’ in chlorideysodium phosphateyEDTAyTriton. accordance with 18 U.S.C. §1734 solely to indicate this fact. *To whom reprint requests should be addressed. e-mail: lsamson@ PNAS is available online at www.pnas.org. sph.harvard.edu. 1486 Downloaded by guest on October 1, 2021 Genetics: Jelinsky and Samson Proc. Natl. Acad. Sci. USA 96 (1999) 1487 with PCR-amplified labeled ORFs (Research Genetics, Hunts- decreased by a factor of .5-fold; 10 decreased by a factor of ville, AL) by using standard methods (17). .7.5-fold; 4 decreased by .10-fold. To establish that the information obtained from GeneChip RESULTS AND DISCUSSION analysis is accurate, we chose 50 ORFs to examine by conven- Global Expression Monitoring After Alkylation Damage. The tional Northern-blot analysis; these included 26 inducible (3.1- to GeneChip methodology developed by Affymetrix was used to 251-fold), 16 repressible (3.1- to 18.1-fold), and 8 nonresponsive monitor global gene expression in S. cerevisiae. The 6,218 ORFs ORFs. Freshly isolated RNA from control and from MMS- of this organism are represented on a set of four high density treated cells was probed with each of the 50 ORFs. Hybridization oligonucleotide arrays (16, 18, 19). Poly(A)1 mRNA was isolated signals, as measured by Northern-blot and GeneChip analysis, are from untreated cells and from cells exposed for1hto0.1% MMS. shown for six of the ORFs (Fig. 2), and data for all 50 ORFs are These conditions were chosen, because they yield optimal MAG1 compiled in Fig. 3 A and B. In terms of whether the ORFs were induction with minimal cell death (11). Poly(A)1 RNA was inducible, repressible, or nonresponsive, the Northern-blot and converted into double-stranded cDNA containing the T7 RNA GeneChip data agreed for 48 of the 50 ORFs. Moreover, the data polymerase promoter, and biotin-labeled cRNA was produced agreed remarkably well in terms of the extent of induction or and hybridized to the GeneChip arrays. The hybridization- repression (Fig. 3 A and B), displaying a correlation coefficient of intensity information was gathered by scanning confocal micros- 0.79 for the complete data set (Fig. 3B). For the majority of the copy and analyzed with GENECHIP software, version 3.0 (16). ORFs, the fold change in transcript levels differed by no more Typical GeneChip-hybridization intensities for control and than a factor of two between the Northern-blot and GeneChip MMS-treated cells are shown in Fig. 1. As a guide, one MMS- data (Fig. 3A). However, for very highly induced ORFs (see Fig. induced, one MMS-repressed, and one nonresponsive ORF are 3B), the correlation weakens slightly, such that the Northern-blot indicated by arrows.