RESEARCH ARTICLE
Nicotinamide induces Fob1-dependent plasmid integration into chromosome XII in Saccharomyces cerevisiae Kaushlendra Tripathi1, Nabil Matmati2, Shamsu Zzaman1, Caroline Westwater3 & Bidyut K. Mohanty1 1
Department of Biochemistry and Molecular Biology, Medical University of South Carolina, Charleston, SC, USA; 2Department of Medicine, Stony Brook University, Stony Brook, NY, USA; and 3Department of Craniofacial Biology, Medical University of South Carolina, Charleston, SC, USA
Correspondence: Bidyut K. Mohanty, Department of Biochemistry and Molecular Biology, Medical University of South Carolina, 173 Ashley Avenue, Charleston, SC 29425, USA. Tel.: +1843 792 0094; fax: +1843 792 8569; e-mail:
[email protected] Received 6 July 2012; accepted 16 August 2012. DOI: 10.1111/j.1567-1364.2012.00844.x Editor: Monique Bolotin-Fukuhara Keywords nicotinamide; Fob1; Sir2; replication arrest; ribosomal DNA recombination.
Abstract In the ribosomal DNA (rDNA) array of Saccharomyces cerevisiae, DNA replication is arrested by the Fob1 protein in a site-specific manner that stimulates homologous recombination. The silent information regulator Sir2, which is loaded at the replication arrest sites by Fob1, suppresses this recombination event. A plasmid containing Fob1-binding sites, when propagated in a yeast strain lacking SIR2 is integrated into the yeast chromosome in a FOB1-dependent manner. We show that addition of nicotinamide (NAM) to the culture medium can stimulate such plasmid integration in the presence of SIR2. Pulsed-field gel electrophoresis analysis showed that plasmid integration occurred into chromosome XII. NAM-induced plasmid integration was dependent on FOB1 and on the homologous recombination gene RAD52. As NAM inhibits several sirtuins, we examined plasmid integration in yeast strains containing deletions of various sirtuin genes and observed that plasmid integration occurred only in the absence of SIR2, but not in the absence of other histone deacetylases. In the absence of PNC1 that metabolizes NAM, a reduced concentration of NAM was required to induce plasmid integration in comparison with that required in wild-type cells. This study suggests that NAD metabolism and intracellular NAM concentrations are important in Fob1-mediated rDNA recombination.
YEAST RESEARCH
Introduction The budding yeast Saccharomyces cerevisiae contains ~150 copies of a ribosomal DNA (rDNA) unit arranged in a tandem array on chromosome XII (Petes, 1979). Each rDNA unit (Fig. 1) contains a 35S and a 5S gene, which are transcribed toward each other. The coding regions of these genes are separated by two intergenic spacers (IGSs) designated IGS1 and IGS2 (Brewer & Fangman, 1988; Ganley et al., 2005). Whereas IGS2 contains a bidirectional origin of replication (Ori), IGS1 contains two closely spaced DNA replication fork barriers (RFBs) called Ter1 (or RFB1) and Ter2 (or RFB2/RFB3), which differ in their DNA sequences (Brewer et al., 1992; Ward et al., 2000; Kobayashi, 2003; Mohanty & Bastia, 2004; Ganley et al., 2005). DNA replication originating from Ori moves bidirectionally, but the fate of the two replication forks is dissimilar. For example, the fork that moves through the 5S gene (right to left in Fig. 1b), is arrested immediately FEMS Yeast Res && (2012) 1–9
at the RFB/Ter sites, whereas the second fork (left to right in Fig. 1b) moves through a small number of rDNA units until it encounters another fork that has been arrested at a RFB/Ter site (Brewer & Fangman, 1988). The replication terminator protein Fob1 binds to the RFB/Ter sites (Kobayashi, 2003; Mohanty & Bastia, 2004) to promote site-specific replication arrest in a unidirectional manner (Brewer et al., 1992; Ward et al., 2000), resulting in the unequal fate of the two replication forks originating from the same Ori. Maintenance of homogeneity among the rDNA units and control of rDNA copy number is dependent on homologous recombination processes, which are primarily mediated by Fob1. Fob1 can generate extra-chromosomal rDNA circles (ERCs), whereby one or more rDNA unit(s) is excised from the rDNA array to form a selfreplicating DNA circle (Defossez et al., 1999; Kaeberlein et al., 1999). Fob1-mediated intrachromatid recombination can lead to a net loss of rDNA repeats or a net gain ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
2
K. Tripathi et al.
(a)
(b)
(c)
(d)
Fig. 1. Fob1-mediated plasmid integration, its suppression by Sir2, and the involvement of other proteins. (a) A section of chromosome XII showing the rDNA array; (b) A single rDNA unit showing 35S and 5S rDNA genes, IGS1 and IGS2, origin of replication (Ori), RFB/Ter sites (black clamps), E-Pro, and Reb1 transcription termination site (light gray clamp); (c) Model showing plasmid integration into rDNA array in a sir2Δ strain; (d) Autoradiogram of a Southern blot showing the intracellular distribution of plasmid pBB3NTS resolved without nicking the DNA in a 0.85% agarose gel; a labeled plasmidspecific probe was used for detection of the plasmid DNA. Chr. XII, chromosome XII; 35S Pr., promoter of 35S gene; Reb1, Reb1-binding site; Ter 1 and 2, Fob1-binding sites; E-Pro, bidirectional promoter; 5S, 5S gene; Ori., origin of replication; SC, supercoiled pBB3NTS; NC, nicked circular pBB3NTS; OC, open circular pBB3NTS; arrowheads, un-integrated plasmid population; long arrow, integrated plasmid.
of the repeats, and these events are greatly suppressed in normally growing fob1 (deleted or mutant) cells (Kobayashi et al., 1998, 2004). Expansion and contraction of the rDNA repeats owing to intrachromatid recombination requires the Fob1 protein and a DNA sequence called EXP, which includes the RFB/Ter sites and an RFB-adjacent sequence containing a bidirectional RNA polymerase II promoter called E-Pro (Fig. 1) (Kobayashi et al., 2001; Kobayashi & Ganley, 2005). Plasmid integration and excision assays have demonstrated that plasmid integration occurs following Fob1-dependent replication arrest. However, plasmid integration does not occur if the EXP region is orientated such that replication is not blocked by Fob1 (Benguria et al., 2003; Mohanty et al., 2009). Therefore, the FOB1 gene is essential for both expansion and contraction of yeast rDNA repeats through fork block-dependent recombination events. We have shown previously that in addition to FOB1, the replication checkpoint genes TOF1 and CSM3 (Katou et al., 2003; Calzada et al., 2005; Tourriere et al., 2005; ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
Mohanty et al., 2006, 2009; Bando et al., 2009), and the homologous recombination gene RAD52 (Mortensen et al., 2009; Thorpe et al., 2011) are also required for plasmid integration when the product of SIR2 is absent (Benguria et al., 2003; Mohanty et al., 2009). The orientation-dependence of the plasmid integration event (Benguria et al., 2003) and the requirement for the fork protection proteins Tof1 and Csm3 strongly suggest that plasmid integration is a replication arrest-dependent process. In this case, Fob1 is required for fork arrest, Tof1 and Csm3 are needed for fork protection at RFB/Ter sites, and Rad52 is necessary for induction of the homologous recombination process. Therefore, Fob1, Tof1, Csm3 and Rad52 are thought to promote plasmid integration, while Sir2 suppresses plasmid integration. How does Sir2 suppress plasmid integration? In the rDNA array, Sir2 is loaded at the RNA polymerase I promoter region and at RFB/Ter sites in a protein complex called RENT (regulator of nucleolar silencing and telophase exit). RENT includes the nucleolar protein, Net1, the NAD-dependent histone deacetylase Sir2, Cdc14 phosphatase, and three other proteins (Tof2, Lrs4, and Csm1) that recruit cohesin to RFB/Ter sites. In the IGS2 region, the RENT complex is loaded by RNA polymerase subunits Rpa190 and Rpa135, whereas at RFB/Ter sites RENT is loaded through Fob1 (Huang & Moazed, 2003). In wild-type yeast cells, Sir2 suppresses transcription from E-Pro; in the absence of Sir2, E-Pro is active and transcription from E-Pro removes cohesin resulting in rDNA expansion/contraction (Kobayashi & Ganley, 2005). The Sir2 activity of NAD metabolism can be regulated by nicotinamide (NAM), a noncompetitive inhibitor that promotes a base-exchange reaction at the expense of deacetylation (Landry et al., 2000; Gallo et al., 2004; McClure et al., 2008). Alterations in Sir2 activity have been shown to affect a wide variety of biological processes, including DNA recombination, transcriptional silencing, and aging (Landry et al., 2000; Gallo et al., 2004; McClure et al., 2008). In this study, we show that when NAM is not present, Fob1-dependent plasmid integration occurs only in the absence of SIR2, but not in the absence of other known histone deacetylases such as HST1-4. We further show that NAM-dependent pharmacological inhibition of Sir2 induces integration of a plasmid containing Fob1-binding sites into the rDNA repeats of S. cerevisiae despite the presence of Sir2 in the cells. NAM-induced plasmid integration occurs exclusively in chromosome XII and requires Rad52, Tof1 or Fob1. Our results suggest that the NAM-induced plasmid integration assay could be used to identify additional factors implicated in Fob1-mediated, fork arrest-dependent recombination. FEMS Yeast Res && (2012) 1–9
3
Nicotinamide and Fob1-dependent recombination
Materials and methods Yeast strains and plasmids
The yeast strains used in the study were BY4741 (MATa his3D1 leu2D0 met15D0 ura3D0; Invitrogen), W303 (MATa leu2-3,112 trp1-1 can1-100 ura3-1 ade2-1 his311,15; Dr. Rodney Rothstein) and LPY11 (W303a sir2D:: HIS3; obtained from Dr. Loraine Pillus (Mohanty et al., 2009)). The S. cerevisiae deletion mutants generated in the parental strain BY4741 were obtained from Invitrogen. Plasmid pBB3NTS (Fig. 1) was used for all integration experiments and has been described previously (Ward et al., 2000; Mohanty et al., 2009).
DNA samples were fractionated in a 0.85% (w/v) agarose gel (25 cm long) containing ethidium bromide (0.33 lg mL 1) for 96 h and transferred to a Nytran membrane (Whatman). Plasmid pBB3NTS was detected using a 32P-labeled pUC18 probe, whereas total chromosomal DNA was identified by probing with a 32P-labeled 1.5-kb rDNA fragment spanning the RFB region (Mohanty et al., 2009). The blots were exposed to a PhosphorImager screen (GE Healthcare) and developed using a Storm scanner (GE Healthcare). In several cases, blots were probed and detected using the non-radioactive ECL labeling/detection kit (GE Healthcare). Pulsed-field gel electrophoresis analysis
Plasmid integration assay
Plasmid pBB3NTS was transformed into the designated yeast strains by the lithium acetate method and transformants were selected on SD/Ura agar plates. Pools of 5-6 transformants were mixed and streaked on SD/Ura and SD/Ura /NAM agar. The plates were incubated at 30 °C for 5-6 days until the colonies reached a diameter of 1-2 mm on the SD/Ura medium. Cells (5-6 colonies) were then mixed, pooled, and restreaked on fresh SD/Ura plates. This process was repeated three times. A small number of colonies (5–6) were then pooled, inoculated into liquid SD/Ura or SD/Ura /NAM media, and the cultures were grown overnight at 30 °C before DNA was extracted. DNA preparation, Southern transfer and hybridization
DNA was prepared by the SDS lysis method as this procedure reduces the incidence of DNA shearing (Mohanty et al., 2009). Briefly, an overnight yeast culture was treated with lyticase (300 units mL 1) in 0.5 mL 0.9 M sorbitol in 100 mM EDTA (pH 7.0) for 1 h at 37 °C and then centrifuged at 13 000 r.p.m. for 1 min. The cell pellet was resuspended in 0.5-mL solution of 25 mM Tris (pH 7.5) and 10 mM EDTA, SDS was added to the solution to a final concentration of 1% (v/v), and the lysates were then stored at 4 °C overnight. Potassium acetate was added to a final concentration of 1.15 M, and the mixture was incubated on ice for 1 h. The lysate was centrifuged (13 000 r.p.m. for 5 min), and the DNA was precipitated with an equal volume of isopropanol. The DNA pellet was washed with 70% ethanol and then stored in TE (10 mM Tris.HCl and 1 mM EDTA) overnight at 4 °C. DNA samples were reprecipitated with sodium acetate and ethanol, washed with 70% ethanol, and finally dissolved in TE with 20 lg mL 1 RNase A. FEMS Yeast Res && (2012) 1–9
Overnight yeast cultures were processed for Pulsed-field gel electrophoresis (PFGE) analysis by a CHEF DR II apparatus according to the manufacturer’s instructions (Bio-Rad). Briefly, cells were harvested, washed with 1 mL of 10 mM Tris (pH 7.2), 20 mM NaCl and 50 mM EDTA, and then treated with 150 units of lyticase for 30 min on ice. The suspension was then mixed with 2% agarose (v/v) in a 1 : 1 ratio and quickly poured into molds (Bio-Rad). The solidified agarose plugs were then removed from the molds and treated with 2 mL of lyticase buffer containing 10 mM Tris (pH 7.2), 50 mM EDTA and 150 units lyticase for 30 min at 37 °C. The plugs were washed twice with 50 mL of 50 mM EDTA and 10 mM Tris.HCl (pH 7.6) and then incubated at 50 °C in 2 mL of 100 mM EDTA (pH 8.0), 0.2% sodium deoxycholate, 1% sodium sarkosyl and 1 mg mL 1 proteinase K. After 24 h, the plugs were washed twice with 50 mL of 50 mM EDTA and 10 mM Tris.HCl (pH 7.6). Plugs were cut to the appropriate size, placed into the wells of agarose gel (1% w/v) and then sealed with molten agarose. Gels were run at 6 v/cm with pulse times of 60 and 120 for 26 h. Gels were stained with ethidium bromide and photographed before Southern analysis was performed. Marker loss assay
The wild-type parental strain BY4741, containing a previously integrated pBB3NTS plasmid (which occurred by propagating the plasmid in the presence of NAM), was grown overnight at 30 °C in SD/Ura /NAM liquid medium. The overnight culture was diluted 10-fold into fresh SC/NAM liquid medium and incubated in the presence or absence of 5 mM of NAM. The cultures were grown for 24 h at 30 °C and then transferred to fresh SD/Ura / NAM medium. After two cycles of growth, cells were serially diluted and plated on yeast peptone dextrose (YPD) and SC/FOA (synthetic complete/5-fluoroorotic ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
4
K. Tripathi et al.
acid) agar plates. The plates were incubated at 30 °C for 3 days, and the colonies were then enumerated.
Results Fob1-dependent plasmid integration in the absence of SIR2
Previous studies have demonstrated that integration of plasmid pBB3NTS into chromosomal DNA occurs in a S. cerevisiae sir2Δ strain but does not occur in the parental wildtype strain (Benguria et al., 2003; Mohanty et al., 2009). In addition, plasmid integration has been shown previously to be dependent on replication arrest because deletion of FOB1, TOF1 or CSM3 in a sir2Δ background abolishes plasmid integration (Benguria et al., 2003; Mohanty et al., 2009). We first confirmed these results by conducting a plasmid integration assay in the S. cerevisiae wild-type strain (W303) and in each of the sir2Δ derivatives that contained single deletions of the replication fork arrest gene FOB1 or the checkpoint genes (CSM3, TOF1). We determined plasmid integration by extracting and resolving the intracellular DNA in agarose gels followed by Southern blotting and hybridization of the blots to a labeled plasmid-specific probe. As shown in Fig. 1d, plasmid integration was not observed in the wild-type strain W303 as the labeled plasmid-specific probe detected only the free form of plasmid pBB3NTS. The intracellular pBB3NTS plasmid DNA in the wild-type cells remained mostly in the supercoiled (SC), nicked circular (NC) and open circular (OC) state. In the sir2Δ strain, almost the entire population of plasmid DNA was integrated, and it migrated as a single band corresponding to the chromosomal DNA (Fig. 1d). As expected, the plasmid DNA remained in the free form in each of the sir2Δ derivatives that contained single deletions of the replication terminator gene FOB1 or the checkpoint genes (CSM3, TOF1) (Fig. 1d). NAM induces plasmid integration
Because the absence of SIR2 induces Fob1-mediated plasmid integration and because NAM is an inhibitor of Sir2, we predicted that plasmid integration would occur in a wild-type S. cerevisiae strain if the culture media were supplemented with NAM. Plasmid pBB3NTS was transformed into the parental strain BY4741 and its sir2Δ derivative, and cells were grown in the presence or absence of 5 mM NAM. As expected plasmid integration did not occur in the wild-type strain in the absence of NAM (Fig. 2, WT – N); however, plasmid integration was observed in the wild-type strain when NAM was present in the media (Fig. 2, WT +N). In contrast, plasmid integration occurred in the sir2Δ strain irrespective of the presence of NAM ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
Fig. 2. Nicotinamide (NAM) induces plasmid integration. Autoradiogram of a Southern blot showing NAM-induced plasmid integration in the wild-type (WT) strain and in the sir2Δ derivative of BY4741. The sir2Δ derivative of W303 was included as a positive control for plasmid integration (see Fig. 1 and Mohanty et al., 2009). In BY4741 (WT), plasmid integration did not occur in the absence of NAM (WT N), but plasmid integration did occur when NAM was present in the media (WT +N). In the sir2Δ derivative of BY4741, plasmid integration occurred regardless of whether NAM was present (sir2Δ N and sir2Δ +N). Arrowheads, un-integrated pBB3NTS plasmid; long arrow, integrated pBB3NTS plasmid.
(Fig. 2, sir2Δ N and sir2Δ +N). Although a small population of un-integrated plasmid is seen in the NAM-treated wild-type cells and in the untreated sir2Δ cells, subsequent propagation of this population resulted in complete integration of the plasmid (Fig. 3). By reprobing the blots with a labeled rDNA probe, we confirmed that the band representing the integrated plasmid corresponded to the chromosomal DNA (data not shown). NAM induces plasmid integration into chromosome XII
Replication fork barrier sites are present only in the rDNA array, which is located in the nucleolus along with Fob1 [8]. Therefore, we predicted that NAM-mediated plasmid integration occurs only in the rDNA array located on chromosome XII. To test this prediction, yeast cells containing plasmid pBB3NTS were grown in the presence or absence of NAM and the yeast chromosomes were then fractionated by PFGE, and pBB3NTS integration was analyzed. Ethidium bromide staining of the agarose gel showed fractionation of the various chromosomes according to size with the largest chromosome XII (~2.4 Mb) located at the top of the gel in both untreated and NAM-treated wild-type cells (Fig. 3a). A Southern blot of the gel was hybridized with a labeled plasmid-specific probe (pUC18) and the resulting autoradiogram showed a single band at an equivalent position to chromosome XII FEMS Yeast Res && (2012) 1–9
5
Nicotinamide and Fob1-dependent recombination
(a)
(b)
(c)
(d)
the PFGE gel, whereas the linear plasmid should exit the gel. As shown in Fig. 3d, the undigested circular plasmids pBB3NTS and pRS315, as well as the yeast chromosomes were visible in the gel. In contrast, the HindIII-digested pBB3NTS (linear ~6 kb DNA fragment) was not visible upon ethidium bromide staining. Together, these experiments show that NAM-induced plasmid integration occurs in chromosome XII. NAM-induced plasmid integration depends on Fob1 and Rad52
Fig. 3. Plasmid integration occurs into chromosome XII as revealed by pulsed-field gel electrophoresis (PFGE) analysis. Agarose plugs prepared from appropriate strains were fractionated by PFGE and stained with ethidium bromide. Southern blots of the resolved DNA samples were probed with a labeled pUC18 probe or a labeled rDNA fragment spanning the RFB region. (a) Ethidium bromide (EB) stained agarose gel shows chromosomes of the wild-type yeast strain containing pBB3NTS that was grown with or without NAM (+N and N, respectively) (b) PhosphorImager analysis of the Southern blot shown in Fig. 3A hybridized with a labeled pUC18 probe; (c) PhosphorImager analysis of the Southern blot shown in Fig. 3a hybridized with a rDNA (present in chromosome XII) probe; (d) Ethidium bromide stained gel showing PFGE-fractionated chromosomes of the wild-type yeast strain containing pBB3NTS grown in the absence of NAM (lane 1) or presence of NAM (lane 2), plasmid pBB3NTS prepared from Escherichia coli (lane 3), HindIII-digested plasmid pBB3NTS (lane 4), and undigested pRS316 plasmid DNA (lane 5). Chr. XII, chromosome XII.
in the NAM-treated sample, but not in the untreated sample (compare Fig. 3b +N and N). By reprobing the blot with a labeled rDNA probe (representing chromosome XII), we observed the presence of a single hybridizing band (Fig. 3c). This result strongly suggests that plasmid pBB3NTS integrated into the rDNA array in chromosome XII only when NAM was present; in the absence of NAM no integration occurred in chromosome XII. Where is the plasmid pBB3NTS in the sample without NAM? The unintegrated plasmid population would be expected to exit the agarose gel owing to its size; however, a small fraction of the plasmid population was retained in the gel (Fig. 3b). To demonstrate that a population of circular plasmid (pBB3NTS) could be retained in the gel under our assay conditions, we performed a second PFGE experiment. Agarose plugs containing yeast chromosomes from untreated and NAM-treated wild-type cells, undigested pBB3NTS (made from Escherichia coli), HindIII-digested pBB3NTS (also made from E. coli and digested with HindIII) and undigested pRS315 (made from E. coli) plasmid were prepared. The samples were fractionated by PFGE, and the gel was stained with ethidium bromide. If our prediction is correct, a population of circular plasmid should remain in FEMS Yeast Res && (2012) 1–9
We wanted to determine whether NAM-induced plasmid integration is a replication arrest-dependent homologous recombination process. A plasmid propagation assay was performed in the fob1Δ and rad52Δ single-deletion strains in the presence and absence of 5 mM NAM. Southern blot analysis showed that the entire plasmid population remained un-integrated in both the fob1Δ and rad52Δ single-deletion strains regardless of the presence of NAM (Fig. 4). In the tof1Δ strain, plasmid integration did not occur regardless of the presence of NAM (data not
Fig. 4. NAM-induced plasmid integration requires Fob1 and Rad52. Autoradiogram of a representative Southern blot showing the effect of sir2Δ, fob1Δ, and rad52Δ single deletions on plasmid integration (pBB3NTS). DNA samples prepared from various strains were fractionated by a 0.85% agarose gel and transferred to a Nytran membrane before probing with a labeled plasmid-specific probe (pUC18). Plasmid integration occurred in the wild-type strain grown in the presence of NAM (WT +N) and also in a sir2Δ single-deletion strain in the presence or absence of NAM (sir2Δ +N and sir2Δ N); however, integration did not occur in the fob1Δ or rad52Δ strain despite the presence of NAM (fob1Δ N, fob1Δ +N, rad52Δ N and rad52Δ +N). Arrowheads, un-integrated pBB3NTS plasmid; long arrow, integrated pBB3NTS plasmid.
ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
6
shown). These data show that NAM-mediated plasmid integration is dependent on replication arrest by Fob1 and is a Rad52-dependent homologous recombination process. SIR2 is the only histone deacetylase that suppresses plasmid integration
Besides Sir2, other NAM-inhibited histone deacetylases such as Hst1, Hst2, Hst3 and Hst4 have been implicated in gene silencing, recombination and longevity (Smith et al., 2007; Hachinohe et al., 2011). Therefore, we wished to determine whether additional histone deacetylases suppress plasmid integration. Plasmid pBB3NTS was transformed into the single-deletion strains hst1Δ, hst2Δ, hst3Δ, hst4Δ, rpd3Δ and umc1Δ, and cells were grown in the presence or absence of NAM. Samples were analyzed for pBB3NTS integration, and it was observed that plasmid integration did not occur in the absence of NAM in any of the six histone deacetylase strains tested (Table 1). Our results demonstrate that Sir2 is the only histone deacetylase tested so far, which when deleted or inactivated by NAM, supports plasmid integration. Role of NAD+ pathway genes in NAM-induced plasmid integration
Because Sir2 is a NAD+-dependent deacetylase that metabolizes NAD+ to deacetylate histones, we wished to analyze additional members of the NAD+ metabolism pathway to explore their role in plasmid integration. The enzymes involved in the NAD+ metabolism pathway (Gallo et al., 2004; Tanny et al., 2004; McClure et al., 2008) are shown in Fig. 5a. NAD+ is generated from nicotinic acid mononucleotide (NAMN) and nicotinamide mononucleotide (NMN) by the action of the nicotinamide mononucleotide adenyltransferase enzymes Nma1 and Nma2. NAD+ is metabolized by Sir2 to generate NAM and O-acetyl-ADP-ribose. NAM can be converted to nicotinic acid (NA) by the nicotinamidase Pnc1; NA is also generated by Tna1 activity. NA is then used as a substrate by the NAD+ salvage pathway enzyme NA phosphosribosyltransferase (Npt1), which converts NA to NAMN; in addition to Npt1, the enzymes Bna1-Bna6 are also involved in the generation of NAMN. We wanted to determine whether deletion of any of the genes involved in the NAD+ metabolism pathway would stimulate plasmid integration. As shown in Table 1, deletion of TNA1, NMA1, BNA1, BNA2, NPT1 or PNC1 did not stimulate plasmid integration unless NAM was present in the media. However, PNC1 is known to metabolize NAM, and in its absence, NAM accumulation occurs (McClure et al., 2008). Therefore, it is possible that the level of NAM in the pnc1Δ strain was ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
K. Tripathi et al.
Table 1. Plasmid integration in the absence and presence of nicotinamide Strains Medium
Plasmid integration SD/Ura
SD/Ura /NAM
WT N Y sir2D Y Y fob1D N N Genes involved in recombination and repair rad52D N N rad57D N Y msc3D N Y mgs1D N Y mre11D N Y Histone deacetylases hst1D N Y hst2D N Y hst3D N Y hst4D N Y rpd3D N Y ume1D N Y Genes involved in NAD metabolism tna1D N Y nma1D N Y bna1D N Y pnc1D N Y npt1D N Y bna2D N Y Genes involved in DNA replication, sister-chromatid cohesion and checkpoints ctf18D N Y ctf4D N Y ctf8D N Y dcc1D N Y psy2D N Y rad9D N Y Helicases srs2D N Y pif1D N Y sgs1D N Y Other pathways isc1D N Y ydr026cD N Y
below the threshold level to induce plasmid integration. To test this possibility, we conducted a plasmid integration experiment at 0, 0.5, 2.5, 5 and 10 mM NAM. In the wild-type strain, integration of the entire plasmid population occurred at 2.5 mM NAM (Fig. 5b), although < 10% of the plasmid population was integrated at 0.5 mM NAM. In npt1Δ cells, a similar level of plasmid integration occurred. Integration of the entire plasmid population was observed at 2.5 mM NAM, whereas < 10% of the population integrated at 0.5 mM NAM (data not shown). In contrast to wild-type and npt1Δ cells, 0.5 mM NAM was sufficient to induce plasmid integration in the pnc1Δ deletion strain (Fig. 5b). FEMS Yeast Res && (2012) 1–9
7
Nicotinamide and Fob1-dependent recombination
(a)
(b)
Fig. 5. Role of PNC1 gene and concentration of NAM needed to induce plasmid integration. DNA samples prepared from the wild-type (WT) and pnc1Δ strains were fractionated by a 0.85% agarose gel and Southern blots were hybridized with a labeled plasmid-specific probe (pUC18). (a) NAD metabolism pathway and genes involved at various steps of the pathway. NAM generated by Sir2 activity inhibits Sir2; (b) PhosphorImager analysis of the Southern blot of WT and pnc1Δ strains containing pBB3NTS and grown in the presence of 0 mM, 0.5 mM, 2.5 mM, and 5 mM NAM. Whereas < 10% of plasmid integrated in the WT strain at 0.5 mM NAM, the entire plasmid population integrated in the pnc1Δ strain when grown in the equivalent concentration of NAM. Un., un-integrated pBB3NTS plasmid; long arrow (Int.), integrated pBB3NTS plasmid.
These results demonstrate that the requirement for NAM in the absence of PNC1 is fivefold less than that required in the wild-type or the npt1Δ strain. Role of DNA replication, recombination and repair genes in NAM-induced plasmid integration
We have examined the possible roles of several genes involved in replication, recombination and repair because plasmid integration is dependent on Fob1, Tof1, and Csm3, as well as on Rad52. Strains containing deletions of various genes involved in DNA replication and sister-chromatid cohesion (CTF4, CTF18, CTF8, DCC1, and PSY1), checkpoint (RAD9), helicase (SRS2, PIF2, and SGS1), and recombination (RAD57, MSC3, MGS1, and MRE11), as well as other pathways such as YDR026C and ISC1 were tested. NAM-induced plasmid integration was not abolished in these strains, suggesting that the gene products do not promote plasmid integration in the presence of NAM (or probably in the absence of SIR2; Table 1). NAM induces loss of the URA3 marker from chromosomes
We have observed that once pBB3NTS is integrated into the genome of a sir2Δ deletion strain, the plasmid can be lost from the chromosome, and ultimately from the cell, if selective pressure for the URA3 marker in the rDNA array FEMS Yeast Res && (2012) 1–9
is not maintained (Mohanty et al., 2009). Marker loss from the rDNA array has been used as a standard assay for rDNA recombination by various laboratories, although the role of Fob1-dependent replication arrest had not been explored. We predicted that the plasmid population that integrated into the genome in the presence of NAM should also be lost in the absence of selective pressure (i.e. inclusion of uracil in the culture media). Saccharomyces cerevisiae BY4741 cells, in which the plasmid was integrated by NAM treatment, were transferred from SD/Ura /NAM to synthetic complete (SC) and SC/NAM media, and the frequency of plasmid loss was calculated (see Materials and methods). We observed that cells grown in SC/NAM medium lost the URA3 marker at a twofold higher frequency than cultures grown in SC medium within 48 h (data not shown). It should be noted that the growth rate of the BY4741::pBB3NTS strain was reduced in SC/NAM when compared to SC media. Therefore, the actual frequency of plasmid loss in SC/NAM medium would be expected to be higher than that observed. These results show that NAM can induce Fob1-mediated, replication arrest-dependent plasmid integration into chromosome XII and can also induce plasmid excision.
Discussion In this work, we have shown that the histone deacetylase inhibitor NAM induces Fob1-dependent integration of a plasmid into chromosome XII. Plasmid integration ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
8
requires the replication arrest protein Fob1, fork protection protein Tof1, and the homologous recombination protein Rad52. The requirement for NAM in the process of plasmid integration is at least fivefold lower in a pnc1Δ strain when compared to the wild-type parental strain. In the absence of NAM, plasmid integration occurred only in a sir2Δ strain, but not in any other histone deacetylase deletion strains tested. Plasmid integration occurred either in the presence of NAM or in the absence of SIR2. Although NAM inhibits several histone deacetylases, plasmid integration occurred only in the absence of Sir2, suggesting that NAM-induced plasmid integration was due to inactivation of Sir2 by NAM. Other histone deacetylases do not function in the Fob1-mediated plasmid integration at RFB/Ter sites. Plasmid integration in the absence of Sir2, as well as in the presence of NAM, requires Fob1, Tof1 and Rad52. These data suggest that NAM inhibits Sir2 resulting in Fob1dependent plasmid integration through homologous recombination by Rad52. The concentration of NAM added to the culture media for plasmid integration in the wild-type strain was found to be 2.5 mM. In contrast, only 0.5 mM NAM was sufficient for plasmid integration in the pnc1Δ genetic background. The PNC1 gene is involved in the clearance of NAM by converting it to nicotinic acid (McClure et al., 2008). In its absence, NAM a non-competitive inhibitor of Sir2 and other histone deacetylases accumulates in the cell; however, NAM accumulation is not sufficient to cause plasmid integration in the pnc1D strain. Our study suggests that the concentration of NAM controlled by Pnc1, determines Sir2 activity, which in turn controls Fob1-mediated plasmid integration. Although our previous work had suggested that plasmid integration occurred into chromosome XII, the PFGE analysis carried out in this study conclusively showed that plasmid integration occurred only into chromosome XII. The RFB sites are present only in rDNA array of chromosome XII and Fob1 is predominantly localized to the nucleolus. Therefore, integration of the entire plasmid population occurred into chromosome XII. The present work can be extended to develop techniques to screen the yeast deletion library to identify genes that promote Fob1-dependent recombination. Currently, we are developing such a technique. We have observed that excision of the integrated plasmid can be induced by the continuous presence of NAM in the growth media and in the absence of selective pressure to maintain the plasmid in the cell. Therefore, this method can be used to quantitatively analyze NAM-mediated recombination at Fob1-binding sites. In addition, as Sir2 blocks extreme lifespan extension (Fabrizio et al., 2005), interplay between Sir2 and other genes involved in this ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
K. Tripathi et al.
phenomenon can be analyzed by NAM inhibition of Sir2. SIR2 deletion induces Spo11p-catalyzed double DNA strand breaks (Mieczkowski et al., 2007), and the gene products that participate in this pathway can also be identified using this technique. Although it has not been tested, a similar approach can be used to conditionally inhibit Sir2 to study functions of Sir2 at other loci such as HML, HMR, telomeres, and the RNA polymerase I region of rDNA. Similar uses of reversible inhibition of mammalian SIR2 homologs can also be made.
Acknowledgements We thank Dr. Deepak Bastia and Dr. Yusuf Hannun for providing facilities and help in various ways for completion of the work. This work was supported in part by the South Carolina COBRE in Lipidomics and Pathobiology (P20 RR17677 from NCRR) and an ACS-IRG grant (IRG 97-219-08) from the Hollings Cancer Center, at the Medical University of South Carolina for B.K.M.
References Bando M, Katou Y, Komata M, Tanaka H, Itoh T, Sutani T & Shirahige K (2009) Csm3, Tof1, and Mrc1 form a heterotrimeric mediator complex that associates with DNA replication forks. J Biol Chem 284: 34355–34365. Benguria A, Hernandez P, Krimer DB & Schvartzman JB (2003) Sir2p suppresses recombination of replication forks stalled at the replication fork barrier of ribosomal DNA in Saccharomyces cerevisiae. Nucleic Acids Res 31: 893–898. Brewer BJ & Fangman WL (1988) A replication fork barrier at the 3′ end of yeast ribosomal RNA genes. Cell 55: 637–643. Brewer BJ, Lockshon D & Fangman WL (1992) The arrest of replication forks in the rDNA of yeast occurs independently of transcription. Cell 71: 267–276. Calzada A, Hodgson B, Kanemaki M, Bueno A & Labib K (2005) Molecular anatomy and regulation of a stable replisome at a paused eukaryotic DNA replication fork. Genes Dev 19: 1905–1919. Defossez PA, Prusty R, Kaeberlein M, Lin S-J, Ferrigno P, Silver PA, Keil RL & Guarente L (1999) Elimination of replication block protein Fob1 extends the life span of yeast mother cells. Mol Cell 3: 447–455. Fabrizio P, Gattazzo C, Battistella L, Wei M, Cheng C, McGrew K & Longo VD (2005) Sir2 blocks extreme lifespan extension. Cell 123: 655–667. Gallo CM, Smith DL Jr & Smith JS (2004) Nicotinamide clearance by Pnc1 directly regulates Sir2-mediated silencing and longevity. Mol Cell Biol 24: 1301–1312. Ganley AR, Hayashi K, Horiuchi T & Kobayashi T (2005) Identifying gene-independent noncoding functional elements
FEMS Yeast Res && (2012) 1–9
Nicotinamide and Fob1-dependent recombination
in the yeast ribosomal DNA by phylogenetic footprinting. P Natl Acad Sci USA 102: 11787–11792. Hachinohe M, Hanaoka F & Masumoto H (2011) Hst3 and Hst4 histone deacetylases regulate replicative lifespan by preventing genome instability in Saccharomyces cerevisiae. Genes Cells 16: 467–477. Huang J & Moazed D (2003) Association of the RENT complex with nontranscribed and coding regions of rDNA and a regional requirement for the replication fork block protein Fob1 in rDNA silencing. Genes Dev 17: 2162–2176. Kaeberlein M, McVey M & Guarente L (1999) The SIR2/3/4 complex and SIR2 alone promote longevity in Saccharomyces cerevisiae by two different mechanisms. Genes Dev 13: 2570–2580. Katou Y, Kanoh Y, Bando M et al. (2003) S-phase checkpoint proteins Tof1 and Mrc1 form a stable replication-pausing complex. Nature 424: 1078–1083. Kobayashi T (2003) The replication fork barrier site forms a unique structure with Fob1p and inhibits the replication fork. Mol Cell Biol 23: 9178–9188. Kobayashi T & Ganley AR (2005) Recombination regulation by transcription-induced cohesin dissociation in rDNA repeats. Science 309: 1581–1584. Kobayashi T, Heck DJ, Nomura M & Horiuchi T (1998) Expansion and contraction of ribosomal DNA repeats in Saccharomyces cerevisiae: requirement of replication fork blocking (Fob1) protein and the role of RNA polymerase I. Genes Dev 12: 3821–3830. Kobayashi T, Nomura M & Horiuchi T (2001) Identification of DNA cis elements essential for expansion of ribosomal DNA repeats in Saccharomyces cerevisiae. Mol Cell Biol 21: 136–147. Kobayashi T, Horiuchi T, Tongaonkar P, Vu L & Nomura M (2004) SIR2 regulates recombination between different rDNA repeats, but not recombination within individual rRNA genes in yeast. Cell 117: 441–453. Landry J, Slama JT & Sternglanz R (2000) Role of NAD(+) in the deacetylase activity of the SIR2-like proteins. Biochem Biophys Res Commun 278: 685–690. McClure JM, Gallo CM, Smith DL Jr et al. (2008) Pnc1pmediated nicotinamide clearance modifies the epigenetic
FEMS Yeast Res && (2012) 1–9
9
properties of rDNA silencing in Saccharomyces cerevisiae. Genetics 180: 797–810. Mieczkowski PA, Dominska M, Buck MJ, Lieb JD & Petes TD (2007) Loss of a histone deacetylase dramatically alters the genomic distribution of Spo11p-catalyzed DNA breaks in Saccharomyces cerevisiae. P Natl Acad Sci USA 104: 3955–3960. Mohanty BK & Bastia D (2004) Binding of the replication terminator protein Fob1p to the Ter sites of yeast causes polar fork arrest. J Biol Chem 279: 1932–1941. Mohanty BK, Bairwa NK & Bastia D (2006) The Tof1p-Csm3p protein complex counteracts the Rrm3p helicase to control replication termination of Saccharomyces cerevisiae. P Natl Acad Sci USA 103: 897–902. Mohanty BK, Bairwa NK & Bastia D (2009) Contrasting roles of checkpoint proteins as recombination modulators at Fob1-Ter complexes with or without fork arrest. Eukaryot Cell 8: 487–495. Mortensen UH, Lisby M & Rothstein R (2009) Rad52. Curr Biol 19: R676–R677. Petes TD (1979) Yeast ribosomal DNA genes are located on chromosome XII. P Natl Acad Sci USA 76: 410–414. Smith DL Jr, McClure JM, Matecic M & Smith JS (2007) Calorie restriction extends the chronological lifespan of Saccharomyces cerevisiae independently of the Sirtuins. Aging Cell 6: 649–662. Tanny JC, Kirkpatrick DS, Gerber SA, Gygi SP & Moazed D (2004) Budding yeast silencing complexes and regulation of Sir2 activity by protein-protein interactions. Mol Cell Biol 24: 6931–6946. Thorpe PH, Alvaro D, Lisby M & Rothstein R (2011) Bringing Rad52 foci into focus. J Cell Biol 194: 665–667. Tourriere H, Versini G, Cordon-Preciado V, Alabert C & Pasero P (2005) Mrc1 and Tof1 promote replication fork progression and recovery independently of Rad53. Mol Cell 19: 699–706. Ward TR, Hoang ML, Prusty R, Lau CK, Keil RL, Fangman WL & Brewer BJ (2000) Ribosomal DNA replication fork barrier and HOT1 recombination hot spot: shared sequences but independent activities. Mol Cell Biol 20: 4948–4957.
ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved
