UV but not X rays stimulate homologous recombination between sister chromatids and homologs in a Saccharomyces cerevisiae mec1 (ATR) hypomorphic mutant

Abstract

MEC1, the essential yeast ATM/ATR homolog, prevents replication fork collapse and is required for the cellular response to DNA damage. We had previously observed higher rates of spontaneous SCE, heteroallelic recombination and translocations in mec1-21 mutants, which still retain some G2 checkpoint function, compared to mec1 null mutants, which are completely defective in checkpoint function, and wild type. However, the types of DNA lesions that are more recombinogenic in mec1-21, compared to wild type, are unknown. Here, we measured DNA damage-associated SCE, homolog (heteroallelic) recombination, and homology-directed translocations in mec1-21, and characterized types of DNA damage-associated chromosomal rearrangements that occur in mec1-21. Although frequencies of UV-associated recombination were higher in mec1-21, the mutant was defective in double-strand break-associated SCE and heteroallelic recombination. Over-expression of Rad53 in mec1-21 reduced UV-associated recombination but did not suppress the defect in X-ray-associated recombination. Both X ray and UV exposure increased translocation frequencies in mec1-21, but the majority of the UV-associated products were non-reciprocal translocations. We suggest that although recombinational repair of double-stand breaks is less efficient in mec1 mutants, recombinants may be generated by other mechanisms, such as break-induced replication.

1. Introduction

MEC1 is the essential yeast paralog [1] of the human ATM (mutated in ataxia telangiectasia, A-T) and the ortholog of ATR (ATM and RAD3 related) (for review, see [2]). Similar to ATR, MEC1 is required for both viability and for preventing replication fork collapse [3]. Both ATR and ATM hypomorphic alleles have been linked to higher incidence of cancer, which correlates with a higher level of genetic instability present in ATR [4] and ATM [5] cell lines. How environmental agents, such as radiation and chemical carcinogens, increase genetic instability in ATR and ATM cell lines is not well understood. Since ATR null mutants are not viable, understanding DNA damage-associated genetic instability phenotypes of well defined yeast mec1 hypomorphic mutants may thus provide insights into genetic instability phenotypes of ATR cells after exposure to DNA damaging agents.

mec1 mutants are hypersensitive to a wide variety of recombinogenic DNA damaging agents, such as X rays, UV, and MMS [6]. MEC1 confers DNA damage resistance by activating downstream kinases, such as those encoded by CHK1 and RAD53 (Chk2), which, in turn, activate other effectors that trigger DNA damage-associated cell cycle arrest, maintain nucleotide balance, and repair double-strand breaks (DSBs). MEC1 controls deoxynucleotide (dNTP) levels by activating Rad53, which in turn, activates the DUN1-mediated induction of ribonucleotide reductase [7]. MEC1 is also required for the induction of Rad51 [8], and for the phosphorylation of Rad55 [9] and Sae2, a protein involved in processing DSBs [10]. Thus, the DNA damage sensitivity of mec1 mutants may be due to deficiencies in multiple checkpoint functions.

mec1 hypomorphic mutants may exhibit different spontaneous hyper-recombination phenotypes than mec1 null mutants. mec1 null mutants exhibit high rates of spontaneous gross chromosomal rearrangements (GCRs, [11]) and loss of heterozygosity (LOH) [12] due to mitotic cross-overs, but only a modest increase in homologous recombination phenotypes. Alternatively, the hypomorphic missense mutant mec1-21[13], which is defective in S phase checkpoint function (14), but retains some G2 checkpoint function [15], exhibits a higher rates of spontaneous sister chromatid exchanges, gene conversion events, and ectopic recombination [16], but lower rates of spontaneous GCRs, compared to the mec1 null [17]. The mec1-21 mutant results from a G to A substitution (G882S) [13], which is outside the kinase domain but conserved with ATR. Although the differences in the recombination phenotypes could be due to multiple functions, one explanation is that DSBs can initiate mitotic, homologous recombination in some hypomorphic mec1 mutants but not in the mec1 null. Consistent with this explanation, higher frequencies of GCRs are observed in mec1 null mutants after exposure to radiation or radiomimetic DNA damaging agents [18], while mec1 mutants are deficient in MMS-associated heteroallelic events [9].

Since mec1 null mutant contains secondary mutation(s) that increase dNTP levels, we asked which DNA damage-associated homologous recombination interactions are triggered in mec1-21, which is viable without secondary mutations. We had demonstrated that, compared to wild type, mec1-21 exhibits higher rates of spontaneous, homologous recombination between sisters, homologs, and repeats on non-homologous chromosomes [16]. Here, we demonstrate that, compared to wild type, mec1-21 exhibits higher frequencies of UV-associated but not X-ray-associated recombination between homologs and sister chromatids. Using a recombination assay that can detect non-reciprocal recombination, we found that the mec1-21 diploid exhibits higher frequencies of UV-associated chromosomal non-reciprocal translocations, compared to wild type. Although we do not know all the mechanisms that stimulate homologous recombination in mec-21, we suggest that one mechanism is break-induced replication (BIR).

2. Materials and methods

2.1. Media and yeast strains

Standard media for the culture of yeast, SC (synthetic complete, dextrose), SC-HIS (SC lacking histidine), SC-LEU (SC lacking leucine), SC-TRP (SC lacking tryptophan), SC-URA (SC lacking uracil), YP (yeast extract, peptone), and YPD (YP, dextrose), are described by Burke et al. [19]. YPL medium contains YP with 2% lactate (pH 5.8); YPGlu medium contains YP medium with 2% ultra-pure glucose; YPGal medium contains YP medium with 2% ultra-pure galactose (Sigma, St Louis, MO). YP(A)D contains YPD with 80 mg/L adenine.

Relevant yeast strains are listed in Table 1. Haploid strains used to measure SCE contain two overlapping his3 fragments [20], positioned in tandem at trp1, and were derived from YB163 [21]. The construction of the mec1-21 rad52 strain was previously described [16]. Diploid strain (YB348) was used to measure translocations that were derived from a cross of one haploid (YB318) that contains the his3 fragments on one copy of chromosomes II and IV [16, 22], and another which did not contain the his3 fragments (YB315). Heteroallelic recombination was measured by selecting for Ade⁺ recombinants between ade2-a and ade2-n [23, 24] in the YB348 diploid strain. The mec1-21 mutation was introduced in strains to measure frequencies of SCE, heteroallelic recombination, and translocations, as previously described [16].

Table 1.

Yeast strains

Strain	Genotype	Source (Synonym)
YA102	MATa ura3-52 his3-Δ200 lys2-801 trp1-Δ1 ade2-101	M. Carlson (MCY727)
YA165	MATα ura3-52 his3-Δ200 trp1-Δ1 leu2-Δ1	F. Winston (FY250)
YA166	MATa ura3-52 his3-Δ200 trp1-Δ1 leu2-Δ1	F. Winston (FY251)
YA184	MATa trp1-1 leu2-3,112 his3-11,15 ura3-1 can1-100 sml1::URA3 rad53::HIS3 RAD5	R. Rothstein (W2105-17B)
YA185	MATa trp1-1 leu2-3,112 his3-11,15 ura3-1 can1-100 mec1-Δ::TRP1 sml1::HIS3 RAD5	R. Rothstein (U963-61A)
YA195	MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 sml1::KanMX	ResGene (512)
YA196	MATa/MATα his3-Δ1/− leu2Δ0/− ura3Δ0/− met15Δ0/+ lys2Δ0/− mec1::KanMX/+	ResGene (23275)
YA197	MATα ade2-1 trp1-1 leu2-3, 112 his3-11,15 ura3-1 can1-100 mec1-21	S. Elledge (Y620)
YB314	MATα ura3-52 his3-Δ200 lys2-801 trp1-Δ1 gal3− mec1-21	Derived from cross of YB312 × YA166
YB315	MATa ura3-52 his3-Δ200 ade2-a lys2-801 trp1-Δ1 gal3−	Derived from YA102
YB316	MATα ura3-52 his3-Δ200 ade2-a lys2-801 trp1-Δ1 gal3− mec1-21	Derived from cross of YB315 × YB314
YB317	MATa ura3-52 his3-Δ200 ade2-a lys2-801 trp1-Δ1 gal3−sml1::KanMX	sml1::KanMX disruption in YB315
YB321	MATa ura3-52 his3-Δ200 ade2-a lys2-801 trp1-Δ1 gal3−sml1::KanMX mec1-Δ::TRP1	mec1-Δ::TRP1 disruption in YB317
Strains to monitor translocations and heteroallelic eventsa
YB109	MATα ura3-52 his3-Δ200 ade2-101 trp1-Δ1 gal3− leu2-3, 112	This laboratory
	GAL1::his3-Δ 5′ trp1::his3-Δ3′::HOcs lys2− (leaky)
YB318	MATα ura3-52 his3-Δ200 ade2-n trp1-Δ1 gal3− leu2-3, 112	Derived from of YB109
	GAL1::his3-Δ 5′ trp1::his3-Δ3′::HOcs lys2− (leaky)
YB319	MATa-inc ura3-52 his3-Δ200 ade2-n trp1-Δ1 gal3− leu2-3, 112	Derived from cross of YB313 × YB318
	GAL1::his3-Δ 5′ trp1::his3-Δ3′::HOcs lys2− (leaky) mec1-21
YB320	MATα ura3-52 his3-Δ200 ade2-n trp1-Δ1 gal3− leu2-3, 112	sml1::KanMX disruption in YB318
	GAL1::his3-Δ 5′ trp1::his3-Δ 3′::HOcs lys2− (leaky) sml1::KanMX
YB322	MATα ura3-52 his3-Δ200 ade2-n trp1-Δ1 gal3− leu2-3, 112	mec1-Δ::TRP1 disruption in YB320
	GAL1::his3-Δ 5′ trp1::his3-Δ 3′::HOcs lys2 (leaky) sml1::KanMX mec1-Δ::TRP1
YB324	YB321 × YB322	This laboratory
YB325	YB316 × YB319	This laboratory
YB348	YB315 × YB318	This laboratory
Strains to monitor SCEa
YB163	MATa-inc ura3-52 his3-Δ200 ade2-101 lys-801 trp1-Δ1 gal3-trp1::[his3-Δ3′::HOcs, his3-Δ5′]	This laboratory
YB204	MATα leu2-Δ1	This laboratory
YB311	MATa-inc mec1-21	10th backcross of Y620 with YB163
YB312	MATα mec1-21	10th back cross of Y620 with YB163
YB326	MATa-inc sml1::KanMX	sml1::KanMX disruption in YB163
YB327	MATa-inc sml1::KanMX mec1-Δ::TRP1	mec1::TRP1 disruption in YB326
YB328	MATα rad52::KanMX	rad52::KanMX disruption in YB204
YB347	MATa-inc mec1-21 rad52:KanMX	from cross of YB329 with YB328
YB338	MATa-inc sml1::URA3	sml1::URA3 disruption in YB163
YB339	MATa-inc sml1::URA3 mec1-Δ::KanMX	mec1::KanMX disruption in YB338
YB352	MATa-inc sml1::URA3 mec1-Δ::TRP1	mec1::TRP1 disruption in YB338
YB368	MATa-inc leu2-Δ1 mec1-21	from cross of YB311 × YA165
YB378	MATa-inc1 mec1-21 + pWL25	Ura+ transformant of YB311
YB381	MATα mec1-21 + pJA98 (GAL10::RAD53)	Ura+ transformant of YB312

^aAll strains listed below YB163 have the same genotype as YB163 unless indicated. Mating type is added for clarity. YB333 and YB334 may contain either ura3-52 or ura3Δ0, and lys2Δ0 or lys2-801.

To make the mec1-Δ mutant, we first introduced the sml1::KanMX or sml1::URA3 allele [25] in the appropriate haploid strains, YB163, YB318, and YB318, by PCR-mediated gene replacement [26], and then introduced the mec1-Δ::TRP1 or mec1-Δ::KanMX allele [25]. The primers used for amplifying sml1::KanMX or sml1::URA3 were: 5′CATATCGTTACTGTTTTGGAACATCGC3′ and 5′TAAAGGGAAAGGAAAATGCACG3′. The primers used for amplifying mec1-Δ::TRP1 were 5′ATTCCTTTTCAAGGCTCCATAACTA3′ and 5′TTTTCCATATCTTCGAGCTCTTCTA3′.

2.3. Measuring rates of spontaneous recombination

The rates (events per cell division) of spontaneous, mitotic events that generate either SCE, heteroallelic recombination, or translocations were determined by the method of the median, [27] as executed by Esposito et al. [28], using 11 independent colonies for each rate calculation. For measuring rates of heteroallelic recombination, we obtained colonies from cells that were inoculated on YP(A)D medium, so that there was no growth selection for Ade⁺ recombinants. At least three independent rate calculations were done for each strain, and statistical significance was determined by the Mann-Whitney U-test [29].

2.4. Determining frequencies of DNA damage-associated recombinants

Protocols used to measure the recombinogenicity of UV and X-rays have been described [30]. At least three independent experiments were done for each DNA-damaging agent. We reported the spontaneous recombination frequencies [number of His⁺ recombinants per colony forming unit (CFU)], recombination frequencies obtained after exposure to DNA-damaging agents (stimulated frequency), and net frequencies for each DNA-damaging agent. The average net frequency of His⁺ recombinants was determined by first subtracting the spontaneous frequency from the stimulated frequency for each experiment and then taking the average.

2.5. Induction of HO endonuclease

pGHOT-GAL3 [22], containing the HO gene under GAL control, was introduced into MEC1, mec1-21 and mec1-Δ strains by selecting for Trp⁺ transformants. After growth in SC-TRP medium, cells were diluted 1:10 in YPLactate and incubated for a minimum of 12 hr. At a density of 10⁷ cells/ml, glucose or galactose was added to a final concentration of 2%, to either repress or induce the expression of HO endonuclease, respectively. After 2 hr, cells were plated directly on YPD medium for viability and on SC-HIS to measure recombination. Colonies appearing on YPD medium were replica plated on SC-TRP to measure the number of Trp⁺ colonies containing the pGHOT-GAL3 plasmid.

2.6. Verification that His⁺ recombinants result from unequal SCE or intrachromatid recombination

Mitotic unequal SCE between his3-Δ5′ and his3-Δ3′ results in His⁺ recombinants that contain HIS3 flanked by his3-Δ5′ and his3-Δ3′ (Figure 1; 21). Three different mechanisms may participate in the repair of the DSB generated by HO endonuclease digestion at the trp1::his3-Δ3′::HOcs locus [31]. These include NHEJ, single-strand annealing (SSA), and homologous recombination. SSA events generate a chromosomal deletion and a single his3 fragment lacking both 3′ and 5′ ends, rendering cells unable to generate His⁺ recombinants (Figure 1). To determine frequencies of chromosomal deletions (SSA), cells were plated on YPD plates after HO induction. The surviving colonies were replica plated onto SC-HIS to measure the number of His⁻ colonies that cannot generate His⁺ recombinants and onto SC-TRP to measure the number of colonies that maintain the pGHOT-GAL3 plasmid. The presence of the single his3 fragment was confirmed by the presence of a 1.4-Kb PCR product using primers 5′-CACGGCAGAGACCAATCAGTA-3′ and 5′-GCACTCCTGATTCCGCTAATA-3′ [31].

Figure 1.

SCE, translocation, and heteroallelic recombination assays used in this study. Ovals represent centromeres and lines represent chromosomes. For simplicity, the left arms of chromosomes are not included. An arrow and feathers together denote HIS3. As indicated in the bottom left of the figure, the 5′ deletion lacks the feather and the 3 ′ deletion lacks the arrow. The two regions of the sequence identity shared by the his3 fragments are indicated by decorated boxes; closely-spaced diagonal-filled boxes indicate a region of 167 bp, and the broadly-spaced diagonal line-filled boxes indicate a region of ~300bp. The 117-bp HO cut site (HOcs), as indicated by an arrowhead, is located between these sequences within the his3-Δ3′::HOcs fragment. The “X” indicates where recombination occurred. The his3-truncated fragments are integrated into the trp1 locus to measure SCE events. His⁺ recombinants resulting from SCE were selected after exposure to radiation or after the induction of HO endonuclease (A, left). Right panel of the figure indicates that HO endonuclease-induced DSB can be repaired by the SSA pathway, which generates intrachromosomal deletions. The deletion renders cells unable to generate His⁺ recombinants, and can be detected by PCR using primers #1 5′CACGGCAGAGACCAATCAGTA3′ and #2 5′GCACTCCTGATTCCGCTAATA3′. (B) Translocation events result from recombination between the same his3 fragments located each on chromosomes II and IV. Positions of the GAL1 and trp1 are shown on chromosomes II, IV, and the reciprocal translocation. The translocated chromosome that contains CEN2 linked to HIS3 and the long arm of chromosome IV is referred to as CEN2::IV, while the translocated chromosome that contains CEN4 linked to the long arm of chromosome II is referred to as CEN4::II. Diploid strains contain one copy of chromosome II and IV that do not contain his3 sequences. (C) Heteroallelic recombination between ade2-a and ade2-n generates ADE2. ADE2 and ade2 alleles are represented as boxes; ade2-a and ade2-n are separated by approximately 1 kb.

2.7. Chromosomal DNA gels

Undigested yeast chromosomal DNA was resolved on contour-clamped homogeneous electric field (CHEF) gels containing 1% agarose [32]. The gels were run at 220 V (6 V/cm) for 26 hr at a 90-sec pulse time [22]. Chromosomal DNA was transferred to nylon after exposure to 60 J/m² of UV radiation for Southern blot analysis [22, 33]. The 1.7-kb BamHI HIS3 fragment was used as a probe.

3. Results

We hypothesized that mec1 mutants defective in the S phase checkpoint could potentially exhibit hyper-recombination phenotypes due to collapsed replication forks or the accumulation of single-stranded DNA. We measured DNA damage-associated recombination in the mec1-21 [14] mutant, which is still proficient at triggering G2 arrest after radiation exposure [15]. The recombination assays are shown in Figure 1. Unequal SCE was measured by selecting for His⁺ recombinants in haploid strains containing two truncated his3 gene fragments [21]. Diploid strains were used to measure heteroallelic recombination between ade2-a and ade2-n [24] and ectopic recombination between GAL1::his3-Δ5′ and trp1::his3-Δ3′ [22].

3.1. Exposure to UV but not X rays stimulates more SCE in mec1-21 mutants than in wild type

We hypothesized that the observed hyper-recombination phenotype resulted from repair of DNA lesions generated in S phase. We therefore expected that higher DNA damage-associated SCE would occur in mec1-21, compared to wild type, in the presence of DNA adducts that block or hinder replication. We measured frequencies of UV-associated SCE in mec1-21 and in wild type (Figure 2). After log phase cells were exposed to UV, we observed eight-fold increased SCE frequencies in wild type, consistent with previous observations [22, 31], but much higher frequencies in mec1-21. After exposure to 90 J/m², the net UV-associated SCE frequency was 2 × 10⁻⁴ in mec1-21, ~fourfold higher than in wild type; net frequencies of UV-associated SCE did not increase at higher doses of UV. Thus, mec1-21 exhibits higher frequencies of UV-associated SCE, compared to wild type.

Figure 2.

SCE frequencies in wild type (MEC1), mec1-Δ and mec1-21 strains after UV and X-ray exposure. Strains used include MEC1(diamond, YB163), mec1-21 (triangle, YB311) and mec1-Δsml1(square, YB327). SCE was measured by selecting for His⁺ recombinants. Log phase cells were exposed to radiation and then plated on SC-HIS to measure recombination and the appropriate dilution was plated on YPD to measure viability. Percent survival is the number of CFU after radiation/number of CFU before radiation × 100%. Top panels show data after cells were exposed to UV and bottom panels show data after cells were exposed to X rays. (A) Percent survival after cells were exposed to UV. (B) Frequency of His⁺ recombinants (number of His⁺ recombinants/c.f.u.) plotted against UV dose. (C) Net frequency of His⁺ recombinants (UV-associated frequency – spontaneous frequency) plotted against UV dose. (D) Percent survival after cells were exposed to X rays. (E) Frequency of His⁺ recombinants (number of His⁺ recombinants/c.f.u.) plotted against X-ray dose. (F) Net Frequency of His⁺ recombinants (X-ray-associated frequency – spontaneous frequency) plotted against X ray dose. 10 Gy = 1 krad.

We also measured the effect of X-ray exposure, and observed that SCE frequencies increased fourfold in wild type after X-ray exposure, but less than twofold in mec1-21 and mec1-Δ mutants (Figure 2). Although wild-type cells in liquid medium appear as X-ray sensitive as mec1-21 cells, we observed that mec1-21 cells were ~50% more X-ray sensitive after being plated and then irradiated [15]. We observed a similar deficiency in X-ray-associated SCE in a mec1-21 diploid mutant (data not shown).

3.2. mec1-21 is deficient in the stimulation of unequal SCE by HO-induced double-strand breaks

Since X-rays generate both SSBs and DSBs, we determined whether unequal SCE frequencies increased after a single HO-induced DSB was targeted to his3-Δ3′::HOcs in mec1 mutants (Table 2). We introduced pGHOT-GAL3, containing the galactose-inducible HO gene, into wild type and mec1 mutants in which the MAT locus (MATa-inc) is not a substrate for HO endonuclease. At trp1:: his3-Δ3::HOcs, the HO-induced DSB can be repaired by SSA or NHEJ mechanisms (Figure 1). After induction of HO in log phase cells, the reduced viability in wild type was similar to what has been previously observed [22, 31]. Viability was reduced over 50% reduction in mec1-Δ cells, consistent with this mutant deficiency in SSA (34). This was supported by observations that unselected mec1-Δ colonies arising after HO induction contain fewer deletions (26%, 35/129), compared to wild type (72%, 36/50). After HO induction, SCE frequencies were increased 10-fold in wild-type and sml1 cells, but only threefold in mec1-21 and twofold in mec1-Δ. After HO induction, SCE frequencies were increased 10-fold in wild type but only threefold in mec1-21. Thus, DSB-associated unequal SCE is MEC1-dependent.

Table 2.

Stimulation of SCE by HO-induced DSBs in mec1 mutants

Genotypea (Strain)	% Viability after HO inductionb	His+recombinants/Trp+CFU×105
		Before HO inductionc	After HO inductiond	Fold increasee
Wild type (YB163)	79 ± 1	8.4 ± 1.0	82 ± 5	10
mec1-21 (YB311)	47 ± 9	11.4 ± 0.4	39 ± 6	3.4
sml1::URA3 (YB338)	81 ± 14	7.8 ± 3.7	94 ± 30	12
mec1–Δ::KanMX	42.1 ± 0.02	9.1 ± 2.2	20 ± 3	2
sml1::URA3 (YB339)

^aFor complete genotype, see Table 1.

^bTrp⁺ CFU after HO induction/Trp⁺ CFU before HO induction × 100%

^cHis⁺ recombinants before HO induction/Trp⁺ CFU before HO induction

^dHis⁺ recombinants after HO induction/Trp⁺CFU after HO induction

^eHis⁺ frequency after HO induction/His⁺ frequency before HO induction

3.3. Suppression of spontaneous and UV-associated SCE hyper-recombination by RAD53 overexpression

Activation of Rad53 is MEC1-dependent, and RAD53 over-expression has been previously shown to suppress the HU and UV sensitivity phenotypes of mec1-21 (7). We determined whether RAD53 over-expression would also suppress the recombination phenotypes of mec1-21(YB311). We introduced the RAD53 gene on pWL25 [35], containing URA3. Ura⁺ transformants were grown in SC-URA medium and Rad53 overproduction was confirmed by Western blots (supplemental Figure 1). Rates of spontaneous SCE in mec1-21 + pWL25 (YB376) were reduced but sill threefold greater than wild type (Table 3). Rates of spontaneous SCE in mec1-21 containing an empty vector (YCp50) were similar than those of mec1-21. Net frequencies of UV-associated SCE were measured after exposure to 90 J/m². In mec1-21, mec1-21 + pWL25, and wild type, the net frequencies were (204 ± 56) × 10⁻⁶, (38 ± 6) × 10⁻⁶, (36 ± 11) × 10⁻⁶, respectively. However, frequencies of X-ray associated SCE did not increase in mec1-21 + pWL25 but did increase sevenfold in MEC1 (YB163) + pWL25 after exposure to 6 krads X rays. Thus, RAD53 over-expression in mec1-21 reduces spontaneous and UV-associated SCE, but does not significantly increase X-ray-associated SCE.

Table 3.

Rates of spontaneous SCE in mec1 mutants.

Strain	Genotypea	Rate (× 106)b	Ratioc
YB163	MEC1	1.1 ± 0.1	1.0
YB326	sml1::KanMX	1.1 ± 0.2	1.0
YB312	mec1-21	6.3 ± 0.9	5.7
YB327	mec1-Δ::TRP1 sml1::KanMX	1.2 ± 0.2	1.1
YB378	mec1-21 + RAD53 (pWL25)	3.3 ± 0.2	3.0

^aFor complete genotype, see Table 1.

^bRate, number of events per cell division; n ≥ 3

^cRatio, rate of SCE in mutant/rate of SCE in wild type

Similarly, mec1-21 cells containing a galactose-inducible RAD53 gene on pJA98, also exhibited lower frequencies of recombination after Rad53 was overproduced. Because the mec1-21 strain contains trp1-Δ1 and lacks the GAL3 promoter, we first introduced pGAL3 (TRP1) and later pJA98 (URA3) [7]. Approximately 10⁶ cells, obtained from independent Ura⁺ Trp⁺ transformants, were inoculated in SC (glucose) -URA-TRP and in SC (galactose) –URA-TRP, and cultures were incubated for 4–5 generations. Recombination frequencies (2.7 ± 0.4 × 10⁻⁵) were twofold lower (P < 0.01, n =11) from cells grown in galactose medium, compared with cells grown in glucose medium (6.8 ± 1.4 × 10⁻⁵, n = 11). There was no difference (P = 0.5) in recombination frequencies between mec1-21 + pGAL3 cells grown in glucose (8.7 ± 0.4 × 10⁻⁵) or in galactose (9 ± 0.6 × 10⁻⁵). Thus, inducible RAD53 expression decreases recombination in mec1-21.

3.4. mec1-21 diploid mutants exhibit UV-associated homolog recombination but are deficient in X-ray-associated homolog recombination

To determine whether homologs were better substrates for recombinational repair than sister chromatids in mec1 mutants, we measured UV and X-ray associated heteroallelic recombination. X-ray exposure increased the frequencies of heteroallelic recombination in mec1-21 and wild type by two-fold (P < 0.05) and 30-fold, respectively (Figure 3). However, UV-associated frequencies of heteroallelic recombination were as high in mec1-21 as in wild type (Figure 3). The mec1-21 diploid is thus deficient in X-ray-associated but not in UV-associated homolog recombination.

Figure 3.

DNA damage associated frequencies of homology-directed translocations and heteroallelic recombination in MEC1(black diamond, YB348), mec1-21(black triangle, YB325), and mec1 sml1 (black square, YB324) diploid strains. To measure translocation frequency, we selected for His⁺ recombinants and to measure heteroallelic recombination, we selected for Ade⁺ recombinants. Top panels show data after cells were exposed to X rays and bottom panels show data after cells were exposed to UV. Percent survival is the number of CFU after radiation/number of CFU before radiation × 100%. (A) Percent survival after cells were exposed to X rays. (B) Frequency of His⁺ recombinants (number of His⁺ recombinants/c.f.u.) plotted against X-ray dose. (C) Frequency of Ade⁺ recombinants (number of Ade⁺ recombinants/c.f.u.) plotted against X-ray dose. (D) Percent survival after cells were exposed to UV. (E) Frequency of His⁺ recombinants (translocations) (number of His⁺ recombinants/c.f.u.) plotted against UV dose. (F) Frequency of Ade⁺ recombinants (heteroallelic recombination) (number of Ade⁺ recombinants/c.f.u.) plotted against UV dose.

3.5. Higher frequencies of radiation-associated translocations occur in mec1-21 compared to wild type

If DSB repair is less efficient between homologs and sister chromatids, we asked whether mec1 mutants exhibit more ectopic recombination between repeated sequences on non-homologs. We selected for His⁺ recombinants in which mitotic recombination between his3 fragments on chromosomes II and IV generates reciprocal translocations. We previously observed a 23-fold increase in spontaneous translocations in mec1-21. Here, we observed higher frequencies of radiation-associated translocations in mec1-21 diploids, compared to wild type (Figure 3). Net frequencies of X-ray associated translocations were between twofold and threefold higher in mec1-21 mutant than in wild type, while net frequencies of UV-associated translocations were between sixfold and 50-fold higher than in wild type. Thus, frequencies of radiation-associated translocations are higher in mec1-21, compared to wild type.

3.6. Electrophoretic karyotypes of chromosome rearrangements from mec1-21 and mec1-Δ mutants

We previously observed that, compared to wild type, the rad9 mutant [22], defective in G2 checkpoint, and the rad51 diploid mutant [21], defective in recombinational repair, exhibit high frequencies of spontaneous and radiation-associated translocations. In diploids, viable His⁺ recombinants can be selected that contain CEN2::IV but lack the CEN4::II translocated chromosome, since CEN4::II does not contain HIS3 (Figure 1). Recombination events that do not generate both reciprocal translocations are referred to as non-reciprocal events. Since, we observed that many recombinants in G2 checkpoint and recombination mutants contain non-reciprocal translocations [21, 22], we compared the electrophoretic karyotypes of spontaneous and UV-associated His⁺ recombinants from mec1-21, mec1-Δ, and wild type (Figure 4). In wild type, most independent, spontaneous and UV-associated His⁺ recombinants contained reciprocal translocations (Table 4). In the mec1-21 (YB325) strain, the majority (5/8) of the UV-associated His⁺ recombinants contained non-reciprocal translocations, but non-reciprocal translocations (1/8) were more difficult to detect among spontaneous recombinants. His⁺ recombinants that contain non-reciprocal translocations typically contain his3 sequences on chromosome II [20]. By CHEF, we determined that in mec1-21, 25% (4/16) of the His⁺ recombinants contained other rearrangements, some of which hybridize to his3 sequences. The majority (6/8) of spontaneous His⁺ recombinants in the mec1-Δ mutant contained non-reciprocal translocations, while UV-associated recombinants contained both non-reciprocal and reciprocal rearrangements, and about half (7/16) of the His⁺ recombinants from the mec1-Δ mutant contained other rearrangements (Figure 4). Thus, non-reciprocal translocations occurred in both mec1-21 and in mec1-Δ, but were seldom observed in wild type.

Figure 4.

Electrophoretic karyotype of spontaneous and DNA damage-associated His⁺ recombinants resulting from ectopic recombination between GAL1::his3-Δ5′ and trp1::his3-Δ 3′ in the mec1-21 (YB325) and mec1-Δ (YB324) diploids. (Left) Southern blot using the ³²P-labeled 1.7-kb BamHI HIS3 fragment. (Right) Picture of the ethidium-bromide-stained gel. Arrows point to the well, wild-type chromosomes II and IV, and the translocations CEN2::IV and CEN4::II. Hybridization to chromosome XV results from the DNA homology between the his3-Δ 200 locus and the sequences that flank HIS3 within the 1.7-kb BamHI fragment. Lanes: (A) His⁻ (parental configuration) of mec1-Δ diploid (YB324), (B) Spontaneous His⁺ nonreciprocal translocation derived from mec1-21diploid (YB325), (C) Spontaneous His⁺ recombinant from mec1-21 diploid (YB325) containing novel rearrangements, (D) Spontaneous His⁺ recombinant from mec1 Δ diploid (YB324) containing novel rearrangements, (E) UV-induced His⁺ recombinant from mec1-Δ diploid (YB324) containing novel rearrangements, (F) UV-induced His⁺ recombinant from mec1 ••• diploid (YB325) containing reciprocal translocation, (G) UV-induced His⁺ recombinant from mec1-Δ diploid (YB324) containing nonreciprocal translocation, (H) His⁻ (parental configuration) of mec1-Δ diploid (YB324).

Table 4.

Electrophoretic karyotypes of translocations stimulated in mec1-21 and mec1 null mutants.

Genotype (Strain)	Agent	No. of His+ recombinants containing indicated rearrangements
		Reciprocal Translocationsa	Reciprocal Translocations And Other	Non-reciprocal translocations	Non-reciprocal translocations and others	Other	Total
Wild type (YB110)	Spontaneous	16					16
	UV	16					16
mec1-21/mec1-21 (YB325)	Spontaneous	5	1		1	1	8
	UV	2		5		1	8
mec1-Δ/mec1-Δ (YB324)	Spontaneous			3	3	2	8
	UV	3		1	3	1	8

^aCEN2::IV and CEN4::II

3.7. Growth phenotypes and UV sensitivities of mec1-21 and mec1-21 rad52 mutants

One explanation for the hyper-recombination in mec1-21, is that more DSBs are spontaneously generated or occur after UV exposure in mec1-21. We had previously observed a synergistic increase in the UV sensitivity of mec1-21 rad52 double mutant, compared to the single mutants [16]. Compared to wild type (YB163), which has a doubling time of 1.9 hrs and a plating efficiency of 91%, the mec1-21(YB312) and rad52 (YB328) mutants exhibited modest increases in doubling time and decreases in plating efficiency (Table 5). The plating efficiency of mec1-21 rad52 (YB347) mutant was less than half that of wild type, and the doubling time of mec1-21 rad52 was twofold higher than mec1-21 and significantly higher than rad52 (P < 0.05). These data suggest that more DSBs spontaneously occur in mec1-21, compared to wild type, and that RAD52 confers UV resistance in mec1-21.

Table 5.

Growth phenotypes of mec1 mutants.

Genotype a (Strain)	Plating efficiency b (%)	Doubling time c (hours)
MEC1 (YB163)	91 ± 5	1.9 ± 0.3
mec1-21 (YB312)	62 ± 15	2.3 ± 0.2
mec1-21rad52 (YB347)	22 ± 4	4.8 ± 1.1
rad52 (YB328 )	64 ± 2	2.9 ± 0.7

^aFor complete genotype, see Table 1.

^bCFU/Total cell number × 100%

^cAverage amount of time for logarithmically growing cells to double its cell population

4. Discussion

The ATR (ATM) yeast homolog MEC1 plays a pivotal role in sensing and transmitting DNA damage signals. Because both ATR an MEC1 are essential [1, 2], understanding the genetic instability phenotypes in mec1 hypomorphic mutants may yield insights into genetic instability of cells defective in ATR. We had previously observed that mec1-21, a missense mutant [13], exhibits higher frequencies of spontaneous homologous recombination, and that the hyper-recombination requires RAD9, CHK1, and PDS1 [16]. However, it was unknown whether exposure to recombinogenic DNA damaging agents would further increase frequencies of homologous recombination. Here, we observed higher frequencies of SCE and heteroallelic recombination in mec1-21 after exposure to UV but not X rays, compared to wild type. Higher frequencies of UV-associated non-reciprocal translocations suggest that in mec1-21 a possible mechanism for DNA damage-associated recombination is BIR. The high frequencies of UV-associated SCE can be partially suppressed by over-expression of Rad53, suggesting that one MEC1 function in suppressing UV-associated recombination is Rad53 signaling. These studies indicate that genetic instability in mec1 (ATR) cells depends on the specific DNA damaging agent.

These conclusions were based on the mec1-21 missense mutation, which does not confer the same G2 checkpoint deficiencies as mec1-Δ[15]. We do not know the kinase activity and substrate specificity of the Mec1 protein encoded by mec1-21. Further studies are necessary to determine whether TEL1 is required for homologous recombination phenotypes in mec1 hypomorphic mutants.

4.1. Comparison with other checkpoint and rad mutants

Other DNA repair and checkpoint mutants also exhibit UV-associated recombination but not X-ray-associated recombination between sister chromatids. For example, both rad9 [22] and rad53 [36] exhibit UV-associated recombination but do not exhibit X-ray associated recombination between sister chromatids. RAD9 is required for multiple functions in yeast, including radiation-associated G2 arrest [37], and transcriptional induction of DNA repair genes after radiation exposure [38], and the tolerance of irreparable UV-induced DNA damage [39]. RAD9-mediated functions are partially conferred by activating Rad53, which, in turn, activates Dun1 and triggers the DNA damage-inducibility of ribonucleotide reductase [40, 41]. Considering that radiation exposure still triggers G2 arrest but poorly activates Rad53 in mec1-21 [15], our data suggest that Rad53 activation may be important for recombinational repair of DSBs, but dispensable for stimulating recombination after exposure to UV. Exactly which downstream effectors of Rad53 are important in X-ray-associated homologous recombination are not known, since dun1 mutants, defective in the induction of RNR, still exhibits X-ray-associated SCE and homolog recombination [42].

Particular rad mutants, defective in DSB repair, share some homologous recombination phenotypes with mec1-21. For example, rad52-c mutants are defective in X-ray-associated heteroallelic recombination but are still proficient at UV-associated recombination [43]. UV-associated recombination has also been observed in rad52 null mutants, but the recombination frequencies are much lower than wild type [44]. rad51 mutants, although defective in all types of DNA damage-associated SCE, still exhibits UV-associated non-reciprocal translocations that result from homologous recombination [21]. Thus, rad mutants defective in recombinational repair of DSBs may still exhibit UV-associated homologous recombination. Further genetic studies are necessary to determine whether rad52-c and mec1-21 mutants are defective in the same pathways for UV and DSB repair.

4.2. Higher levels of DNA damage-associated recombination observed in mec1-21 may be initiated by ssDNA or occur by BIR

Although the well known pathway for initiating homologous recombination is the DSB pathway (gap repair) described by Szostak et al.[45], we observed that both HO-induced DSBs and X rays were less efficient in stimulating recombination in mec1-21, compared to wild type. It is thus unlikely that repair of DSBs by gap repair can account for all the hyper-recombination observed in mec1-21. We suggest two possible recombination mechanisms that explain why UV but not X ray exposure stimulates recombination in mec1-21. First, DSBs may initiate BIR, which could then generate recombinants. Second, recombinational repair may be initiated by single-stranded DNA, which may accumulate due to incomplete DNA replication.

BIR could occur when a replication fork collapses, thus yielding a DSB. Restoration of the DNA replication fork occurs after strand invasion of the broken chromatid with the incompletely replicated sister chromatid. Using the unbroken chromatid as a template, the recombination intermediate can thus prime DNA synthesis and replication could continue to the telomere of the unbroken chromatid [46]. Our data supports the idea that the broken chromatid could serve as a substrate for strand invasion of a repeated sequence on a non-homologous chromosome, which would generate non-reciprocal translocations. The notion that DSBs arise spontaneously in mec1-21 is supported by the observation that the mec1-21 rad52 mutant exhibits a very low plating efficiency. We speculate that mec1 deficiency in signaling Rad53 may contribute to spontaneous or UV-induced replication fork collapse, since RAD53 is required to stabilize replication forks after exposure to DNA damaging agents [47].

However, we cannot exclude the possibility that ssDNA initiates homologous recombination in mec1-21. This possibility is supported by observations that ssDNA accumulates in other mec1 hypomorphs, such as mec1-srf [48], and that ssDNA is postulated to initiate recombination in other DNA repair mutants, such as in sgs1 [49] and in rad52-C [43]. SGS1 and MEC1 have been previously suggested to participate in different pathways for stabilizing replication forks [50]. Further studies are necessary to determine whether ssDNA spontaneously accumulate in mec1-21 or during the course of UV excision repair.

In summary, mec1-21 exhibits both spontaneous and UV-associated hyper-recombination. Since DSB repair mechanisms mediated by homologous recombination are less efficient in mec1-21, recombination initiated by BIR or ssDNA may explain some of the hyper-recombination phenotypes. It would thus be interesting to know whether SGS1 and MEC1 participate different pathways for suppressing homologous recombination in response to DNA damaging agents. These studies demonstrate that mec1 hypomorphic mutants that retain some G2 checkpoint functions can still exhibit significant hyper-recombination phenotypes that depend on the DNA damaging agent.