Expression of human poly (ADP-ribose) polymerase 1 in Saccharomyces cerevisiae: Effect on survival, homologous recombination and identification of genes involved in intracellular localization

Mutation Research 774 (2015) 14–24

Contents lists available at ScienceDirect

Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis journal homepage: www.elsevier.com/locate/molmut Community address: www.elsevier.com/locate/mutres

Expression of human poly (ADP-ribose) polymerase 1 in Saccharomyces cerevisiae: Effect on survival, homologous recombination and identification of genes involved in intracellular localization Marco La Ferla a,1 , Alberto Mercatanti a,1 , Giulia Rocchi a,1 , Samuele Lodovichi a , Tiziana Cervelli a , Luca Pignata a , Maria Adelaide Caligo b , Alvaro Galli a,∗ a b

Yeast Genetics and Genomics, Institute of Clinical Physiology, National Council of Research (CNR), via Moruzzi 1, 56122 Pisa, Italy Section of Genetic Oncology, University Hospital and University of Pisa, via Roma 57, 56125 Pisa, Italy

a r t i c l e

i n f o

Article history: Received 30 October 2014 Received in revised form 3 February 2015 Accepted 26 February 2015 Available online 6 March 2015 Keywords: Poly (ADP-ribose) polymerase 1 Homologous recombination UV Saccharomyces cerevisiae Yeast genome wide screening PARP-1 localization

a b s t r a c t The poly (ADP-ribose) polymerase 1 (PARP-1) actively participates in a series of functions within the cell that include: mitosis, intracellular signaling, cell cycle regulation, transcription and DNA damage repair. Therefore, inhibition of PARP1 has a great potential for use in cancer therapy. As resistance to PARP inhibitors is starting to be observed in patients, thus the function of PARP-1 needs to be studied in depth in order to find new therapeutic targets. To gain more information on the PARP-1 activity, we expressed PARP-1 in yeast and investigated its effect on cell growth and UV induced homologous recombination. To identify candidate genes affecting PARP-1 activity and cellular localization, we also developed a yeast genome wide genetic screen. We found that PARP-1 strongly inhibited yeast growth, but when yeast was exposed to the PARP-1 inhibitor 6(5-H) phenantridinone (PHE), it recovered from the growth suppression. Moreover, we showed that PARP-1 produced PAR products in yeast and we demonstrated that PARP-1 reduced UV-induced homologous recombination. By genome wide screening, we identified 99 mutants that suppressed PARP1 growth inhibition. Orthologues of human genes were found for 41 of these yeast genes. We determined whether the PARP-1 protein level was altered in strains which are deleted for the transcription regulator GAL3, the histone H1 gene HHO1, the HUL4 gene, the deubiquitination enzyme gene OTU1, the nuclear pore protein POM152 and the SNT1 that encodes for the Set3C subunit of the histone deacetylase complex. In these strains the PARP-1 level was roughly the same as in the wild type. PARP-1 localized in the nucleus more in the snt1 than in the wild type strain; after UV radiation, PARP-1 localized in the nucleus more in hho1 and pom152 deletion strains than in the wild type indicating that these functions may have a role on regulating PARP-1 level and activity in the nucleus. © 2015 Elsevier B.V. All rights reserved.

1. Introduction The poly(ADP-ribose)polymerase 1 (PARP-1) is an ubiquitous and abundant nuclear protein involved in DNA repair and transcription [1]. The PARP-1 protein has a catalytic site for NAD+ -dependent

Abbreviations: PARP-1, poly (ADP-ribose) polymerase 1; PAR, poly ADP ribose; DSB, double-strand breaks; PHE, 6(5-H) phenantridinone; 3PGK, 3 phosphoglycerate kinase; SC-URA-LEU, synthetic complete medium lacking uracil and leucine; UV, ultraviolet radiation. ∗ Corresponding author. Tel.: +39 0503153094. E-mail address: [email protected] (A. Galli). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.mrfmmm.2015.02.006 0027-5107/© 2015 Elsevier B.V. All rights reserved.

activity that synthesizes for a negatively charged polymer that binds to target proteins. This polymer is known as “poly (ADPribose)” or PAR [1]. PARP-1 actively participates in a series of functions within the cell that include: mitosis, intracellular signaling, cell cycle regulation, transcription and DNA damage repair. PARP-1 binds to a variety of DNA structures that can be formed, such as double-strand breaks (DSB) and Holliday junctions [2,3]. Finally, PARP-1 can interact with a wide variety of proteins that act at the DNA level which include components of the transcription machinery, specific factors in binding specific sequences on DNA and histone variants [2–4]. PARP-1 binds the promoter of many genes that are highly transcribed [4,5]. In case of extensive damage of the DNA, there is

M. La Ferla et al. / Mutation Research 774 (2015) 14–24

an over-activation of PARP-1 that depletes cell of NAD+ substrate resulting in the decrease of glycolysis and the Krebs cycle, thus giving rise to cell dysfunction and death by necrosis [6]. On the other hand, PARP-1 over-activation modulates important inflammatory pathways [7,8]. Many drugs have been produced that act as inhibitors of PARP1. Thus, with PARP1 inhibition, cells with DNA damage that are in the S phase of the cell cycle, transform DNA lesions into cytotoxic double-strand breaks (DSB). The basic problem is that many tumor cell lines have defects and mutations in proteins involved in the inspection and repair of DNA damage, such as p53, ATM, MRE11 and BRCA1-2. Studies showed, however, that these tumor lines, if treated with PARP inhibitors, appear to be much more sensitive to cytotoxic agents used as chemotherapeutic [9,10]. This is the main reason why most inhibitors are used as adjuvant in cancer chemotherapy [11]. Moreover, PARP-inhibitors are often used to treat breast, ovarian and prostate cancer with mutation in BRCA1 and/or BRCA2 genes [12–14]. Recently, clinical trials with PARP inhibitors in combination with DNA damage drugs have shown promising results against different types of cancer [15,16]. However, resistance of cancer cells to the PARP inhibitors is starting to be observed [17]. Additionally, not all BRCA-defective cancers seem to respond to PARP inhibitor therapy [18]; this could be due to reversion of the BRCA mutated gene to wild type or to an up-regulation of the ABCB1a/b genes which encode for P-glycoprotein multidrug resistance drug efflux pumps [17,19]. Basically, the potential mechanisms of resistance to PARP inhibitors are: increased DNA repair by homologous recombination, mis-regulation of non-homologous end joining, decreased levels of PARP-1 and decreased intracellular availability of PARP inhibitors [20]. The expression of PARP-1 in the yeast Saccharomyces cerevisiae causes interference on cell growth [21]. This phenotype has been very helpful to identify novel PARP inhibitors and PARP-1 substrates using yeast as model system [22,23]. Due to its ease of manipulation and genetic tractability, the yeast S. cerevisiae has been used to analyze the function of many proteins from mammalian cells [24]. Moreover, yeast is a valuable predictor of human gene functions because 31% of proteins encoded by yeast genes have human homologs and conversely, almost 50% of human genes implicated inheritable diseases have yeast homologs [25,26]. Over the past years, yeast improved the knowledge of the fundamental pathways in humans and facilitated the identification of many disease genes [27–29]. In this report, we first verified the phenotype that occurs when haploid and diploid yeast cells express human PARP-1 gene. After, we set up a genome wide genetic screening using a “pool” of 4746 diploid yeast strains each carries a specific barcoded-deletion of a not essential gene in order to identify possible candidate genes affecting PARP activity. 2. Materials and methods 2.1. Plasmids The plasmids pYES2-PARP1 and pYES2-PARP1GFP that allow PARP-1 expression in culture medium containing galactose because of the GAL1 promoter were obtained from Ed Perkins and Hungwen Liu.

15

HIS3::pRS6 LYS2/MAT˛ ura3-52 leu298 TRP5 ade2-101 ilv1-92 ARG4 his3-200 lys2-801) were used to assess the effect of human PARP-1 expression on growth, survival and homologous recombination. The genotype of the parental diploid BY4743 strain is MAT a/˛; his31/his31; leu20/leu20; LYS2/lys20; met150/MET15; ura30/ura3. The strains were transformed with pYES2, pYES2-PARP-1 or pYES2-PARP-1GFP following the procedure described in Gietz and Schiestl [30]. Complete media (YPAD), synthetic complete (SC), and drop-out media (SC-HIS, SCURA, SC-LEU and SC-ADE) were prepared according to standard procedures. 2.3. Growth, survival and homologous recombination assay The effect of PARP-1 expression on cell growth was evaluated in the BY4743 strain carrying the pYES2-PARP1 plasmid. Single colonies were grown in 5–10 ml SC-URA for 24 h at 30 ◦ C. Then, aliquots corresponding to 3–5 × 105 cells/ml were incubated for additional 24 h at 30 ◦ C in SC-URA glucose or 5% galactose with or without the inhibitor 6(5-H) phenantridinone (PHE). Cell numbers were determined by measuring the OD at 600 nm as already reported [30]. The effect of PARP-1 on cell survival was assessed in the strain RSY6, RS112, RSY12 and BY4743 as colony forming efficiency by plating directly 200–300 cells from a fresh glucoseculture in SC-URA glucose and SC-URA 5% galactose. The effect of PARP-1 on yeast homologous recombination was investigated in the RS112 strain that allows to measure two recombination events. In fact, RS112 carries an intra-chromosomal recombination substrate that consists of two his3 alleles, one with a deletion at the 3 end and the other with a deletion at the 5 end, that share 400 bp of homology. These two alleles are separated by the LEU2 marker and by the plasmid DNA sequence. An intrachromosomal recombination event leads to HIS3 reversion and loss of LEU2 [31]. The diploid RS112 strain also contains the two alleles ade2-40 and ade2-101, located in two homologous chromosomes that allow the measurement of inter-chromosomal recombination events [32]. To determine the frequency of intra-chromosomal and inter-chromosomal recombination, single colonies were inoculated into 5 ml of SC-URA-LEU medium and incubated at 30 ◦ C for 24 h. Thereafter, cultures were induced in fresh SC-URA-LEU containing 5% galactose for 24 h at 30 ◦ C under constant shaking with or without 100 ␮m PHE. In parallel, yeast cells carrying the empty pYES2 vector were grown in 5% galactose medium. Then, cells were washed twice, counted and appropriate numbers were plated onto complete medium to determine the number of vital cells and onto solid SC-HIS and SC-ADE to determine the frequency of intra-chromosomal and inter-chromosomal recombination, respectively. Plates were incubated at 30 ◦ C for 3–5 days. UV irradiation was performed as follows: after galactose induction, 10 ml aliquots containing 3 × 107 cells/ml, were distributed in dishes and irradiated in distilled sterile water using an UV radiator with two germicide lamps. Irradiation was performed at the dose ratio of 2–4 J/m2 /s, as determined by a digital radiometer with J-260-1A sensor (Ultraviolet Product, San Gabriel, CA, USA). Then, cells were washed and plated as described [33]. Results are reported as mean of at least four independent experiments ± standard deviation. Statistics was performed using the Student’s “t” test. 2.4. Cell fractionation and Western blot analysis

2.2. Yeast strains The haploid strains RSY12 (MAT aleu2-3,112 his3-11,15 URA3::HIS3) and RSY6 (MAT aura3-52 leu2-3,112 trp5-27 ade2-40 ilv1-92 arg4-3 HIS3::pRS6), and the diploid RS112 of S. cerevisiae (MAT aura3-52 leu2-3,112 trp5-27 ade2-40 ilv1-92 arg4-3

Yeast cells carrying the pYES2 or pYES2-PARP-1 plasmids were grown at 30 ◦ C under constant shaking for 17–24 h and then shifted in 5% galactose medium for 24 h. After this period of incubation, we extracted the total proteins from the yeast strains expressing PARP-1 and from the other strains that did not express this

16

M. La Ferla et al. / Mutation Research 774 (2015) 14–24

protein. The cells were washed twice in sterile bi-distilled water. The pellets obtained were resuspended in a mixture consisting of lysis buffer (Sorbitol 2 M, Tris–HCl 2 M pH 7.4, NaCl 5 M, MgCl2 1 M, ethylene-di-amine-tetra-acetic acid 0.5 M pH 8), phenyl-methanesulfonyl-fluoride (PMSF), necessary to inhibit the protease, and 1/4 tablet of Protease Inhibitor Cocktail (Roche). The whole mix was subjected to agitation by a vortex for at least four times for 1 min, alternating 1 min of vortexing to a minute of ice. After these steps we recovered the supernatant, which contains the protein fraction and we added 5 ␮l of TRITON X-100 1%. The whole was subjected to a centrifugation at 4 ◦ C for 15 min. From each extraction we loaded about 30 ␮g of protein sample in the polyacrylammide gel and hybridized with monoclonal antibodies: anti-PARP-1 monoclonal mAb (Santa Cruz Biotechnology) and anti-PAR (Trevigen). Anti-mouse Ab IgG HRP (Santa Cruz Biotechnology) and anti-3 phosphoglycerate kinase (3PGK, Molecular Probes) were used as secondary antibody and as loading control respectively. Nuclear extracts were carried out as follows: cells were grown to mid-log phase at 30 ◦ C with vigorous shaking and pelleted at 4 ◦ C for 5 min at 1500 g. Cells were washed in distilled water, weighted and resuspended (1 g pellet in 4 ml) in pre-treatment buffer (50 mM Tris–HCl, 20 mM dithiothreitol (DTT), 5 mM EDTA, pH 8); then, pellets were washed with 10 volumes of Ki buffer (2 mM K2 HPO4 , 3 mM KH2 PO4 , pH 6.5), centrifuged at 4 ◦ C and washed in 10 volumes of 20% sorbitol/Ki buffer with 0.5 mg/g pellet Zymolyase 100T. After 30 min at 30 ◦ C, an equal volume of 40% sorbitol/YPD (400 g/l sorbitol, 10 g/l yeast extract, 20 g/l bacto-peptone, 20 g/l glucose, pH 6.5) was added; spheroplasts were incubated for 30 min in 40% sorbitol/YPD, at room temperature. Thereafter, spheroplasts were washed in 40% sorbitol/YPD and resuspended in 20 ml cold Ficoll buffer 18% (w/v) (Ficoll 400, 10 mM Tris–HCl pH 7.5, 20 mM KCL, 5 mM MgCl2 , 3 mM DTT, 1 mM EDTA, proteinase inhibitors (1/4 tablet in 2.5 ml 0.2 M NaOH), 1 mM PMSF). Cell suspensions were transferred to type A tight dounce and subjected to 15 strokes. The dounce suspensions were centrifuged for 5 min at 3000 × g at 4 ◦ C. The supernatant was centrifuged at 13,000 × g for 20 min at 4 ◦ C. At this point, the supernatant contains the cytoplasmic proteins and the pellet the nuclei. Nuclear extract was obtained by resuspending the pellet in 0.2 ml NaOH plus proteinase inhibitor, incubating for 5 min at room temperature and finally centrifugating at 4 ◦ C at maximum speed. Pellets were resuspended in SDS sample buffer (0.06 Tris–HCl pH 6.8, 8.5% glycerol, 2% SDS, 8.5% ␤-mercaptoethanol, 0.0025% bromophenol blue), boiled for 3 min, centrifuged and further analyzed. Nuclear and cytoplasmic extract were loaded in the polyacrylammide gel and hybridized with anti-fibrillarin (Abcam) as nuclear marker [34].

Genomic DNA from single PARP-1 resistant clones was extracted using the MasterPure Yeast DNA purification kit (Epicentre, Madison, WI). DNA barcoded regions from each yeast deletion clone were amplified by PCR using the primers: U1 5 -GAT GTC CAC GAG GTC TCT-3 , Kan B 5 -CTG CAG CGA GGA GCC GTA AT3 ; and D1 5 -CGG TGT CGG TCT CGT AG-3 , Kan C 5 -TGA TTT TGATGA CGA GCG TAA T-3 . The resulting PCR fragment consists of a 600 bp region that contains the barcode and part of the G418 resistance gene. The information about the gene identified in this screening was obtained using the FASTA barcodes database (http://www.ttuhsc.edu/som/cbb/FASTAbarcodes/).

2.6. Bioinformatics We relied on the functional classification of the yeast pool genes and their human orthologous genes, and on comparison between S. cerevisiae and Homo sapiens; therefore, we were able to extrapolate equivalent biological pathways and processes [35]. Results are analyzed according to statistical significance. The approach used is based on the Gene Enrichment Analysis, applied by an algorithm implemented by Bioconductor gProfileR package [36–40]. Furthermore, the human orthologous genes were analyzed for searching association to human cancers. For our goal, a bioinformatics pipeline was developed in R language. Here we reported the steps followed by the analysis performed: Step 1: conversion of the yeast gene names to follow the official nomenclature. Multiple names for the same gene were corrected by bioDBnet tools (http://biodbnet.abcc.ncifcrf.gov/) [41]. Step 2: human orthologous genes were retrieved from remote updated database (g:ProfileR database server). Step 3: functional enrichment, based on hyper-geometric test in gProfileR package, was carried out on orthologous genes list for discovering same gene ontology profiles and pathways. This algorithm walks into deeper levels of gene ontology tree retrieving data from updated remote database. A false discovery rate correction was added to analysis results. Each subset of genes was identified in according to statistical limit of p-value ≤0.05. Step 4: by SQL language and R scripts, human orthologous genes were searched for into “Human Protein Atlas” database for discovering which genes were differentially expressed in several kinds of tumors. The database hosts the expression profiles of human proteins in tissues and cells; moreover, for each tumor type, the number of patients and the expression levels, categorized into “high”, “medium”, “low” and “not detected”, are annotated. The cancer association was hypothesized when the sum of “high” and “medium” occurrence was greater than 60% of the total amount of patients for the tumor type.

2.5. Genome-wide screening A pool of 4746 nonessential diploid deletion strains (yeast deletion pool http://tools.invitrogen.com/content/sfs/manuals/yeast) each containing a unique 20 base pair genetic tag was obtained from Invitrogen (Carlsbad, CA). Yeast deletion strains were thawed from a frozen aliquot of the pooled strains (200 ␮l, at room temperature) and grown by inoculating the thawed cells into a 50 ml volume of YPD liquid medium maintained at 30 ◦ C with shaking until a cells count of 1–2 × 107 cells/ml was obtained. Cells were transformed with 5 ␮g of pYES-PARP-1 and plated directly to SC-URA containing 5% galactose. Rapidly growing colonies were picked following 3–5 days incubation at 30 ◦ C. Individual colonies were streaked again on SC-URA 5% galactose plates to confirm the phenotype. The transformation efficiency was checked by plating in SC-URA glucose. The average transformation efficiency was 1–5 × 104 URA3 transformants per ␮g plasmid DNA. The BY4743 strain was used as parental control for the genetic screening.

2.7. Fluorescence microscopy Cells expressing PARP1-GFP were grown in glucose-containing medium for 24 h at 30 ◦ C under constant shaking. Cells were diluted 1:4 in galactose-containing medium and allowed to grow for an additional 20 h. To visualize PARP-1-GFP living cells, 400 ␮l of yeast culture was collected by centrifugation, washed with water and resuspended in 400 ␮l of water. The DNA was stained by incubating the cells with 4 ␮M Hoescht 33342 at room temperature for 30 min. Twenty microliter aliquots of cells were then spotted on glass slides and observed under the microscope Zeiss Observer.Z1. GFP was excited with 488 nm light, and the emission was collected using a 495–527 nm band-pass filter. Hoescht was excited with UV light, and the emission was collected using a 418–466 nm band-pass filter. Further processing of images was performed using ImageJ1.47.

M. La Ferla et al. / Mutation Research 774 (2015) 14–24

3. Results 3.1. The expression of human PARP-1 affects cell growth and survival in yeast To determine the effect of PARP1 on yeast growth and viability, the haploid strains RSY12 and RSY6, and the diploid strains RS112 and BY4743 of S. cerevisiae were transformed with the pYES2PARP1 vector. The same strains were transformed with the empty vector pYES2 as negative control. To determine if the yeast strains were able to sustain human PARP1 expression, single clones from pYES2-PARP1 transformed RSY12, RS112 and BY4743 strains were grown in 5 ml SC-URA containing 2% glucose for 24 h at 30 ◦ C under constant shaking. Then, cells were washed twice in distilled water and dilute in 20 ml SC-URA with 5% galactose and incubated for 24 h at 30 ◦ C under shaking to induce the PARP1 expression. Total protein extract from PARP1 expressing yeast were prepared and analyzed by Western blot as described in Section 2. Results indicated that all the strains were able to express the human PARP1 (Fig. 1A, lane 2–5). In parallel, yeast extracts from yeast grown in glucose were analyzed by Western blot and no PARP-1 band was detected (Fig. 1A, lane 6 and data not shown). Then, we would like to demonstrate whether PARP-1 were able to produce PARylated proteins in yeast. Western blot analysis showed that PARP1 expressing

17

cells (BY4743 strain) contain a high level of PAR products as compared to control cells (Fig. 1B, lanes 1 and 2 compared to lane 5) that decreased in the presence of PHE (Fig. 1B lanes 3 and 4). Finally, PARP-1 level was not affect by PHE treatment (data not shown). To determine the effect of PARP-1 on growth, BY4743 strain carrying pYES2-PARP-1 plasmid was grown for 24 h in galactose or in glucose. Then, cell number was measured by checking the OD at 600 nm assuming that 0.1 OD corresponds to 1.0 × 106 cells/ml [30]. When cells were grown without expressing PARP-1 (glucose or pYES2 galactose), cells number increased from 4.0 × 105 cells/ml to 9.5–13 × 107 cells/ml (Table 1). In galactose, when PARP-1 was expressed, the strain grew less than the negative control (Table 1). Interestingly, when cells were incubated in the presence of increasing concentration of the PARP-1 inhibitor PHE, growth was restored and the cell number significantly increased (Table 1). Particularly, at the highest PHE concentration, the cell number reached 1.27 × 107 cells/ml, that is comparable to the cell number obtained when PARP-1 was not expressed (Table 1). We have also evaluated the effect of PARP-1 expression on cell survival by plating directly onto galactose and glucose medium. The cell survival was determined as percent colony forming efficiency calculated by dividing the number of colonies grown in galactose to the number of colonies counted in glucose plates. In all the strains, the expression of PARP-1 reduced cell survival to 2.0–5.6% confirming that

Fig. 1. Expression of human PARP-1 and PAR product formation in the haploid strain RSY12 and RSY6, and diploid strain RS112 and BY4743 of Saccharomyces cerevisiae. (A) Human PARP-1 was detected in the total protein extracts from yeast strains grown in glucose and 24 h in galactose by Western blot analysis with anti-PARP-1 antibody. Extracts were loaded as follows: lane 1, RSY12-pYES2 grown in galactose; lane 2, RSY12-pYES2-PARP-1 grown in galactose; lane 3, RS112-pYES2-PARP-1 grown in galactose; lane 4, RSY6-pYES2-PARP-1 grown in galactose; lane 5, BY4743-pYES2-PARP-1 grown in galactose; lane 6, BY4743-pYES2-PARP-1 grown in glucose. In the lower part of the figure, loading control evaluated by detecting the 3PGK band is shown. (B) PAR product was detected in the total protein extracts by Western blot analysis with anti-PAR antibody. Extracts were loaded as follows: lanes 1 and 2, BY4743-pYES2-PARP-1 grown in galactose; lanes 3 and 4, BY4743-pYES2-PARP-1 grown in galactose in the presence of 100 ␮M PHE; lane 5, BY4743-pYES2 grown in galactose. In the lower part of the figure, loading control evaluated by detecting the 3PGK band is shown. (C) Effect of PARP1 expression on colony forming efficiency in RS112 strain. Cells carrying pYES2 or pYES2-PARP-1 were grown in SC-URA glucose for 24 h. Then same amount of cells were plated onto SC-URA glucose and SC-URA galactose.

18

M. La Ferla et al. / Mutation Research 774 (2015) 14–24

Table 1 Effect of the PARP-1 inhibitor PHE on PARP-1 induced cell growth in yeast.

0 0.5 ␮M 5 ␮M 25 ␮M 50 ␮M

pYES2 cells/ml (×105 )

pYES2 PARP-1 cells/ml (×105 )

24 h in GLUCOSE 103.2 ± 22.6 120.3 ± 25.3 127.8 ± 9.3 123.1 ± 8.2 126.9 ± 3.4

95.1 108.3 102.6 125.4 130.1

± ± ± ± ±

25.9 22.9 19.1 27.8 25.4

pYES2 cells/ml (×105 )

pYES2 PARP-1 cells/ml (×105 )

24 h in GALACTOSE 99.1 ± 29.1 119.0 ± 8.9 107.5 ± 16.3 121.5 ± 25.7 130.2 ± 22.9

54.4 63.3 109.1 112.4 127.0

± ± ± ± ±

6.1 9.9 38.4* 7.1*** 31.9**

Yeast cells carrying empty pYES2 or the pYES2-PARP-1 were pre-grown in SC-URA for 24 h. Thereafter, aliquots corresponding to 4.0 × 105 cells/ml were incubated for additional 24 h in glucose or galactose with different PHE concentrations. Then, cell number was determined as OD at 600 nm and counted under the microscope as described in Section 2. Experiments were performed in triplicate and results are reported as mean of 3 independent experiments ± standard deviation. Statistics was performed by comparing different PHE concentrations to the untreated cells (0) using the Student’s “t” test. * p ≤ 0.05. ** p ≤ 0.01. *** p ≤ 0.001.

this protein has a lethal effect in yeast (Table 2, Fig. 1C) [22,42]. In conclusion, the expression of human PARP-1 has a lethal effect after 3–5 days of continuous induction onto galactose plates, whereas, in liquid medium, cells just decreased growth without lethality. 3.2. PARP-1 expression decreased UV-induced homologous recombination In mammalian cells, PARP-1 overexpression decreases alkylation-induced sister chromatid exchange and the inhibition of PARP-1 increases intra-chromosomal recombination [43,44]; on the other hand, cells with non-functional PARP-1 show increased levels of sister chromatid exchange [45]. Moreover, in parp1−/− mice, spontaneous homologous recombination is significantly higher than in the wild type [46]. Therefore, we wanted to address whether PARP-1 expression affects spontaneous and UV-induced homologous recombination in yeast. After 24 h in 5% galactose, cells were plated as described in Section 2 to determine spontaneous homologous recombination. We observed a weak but not significant increase of intra-chromosomal recombination in the RS112 strain carrying pYES2- PARP-1 in galactose as compared to the frequency in glucose (Table 3). However, in galactose the strain not expressing PARP-1 recorded the same frequency of intra-chromosomal recombination as the PARP-1 expressing strain (Table 3). These results clearly show that PARP-1 had no effect on yeast spontaneous homologous recombination. To determine the effect of PARP-1 expression on UV induced homologous recombination, after 24 h in galactose, cells were irradiated and recombination measured as described. In the PARP-1 non-expressing conditions, such as: pYES2 glucose, pYES2 galactose and pYES2-PARP-1 glucose, 200 and 400 J/m2 UV induced a statistically significant increase of intra- and interchromosomal recombination as compared to the untreated control

Table 2 Effect of PARP-1 expression on survival in the haploid RSY12 and RSY6, and diploid strains RS112 and BY4743 of Saccharomyces cerevisiae. Yeast Strain

pYES2

RSY12 RSY6 RS112 BY4743

79.9 63.3 90.0 89.3

± ± ± ±

pYES2-PARP-1 5.8 6.2 0.4 6.9

4.3 3.7 5.6 2.0

± ± ± ±

0.6 0.1 0.5 0.4

Yeast cells of the strains listed in the table were grown in SC-URA glucose for 24 h. Thereafter, cells were washed, diluted and were plated (200 cells each plate) onto SC-URA glucose and galactose. Results are expressed as percent survival calculated by dividing the number of colonies counted in galactose by the total colonies counted in glucose and normalized. Results are reported as mean of 4 independent experiments ± standard deviation.

(Table 4). Particularly, in the not expressing strains exposed to the highest UV dose, inter-chromosomal recombination increased by 71–177-fold and intra-chromosomal recombination increased by 13–15-fold confirming that intra- is less UV inducible than interchromosomal recombination [32,47]. In PARP-1 expressing cells, no or only a weak increase in intra-chromosomal recombination was observed (Table 4). Moreover, in PARP-1 expressing cells, UV induced a significant increase (up to 20-fold) in inter-chromosomal recombination than in the control PARP-1 negative cells (Table 4). In conclusion, PARP-1 did not affect spontaneous recombination, but has an effect on DNA damage-induced recombination in yeast. However, PARP-1 had no or little effect on UV survival (Table 4). In parallel, we checked the level of PARP-1 after UV radiation by Western blot analysis. After 400 J/m2 UV, PARP-1 was still detected in yeast cells at comparable level as the not irradiated control (data not shown). To understand if the reduction on UV-induced recombination is related to PARylation activity, we determined the effect of PARP1 inhibitor on UV-induced homologous recombination. PARP1 expression was induced in galactose for 24 h in the presence of 100 ␮m PHE; then cells were irradiated and plated as previously described. When 100 ␮m PHE was added to the control cells prior 200 J/m2 UV radiation, intra and inter-chromosomal recombination increased 9 and 33-fold respectively (Table 4). In PARP-1 expressing yeast cells and in the presence of PHE, both 200 and 400 J/m2 UV radiation induced a significant dose dependent increase of intraand inter-chromosomal recombination that was higher than in the absence of PHE (Table 4). This suggests that the PARylation activity is able to modulate UV induced recombination in yeast.

Table 3 Effect of PARP-1 expression on spontaneous homologous recombination in the RS112 strain of Saccharomyces cerevisiae. Intra-chromosomal recombination HIS3 colonies/104 cells pYES2 glucose pYES2 galactose pYES2-PARP-1 glucose pYES2-PARP-1 galactose

0.64 0.76 0.66 0.69

± ± ± ±

0.19 0.23 0.16 0.21

Inter-chromosomal recombination ADE2 colonies/105 cells 0.59 1.09 0.78 1.01

± ± ± ±

0.22 0.29 0.17 0.32

Yeast cells of the RS112 carrying pYES2 or pYES2-PARP-1 were grown in SC-URA glucose for 24 h. Thereafter, cell cultures were incubated in glucose or 5% galactose for 20 h to allow PARP-1 expression. Then, cells were washed diluted and plated in SC-HIS and SC-ADE to determine intra- and inter-chromosomal recombination events, respectively. The fraction of vital cells was determined by plating in complete medium as described in Section 2. Frequency of intra- and inter- chromosomal recombination is expressed as number of HIS3 colonies/104 cells and ADE2 colonies/105 cells, respectively. Results are reported as mean of 4–6 independent experiments ± error bars.

M. La Ferla et al. / Mutation Research 774 (2015) 14–24

19

Table 4 Effect of PARP-1 expression on UV induced intra- and inter-chromosomal recombination in the RS112 strain of Saccharomyces cerevisiae. Strain

UV (J/m2 )

% survival

Intra-chromosomal recombination HIS3 colonies/104 survivors (fold increase)

pYES2 Glucose

0 100 200 400 0 100 200 400 PHE+0 PHE+200

100 55.0 ± 4.3 22.3 ± 6.7 3.3 ± 1.8 100 24.7 ± 6.7 15.1 ± 0.8 2.3 ± 1.0 100 16.8 ± 2.4

0.81 2.81 6.54 10.50 1.12 3.43 8.26 16.29 1.58 14.21

± ± ± ± ± ± ± ± ± ±

0.06 (1) 0.21** (3) 2.85** (5) 3.33** (13) 0.27 (1) 1.47 (3) 0.61*** (7) 1.91*** (14) 0.63 (1) 0.58*** (9)

0.41 ± 0.21 (1) 33.82 ± 4.55*** (83) 52.23 ± 3.40*** (125) 72.94 ± 17.41*** (177) 0.50 ± 0.13 (1) 20.04 ± 5.82** (40) 32.43 ± 3.73*** (65) 59.13 ± 12.79*** (109) 0.83 ± 0.07 (1) 27.49 ± 3.19*** (33)

0 100 200 400 0 100 200 400 PHE+0 PHE+100 PHE+200

100 37.5 ± 1.5 10.5 ± 1.3 2.9 ± 0.6 100 46.6 ± 1.9 8.0 ± 2.4 2.7 ± 0.5 100 53.5 ± 10.4 14.6 ± 4.5

0.74 3.48 5.06 10.95 1.43 2.86 1.60 1.41 2.04 6.68 12.70

± ± ± ± ± ± ± ± ± ± ±

0.57 (1) 1.85* (5) 0.93** (7) 0.80*** (15) 0. 3 (1) 0.55* (2) 0.27 (1) 0.62 (1) 0.94 (1) 2.31** (3) 1.32 (6)

0.54 16.25 22.36 38.28 0.66 12.82 13.94 13.30 1.21 36.96 65.70

Galactose

Galactose pYES-PARP-1 Glucose

Galactose

Galactose

Inter-chromosomal recombination ADE2 colonies/105 survivors (fold increase)

± ± ± ± ± ± ± ± ± ± ±

0.08 (1) 0.91*** (30) 6.84** (41) 5.29*** (71) 0.42 (1) 4.80** (19) 4.03** (21) 2.28 ** (20) 0.52 (1) 7.06*** (30) 6.56*** (54)

Yeast cells of the RS112 carrying pYES2 or pYES2-PARP-1 were grown in SC-URA glucose for 24 h. Thereafter, cell cultures were incubated in glucose or galactose for 20 h to allow expression of PARP-1. As described in Section 2, cells were irradiated, post-incubated, washed, diluted and plated in SC-HIS and SC-ADE to determine intra- and interchromosomal recombination events, respectively. The fraction of survival was determined by plating in YPAD. Frequency of intra- and inter- chromosomal recombination is expressed as number of HIS3 colonies/104 cells and ADE2 colonies/105 cells, respectively. Results are reported as mean of 4–6 independent experiments ± standard deviation. In the brackets, the fold increase over the untreated control was reported. Results were statistically analyzed using the Student’s “t” test. The probabilities refer to the comparison between the exposure and the untreated control. * p ≤ 0.05. ** p ≤ 0.01. *** p ≤ 0.001.

3.3. A genome-wide screen identifies genes that suppress PARP-1-induced lethality in yeast Although the PARP-1 orthologous gene is not found in the yeast S. cerevisiae, PARP-1 expression in yeast resulted in slow growth and lethality. Moreover, exposure to the PARP-1 inhibitor PHE restored the growth. In addition, we have shown here that PARP1 decreased DNA damage induced recombination in yeast leading to the conclusion that yeast presumably possesses cellular PARP-1 targets. Therefore, we reasoned that an interaction of PARP-1 with a conserved yeast protein(s) could interfere with a basic cellular function essential for yeast survival and that these proteins would be involved in the modulation of PARP-1 activity. As the survival fraction after prolonged 5% galactose exposure of the BY4743 strain is 2% (Table 2, Fig. 1C), theoretically, at least 98% strains acquired the ability to form colonies in galactose as consequence of the deletion that, therefore, suppressed the PARP-1 induced growth inhibition. By screening for rapid growth on galactose medium immediately following transformation of the PARP-1 plasmid into a pool of genetically tagged deletion strains, we have characterized a total of 99 deletion strains that suppressed PARP-1 growth inhibition. All the identified genes were listed and described in Table S1. The human homologous genes were found for a total of 41 yeast genes; the complete list is reported in Table S2. Notably, 22 out of 41 yeast genes have more than one human orthologous; totally, we counted 102 human genes homologous to the newly identified yeast genes (Table S2). When yeast and human genes were grouped according to cellular process or biological function, results showed a significant overlapping between yeast and human was observed (Table S3). Interestingly, according to Gene Ontology classification, the genes CLF1, CRF1, HHO1, SNT1 and VPS72 belong to the “chromatin” cellular component and, therefore, may affect the PARP-1 activity or protein level. Moreover, we found that a total of 20 genes are involved in RNA processing, transcription, histone

modification, chromatin remodeling, protein degradation or localization; these gene products may interfere with PARP-1 activity and affect the intracellular level or nuclear localization; therefore, they could have role in the sensibility to PARP-1 inhibitors (Table S4). As PARP-1 is considered a molecular target in cancer therapy, we asked whether the homologous human genes were differential expressed in several kinds of tumor by interrogating the “human protein atlas” database (http://www.proteinatlas.org/) [48]. Among the 102 human orthologous genes identified, a total 91 were reported in the database (Table S5); 52 of them were found to be differentially expressed in at least one cancer (Table S5). Particularly, the genes ARGLU1, H1F0, RCOR1, RCOR2, TALDO1 and ZNF207 are differentially expressed in all the types of tumors (Table S5). Supplementary Table S1 related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mrfmmm. 2015.02.006. Supplementary Table S2 related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mrfmmm. 2015.02.006. Supplementary Table S3 related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mrfmmm. 2015.02.006. Supplementary Table S4 related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mrfmmm. 2015.02.006. Supplementary Table S5 related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mrfmmm. 2015.02.006. Here, we decided to further investigate the strains listed in Table 5. All the genes deleted in the strains selected have one or more human homologous gene (Table 5). Then, we determined whether the PARP-1 protein level was altered in the strain which is deleted for the transcription regulator GAL3, the histone H1 gene HHO1, the HUL4 gene, the deubiquitination enzyme gene OTU1,

20

M. La Ferla et al. / Mutation Research 774 (2015) 14–24

Table 5 Yeast deletion pool library genes identified as resistant to PARP-1 growth inhibition; their function and human homologous genes are reported. Yeast Gene

Description

Human Gene

Description

GAL3

Transcriptional regulator involved in activation of the GAL genes in response to galactose Histone H1, a linker histone required for nucleosome packaging

GALK2 GALK1 H1F0 H1FNT H1FX HIST1H1T HERC6

Galactokinase 2 Galactokinase 1 H1 histone family H1 histone family, member N H1 histone family, member Histone cluster 1, H1t HECT and RLD domain containing E3 ubiquitin protein ligase family member 6 Ubiquitin protein ligase E3A HECT and RLD domain containing E3 ubiquitin protein ligase 4 HECT domain containing E3 ubiquitin protein ligase 2 HECT and RLD domain containing E3 ubiquitin protein ligase 5 HECT and RLD domain containing E3 ubiquitin protein ligase OTU deubiquinating enzyme 1 homolog Nucleoporin 210 kDa REST (repressor element-1 silencing transcription factor, a component of the histone deacetylase) corepressor 1 REST corepressor REST corepressor 3 Nuclear receptor corepressor 1 Nuclear receptor corepressor 2

HHO1

HUL4

OTU1 POM152 SNT1

Protein with similarity to HECT domain E3 ubiquitin-protein ligases

De-ubiquitylation enzyme Nuclear pore membrane glycoprotein Subunit of the Set3C histone deacetylase complex

UBE3A HERC4 HECTD2 HERC5 HERC3 YOD1 NUP210 RCOR1 RCOR2 RCOR3 NCOR1 NCOR2

Human orthologous genes were retrieved from g:ProfileR database.

the nuclear pore protein POM152 and SNT1 that encodes for the Set3C subunit of the histone deacetylase complex. PARP-1 expression was induced by 24 h growth in galactose as described in Section 2. Western blot analysis showed that in the strains hho1, hul4, otu1, pom152 and snt1, the PARP-1 level was roughly the same as in the wild type (Fig. 2A). No PARP-1 specific band was obeserved in the protein extract from the gal3 strain confirming the defect of galcose induction of this strain [49,50] (Fig. 2A). Similarly, no PARP-1 was detected in the strains after growth in glucose or carrying empty vectors (data not shown). We have also determined the growth in galactose by plating serial dilutions of yeast cultures grown in SC-URA medium. All the strains, except for the BY4743 wild type, are able to growth in galactose (Fig. 2B). The gal3 strain was able to grow in 5% galactose, but did not express

PARP-1 at detectable level in the Western blot; therefore, it could be considered negative control. 3.4. PARP-1 localization following UV radiation is altered in the hho1 and pom152 strain One of the mechanisms that can be suggested to explain the lack of PARP-1 sensitivity is a change in protein localization. As PARP-1 is mainly involved in DNA repair and interacts with DNA [2,3], it should be mainly localized in the nucleus after DNA damage. Therefore, we thought to investigate the PARP-1 localization following UV radiation by expressing PARP-1 fused to GFP gene. To this purpose, cells of the gal3, hho1, hul4, otu1, pom152 and stn1 strain was first selected for loss of PARP-1 plasmid by

Fig. 2. Expression of human PARP-1 in the BY4743 wt and in the hho1, hul4, stn1, pom152, otu1 and gal3 deleted strains of Saccharomyces cerevisiae. (A) Western blot analysis of total protein extracts from yeast strains grown in glucose and 20 h in 5% galactose. Extracts were loaded as follows: lane 1, BY4743 wt pYES2 grown in galactose; lane 2, BY4743 wt pYES2-PARP-1 grown in glucose; lane 3, BY4743 wt pYES2-PARP-1 grown in galactose; lanes 4–8, hho1, hul4, stn1, pom152, otu1 deleted strains carrying pYES2PARP-1 grown in galactose; lane 9, gal3 deleted strain carrying pYES2-PARP-1 grown in galactose. In the lower part of the figure, loading control evaluated by detecting the PGK band is shown. (B) Effect of PARP1 expression on growth BY4743 wt and hho1, hul4, stn1, pom152, otu1 and gal3 deleted strains. Cells of the BY5743 wt and hho1, hul4, stn1, pom152, otu1 and gal3 deleted strains carrying pYES2 or pYES2-PARP-1 were grown in SC-URA glucose for 24 h. Then, serial dilutions (from 100 to 10−5 ) were spotted onto SC-URA glucose and SC-URA galactose plates, and incubated for 4 days at 30 ◦ C.

M. La Ferla et al. / Mutation Research 774 (2015) 14–24

21

Table 6 Effect of UV radiation on PARP-1 localization in the BY4743 wild type and deletion strains. 24 h in galactose + UV 400 J/m2

24 h in galactose % GFP positive cells BY4743 wild type gal3 hho1 hul4 pom152 snt1 otu1

51.5 26.5 54.8 45.6 46.1 49.1 46.3

± ± ± ± ± ± ±

2.9 2.6 2.9 2.3 3.1 3.0 1.9

% cells with nuclear PARP-1

% GFP positive cells

1.8 ± 1.1 0 4.5 ± 1.7 4.3 ± 1.1 0 6.1 ± 1.8 2.8 ± 1.9

30.1 15.6 52.3 48.1 47.5 43.6 47.3

± ± ± ± ± ± ±

2.8 2.5 6.2 2.5 3.2 2.3 2.3

% cells with nuclear PARP-1 1.7 ± 0.9 0 11.1 ± 1.6 4.8 ± 2.4 3.1 ± 0.9 9.6 ± 2.5 1.5 ± 0.7

Yeast cells of the BY4743 and deletion strains carrying pYES2-PARP-1GFP were grown in SC-URA glucose for 24 h. Thereafter, cell cultures were incubated in 5% galactose for 20 h to allow PARP-1 expression. As described in Section 2, cells were irradiated, treated with Hoescht and analyzed by fluorescence microscopy. For each slide, at least 900 cells were counted. The cells with nuclear localization were counted as percent of the GFP+ cells. Results are reported as mean of at least 3 independent experiments ± error bars.

plating in YPAD and replicating them in SC-URA. Then, PARP-1-GFP carrying strains were obtained by transforming the URA3− single colonies with the pYES2 PARP-1-GFP plasmid. Single URA3 colonies were inoculated for 24 h in liquid SC-URA glucose. Then, the expression of PARP-1-GFP was induced in SC-URA 5% galactose for 24 h, irradiated with 400 J/m2 UV and analyzed by fluorescence microscopy. As already reported, human PARP-1 localized in the nucleus when overexpressed in yeast [42]. In the BY4743 wild type strain, 51.5% cells were GFP positive with only 1.8% localized in the nucleus (Table 6, Fig. 3A). After UV, the GFP positive cells weakly decreased to 30.1%, whereas the percentage of cells where PARP-1 is distributed in the nucleus is not affected (Table 6, Fig. 3B). In the galactose-induction defective strain gal3, 26.5% cells were GFP positive and decreased to 15.6% after UV radiation. No nuclear localization was observed in this strain (Table 6). The gal3 strain has been reported to have defect in galactose induction that could determine the low level of GFP positive cells that correlates with

the negative results in Western blot analysis, and, consequently, no nuclear localization [49]. In the hho1, hul4, otu1, pom152 and stn1 strain, the amount of GFP positive cells ranged from 54.8 to 46.1% and did not change after UV radiation (Table 6, Fig. 3A and B). In the hho1 strain, the number of cells where PARP-1 was localized in the nucleus is 4.5% that increased to 11.1% after UV treatment; similarly, in the pom152 strain, UV increased nuclear localization up to 3.1%. Moreover, when the hho1, hul4, otu1 and stn1 strains were not exposed to UV, showed a higher level of nuclear localization than the wild type; after UV radiation, hho1, hul4, otu1, pom152 and stn1 strain showed higher level of nuclear localization than the wild type (Table 6). To finally confirm that in the pom152 strain PARP1 localized in the nucleus after UV radiation, we prepared nuclear and cytoplasmic protein extracts and performed Western blot analysis. In the BY4743 strain, PARP-1 was detected in nuclear and cytoplasmic extracts from UV exposed and not exposed cells (Fig. 4A and B, lanes 1 and 3). Interestingly,

Fig. 3. Expression and localization of human PARP-1 in the BY4743 wt and hho1, hul4, stn1, pom152, otu1 and gal3 deleted strains of Saccharomyces cerevisiae. (A) Cells carrying pYES2-PARP-1-GFP were grown in glucose for 24 h and then, transferred in 5% galactose for 20 h to allow the expression of PARP-1-GFP. (B) After 20 h in galactose cell were exposed to 400 J/m2 as described. Then, aliquots of 400 ␮l was collected by centrifugation, washed with distilled water and resuspended in 400 ␮l of distilled water. The DNA was stained by incubating the cells with 4 ␮M Hoescht at room temperature for 30 min. Cells were then spotted on glass slides and observed under the microscope Zeiss Observer.Z1. GFP was excited with 488 nm light, and the emission was collected using a 495–527 nm band-pass filter. Hoescht was excited with UV light, and the emission was collected using a 418–466 nm band-pass filter. White and red arrow indicates nuclear and diffused localization, respectively.

22

M. La Ferla et al. / Mutation Research 774 (2015) 14–24 Table 7 Human proteins classified as PARP-1 target. Swissprot

Human gene

Description

Q9Y2I7 P51570 Q9NQS7 Q96HL8 Q9BZJ0 P52789 Q2TB90 Q8N3Z6 Q9BQ24 Q92522

FYV1 GALK1 INCE SH3Y1 CRNL1 HXK2 HKDC1 ZCHC7 ZFY21 H1FX

FYVE finger-containing phosphoinositide kinase Galactokinase 1 Inner centromere protein SH3 domain-containing YSC84-like protein 1 Crooked neck-like protein 1 Hexokinase-2 Putative hexokinase HKDC1 Zinc finger CCHC domain-containing protein 7 Zinc finger FYVE domain-containing protein 21 Histone H1x

These genes/proteins were identified as homologous to those ones deleted in the yeast deletion pool that resulted resistant to PARP-1 growth defect. These genes were previously reported as PARP-1 target because contain the PAR binding site [58].

Fig. 4. Effect of UV exposure on PARP-1 nuclear and cytoplasmic localization in the strain BY4347 wt and pom152. Cells were exposed to UV radiation and fractionated as reported: nuclear and cytoplasmic fractions from PARP-1 expressing cells were loaded and analyzed by Western blot with anti-PARP-1 antibody. Anti-fibrillarin antibody was used as nuclear specific marker; anti-PGK antibody was used as cytoplasmic marker. (A) Nuclear and cytoplasmic fractions from not exposed BY4743 wt cells (lanes 1 and 3) or BY4743 pom152 (lanes 2 and 4) were loaded. (B) Nuclear and cytoplasmic fractions from UV exposed BY4743 wt cells (lanes 1 and 3) or BY4743 pom152 (lanes 2 and 4) were loaded. Fibrillarin was detected only in the nuclear extract and PGK much more in the cytoplasmic extracts.

in the pom152 strain, no PARP-1 was detected in the nuclear extract from untreated cells (Fig. 4A, lane 2) confirming results obtained with the fluorescence microscopy. Moreover, PARP-1 was detectable in the nuclear extract from the pom152 strain exposed to UV (Fig. 4B, lane 2). Finally, in the pom152 strain, PARP-1 is always present in the cytoplasmic protein extract from UV exposed or control cells (Fig. 4A and B, lane 4). 4. Discussion The inhibition of PARP-1 is a potential synthetic lethal therapeutic strategy for the treatment of cancers with specific DNA-repair defects, including those arising in carriers of a BRCA1 or BRCA2 mutation [18]. As PARP-1 is one of the most important responders to DNA damage, the therapeutic effect of PARP inhibitors on cancer cells is observed also in combination with DNA damaging chemotherapeutic agents [51]. In the presence of DNA damage, PARP-1 reaches the DNA damaged sites and regulates a catalytic activation response that globally affects the DNA damage response and repair [11,52]. Among DNA repair pathways, PARP-1 has an impact on base and nucleotide excision repair, and DNA double strand breaks repair via homologous recombination and non-homologous end-joining [53,14,54–56]. Recently, in cancer therapy, the resistance to PARP inhibitors is starting to be observed [17]. Therefore, it would be important to find new factors controlling the PARP-1 expression or activity in order to a better understanding of different mechanisms of resistance to PARP inhibitors for the “personalization” of therapy [20,51]. Potential mechanisms by which resistance to PARP inhibitors occurs in cancer cells, may include altered DNA repair/recombination capacity, altered PARP-1 expression level or activity and defect in intracellular availability or localization [17,20]. In the present study, we thought to use the yeast S. cerevisiae to gain some clues on PARP-1 nuclear localization and activity. We confirmed that growth defect induced in yeast by PARP-1 expression is due to the

poly-ADP ribose polymerase activity because it reversed, when the PARP inhibitor PHE is used [42,22]. Moreover, we showed the PAR products are formed when PARP-1 is expressed and decreased when PARP-1 is inhibited by PHE, confirming that yeast may possess PARP-1 target [42,22]. Interestingly, we found that PARP-1 expression reduced the UV-induced homologous recombination in yeast; as this effect is affected by the PARP inhibitor PHE, it suggests that in yeast as in mammalian cells, PARP-1 may interfere in DNA repair pathways [12,43,46,57]. As previously reported in our strain, intra-chromosomal recombination events are due to several mechanisms that could involve sister chromatids [47]. In agreement with our results, PARP-1 has been demonstrated to reduce the frequency of DNA damage-induced sister chromatid exchange in Chinese hamster cell lines [43]. Our results clearly indicate that in S. cerevisiae, PARP-1 may have the same targets as in mammalian cells; this observation makes yeast a very attractive system to investigate for functionally related partners. In trying to identify new factors affecting in PARP-1 activity and presumably involved in PARP-1 inhibition, we took advantage of the evolutionary distance between humans and yeast to screen for proteins and/or processes that have the ability to interact functionally with PARP-1; these PARP-1-interacting partners have been identified by finding yeast deletions that suppress PARP1-induced growth defect. By screening a deletion barcoded yeast pool, we identified a total of 99 gene deletions that suppress PARP-1-growth inhibition. Previously, utilizing a yeast proteome microarray screening that is based on detection poly ADP ribosylated proteins, 33 putative PARP-1 substrates have been identified [42]. The protein encoded by DOT5 is a potential PARP-1 substrate and here, we found that its deletion suppressed the PARP-1 induced growth defect. Interestingly, out of 33 substrates only DOT5 came out in our genetic screening indicating that we identified factors presumably involved in PARP-1 activity [42]. In addition, we found 10 proteins that are likely to be potential target of human PARP-1 (Table 7) [58]. These proteins have been previously identified as poly (ADP-ribose)-associated protein complexes and contain the poly (ADP-ribose) binding site [58]. Bioinformatics analysis showed that human orthologous genes belong to the same pathways as yeast, confirming again the usefulness of this model to identify new functionally interacting factors of human proteins involved in DNA repair [59]. However, no protein directly involved in DNA repair has been identified. The histone H1 encoded by the HHO1 gene is the only gene identified by this screening that affects homologous recombination in yeast [60]. We found that 20 genes encoding for proteins that may affect the level and localization of human PARP1 in yeast (Table S4). Two of them, HUL4 and OTU1 are involved in post-transduction modification that may affect the intracellular protein level and/or localization [61,62]. HHO1 and SNT1 could also be involved in transcription and, therefore, may have an impact

M. La Ferla et al. / Mutation Research 774 (2015) 14–24

on the intracellular level of PARP-1 [5,63]. POM152 encodes for a subunit of transmembrane ring of nuclear pore complex and contributes to nucleus–cytoplasmic transport [64]. With no DNA damage, PARP-1 localized in the nucleus more in the snt1 than in the wild type strain; after UV radiation, PARP-1 localized in the nucleus more in hho1 and pom152 deletion strains than in the wild type indicating that these functions may have a role on regulating PARP-1 level and activity in the nucleus. We can hypothesize that the human homologs to the yeast HHO1, POM152 or SNT1 genes could affect the PARP-1 localization; therefore, if these genes were overexpressed in human cancers, the PARP-1 “nuclear localization could be actually prevented”. On the other hand, if these genes were mutated or poorly expressed in cancer patients the PARP-1 “nuclear localization could be enhanced. This could be relevant for using PARP inhibitors in cancer therapy. By interrogating the “protein atlas” database, we found that H1F0, RCOR1, RCOR2 and RCOR3 that are homologous to the yeast genes HHO1 and SNT1, were found overexpressed in human cancers. PARP-1 functionally related genes can be differentially expressed in human cancers and impact the sensitivity to PARP-1 inhibition. Thus, our screening identified new factors that may be relevant for the respond to PARP-1 inhibition. Moreover, the genes ARGLU1, TALDO1 and ZNF207 are differentially expressed in all the types of tumors. ARGLU1, the SEC3 human counterpart, localizes to estrogen receptor target gene promoter upon estrogen induction and its depletion impairs breast cancer cell growth [65]. The expression of the histone H1 human homologous H1F0 has been found to vary in response of therapy to acute leukemia [66]; similarly, the expression of RCOR1 (homologous to the yeast SNT1) was found altered in carboplatin sensitive ovarian cancer cells as compared to resistant cells [67]. RCOR2 (homologous to the yeast gene SNT1) was found to be a target of the hypoxia inducible factor HIF that is important in several pathologies including cancer [68]. TALDO1 that is homologous to the yeast gene NQM1 has been demonstrated to be differentially expressed in colon cancer cells in response to chemotherapy [69]. Therefore, these proteins may have a role in cancer therapy response and also, in the response to PARP-1 inhibitors. It would be interesting to investigate whether these proteins are differentially expressed in PARP-1 insensitive patients or in cancer cells treated with PARP inhibitors. 5. Conclusions Although PARP-1 is absent in yeast, it may interact with DNA recombination/repair pathways as in human cells. This suggests that there are certain functional similarities between PARP-1 activity in humans and yeast. We showed that the yeast genes SNT1, HHO1 and POM152 have a role in regulating the intracellular PARP1 nuclear localization and, consequently, PARP activity. Moreover, the human homologous genes has been found to be overexpressed in several kinds of cancer indicating that these functions should be further investigated to evaluate theirs role in therapy. The findings suggest that nuclear localization following DNA damage may have a role in the response to PARP inhibitors in cancer therapy. Conflict of interest statement The authors declare that they have no competing interests. Authors’ contributions MLF and GR performed all the experiments concerning PARP1 expression, yeast screening and localization. AM performed bioinformatics analyses. SL, TC and LP carried out the UV experiments and Western blot analysis. MAC was involved in the evaluation and

23

discussion of the genetic data. AG coordinated the whole work, evaluated all the data and wrote the manuscript. All authors read and approved the final manuscript. Acknowledgements The authors thank Robert Schiestl, Hung-wen Liu and Teri Melese for yeast strains and plasmids. Special thanks to Laura Spugnesi for technical assistance in protein expression and Western blot experiments. The work was supported by a grant from “Istituto Toscano Tumori” Regione Toscana-(ITT-grant number AOOGRT/0325418/Q.08.110) and partially, by a grant for the Italian Association for Cancer Research (AIRC, Grant IG 2013 n. 14477) to A.G. References [1] D. D’Amours, S. Desnoyers, I. D’Silva, G.G. Poirier, Poly(ADP-ribosyl)ation reactions in the regulation of nuclear functions, Biochem. J. 342 (Pt 2) (1999) 249–268. [2] W.L. Kraus, Transcriptional control by PARP-1: chromatin modulation, enhancer-binding, coregulation, and insulation, Curr. Opin. Cell Biol. 20 (2008) 294–302. [3] W.L. Kraus, J.T. Lis, PARP goes transcription, Cell 113 (2003) 677–683. [4] M.Y. Kim, S. Mauro, N. Gevry, J.T. Lis, W.L. Kraus, NAD+-dependent modulation of chromatin structure and transcription by nucleosome binding properties of PARP-1, Cell 119 (2004) 803–814. [5] R. Krishnakumar, M.J. Gamble, K.M. Frizzell, J.G. Berrocal, M. Kininis, W.L. Kraus, Reciprocal binding of PARP-1 and histone H1 at promoters specifies transcriptional outcomes, Science 319 (2008) 819–821. [6] J.B. Kirkland, Poly ADP-ribose polymerase-1 and health, Exp. Biol. Med. (Maywood) 235 (2010) 561–568. [7] P. Pacher, Poly(ADP-ribose) polymerase inhibition as a novel therapeutic approach against intraepidermal nerve fiber loss and neuropathic pain associated with advanced diabetic neuropathy: a commentary on PARP Inhibition or gene deficiency counteracts intraepidermal nerve fiber loss and neuropathic pain in advanced diabetic neuropathy, Free Radic. Biol. Med. 44 (2008) 969–971. [8] P. Pacher, C. Szabo, Role of poly(ADP-ribose) polymerase 1 (PARP-1) in cardiovascular diseases: the therapeutic potential of PARP inhibitors, Cardiovasc. Drug Rev. 25 (2007) 235–260. [9] D.G. Bebb, S.P. Lees-Miller, Predicting PARP inhibitor sensitivity and resistance, Cell Cycle 11 (2012) 4110. [10] A.J. Chalmers, Overcoming resistance of glioblastoma to conventional cytotoxic therapies by the addition of PARP inhibitors, Anticancer Agents Med. Chem. 10 (2010) 520–533. [11] B.A. Gibson, W.L. Kraus, New insights into the molecular and cellular functions of poly(ADP-ribose) and PARPs, Nat. Rev. Mol. Cell Biol. 13 (2012) 411–424. [12] P. Gottipati, B. Vischioni, N. Schultz, J. Solomons, H.E. Bryant, T. Djureinovic, N. Issaeva, K. Sleeth, R.A. Sharma, T. Helleday, Poly(ADP-ribose) polymerase is hyperactivated in homologous recombination-defective cells, Cancer Res. 70 (2010) 5389–5398. [13] M.K. Weil, A.P. Chen, PARP inhibitor treatment in ovarian and breast cancer, Curr. Probl. Cancer 35 (2011) 7–50. [14] M. De Vos, V. Schreiber, F. Dantzer, The diverse roles and clinical relevance of PARPs in DNA damage repair: current state of the art, Biochem. Pharmacol. 84 (2012) 137–146. [15] C.J. Lord, A. Ashworth, Targeted therapy for cancer using PARP inhibitors, Curr. Opin. Pharmacol. 8 (2008) 363–369. [16] C.J. Lord, A. Ashworth, Mechanisms of resistance to therapies targeting BRCAmutant cancers, Nat. Med. 19 (2013) 1381–1388. [17] A. Chiarugi, A snapshot of chemoresistance to PARP inhibitors, Trends Pharmacol. Sci. 33 (2012) 42–48. [18] P.C. Fong, D.S. Boss, T.A. Yap, A. Tutt, P. Wu, M. Mergui-Roelvink, P. Mortimer, H. Swaisland, A. Lau, M.J. O’Connor, A. Ashworth, J. Carmichael, S.B. Kaye, J.H. Schellens, J.S. de Bono, Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers, N. Engl. J. Med. 361 (2009) 123–134. [19] A. Lovato, L. Panasci, M. Witcher, Is there an epigenetic component underlying the resistance of triple-negative breast cancers to parp inhibitors? Front. Pharmacol. 3 (2012) 202. [20] A. Montoni, M. Robu, E. Pouliot, G.M. Shah, Resistance to PARP-inhibitors in cancer therapy, Front. Pharmacol. 4 (2013) 18. [21] P. Kaiser, B. Auer, M. Schweiger, Inhibition of cell proliferation in Saccharomyces cerevisiae by expression of human NAD+ ADP-ribosyltransferase requires the DNA binding domain (zinc fingers), Mol. Gen. Genet. 232 (1992) 231–239. [22] E. Perkins, D. Sun, A. Nguyen, S. Tulac, M. Francesco, H. Tavana, H. Nguyen, S. Tugendreich, P. Barthmaier, J. Couto, E. Yeh, S. Thode, K. Jarnagin, A. Jain, D. Morgans, T. Melese, Novel inhibitors of poly(ADP-ribose) polymerase/PARP1 and PARP2 identified using a cell-based screen in yeast, Cancer Res. 61 (2001) 4175–4183. [23] G.H. Tao, L.Q. Yang, C.M. Gong, H.Y. Huang, J.D. Liu, J.J. Liu, J.H. Yuan, W. Chen, Z.X. Zhuang, Effect of PARP-1 deficiency on DNA damage and repair in human

24

M. La Ferla et al. / Mutation Research 774 (2015) 14–24

[24] [25]

[26] [27]

[28] [29] [30] [31] [32] [33]

[34] [35] [36] [37]

[38] [39]

[40] [41] [42]

[43]

[44]

[45]

[46] [47]

[48]

[49]

bronchial epithelial cells exposed to benzo(a)pyrene, Mol. Biol. Rep. 36 (2009) 2413–2422. D. Botstein, S.A. Chervitz, J.M. Cherry, Yeast as a model organism, Science 277 (1997) 1259–1260. B. Suter, D. Auerbach, I. Stagljar, Yeast-based functional genomics and proteomics technologies: the first 15 years and beyond, Biotechniques 40 (2006) 625–644. L.H. Hartwell, Yeast and cancer, Biosci. Rep. 24 (2004) 523–544. J.S. McKinney, S. Sethi, J.D. Tripp, T.N. Nguyen, B.A. Sanderson, J.W. Westmoreland, M.A. Resnick, L.K. Lewis, A multistep genomic screen identifies new genes required for repair of DNA double-strand breaks in Saccharomyces cerevisiae, BMC Genomics 14 (2013) 251. D. Petranovic, J. Nielsen, Can yeast systems biology contribute to the understanding of human disease? Trends Biotechnol. 26 (2008) 584–590. W.H. Mager, J. Winderickx, Yeast as a model for medical and medicinal research, Trends Pharmacol. Sci. 26 (2005) 265–273. R.D. Gietz, R.H. Schiestl, High-efficiency yeast transformation using the LiAc/SS carrier DNA/PEG method, Nat. Protoc. 2 (2007) 31–34. R.H. Schiestl, S. Igarashi, P.J. Hastings, Analysis of the mechanism for reversion of a disrupted gene, Genetics 119 (1988) 237–247. R.H. Schiestl, R.D. Gietz, R.D. Mehta, P.J. Hastings, Carcinogens induce intrachromosomal recombination in yeast, Carcinogenesis 10 (1989) 1445–1455. T. Cervelli, A. Galli, Effects of HDF1 (Ku70) and HDF2 (Ku80) on spontaneous and DNA damage-induced intrachromosomal recombination in Saccharomyces cerevisiae, Mol. Gen. Genet. 264 (2000) 56–63. J.P. Aris, G. Blobel, Identification and characterization of a yeast nucleolar protein that is similar to a rat liver nucleolar protein, J. Cell Biol. 107 (1988) 17–31. H. Karathia, E. Vilaprinyo, A. Sorribas, R. Alves, Saccharomyces cerevisiae as a model organism: a comparative study, PLoS ONE 6 (2011). J. Reimand, T. Arak, J. Vilo, g:Profiler – a web server for functional interpretation of gene lists (2011 update), Nucleic Acids Res. 39 (2011) W307–W315. J. Reimand, M. Kull, H. Peterson, J. Hansen, J. Vilo, g:Profiler – a web-based toolset for functional profiling of gene lists from large-scale experiments, Nucleic Acids Res. 35 (2007) W193–W200. S. Draghici, P. Khatri, R.P. Martins, G.C. Ostermeier, S.A. Krawetz, Global functional profiling of gene expression, Genomics 81 (2003) 98–104. R.C. Gentleman, V.J. Carey, D.M. Bates, B. Bolstad, M. Dettling, S. Dudoit, B. Ellis, L. Gautier, Y. Ge, J. Gentry, K. Hornik, T. Hothorn, W. Huber, S. Iacus, R. Irizarry, F. Leisch, C. Li, M. Maechler, A.J. Rossini, G. Sawitzki, C. Smith, G. Smyth, L. Tierney, J.Y.H. Yang, J. Zhang, Bioconductor: open software development for computational biology and bioinformatics, Genome Biol. 5 (2004). R. Ihaka, R.R. Gentleman, A language for data analysis and graphics, J. Comput. Graph. Stat. 5 (1996) 299–314. U. Mudunuri, A. Che, M. Yi, R.M. Stephens, bioDBnet: the biological database network, Bioinformatics 25 (2009) 555–556. Z. Tao, P. Gao, H.W. Liu, Studies of the expression of human poly(ADPribose) polymerase-1 in Saccharomyces cerevisiae and identification of PARP-1 substrates by yeast proteome microarray screening, Biochemistry 48 (2009) 11745–11754. R. Meyer, M. Muller, S. Beneke, J.H. Kupper, A. Burkle, Negative regulation of alkylation-induced sister-chromatid exchange by poly(ADP-ribose) polymerase-1 activity, Int. J. Cancer 88 (2000) 351–355. A.S. Waldman, B.C. Waldman, Stimulation of intrachromosomal homologous recombination in mammalian cells by an inhibitor of poly(ADP-ribosylation), Nucleic Acids Res. 19 (1991) 5943–5947. N. Schultz, E. Lopez, N. Saleh-Gohari, T. Helleday, Poly(ADP-ribose) polymerase (PARP-1) has a controlling role in homologous recombination, Nucleic Acids Res. 31 (2003) 4959–4964. A. Claybon, B. Karia, C. Bruce, A.J. Bishop, PARP1 suppresses homologous recombination events in mice in vivo, Nucleic Acids Res. 38 (2010) 7538–7545. A. Galli, R.H. Schiestl, On the mechanism of UV and gamma-ray-induced intrachromosomal recombination in yeast cells synchronized in different stages of the cell cycle, Mol. Gen. Genet. 248 (1995) 301–310. M. Uhlen, P. Oksvold, L. Fagerberg, E. Lundberg, K. Jonasson, M. Forsberg, M. Zwahlen, C. Kampf, K. Wester, S. Hober, H. Wernerus, L. Bjorling, F. Ponten, Towards a knowledge-based human protein atlas, Nat. Biotechnol. 28 (2010) 1248–1250. C.Q. Diep, X. Tao, V. Pilauri, M. Losiewicz, T.E. Blank, J.E. Hopper, Genetic evidence for sites of interaction between the Gal3 and Gal80 proteins

View publication stats

[50] [51]

[52] [53]

[54]

[55]

[56]

[57]

[58]

[59]

[60]

[61]

[62] [63] [64]

[65]

[66]

[67]

[68]

[69]

of the Saccharomyces cerevisiae GAL gene switch, Genetics 178 (2008) 725–736. K. Weinhandl, M. Winkler, A. Glieder, A. Camattari, Carbon source dependent promoters in yeasts, Microb. Cell Fact. 13 (2014) 5. E.S. Ratner, A.C. Sartorelli, Z.P. Lin, Poly (ADP-ribose) polymerase inhibitors: on the horizon of tailored and personalized therapies for epithelial ovarian cancer, Curr. Opin. Oncol. 24 (2012) 564–571. M. Javle, N.J. Curtin, The potential for poly (ADP-ribose) polymerase inhibitors in cancer therapy, Ther. Adv. Med. Oncol. 3 (2011) 257–267. J. Yelamos, J. Farres, L. Llacuna, C. Ampurdanes, J. Martin-Caballero, PARP-1 and PARP-2. New players in tumour development, Am. J. Cancer. Res. 1 (2011) 328–346. B.S. King, K.L. Cooper, K.J. Liu, L.G. Hudson, Poly(ADP-ribose) contributes to an association between poly(ADP-ribose) polymerase-1 and xeroderma pigmentosum complementation group A in nucleotide excision repair, J. Biol. Chem. 287 (2012) 39824–39833. M. Robu, R.G. Shah, N. Petitclerc, J. Brind’Amour, F. Kandan-Kulangara, G.M. Shah, Role of poly(ADP-ribose) polymerase-1 in the removal of UV-induced DNA lesions by nucleotide excision repair, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 1658–1663. A. Pines, M.G. Vrouwe, J.A. Marteijn, D. Typas, M.S. Luijsterburg, M. Cansoy, P. Hensbergen, A. Deelder, A. de Groot, S. Matsumoto, K. Sugasawa, N. Thoma, W. Vermeulen, H. Vrieling, L. Mullenders, PARP1 promotes nucleotide excision repair through DDB2 stabilization and recruitment of ALC1, J. Cell Biol. 199 (2012) 235–249. H.E. Bryant, E. Petermann, N. Schultz, A.S. Jemth, O. Loseva, N. Issaeva, F. Johansson, S. Fernandez, P. McGlynn, T. Helleday, PARP is activated at stalled forks to mediate Mre11-dependent replication restart and recombination, EMBO J. 28 (2009) 2601–2615. J.P. Gagne, M. Isabelle, K.S. Lo, S. Bourassa, M.J. Hendzel, V.L. Dawson, T.M. Dawson, G.G. Poirier, Proteome-wide identification of poly(ADP-ribose) binding proteins and poly(ADP-ribose)-associated protein complexes, Nucleic Acids Res. 36 (2008) 6959–6976. C.B. Bennett, T.J. Westmoreland, C.S. Verrier, C.A. Blanchette, T.L. Sabin, H.P. Phatnani, Y.V. Mishina, G. Huper, A.L. Selim, E.R. Madison, D.D. Bailey, A.I. Falae, A. Galli, J.A. Olson, A.L. Greenleaf, J.R. Marks, Yeast screens identify the RNA polymerase II CTD and SPT5 as relevant targets of BRCA1 interaction, PLoS ONE 3 (2008) e1448. J.A. Downs, E. Kosmidou, A. Morgan, S.P. Jackson, Suppression of homologous recombination by the Saccharomyces cerevisiae linker histone, Mol. Cell 11 (2003) 1685–1692. K.S. Makarova, L. Aravind, E.V. Koonin, A novel superfamily of predicted cysteine proteases from eukaryotes, viruses and Chlamydia pneumoniae, Trends Biochem. Sci. 25 (2000) 50–52. G. Wang, J. Yang, J.M. Huibregtse, Functional domains of the Rsp5 ubiquitinprotein ligase, Mol. Cell. Biol. 19 (1999) 342–352. T. Kim, S. Buratowski, Dimethylation of H3K4 by Set1 recruits the Set3 histone deacetylase complex to 5’ transcribed regions, Cell 137 (2009) 259–272. R.W. Wozniak, G. Blobel, M.P. Rout, POM152 is an integral protein of the pore membrane domain of the yeast nuclear envelope, J. Cell Biol. 125 (1994) 31–42. D. Zhang, P. Jiang, Q. Xu, X. Zhang, Arginine and glutamate-rich 1 (ARGLU1) interacts with mediator subunit 1 (MED1) and is required for estrogen receptormediated gene transcription and breast cancer cell growth, J. Biol. Chem. 286 (2011) 17746–17754. S. Chiaretti, X. Li, R. Gentleman, A. Vitale, M. Vignetti, F. Mandelli, J. Ritz, R. Foa, Gene expression profile of adult T-cell acute lymphocytic leukemia identifies distinct subsets of patients with different response to therapy and survival, Blood 103 (2004) 2771–2778. D. Peters, J. Freund, R.L. Ochs, Genome-wide transcriptional analysis of carboplatin response in chemosensitive and chemoresistant ovarian cancer cells, Mol. Cancer Ther. 4 (2005) 1605–1616. A. Ortiz-Barahona, D. Villar, N. Pescador, J. Amigo, L. del Peso, Genome-wide identification of hypoxia-inducible factor binding sites and target genes by a probabilistic model integrating transcription-profiling data and in silico binding site prediction, Nucleic Acids Res. 38 (2010) 2332–2345. Y. Yao, X.Y. Jia, H.Y. Tian, Y.X. Jiang, G.J. Xu, Q.J. Qian, F.K. Zhao, Comparative proteomic analysis of colon cancer cells in response to oxaliplatin treatment, Biochim. Biophys. Acta 1794 (2009) 1433–1440.