Rap1p telomere association is not required for mitotic stability of a C(3)TA(2) telomere in yeast

Abstract

Telomeric DNA usually consists of a repetitive sequence: C(1-3)A/TG(1-3) in yeast, and C(3)TA(2)/T(2)AG(3) in vertebrates. In yeast, the sequence-specific DNA- binding protein Rap1p is thought to be essential for telomere function. In a tlc1h mutant, the templating region of the telomerase RNA gene is altered so that telomerase adds the vertebrate telomere sequence instead of the yeast sequence to the chromosome end. A tlc1h strain has short but stable telomeres and no growth defect. We show here that Rap1p and the Rap1p-associated Rif2p did not bind to a telomere that contains purely vertebrate repeats, while the TG(1-3) single-stranded DNA binding protein Cdc13p and the normally non-telomeric protein Tbf1p did bind this telomere. A chromosome with one entirely vertebrate-sequence telomere had a wild-type loss rate, and the telomere was maintained at a short but stable length. However, this telomere was unable to silence a telomere-adjacent URA3 gene, and the strain carrying this telomere had a severe defect in meiosis. We conclude that Rap1p localization to a C(3)TA(2) telomere is not required for its essential mitotic functions.

Fig. 1. A telomere consisting solely of vertebrate telomeric DNA can be stably maintained in yeast. (A) To generate a completely vertebrate-sequence telomere at chromosome VII-L, EcoRI–SalI-digested pUT-H was integrated at the ADH4 locus on chromosome VII-L in a tlc1h strain, in a manner that deletes the terminal ∼20 kb from the chromosome (Gottschling et al., 1990). The 60 bp C₃TA₂ sequence (black box) acts as a seed for telomere formation. The tlc1h telomerase can add only vertebrate repeats to this end, resulting in a VII-L telomere that contains no yeast telomeric repeats. A unique sequence (striped block) between the telomere seed and the URA3 gene was used as a PCR primer site for telomere sequencing and chromatin immunoprecipitations. The arrow above the spotted box indicates the URA3 promoter. The indicated PstI site is located upstream of the URA3 start codon. (B) Sequencing results for modified VII-L telomere. VII-L telomeres from the 499UT-H strain were cloned and sequenced. Clones were made ∼125 cell divisions after transformation in three independent integrants of the UT-H construct. Representative sequences for three clones, one from each integrant, are shown; the centromere-proximal end of the sequence is to the right. Seven modified telomeres were sequenced; none contained yeast telomeric DNA. While none of the sequenced telomeres contained the heterogeneous C_1–3A repeats characteristic of yeast telomeres, the VII-L telomere in the 499UT-H2 transformant contained one copy each of two variants of the vertebrate repeat: C₄TA₂ and C₂TA₂. The 499UT-H1 strain, with 245 bp of pure vertebrate repeats, was used for all subsequent experiments. (C) Telomere length is stable in the 499UT-H strain over many generations. Telomere length in the 499UT-Y and 499UT-H strains was assayed after one and ten serial restreaks on YC plates. Genomic DNA was digested with PstI and XhoI; the Southern blot was probed sequentially with URA3, C_1–3A and C₃TA₂. The upper band in the URA3-probed blot (*) is the endogenous ura3-52 allele. Two to three independent colonies from each strain are shown. Black arrowheads indicate the 499UT-Y VII-L telomere; white arrowheads indicate the 499UT-H C₃TA₂ VII-L telomere. (D) Effect of rad52Δ on telomere length. Southern blots of PstI-digested genomic DNA from four serial restreaks of rad52 and RAD52 versions of the 499UT-Y and 499UT-H strains were probed with URA3 which detects the VII-L telomere fragment. The upper band in the URA3-probed blot (*) is the endogenous ura3-52 allele. Arrows indicate the VII-L (URA3 probe) telomere band.

Fig. 2. The C3TA2 telomere has a wild-type loss rate. (A) Disome strain for chromosome stability assay. Loss of the test copy of chromosome VII was measured in a chromosome VII disome strain with the tlc1h mutation by selecting for white, cycloheximide-resistant colonies. URA3 with a vertebrate telomere seed was integrated at the VII-L telomere on the test copy of chromosome VII, generating a telomere containing solely vertebrate telomeric DNA. The control strain was a VII-L disome with URA3 at the VII-L telomere on the test copy of the chromosome and wild-type TLC1 (Sandell and Zakian, 1993). C₃TA₂ repeats are in black portion of triangles; C_1–3A DNA is in white. (B) Rates of chromosome loss. The average rates of chromosome loss for UT-Y (TLC1, wt telomeres) and UT-H (tlc1h, C₃TA₂ VII-L telomere) disome strains. Error bars are standard deviation.

Fig. 3. TPE and nuclear organization of a tlc1h strain. (A) TPE of 499UT-Y and 499UT-H strains. Ten-fold serial dilutions of 499UT-Y (Y) and 499UT-H (H) were plated on YC plates (to control for viability) and FOA plates (to select for cells that did not express the telomeric URA3 gene). TPE was also quantitated by comparing plating efficiency of the 499UT-Y and 499UT-H strains on FOA and non-selective medium. (B) Localization of myc-Tbf1p and Rap1p by immunofluorescence. 499UT-Y (Y) and 499UT-H (H) strains expressing myc-tagged Tbf1p were stained with α-Rap1p polyclonal antibodies and DAPI stain or with α-myc monoclonal antibodies.

Fig. 4. The tlc1h strain has altered meiosis. (A) Diploid strains used in meiosis experiments. Diploid strains containing URA3 on both copies of chromosome VII-L with either solely yeast telomere sequence and wild-type TLC1 (Y/Y) or homozygous for both the C₃TA₂ telomere and the tlc1h mutation (H/H), as well as diploids homozygous for the tlc1h mutation and carrying URA3 with the C₃TA₂ telomere on only one (H/YH) or neither (YH/YH) copy of chromosome VII-L were sporulated (Y indicates solely yeast repeats at VII-L telomere; H indicates solely human telomeric repeats on the VII-L telomere; YH indicates a mixture of yeast and human telomeric DNA at the VII-L telomere; the rest of the telomeres in the tlc1h strains were a mixture of yeast and C₃TA₂ repeats). Yeast C_1–3A telomeric repeats are in gray; vertebrate C₃TA₂ telomeric repeats are in black. (B) The number of asci was divided by the number of asci plus unsporulated cells to give the percent sporulated cells. The number of spores per ascus was also determined, and used to calculate the percentage of asci containing dyads and tetrads. Graphs display averages ± standard deviation.

Fig. 5. Protein composition of the C₃TA₂ telomere. (A) Sequences assayed in ChIPs. Sequences amplified by PCR are indicated by a black bar below the chromosome. Multiplex PCR was performed on the immunoprecipitated samples as well as input DNA dilutions, amplifying four sequences: the internal control sequence ARO, the sub- telomeric sequence ADH and the telomere-adjacent sequences VII-L TEL and VI-R TEL (representing the completely vertebrate VII-L telomere and the mixed-sequence VI-R telomere, respectively). (B) ChIP results. ChIPs were performed using polyclonal antibodies to Rap1p and Rif2p, and myc-tagged Tbf1p and Cdc13p. An N-terminal 9-myc tag was integrated at the chromosomal copy of Tbf1p; the tagged protein was expressed from the endogenous TBF1 promoter. Since myc-TBF1 cells grew nearly as well as wild-type, the tagged protein fulfilled the essential function of Tbf1p. Cdc13p with a C-terminal 9-myc tag was expressed from the CDC13 promoter on a centromere plasmid in a strain that also expressed wild-type Cdc13p from the endogenous locus; when the 9-Myc-Cdc13p is the only Cdc13p in the cell, it supports wild-type telomere length and growth rate (Taggart et al., 2002). Input DNA represents the total chromatin isolated, before immunoprecipitation; two-fold serial dilutions of the input DNA were amplified by PCR to establish the linear range of the PCR amplifications. Negative control samples were immunoprecipitated from untagged strains with either polyclonal antibodies to the non-telomere protein Ypt1 (this was used to define background levels for the polyclonal Rap1p and Rif2p sera) or α-myc monoclonal antibodies (the control for the myc-Tbf1 and Cdc13-myc strains). (C) Quantitation of ChIP data. The fold enrichment of the VII-L and VI-R TEL bands over background levels are shown; error bars indicate standard deviations. The fold enrichment was normalized to the ARO1 control band. Note that because the fold enrichments of Rap1p, Tbf1p and Cdc13p at the VII-L telomere are considerably higher than the other signals obtained, the leftmost plot is on a different scale than the center and right plots.