Fig. 6. Rad50, but not Rad51, is required for homologous recombination-based survival in checkpoint-deficient cdc13-1 cells. (A) The indicated strains were propagated at 25°C before being grown at 29°C for the indicated numbers of passages. (B) Telomere structure of the strains shown above. Genomic DNA was prepared after growth at 29°C during the indicated periods of time, cut with XhoI and revealed with a TG_1–3 probe. A cdc13-1 rad51Δ mec3Δ triple mutant behaved exactly like a cdc13-1 mec3Δ strain in terms of growth (A) and generation of type II survivors (B), while a cdc13-1 rad50Δ mec3Δ triple mutant failed to generate survivors after the second passage (A).

Fig. 2. Continuous proliferation of cdc13-1 mec3-2 cells at 29°C requires the presence of Rad52 and is accompanied by the acquisition of TG_1–3 elongated telomeres. In (B), telomere lengths (XhoI digestion, TG_1–3 ³²P-labeled probe) of the cells shown in (A) are displayed at various time points. A and B represent two isogenic cdc13-1 mec3-2 strains with different kinetics of senescence. In strain A, the crisis began during the second passage, attested by slowing down of the growth rate (A), followed by dramatic telomere elongation during the third passage (B), while strain B has entered crisis during the third passage (A) but has not yet undergone recombination even during the fourth passage (B). In strain A, passage through the crisis and acquisition of elongated telomeres coincided with the acquisition of the capacity of the cdc13-1 mec3-2 cells to grow at temperatures up to 34°C. Strain B, which has not attained survival yet, is more temperature-sensitive than strain A. Strains C and D, two isogenic cdc13-1 mec3-2 rad52Δ triple mutant cells, totally ceased growth during the second passage. (C) Genomic DNA from a wild-type strain (lane 1) or from a cdc13-1 mec3-2 survivor (lane 2), grown on plates (‘streak assay’ described in Materials and methods), were digested with a mixture of restriction enzymes (AluI, HaeIII, HinfI and MspI) using a 4 bp recognition sequence, and the Southern revealed with a TG_1–3 probe. Since these enzymes cut within telomeric Y′ sequences but not within the TG_1–3 sequences (Teng and Zakian, 1999), the present data show that homologous recombination-induced telomere elongation in cdc13-1 mec3-2 concerns TG_1–3 but not Y′ sequences.

Fig. 4. Accumulation of single-stranded DNA in cdc13-1 mec3Δ cells. Non-denaturing (native, right panels in A, B, C and D) or denaturing (denatured, left panels in A, B, C and D) Southern hybridization of genomic DNA to a TG_1–3 ³²P-labeled probe following digestion with either XhoI, XbaI or EcoRI, as indicated. These enzymes cut in subtelomeric regions of Y′ chromosomes, as represented schematically (upper left). (A and B) Levels of single-stranded DNA in cdc13-1 mec3Δ cells increased with both time (1.5 or 3 h of incubation in A and B, respectively) and temperature, as indicated. Meanwhile, the amount of double-stranded telomeric DNA diminished concomitantly (denatured, left panels in A and B), a reaction that progressed centripetally from the telomere end (for instance, in B, degradation at 3 h at 25–34°C in XhoI-digested samples, at 32–34°C in XbaI-digested samples and no degradation in EcoRI-digested samples or in non-telomeric DNA, CDC15). (C) Accumulation of single-stranded DNA and degradation of double-stranded telomeric DNA were not due to homologous recombination because they still took place in cdc13-1 mec3Δ rad52Δ cells. (D) During the pre-senescence period (lanes 1–3, 29°C), DNA became degraded in telomeric regions, as evidenced by a decrease in the amount of double-stranded DNA exhibiting the XhoI site (1.2 kb band in middle left panel), while more internal telomeric DNA was not degraded (3.7 kb band after EcoRI digestion in upper left panel). In non-telomeric regions, DNA also remained intact (CDC15 probe, lower left panel). Post-senescence survivors underwent homologous recombination (lanes 4 and 5, 29°C), indicated by dramatic telomere elongation (middle left panel). Quantitations, indicated as numbers under the lanes, were made with a PhosphorImager using ImageQuant. Measurement of single-stranded DNA in native gels (middle right panel) presumably was not sensitive enough to detect the telomeric DNA damage described above.

Fig. 1. Checkpoint-deficient cdc13-1 cells undergo a senescence program followed by survival. (A) The cdc13-1 mec3-2 and cdc13-1 mec3Δ double mutant strains could grow at temperatures up to 29°C, whereas the cdc13-1 single mutant stopped growing above 25°C. A cdc13::TRP1 YCp111-CDC13 strain was used as a positive control. All cells were first grown at 25°C, and 10-fold serial dilutions (from left to right in each row) were incubated for 2 days at the indicated temperatures and photographed. (B) At 25°C, telomeres in the cdc13-1 mutant cells were slightly longer than those in a wild-type isogenic strain (left panel, compare lanes 1 and 2), while cdc13-1 mec3Δ and cdc13-1 mec3-2 double mutants exhibited wild-type length telomeres at either 25 or 29°C, as indicated. Similarly, mec3::TRP1 (middle panel) and mec3-2 mutants (right panel) had telomeres of wild-type length at 25 and 29°C. Total DNA was digested with XhoI and a TG_1–3 ³²P-labeled probe was used to detect telomeric sequences (see Materials and methods). (C) Double mutant strains of the indicated relevant genotype were incubated either at 25°C (upper left panel, photographed here 3 days after re-streaking) or at 29°C (all other panels, with the exception of cdc13-1 rad9 cells incubated at 28°C) for the indicated number of passages (typically, one passage or re-streak is performed after 20–25 generations). Survivors appeared in all strains, at different times, as rare isolated colonies. cdc13-1 rad9Δ cells generated survivors at 28°C but not at 29°C. During passages 1 and 4, cells were prepared for telomere length measurement (XhoI digestion, TG_1–3 probe). All five cdc13-1 strains bearing a different checkpoint mutation exhibited recombination of telomeric sequences.

Fig. 3. The cdc13-1-induced senescence/survival process is reversible; it also decreases cdc13-1 temperature sensitivity. (A) Growth characteristics of various mutants combining the temperature-sensitive cdc13-1 mutation and non-temperature-sensitive mutations in MEC3, TLC1, TEN1 or RIF2 (see text for explanations). Ten-fold serial dilutions (from left to right in each row) of transformants were grown for 2 days at the indicated temperatures and photographed. The letter ‘R’ indicates those of the strains that have overcome crisis and undergone recombination-dependent survival. Only the checkpoint-deficient cdc13-1 survivors displayed an increase in the permissive temperature of growth from 25°C (in cdc13-1 cells, row 1) or 29°C (in cdc13-1 mec3-2 cells prior to survival, row 2) to 34–37°C (rows 3, 4 and 6), while checkpoint-proficient, telomerase-deficient cdc13-1 cells (cdc13-1 tlc1::LEU2 ‘R’, row 5) could not grow above 25°C. In the control experiment shown in row 11, cells disrupted for CDC13 and harboring a LEU2-based single-copy plasmid expressing cdc13 sequences amplified from genomic DNA from three different cdc13-1 mec3-2 survivors (only one is shown here) can live only at 25°C after the plasmid containing wild-type CDC13 had been shuffled out on 5-fluoro-orotic acid-containing medium (cdc13::TRP1 YCp-cdc13-1), thus indicating that the temperature-sensitive cdc13-1 allele has not reverted to a less temperature-sensitive allele in the original strain. (B) Another control experiment shows that reintroduction of MEC3 on a single-copy plasmid (cdc13-1 mec3-2 ‘R’ pRS-MEC3, row 2) restored the arrest at 29°C in cdc13-1 mec3-2 cells having acquired the ability to grow at temperatures higher than 34°C following survival (cdc13-1 mec3-2 ‘R’ pRS, transformed with plasmid alone, row 1), which also suggests that the cdc13-1 mutation is still present in these cells. (C) When survivors of cdc13-1 mec3Δ were brought back from 29°C to permissive temperature (25°C), they slowly re-acquired wild-type length telomeres, during the indicated periods of time, as shown here for three independent clones (XhoI digestion, TG_1–3 probe).

Fig. 5. Post-senescence survival in cdc13-1 mec3-2 occurs without prior telomere shortening. Growth kinetics (A) and corresponding telomere length (B) of cdc13-1 mec3-2 and cdc13-1 mec3-2 tlc1Δ cells at 25 or 29°C. Each lane in (B) has been labeled with a letter corresponding to the cell patches shown in (A). At 25°C (A, top panel), cdc13-1 mec3-2 never experienced senescence, in contrast to cdc13-1 mec3-2 tlc1Δ cells which started to senesce by the end of the third passage (patch B), continuing for two additional passages (patch D), accompanied by clear telomere shortening (B, lanes B and D) prior to recombination-induced telomere elongation (lane F). At 29°C (A, bottom panel), cdc13-1 mec3-2 cells senesced less rapidly than cdc13-1 mec3-2 tlc1Δ cells. There was no telomere shortening prior to post-senescence survival in cdc13-1 mec3-2 cells (B, lane G). Note that type II homologous recombination was observed in both cdc13-1 mec3-2 (B, lanes I, K and M) and cdc13-1 mec3-2 tlc1Δ cells (B, lanes H, J, L and N). Cells were grown at 25 or 29°C for the indicated number of passages (∼25 generations per passage) prior to preparation of genomic DNA for telomere structure analysis (XhoI digestion; TG_1–3 probe). (C) In telomerase-deficient cells, the type II pattern induced by the presence of the cdc13-1 mutation (even at permissive temperature for growth) is dominant over type I. Cells were grown at 25°C for ∼75 generations prior to preparation of genomic DNA (XhoI digestion; TG_1–3 probe). This also illustrates the very distinctive nature of type II (lanes 3–5, cdc13-1 tlc1::TRP1) compared with type I recombination (lane 1, tlc1::TRP1) attested here by the disappearance of the non-Y′ bands (arrows), well visible in wild-type cells (lane 2, wt), as described in Materials and methods. The different appearance of type II in (B) and (C) is due to the fact that in (B) the whole patch of cells had to be processed in order to illustrate each time point accurately, hence the smear corresponding to the superimposition of different banding patterns from different colonies, whereas in (C) each type II illustrates a single clone.