ATR Homolog Mec1 Promotes Fork Progression, Thus Averting Breaks in Replication Slow Zones Rita S. Cha and Nancy Kleckner

Mutagenesis and strain information. The entire coding region of MEC1 was mutagenized by hydroxylamine (S1). The mec1-4 and mec1-40 alleles were identified by plasmid shuffle method (S1). The following strains are isogenic SK1 derivatives of RCY22 (MATa/MAT, mec1::LEU2/+): RCY301 and RCY307 (MEC1); RCY415 and RCY419 (mec1-4); RCY423 and RCY426 (mec1-40); RCY1288 (mec1-4, chrIII dup*), RCY1392 (mec1-4, ars305), RCY1396 (mec1-4, ars306, RCY1443 (mec1-4, chrIII dup* ars310), RCY1315 (mec1-4, sml1) and RCY1318 (mec1-40, sml1). RCY1938 (mec1-4, ars305 306307-T108676G) is a spore clone of a diploid generated by mating RCY415 and MJL2533-4C (S2). dup* contains duplication of a region between nucleotide (nt) positions 165,317-179,251. dup* 310 is dup* with an internal deletion of the 3.7 kb NhiI fragment containing ARS310. Origin activities in ars305 and ars306 are eliminated by deletion of minimal origin sequence (S3) whereas that of ars307 was inactivated by a single base pair substitution (T G) at 108676 (S2).

Pulse-field gel electrophoresis (PFGE)/Southern analysis. Chromosome sized DNA in agarose plugs for PFGE was prepared as described (S2). Electrophoresis was performed at 14°C in a Bio-Rad CHEF Mapper under the following condition: a voltage gradient of 5.5V/cm, switch times of 5- to 30-sec, a switch angle of 115°, in 1 % agarous gel in 0.5X TBE for 30 hours. DNA in gels was transferred to nylon membranes, hybridized with ³²P-labeled probe and quantified using a BioRad Molecular Imager FX. Probes for Southern analysis were generated by PCR using commercially available yeast ORF primer pairs (Research Genetics).

Two-dimensional gel analysis. Specific restriction enzymes used for releasing each segment and its coordinates along chrIII are as followed: Pst1 for B-a (48,147 - 54,787) and NB-c (215,922 - 222,826); BglII for B-c (249,984 – 257,668) and NB-b (66,613 – 69,591); XhoI for NB-a (43,129 – 46,202); XbaI/XhoI for B-b (77,614 – 82,767). Probes: YCL41 for B-a, YCL22 for B-b, YCR79 for B-c, YCL46 for NB-a, YCL29 for NB-b, and YCR56 for NB-c. 10 g of yeast genomic DNA is digested with 100 units of specific restriction enzymes, electrophoresed as described (S4), transferred to nylon membranes, hybridized with ³²P-labeled probe and quantified.

SUPPORTING TEXT

Resumption of DNA replication following replication stalling. The relative amounts of DNA in a given culture at various times following -factor arrest/release were estimated by measuring total chromosome-associated EthBr signals in a PFG using a BioRad Molecular Imager FX. In a wild type culture, the amount of DNA increases steadily, doubling by t=180 minutes. In a mec1-4 culture, the amount of DNA increases by ~50% during the first 40-60 minutes, remains constant for the next ~ 60 minutes, and then increases again, eventually achieving twice the starting level.

In depth discussion on chromosome III structural variant analysis. Chromosome fragmentation could occur in response to specific genetic determinants located within the breakage zones themselves. Alternatively, since mec1-ts break zones occur in alternation with highly active replication origins, the positions of breaks could be determined indirectly, in relation to the positions of replication initiation. For example, breaks could occur at positions where forks from adjacent origins converge, in the absence of any specific determinants at the convergence point, or at positions that are some fixed distance from origins.

We first examined chromosomes in which the distance between two break zones was increased by virtue of a tandem duplication with or without an encoded ARS [Fig. 2E(a) dup* and dup*310]. If the position of breaks is genetically determined, the presence of additional sequence between two break zones should not change their positions along the chromosomal (genetic) map. In this case, in PFGE analysis, the distance of between a probed chromosome end and the various mec1-ts break zones will be unchanged for zones located proximal to the added sequence and, for zones located distal to the added sequence, will be increased by the length of the addition. Both duplication chromosomes exhibit the same break zones, at the same genetic positions along the chromosome, as does the wild type chromosome [Fig. 2E(b) and (c)]. These results exclude specifically the possibility that break zones arise at positions where forks happen to converge, as is known to be possible in yeast (S5): the duplicated region increases the distance between origins to either side of the insert and thus should have moved any fortuitous points of convergence resulting from forks entering the duplicated region from opposite sides. These results also exclude the possibility that the break zone immediately distal to the insert arises at a fixed distance from the first distal origin.

Next, we investigated directly the possibility that positions of mec1 breaks might be defined in relation to nearby ARS elements. In that situation, addition or deletion of an ARS should alter the pattern of mec1-ts break zones. One such case is provided by the ARS-containing duplication [Fig. 2E(b)]. This chromosome exhibits no new break zones, in accord with the lack of a determining role for ARSs; however, we did not confirm that the ARS element on the duplication was active. More importantly, we examined three altered chromosomes from which either an individual ARS (ars305 or ars306) or all three ARS elements in the left arm of chromosome III (ars305, ars306, ars307T108676G) have been deleted [Fig. 2E(a)]. In the latter case, replication of the left arm occurs passively and unidirectionally via forks that originate in the right arm and progress through break zone-II and break zone-I to the end of the chromosome (S2). All three deletion chromosomes exhibited the same break zones as the wild type chromosome [Fig. 2E(d), fig. S3]. This result shows that the positions of ARS elements do not determine the positions of mec1-ts breaks. The triple ars mutant chromosome demonstrates specifically that breaks can occur when forks are encountering break zones from only one direction and thus that breaks to do not result from unprogrammed convergence of opposing forks.

RSZs, relative to previously mapped termination sites and pause sites along chromosome III. Genetically encoded replication "slow zones" should not only cause slower fork progression but also will tend to accumulate convergent forks and thus should be positions of preferential fork termination. A recent genomic analysis examined fork progression on chromosome III (S6). The results of that analysis, and earlier studies, fit well with the current data. First, of the nine fork termination sites identified on chrIII (S6, S7), six occur within or at the edge of the six corresponding RSZs (fig. S2B). One of the three termination sites that does not correspond to a slow zone occurs at centromere; the other two (“T: 156” and “T: 210”) occur half-way between a minor secondary replication origin and one of the major origins with no slow zone in between and thus can be explained by stochastic fork convergence. Second, the three nonbreak zone fragments analyzed in Fig. 3 correspond in location to regions in chrIII where fork movement is relatively fast while break zone-II emerges as a prominent region of slow fork progression (S6). Other break zones are correlated either with fork termination (above) and/or with regions of abrupt changes in fork rate which are consistent with, but also tend to obscure detection of, slow fork movement (S6). Third, previous studies have also identified specific localized pause "sites" whose pausing activity has been associated with tRNA genes (S5, S8), centromeres (S9), or inactive ARS elements (S10)(fig. S2B). A discrete pause site is also detected within one of the break zones probed here (fig. S4C). However, the remaining two break zone fragments probed do not contain such a site, suggesting that this may not be a general defining feature of replication slow zones.

Roles of RSZs: further discussion of pausing and termination in MEC1 and mec1-ts strains. The existence of genetically encoded replication slow zones, and their occurrence specifically between active replication origins, could confer two important advantages to wild type cells. First, as discussed in the text, periodic pausing will provide time for development of important chromosomal features behind the fork and could also play a key role in regulation of S phase. Second, inter-origin slow zones will focus both pausing and termination to neutral positions, thus minimizing the risk that these events will interfere with replication initiation. Programmed fork stalling zones have not previously been identified in a eukaryotic organism but are known to exist in bacteria (S11). Perhaps such regions are a general feature of all types of genomes.

In mec1-ts strains, breaks could in principle arise from stalling of a single fork in an RSZ or they could arise only on those chromosomes where an RSZ happened to accumulate a pair of convergent forks. We favor the former possibility, for two reasons. First, breaks were not detected at any of the three termination sites identified from genomic analysis that did not correspond to break zones (S6). Second, two-dimensional gel analysis did not reveal any "X-forms" diagnostic of fork termination in any of the three break zone fragments analyzed; however, we cannot exclude the possibility that breaks require convergent forks but that they became stalled at some distance from one another such that they do not appear within the same probed fragment.

Prolongation of mitotic and meiotic S-phases in rad53 mutant. The length of S-phase in WT and rad53K277A cells is determined as described (S12). The rad53 mutation confers ~25% lengthening of S phases in both mitosis (39.7 ± 4.04 min vs. 32.7 ± 0.07; n=3, p < 0.05) and meiosis (105.5 ± 8.9 min vs. 82.9 ± 11.0 min; n = 4, p < 0.005).

This file is in Adobe Acrobat PDF format. If you have not installed and configured the Adobe Acrobat Reader on your system, please see Help with Printing for instructions.

Download Figures   size = 2,961KB

Supplemental Fig. S1. mec1-ts cells lose viability concomitant with the formation of DSBs, not during prolonged replication stalling. (A) Commitment-To-Inviability (CTI) measures the fraction of mec1-4 cells that is unable to resume growth following restoration of Mec1 function by temperature down shift to 23°C. CTI is calculated as followed; First, “% Cell viability” is determined by removing an aliquot of cells at each time point, plating out YPD at 23°C, and counting colony forming units (CFU) 2-3 day later. “% Cell viability” is defined by normalizing the CFU at each time point to CFUs at t = 0 (100%) and t = 240 (0%). CTI at each time point is obtained by subtracting “% Cell viability” at the corresponding time point from 100%. (B) Status of chrIII assessed by PFGE/Southern analysis. Probe: CHAI. (C) The amount of CHAI probe hybridizing to different chr III species are quantified, and expressed as the fraction of total chrIII ends detected.

Supplemental Fig. S2. Lack of correlation between mec1-ts break zones and meiotic DSBs, base composition, and cohesion distribution. (A) Positions of mec1-ts break zones, centromere, and ARS elements along chr III (see text). (B) Positions of replication termination [“T” (S6)], previously identified discrete fork pause sites [“P”; (S5, S10, S13)], and other notable features. P*: Discrete fork pause site identified in this study (fig. S4C). (C) Base composition isochores revealed by a sliding window of 30 kb. All yeast chromosomes, including chromosome III, exhibit broad regional biases comprising GC-rich and AT-rich isochores that average around 50 kb (S14, S15, S16). (D) Locations of, and levels of DSBs at meiotic DSB hotspots (S17). (E) Local fluctuations in base composition in chromosome III revealed by a sliding window of 5 kb. Local average base composition oscillates between GC-rich peaks and GC-poor valleys (i.e., AT-rich peaks or AT-queue) with a regular periodicity of ~15 kb (S14). (F) Relative distribution of mitotic cohesin Mcd1/Scc1 in cells arrested at early S phase with HU. Distribution of cohesin correlates with local base composition – i.e., cohesin binding peaks in general correspond to peaks of high local AT composition (S14).

Supplemental Fig. S3. Break distribution in mec1-4 strains containing wild type or varient chrIII. Breakage patterns in wild type and chrIII variants in which either ARS305 or ARS306 are inactivated by small deletions. Probe: YCR098

Supplemental Fig. S4. Two-dimensional gel analysis of mec1 break- (B-) and nonbreak (NB-) zones. Two additional B/NB segment pairs are analyzed in MEC1 (A) and mec1-ts (B) cells following -factor arrest/release (Fig. 3, A to C). Signals associated with replication intermediates (RI’s) in these gels are quantified and expressed as the fraction of total signal in Fig. 3, B and C, panels (c) and (d). Results in panels (b) and (c) are from the same -factor arrest/release experiment. Results in panel (d) are from an independent experiment. In each panel, both segments were assayed in the same genomic DNA sample. (C) A discrete fork pause site (*) in the B-c segment is revealed under the following condition: a 30 hour electrophoresis in 0.4% agarous gel at 1V/cm followed by 8 hour in 0.8% agarous gel containing 0.3ug/ml EthBr at 5V/cm. Locations of previously identified discrete fork pause sites correlate with that of tRNA genes (S5, S13), silent ARS element (S10), or centromere (S5, S9) (fig. S2B). The discrete pause site in break-zone V is located approximately 1 kb away from the mid-point of the analyzed Bgl II fragment (249,984 – 257,668) - that is, either around ~ 252,800 or ~254,800 – where none of the above mentioned notable features is found. Also note that the same pausing pattern is observed in both MEC1 and mec1-ts cultures.

REFERENCES

S1. C. Guthrie, G. Fink, Eds., Guide to Yeast Genetics and Molecular Biology (Academic Press Inc., San Diego, CA, 1991), vol. 194.

S2. V. Borde, A. S. H. Goldman, M. Lichten, Science 290, 806 (2000).

S3. C. Newlon et al., Genetics 129, 343 (1991).

S4. B. J. Brewer, W. L. Fangman, Cell 51, 463 (1987).

S5. S. Greenfeder, C. Newlon, Mol. Biol. Cell 3, 999 (1992).

S6. M. K. Raghuraman et al., Science 294, 115 (2001).

S7. J. Zhu, C. S. Newlon, J. A. Huberman, Mol. Cell. Biol. 12, 4733 (1992).

S8. A. M. Deshpande, C. S. Newlon, Science 272, 1030 (1996).

S9. S. A. Greenfeder, C. S. Newlon, Mol. Cell. Biol. 12, 4056 (1992).

S10. Y. Wang, M. Vujcic, D. Kowalski, Mol. Cell. Biol. 21, 4938 (2001).

S11. T. Hill, in Excherichia coli and Salmonella: Cellular and Molecular Biology F. Neidhardt, Ed. (ASM Press, Washington DC, 1996), pp. 1602.

S12. R. Cha, B. Weiner, S. Keeney, J. Dekker, N. Kleckner, Genes Dev. 14, 493 (2000).

S13. J. Theis, C. Newlon, Mol. Cell. Biol. 21, 2970 (2001).

S14. Y. Blat, N. Kleckner, Cell 98, 249 (1999).

S15. P. Sharp, A. Lloyd, Nucleic Acids Res. 21, 179 (1993).

S16. B. Dujon, Trends Genet. 12, 263 (1996).

S17. F. Baudat, A. Nicolas, Proc. Natl. Acad. Sci. U.S.A. 94, 5213 (1997).