A Novel Tel1/ATM N-Terminal Motif, TAN, Is Essential for Telomere Length Maintenance and a DNA Damage Response

Abstract

Tel1/ATM, a conserved phosphatidylinositol 3-kinase-related kinase (PIKK), acts in the response to DNA damage and regulates telomere maintenance. PIKK family members share an extended N-terminal region of low sequence homology. Sequence alignment of the N terminus of Tel1/ATM orthologs revealed a conserved, novel motif we term TAN (for Tel1/ATM N-terminal motif). Point mutations in conserved residues of the TAN motif resulted in telomere shortening, and its deletion caused the same short telomere phenotype as complete deletion of Tel1 did. Overexpressing Tel1 TAN mutants did not rescue telomere shortening. The TAN motif was also essential for the function of Tel1 in the response to DNA damage, as TAN-deleted Tel1 was indistinguishable from the complete lack of Tel1 in causing reduced viability and signaling through Rad53 upon DNA damage. Strikingly, TAN deletion reduced recruitment of Tel1 to a double-strand DNA break. Together, these results define a conserved sequence motif within an otherwise poorly defined region of the Tel1/ATM kinase family proteins that is essential for normal Tel1 function in Saccharomyces cerevisiae.

ATM is a large protein kinase mutated in patients with ataxia telangiectasia (AT), a rare autosomal recessive disorder characterized by neurodegeneration, immunodeficiency, and cancer predisposition (33). In humans, ATM is critical for responding to DNA double-strand breaks (DSBs). While in budding yeast (Saccharomyces cerevisiae), the major kinase responsible for regulating the response to DNA damage is Mec1, an ortholog of the related mammalian kinase ATR, multiple lines of evidence indicate that the ATM ortholog in budding yeast, Tel1, also regulates responses to DSBs. Tel1 is one of the first proteins detected to localize to a DSB (23), and this recruitment requires an evolutionarily conserved interaction with Xrs2 (the ortholog of mammalian NBS1) (11, 29, 42). One setting in which a functional role for Tel1 in the DNA damage response is uncovered is in cells lacking Mec1 and another protein, Sae2. When these cells are exposed to the DNA-damaging agent methyl methane sulfonate (MMS), Tel1 is important for maintaining viability and for phosphorylation of the DNA damage signal transducer kinase Rad53 (an ortholog of mammalian CHK2) (41).

In addition to functioning in the response to DSBs, numerous findings indicate that Tel1/ATM regulates telomeres. Yeast cells lacking Tel1, like human AT cells, display telomere shortening and alterations in telomeric chromatin structure (5, 17, 25, 28, 35). Mutating both Tel1 and Mec1 together causes a growth defect and more extensive telomere shortening than in tel1Δ cells (10, 31, 39). Tel1 is preferentially recruited to short telomeres (6, 18, 32), and as at DSBs, this recruitment requires the Xrs2 C terminus (32). Tel1 is required for efficient recruitment of telomerase to short telomeres (16, 32). Furthermore, the kinase activity of Tel1 is important for telomere maintenance, as mutations in the catalytic residues of the kinase domain of Tel1 result in telomere shortening (27).

Insights into the mechanism of function of Tel1/ATM have been generated by studying its domain structure. The overall domain structure of Tel1/ATM is shared by proteins of the phosphatidylinositol 3-kinase (PI3K)-related kinase (PIKK) family (reviewed in reference 1). Thus, Tel1/ATM, like other PIKKs, contains a C-terminal region composed of a PI3K-like kinase domain, a FAT (FRAP [TOR], ATM, and TRAPP) domain, and a short FATC (FAT C-terminal) domain (7, 19). Another feature shared among PIKKs is an extended N-terminal region of over 1,500 amino acids that has low primary sequence conservation and is predicted to be composed of helical repeats (3, 9, 22, 30). The mammalian ATM N terminus is important for regulating the response to DNA damage (12, 40), and regions of it have been reported to interact with protein partners (4, 11, 15, 20, 37), associate with chromatin (43), and regulate the nuclear localization of ATM (43). However, the role of this Tel1/ATM sequence in telomere maintenance has not been characterized, and the function of this region in Tel1 has not been extensively studied.

Here we report an analysis of the region of Tel1 N terminal to the kinase domain in telomere maintenance and the DNA damage response. We identified a novel sequence motif, which we term TAN, that is located near the N terminus of this larger region and that is conserved specifically in the Tel1/ATM subclass of the PIKKs. The TAN motif was essential for both telomere length maintenance and for Tel1 action in response to DNA damage. While mutations in the TAN motif reduced Tel1 protein levels, overexpression of TAN-mutated Tel1 proteins did not rescue the telomere length defect. Furthermore, deletion of the TAN motif impaired localization of Tel1 to a DSB, even in cells expressing increased levels of the mutant protein. Hence, the TAN motif is a previously unrecognized conserved domain of the Tel1/ATM PIKK subclass essential for the known functions of Tel1 in S. cerevisiae.

MATERIALS AND METHODS

Plasmids and strains.

All strains, unless indicated, were in the Saccharomyces cerevisiae BY4741 background (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0; this strain is renamed here 12178) (8). The S. cerevisiae strains used in this study are shown in Table 1. Standard techniques were used for growth and sporulation of yeast.

TABLE 1.

S. cerevisiae strains used in this study

Figurea	Strain	Relevant genotypeb	Comment	Referencec
Fig. S1B in the supplemental	12178 (BY4741)	MATahis3Δ1 leu2Δ0 met15Δ0 ura3Δ0		8
material	12182A	3xFLAG-TEL1
	12182B	3xFLAG-TEL1
	12196A	3xFLAG-tel1Δ(419-649)
	12196B	3xFLAG-tel1Δ(419-649)
	12195A	3xFLAG-tel1Δ(650-939)
	12195B	3xFLAG-tel1Δ(650-939)
	12197A	3xFLAG-tel1Δ(940-1383)
	12197B	3xFLAG-tel1Δ(940-1383)
	12194A	3xFLAG-tel1Δ(1384-1859)
	12194B	3xFLAG-tel1Δ(1384-1859)
	12193A	3xFLAG-tel1Δ(1860-2405)
	12193B	3xFLAG-tel1Δ(1860-2405)
	12181A	tel1Δ::kanMX6
	12181B	tel1Δ::kanMX6
Fig. S1C in the supplemental	12178 (BY4741)	MATahis3Δ1 leu2Δ0 met15Δ0 ura3Δ0		8
material	12182A	3xFLAG-TEL1
	12182B	3xFLAG-TEL1
	12183A	3xFLAG-tel1-ΔN567
	12184A	3xFLAG-tel1-ΔN1729
	12185A	3xFLAG-tel1-ΔN2458
	12181A	tel1Δ::kanMX6
	12181B	tel1Δ::kanMX6
Fig. 2B	12178 (BY4741)	MATahis3Δ1 leu2Δ0 met15Δ0 ura3Δ0		8
	12182A	3xFLAG-TEL1
	12182B	3xFLAG-TEL1
	12929A	3xFLAG-tel1-ΔN10
	12929B	3xFLAG-tel1-ΔN10
	12512A	3xFLAG-tel1-ΔN20
	12512B	3xFLAG-tel1-ΔN20
	12944A	3xFLAG-tel1-L13A K17A E20A R21A
	12944B	3xFLAG-tel1-L13A K17A E20A R21A
	12928A	3xFLAG-tel1-ΔN30
	12928B	3xFLAG-tel1-ΔN30
	12513A	3xFLAG-tel1-ΔN40
	12513B	3xFLAG-tel1-ΔN40
	12181A	tel1Δ::kanMX6
	12181B	tel1Δ::kanMX6
Fig. 2C	12178 (BY4741)	MATahis3Δ1 leu2Δ0 met15Δ0 ura3Δ0		8
	12351A	kanMX6::PGAL1-TEL1
	12351B	kanMX6::PGAL1-TEL1
	12354A	kanMX6::PGAL1-tel1-L13A K17A E20 R21A
	12354B	kanMX6::PGAL1-tel1-L13A K17A E20 R21A
	12947A	kanMX6::PGAL1-tel1-ΔN40
	12947B	kanMX6::PGAL1-tel1-ΔN40
	12181A	tel1Δ::kanMX6
	12181B	tel1Δ::kanMX6
Fig. 2D	12178 (BY4741)	MATahis3Δ1 leu2Δ0 met15Δ0 ura3Δ0		8
	12181A	tel1Δ::kanMX6
	12181B	tel1Δ::kanMX6
	12362A	tel1-ΔN40
	12362B	tel1-ΔN40
	12220A	tel1-L13A K17A E20A R21A
	12220B	tel1-L13A K17A E20A R21A
	12222A	tel1-L13A
	12222B	tel1-L13A
	12234A	tel1-K17A E20A R21A
	12234B	tel1-K17A E20A R21A
	12224A	tel1-K17A
	12939A	tel1-K17A
	12226A	tel1-E20A
	12226B	tel1-E20A
	12228A	tel1-R21A
	12228B	tel1-R21A
Fig. 2E	12178 (BY4741)	MATahis3Δ1 leu2Δ0 met15Δ0 ura3Δ0		8
	12230A	tel1-R21K
	12230B	tel1-R21K
Fig. 3	12308	MATa/MATα his3Δ1/his3Δ1 leu2Δ0/leu2Δ0 met15Δ0/met15Δ0 ura3Δ0/ura3Δ0 mec1Δ::LEU2/MEC1 sae2Δ::HIS3/SAE2 sml1Δ::MET15/SML1 tel1Δ::kanMX6/TEL1	Used for sporulation
	12311	MATa/MATα his3Δ1/his3Δ1 leu2Δ0/leu2Δ0 met15Δ0/met15Δ0 ura3Δ0/ura3Δ0 mec1Δ::LEU2/MEC1 sae2Δ::HIS3/SAE2 sml1Δ::MET15/SML1 tel1-L13A K17A E20A R21A/TEL1	Used for sporulation
	12514	MATa/MATα his3Δ1/his3Δ1 leu2Δ0/leu2Δ0 met15Δ0/met15Δ0 ura3Δ0/ura3Δ0 mec1Δ::LEU2/MEC1 sae2Δ::HIS3/SAE2 sml1Δ::MET15/SML1 tel1-ΔN40/TEL1	Used for sporulation
Fig. 3A	12976A	mec1Δ::LEU2 sae2Δ::HIS3 sml1Δ::MET15	Spore colony from strain 12308
Fig. 3B	12977A	mec1Δ::LEU2 sae2Δ::HIS3 sml1Δ::MET15 tel1-4A	Spore colony from strain 12311
Fig. 3C	12978A	mec1Δ::LEU2 sae2Δ::HIS3 sml1Δ::MET15 tel1-ΔN40	Spore colony from strain 12514
Fig. 3D	12979A	mec1Δ::LEU2 sae2Δ::HIS3 sml1Δ::MET15 tel1Δ::kanMX6	Spore colony from strain 12308
Fig. 4	12514	MATa/MATα his3Δ1/his3Δ1 leu2Δ0/leu2Δ0 met15Δ0/met15Δ0 ura3Δ0/ura3Δ0 mec1Δ::LEU2/MEC1 rad52Δ::HIS3/RAD52 sml1Δ::MET15/SML1 tel1-ΔN40/TEL1	Used for sporulation
	12314	MATa/MATα his3Δ1/his3Δ1 leu2Δ0/leu2Δ0 met15Δ0/met15Δ0 ura3Δ0/ura3Δ0 mec1Δ::LEU2/MEC1 rad52Δ::HIS3/RAD52 sml1Δ::MET15/SML1 tel1Δ::kanMX6/TEL1	Used for sporulation
	12317	MATa/MATα his3Δ1/his3Δ1 leu2Δ0/leu2Δ0 met15Δ0/met15Δ0 ura3Δ0/ura3Δ0 mec1Δ::LEU2/MEC1 rad52Δ::HIS3/RAD52 sml1Δ::MET15/SML1 tel1-L13A K17A E20A R21A/TEL1	Used for sporulation
Fig. 4A and B	12314-1-1c	mec1Δ::LEU2 sml1Δ::MET15
	12317-16c	mec1Δ::LEU2 sml1Δ::MET15 tel1-L13A K17A E20A R21A
	12514-5a	mec1Δ::LEU2 sml1Δ::MET15 tel1-ΔN40
	12314-2-6b	mec1Δ::LEU2 sml1Δ::MET15 tel1Δ::kanMX6
Fig. 5	12182A	3xFLAG-TEL1
	12928A	3xFLAG-tel1-ΔN30
Fig. 6	12981A	XRS2-13MYC::kanMX6
	12182A	3xFLAG-TEL1
	12200	MATaade2Δ::hisG his3Δ200 leu2Δ0 lys2Δ0 met15Δ0 trp1Δ63 ura3Δ0 3xFLAG-TEL1 XRS2-13MYC::kanMX6
	12982A	3xFLAG-tel1-ΔN30 XRS2-13MYC::kanMX6
Fig. 7	13257	MATa-inc ade2Δ::hisG his3Δ200 met15Δ0 trp1Δ63 ura3Δ0 trp5::HO recognition site-TRP1 pGAL-HO		2
	13258	MATa-inc ade2Δ::hisG his3Δ200 met15Δ0 trp1Δ63 ura3Δ0 trp5::HO recognition site-TRP1 pGAL-HO 3xFLAG-TEL1		2
	12991A	MATa-inc ade2Δ::hisG his3Δ200 met15Δ0 trp1Δ63 ura3Δ0 trp5::HO recognition site-TRP1 pGAL-HO 3xFLAG-tel1-ΔN30
	13375	MATa-inc ade2Δ::hisG his3Δ200 met15Δ0 trp1Δ63 ura3Δ0 trp5::HO recognition site-TRP1 pGAL-HO 3xFLAG-TEL1 pRS413
	13406	MATa-inc ade2Δ::hisG his3Δ200 met15Δ0 trp1Δ63 ura3Δ0 trp5::HO recognition site-TRP1 pGAL-HO 3xFLAG-tel1-ΔN30 pRS413
	13414	MATa-inc ade2Δ::hisG his3Δ200 met15Δ0 trp1Δ63 ura3Δ0 trp5::HO recognition site-TRP1 pGAL-HO 3xFLAG-tel1-ΔN30 p-3xFLAG-tel1-ΔN30

^aFor ease of reference, strains are listed by the figure in which they were used. Some strains were used in multiple figures.

^bAll strains are isogenic with strain 12178 (BY4741) (8) unless indicated. Diploids that were sporulated to obtain haploid cells are listed.

^cUnless noted otherwise, strains were generated during the course of the present study.

Deletion mutants.

S. cerevisiae strains with deletion mutations were constructed by one-step allele replacement. The tel1Δ::kanMX6 strains were made by transformation of a PCR product generated using pFA6a-kanMX6 as a template (24). Strains containing sml1Δ::MET15, sae2Δ::HIS3, and mec1Δ::LEU2 were made by transformation of PCR products generated using pRS401, pRS403, and pRS405, respectively, as templates (8).

3xFLAG proteins.

To introduce the 3xFLAG tag, a pMPY-3xFLAG plasmid (p12009) was constructed (34). The 3xFLAG-TEL1, 3xFLAG-tel1-ΔN10, 3xFLAG-tel1-ΔN20, 3xFLAG-tel1-4A, 3xFLAG-tel1-ΔN30, 3xFLAG-tel1-ΔN40, 3xFLAG-tel1-ΔN567, 3xFLAG-tel1-ΔN1729, and 3xFLAG-tel1-ΔN2458 strains were generated by transformation of a PCR fragment generated using p12009 as a template, followed by mitotic recombination, as previously described (34). 3xFLAG-tel1-Δ(419-649), 3xFLAG-tel1-Δ(650-939), 3xFLAG-tel1-Δ(940-1383), 3xFLAG-tel1-Δ(1384-1859), and 3xFLAG-tel1-Δ(1860-2405) were generated by first targeting a URA3 PCR product made using pRS406 as a template to different regions of the 3xFLAG-TEL1 gene in strain 12182A, transforming PCR products containing the TEL1 sequence but lacking sequence encoding the desired deleted residues, and selecting for recombinants on 5-fluoroorotic acid (5-FOA).

Untagged Tel1 mutant proteins.

The tel1-ΔN40 allele was generated by targeting a URA3 PCR cassette to the 5′ region of TEL1 in strain 12178, transforming with a TEL1 PCR product lacking sequence encoding amino acids 2 to 40, and selecting for recombinants on 5-FOA. To introduce point mutations into sequence encoding the TEL1 TAN motif, plasmids for two-step allele replacement were made. First, overlapping PCR was used to make 1,038-bp fragments (positions −475 to +562) of TEL1 harboring the desired mutation(s). The fragments were cloned into the XhoI and SpeI sites of pRS406 (8). The following plasmids were digested with BsiWI for linearization: p12056 (tel1-L13A), p12058 (tel1-K17A), p12059 (tel1-E20A), p12060 (tel1-R21A), p12061 (tel1-R21K), p12110 (tel1-K17A E20A R21A[3A]), and p12055 (tel1-L13A K17A E20A R21A [4A]). Digested plasmids were transformed into strain 12178, and integrants were selected on media lacking uracil. Following PCR confirmation of integration, cells were grown overnight in yeast extract-peptone-dextrose (YEPD), washed in water, and plated on 5-FOA to select for mitotic recombinants. To generate the kinase-deficient TEL1 allele, a fragment of TEL1 (positions 7213 to 8364) mutated to encode D2612A N2617K was cloned into the XbaI and XhoI sites of pRS406 to form plasmid p12036. The p12036 plasmid was digested with HindIII, and two-step allele replacement was performed as described above.

PGAL1 strains.

The PGAL1 promoter was integrated by transforming strain 12178, 12220A, or 12362A with PCR products generated using pFA6a-kanMX6-PGAL1 as a template (24).

Xrs2-13MYC.

Xrs2-13MYC strains were made by transformation of a PCR product generated using pFA6a-13Myc-kanMX6 as a template (24).

Strains for chromatin immunoprecipitation (ChIP).

Construction of strains 13257 and 13258, containing an HO recognition site at the TRP1 locus and the plasmid pGal-HO for galactose-inducible expression of HO endonuclease, was described in reference 2. The 3xFLAG-tel1-ΔN30 allele was introduced into this strain background using a PCR product from p12009 as described above, generating strain 12991A. An extra copy of 3xFLAG-tel1-ΔN30 was introduced into strain 12991A by transforming it with a BsaBI digestion product of the CEN plasmid p13068 (2), which contains 3xFLAG-TEL1 downstream of the TEL1 promoter. The BsaBI digestion removed the 5′ end of the full-length 3xFLAG-TEL1 gene, which was then replaced with the 5′ end of the deletion mutant 3xFLAG-tel1-ΔN30 by gap repair in vivo. For controls, strains 13257 and 12991A were transformed with the empty vector pRS413(8).

Southern blotting and telomere length detection.

Genomic DNA was isolated from cells treated with zymolyase and then digested with BanI (New England Biolabs) for 3 hours at 37°C. Digested genomic DNA was electrophoresed on 0.8% agarose gels for 16 h at 65 V. Following electrophoresis, gels were rinsed in Milli-Q water and then incubated in depurination buffer (0.25 N HCl) for 30 min. DNA was transferred to Hybond-XL membrane (GE Healthcare) by Southern blotting in denaturation buffer (1.5 M NaCl and 0.5 M NaOH) for at least 7 h and was cross-linked to the membrane using a UV Stratalinker 1800 (Stratagene). The membrane was washed in hybridization buffer (0.5 M Na₂HPO₄, 1 mM EDTA, and 7% sodium dodecyl sulfate [SDS]) at 55°C. The KL1 DNA fragment for making the probe was generated by PCR on yeast genomic DNA with primers oEHB15-060 (CCCACACTTTTCACATCTACCTCTACTCTCGCTGTCACTCCTTACCCGGC) and oEHB15-106 (CCCCGAATTCCGGCATTCCTGTCGATGCTGATAGGG) (26). KL1 was used for making the probe using [α-³²P]dCTP (Perkin Elmer) and the Rediprime II random prime labeling system (GE Healthcare). The radiolabeled probes were incubated with Southern blots for approximately 12 h in hybridization buffer at 55°C. Blots were rinsed with wash buffer (0.1 M Na₂HPO₄, 0.068% H₃PO₄, and 2% SDS) and washed two times in wash buffer for 30 min each time at 55°C. The probed Southern blot was exposed to a phosphor screen (Molecular Dynamics), and the exposed phosphor screen was scanned using a Storm or Typhoon PhosphorImager (Molecular Dynamics).

Sequence alignment.

Sequences were aligned using ClustalW (http://www.ebi.ac.uk/clustalw/). The species used in the lineup (GenBank accession numbers for proteins from the species are shown in parentheses) include the following: Aspergillus clavatus NRRL 1 (XP_001274554), Aspergillus nidulans (Q5BHE2), Aspergillus oryzae (Q2U639), Bos taurus (XP_605200), Candida albicans (Q5ABX0), Candida glabrata (Q6FRZ9), Canis familiaris (XP_862768), Danio rerio (XP_001334561), Debaryomyces hansenii (Q6BV76), Drosophila melanogaster (Q5EAK6), Drosophila pseudoobscura (EAL29042.1), Eremothecium gossypii (Q751J3), Fibobasidiella neoformans (Q5KFE0), Gallus gallus (XP_417160), Gibberella zeae (Q4IB89), Homo sapiens (AAB65827), Kluyveromyces lactis (Q6CP76), Mus musculus (Q62388), Neosartorya fischeri NRRL 181 (XP_001259432), Neurospora crassa (Q7RZT9), Oryza sativa (NP_001041778), Ostreococcus tauri (CAL57286), Pan troglodytes (XP_001139487), Pichia stipitis CBS 6054 (XP_001384387), Pichia guilliermondii ATCC 6260 (XP_001481987), Rattus norvegicus (XP_236275), Saccharomyces bayanus (file name MIT_Sbay_c434_1214; obtained from http://db.yeastgenome.org/fungi/YBL088C.html [10a, 19a]), Saccharomyces cerevisiae (P38110), Saccharomyces paradoxus (file name MIT_Spar_c450_679; obtained from http://db.yeastgenome.org/fungi/YBL088C.html [19a]), Schizosaccharomyces pombe (O74630), Sus scrofa (AAT01608), and Xenopus laevis (AAT72929).

Immunoblotting.

Cells equivalent to 1 ml of cells at an A₆₀₀ of 1.5 from liquid cultures were pelleted. Cells were resuspended in 150 μl of 1.85 M NaOH and 7.4% 2-mercaptoethanol and incubated on ice for 10 min. One hundred fifty microliters of 50% trichloroacetic acid was added, and the resulting suspension was incubated on ice for 10 min. Precipitated proteins were pelleted by spinning at 20,800 × g for 2 min at 4°C. Pellets were washed with 1 ml of acetone at 4°C and then resuspended in SDS sample buffer. Samples were electrophoresed on SDS-polyacrylamide gels (6.5% or 7.5%). The Full-Range Rainbow molecular weight markers (GE Healthcare) were used to estimate protein molecular mass. Proteins were transferred to Hybond-P (GE Healthcare) by Western blotting. Western blots were blocked with 5% nonfat dried milk (5% NFDM) in TBST (Tris-buffered saline [TBS] plus 0.1% Tween 20 [Sigma]). Blots were probed with primary antibodies in 5% NFDM-TBST at a 1:1,000 dilution. Primary antibodies included mouse monoclonal anti-FLAG M2 (Sigma), mouse monoclonal anti-c-myc 9E10 (Covance), goat polyclonal Rad53 (yC-19) (Santa Cruz Biotechnology), and rabbit polyclonal anti-Tel1 2973, which was generated in a rabbit injected with Tel1(961-976) peptide (NH₂-GCQNHDLSHGSIRGGKQR-OH) (Bio-Synthesis) and affinity purified (Rockland Immunochemicals). Following two brief washes in TBST, the following secondary antibodies were added in TBST containing 5% NFDM: horseradish peroxidase (HRP)-conjugated sheep anti-mouse, HRP-conjugated donkey anti-rabbit, or HRP-conjugated donkey anti-goat (Jackson ImmunoResearch) antibodies. These antibodies were added at a dilution of 1:10,000 (1:20,000 for HRP-conjugated donkey anti-goat antibodies). Probed blots were washed three times for 10 min each time in TBST, incubated with the ECL Plus Western blotting detection system (GE Healthcare), and exposed to BioMax light film (Kodak).

DNA damage studies.

Cells were grown to an A₆₀₀ of 0.39 to 0.41. The cultures were split, MMS was added to a concentration of 0.01% to one culture for each strain, and cells were grown for 105 min at 30°C. One-milliliter aliquots of MMS-treated or untreated cells were washed three times with 1-ml portions of YEPD. Cell densities were normalized to 4 × 10⁶ cells/ml, as calculated using counts enumerated using a hemacytometer, five serial 1:5 dilutions were made, dilutions were plated on YEPD, and growth at 30°C was examined after approximately 2 days. Additional aliquots of MMS-treated or untreated cells were taken, equal numbers of cells of each strain were pelleted, and Rad53 phosphorylation was examined.

Cell fractionation.

Cells were spheroblasted essentially as described previously (21). Spheroblasts were pelleted at 1,500 × g for 5 min and resuspended in PVP lysis buffer, which consisted of 8% polyvinylpyrrolidone-40 (PVP-40), 11.5 mM KH₂PO₄, 8.3 mM K₂HPO₄, 7.5 mM MgCl₂, 0.025% Triton X-100, 5 mM dithiothreitol (pH 6.5), and complete mini, EDTA-free protease inhibitor cocktail tablet (Roche). Spheroblasts were lysed with 15 strokes in a 1-ml type A Dounce homogenizer. Unlysed cells were removed by spinning 500 × g for 1 min four times. Nuclei were pelleted at 1,000 × g for 10 min. The cytoplasmic fraction was removed, and the pellet was resuspended in an equal volume of PVP lysis buffer. Proteins in the whole-cell lysate, cytoplasmic, and nuclear fractions were precipitated with methanol and prepared for SDS-polyacrylamide gel electrophoresis as described previously (21). Samples were electrophoresed on a 6.5% SDS-polyacrylamide gel, subjected to Western transfer, probed for Tel1 (anti-FLAG M2; Sigma), stripped and reprobed for nuclear pore complex proteins (Mab414; Covance), and stripped and reprobed for glucose-6-phosphate dehydrogenase (G6PD) (HRP-conjugated anti-G6PD; Biodesign).

Tel1 and Xrs2 coimmunoprecipitation.

Cultures (125 ml) of strains 12182A, 12200, 12981A, and 12982A were grown in YEPD to an A₆₀₀ of approximately 0.6 and harvested. Cell pellets were transferred to 0.5-ml screw-cap tubes and frozen at −80°C. Cell pellets were thawed, 450 μl of lysis buffer (25 mM HEPES-KOH [pH 7.5], 10% glycerol, 150 mM KCl, 1 mM EDTA, 1 mM EGTA, 1 mM MgCl₂, 0.1% NP-40, 0.2 mM phenylmethylsulfonyl fluoride [PMSF], complete mini, EDTA-free protease inhibitor cocktail tablet [Roche]) was added, 0.5-mm glass beads (Biospec) were added to the top of the tube, and the mixtures were subjected to bead beating in a MiniBeadBeater (Biospec) at the maximum setting five times (30 s each time), with 1 min on ice between rounds of bead beating. Cell lysate was retrieved, and debris was pelleted at 4°C at 20,800 × g for 15 min and again for 5 min. The protein concentration of the samples was normalized to 11.3 mg/ml using protein assay dye reagent concentrate (Bio-Rad). Extracts were precleared using 100 μl of protein G Dynabeads (Invitrogen) equilibrated with lysis buffer. Anti-c-myc 9E10 antibody (Covance) was added to the precleared lysate at a dilution of 1:150 in 150 μl of lysate, and the mixture was rotated for 1 h at 4°C. The extract-antibody mixture was added to 25 μl of protein G Dynabeads equilibrated in lysis buffer and then rotated for 1 h at 4°C. The beads were washed three times with 250 μl of lysis buffer and resuspended in 2× SDS sample buffer, and the samples were subjected to electrophoresis on a 6.5% gel, transferred to Hybond-P, and probed essentially as described above to detect 3xFLAG-Tel1 and Xrs2-13MYC proteins.

Chromatin immunoprecipitation.

Growth of cultures, induction of HO endonuclease expression, ChIP, and quantitative real-time PCR were performed essentially as described previously (2, 36). The DNA DSB was induced essentially as described previously (2) with the following alterations: cells were grown to an A₆₀₀ of 0.3 before being split, G₁ arrest was initiated by adding α-factor to a final concentration of 10 μg/ml, and the ChIP lysis buffer used for washing cells before storage lacked PMSF. ChIP was performed essentially as described previously (2, 36) with the following changes: the lysis buffer lacked PMSF, and the extracts were sonicated seven times (17 pulses each time at power level 3, 50% duty) with a Branson 450 sonifier. Strains shown in Fig. 7A were grown in medium lacking uracil; strains shown in Fig. 7B were grown in medium lacking both uracil and histidine.

FIG. 7.

Deletion of the TAN motif reduces the association of Tel1 with DNA near a DNA double-strand break. Bars depict the average relative amount of DNA near a HO cleavage site that immunoprecipitated with anti-FLAG antibody in three independent experiments. The error bars depict the standard deviations. Strains contained untagged Tel1, 3xFLAG-Tel1, or 3xFLAG-Tel1-ΔN30, an HO cleavage site on chromosome VII, and a plasmid containing the HO endonuclease gene under the control of a galactose promoter. Cultures grown in raffinose were arrested with α-factor in the G₁ phase of the cell cycle. Galactose was added to induce expression of the HO endonuclease for 1 hour, and protein-DNA complexes were cross-linked (see Materials and Methods). Anti-FLAG antibody was used to immunoprecipitate cross-linked complexes; the cross-links were reversed, and quantitative PCR was used to calculate immunoprecipitated DNA. Values shown are normalized to input DNA, as well as to a control locus on a different chromosome (ARO1); the change in enrichment for each sample was calculated as [HO_IP/HO_input]/[ARO_IP/ARO_input]. (A) The value for cells with untagged Tel1 grown in raffinose was set at 1. A Western blot shows the relative levels of 3xFLAG-Tel1 and 3xFLAG-Tel1ΔN30 in asynchronous cells. The asterisk denotes a background band. (B) The value for full-length 3xFLAG-Tel1 cells grown in raffinose was set at 1. Each strain contained either an empty CEN plasmid or a CEN plasmid expressing 3xFLAG-Tel1ΔN30 under the control of the TEL1 promoter. A Western blot shows the expression levels of in cells collected just before cross-linking during the ChIP experiment; for a loading control, the same samples were run on a separate gel and probed for tubulin.

RESULTS AND DISCUSSION

Identification of a novel N-terminal motif in Tel1/ATM orthologs.

To aid in demarcating functional regions in the Tel1 N terminus, we aligned the sequences of 32 Tel1/ATM orthologs. Strikingly, this analysis revealed a novel motif within the first approximately 30 amino acids of Tel1/ATM proteins (Fig. 1A). The motif contains a highly conserved (L/V/I)XXX(R/K)XX(E/D)RXXX(L/V/I) signature (Fig. 1A), with arginine 21 of S. cerevisiae Tel1 being 100% conserved in proteins with this motif. The motif is found in Tel1/ATM orthologs from vertebrates, flies, yeasts, and plants. However, it is unclear whether the TAN motif is conserved in the ATM ortholog from Arabidopis thaliana (Fig. 1B), which has added N-terminal sequence relative to other Tel1/ATM orthologs (14, 38). Using PHI-BLAST searches, the consensus motif was not found to be conserved in the extreme N terminus of other PIKK subclasses (Mec1/ATR, Tor1/Tor2/mTOR, Tra1/TRAPP, DNA-PK, and hSMG-1). Consistent with this finding, attempts to align the first 50 amino acids of apparent Mec1/ATR orthologs from multiple species did not reveal conservation of a similar sequence motif (data not shown). Thus, this motif appears to be specific to the extreme N terminus of Tel1/ATM orthologs; we term this sequence TAN, for Tel1/ATM N-terminal motif.

FIG. 1.

Sequence alignment of the N termini of Tel1/ATM orthologs reveals the conserved TAN motif. (A) Alignment of amino acids 1 to 50 of Tel1/ATM orthologs. The alignment was made using ClustalW; conserved residues are indicated by shading. Phylum and species names are indicated (see Materials and Methods). The TAN motif is indicated by the bar at the top; a minimal consensus motif is depicted at the bottom. (B) Alignment of the ATM orthologs from the plants O. sativa (rice) and A. thaliana (thale cress). Amino acids 781 to 991 of the A. thaliana ATM ortholog (Q9M3G7) and amino acids 1 to 98 of the O. sativa ATM ortholog (NP_001041778) were aligned using ClustalW. The consensus TAN sequence and a bar over the TAN sequence for the O. sativa ATM ortholog are shown. The A. thaliana ATM ortholog has sequence that is conserved with several of the residues in the TAN motif of the O. sativa ATM ortholog.

The TAN motif is required for normal telomere length maintenance.

To determine whether the TAN motif plays a role in telomere length regulation, mutations were made in the extreme N terminus of 3xFLAG-Tel1 encoded by the endogenous TEL1 locus (Fig. 2A) (see Fig. S1 in the supplemental material). Deletion of amino acids 2 to 10 (ΔN10), which lie N terminal to the conserved TAN motif, did not result in significant telomere shortening relative to 3xFLAG-Tel1 strains (Fig. 2B, top panel). However, deletion of amino acids 2 to 20 (ΔN20), which includes part of the conserved TAN sequence, resulted in telomeres maintained at lengths shorter than in wild-type cells but longer than in tel1Δ cells (Fig. 2B, top panel), as did alanine substitutions of four highly conserved TAN residues (L13A K17A E20A R21A; 4A) (Fig. 2B, top panel). Strikingly, deletion of amino acids 2 to 30 (ΔN30) or 2 to 40 (ΔN40) was sufficient to shorten telomeres as much as deletion of the entire 2,787-amino-acid Tel1 protein did (Fig. 2B, top panel). In a mec1Δ sml1Δ background, the ΔN40 mutation also caused telomere shortening similar to that caused by complete deletion of TEL1 (Fig. 3). We conclude that deletion of the TAN eliminates the function of Tel1 in telomere length maintenance.

FIG. 2.

Mutations in the Tel1 TAN motif result in telomere shortening. (A) Diagram of Tel1 mutants shown in Fig. 2B to E. Amino acids 2 to 50 of Tel1 are shown. Conserved TAN residues L13, K17, E20, and R21 are underlined. Bars underline residues encoded by the mutants. A rectangle at the end of a bar represents the 3xFLAG tag (3xF). M represents an initiator methionine for untagged proteins, A represents an alanine substitution, and K represents a lysine substitution. (B) Deletion of the TAN motif results in telomere shortening. (Top) Southern blot of genomic DNA digested with BanI and probed for Y' telomeres. Two isolates with each genotype are shown. (Bottom) Western blot of cell extracts probed with anti-FLAG antibody. Asterisks denote background bands that serve as internal loading controls. (C) Overexpression does not rescue the telomere length defect of TAN-deleted Tel1. (Top) Southern blot of BanI-digested genomic DNA from streak 6 of strains passaged on galactose plates every 3 days. The blot was probed for Y' telomeres. Two isolates with each genotype are shown, except for the TEL1 strain. The positions (in kilobase pairs) of molecular size markersare shown to the left of the gel. (Bottom) Western blot of cell extracts probed with anti-Tel1 antibody (see Materials and Methods). The antibody did not recognize endogenously expressed levels of Tel1 in this assay. An asterisk denotes a background band that serves as an internal loading control. The positions (in kilodaltons) of molecular mass markers are shown to the left of the gel. (D and E) Point mutations in conserved TAN residues result in telomere shortening. Southern blots were probed for Y' telomeres. Two isolates with each genotype are shown.

FIG. 3.

The effect of deletion of the TAN motif was similar to that of deletion of full-length Tel1 in mec1Δ sml1Δ cells. Southern blots of genomic DNA digested with BanI and probed for Y' telomeres. The positions (in kilobase pairs) of molecular size markers are indicated to the left of the gels. Following dissection of tetrads, yeast were streaked every 2 days and grown at 30°C on YEPD. Genomic DNA was isolated following each streak (streak numbers are shown below the lanes). (A) mec1Δ sml1Δ TEL1 cells. (B) The mec1Δ sml1Δ tel1-4A cells had a slight growth defect relative to mec1Δ sml1Δ TEL1 cells (data not shown). (C and D) The mec1Δ sml1Δ tel1-ΔN40 (C) and mec1Δ sml1Δ tel1Δ (D) cells had a more severe growth defect (data not shown), and in both strains, the telomeres shortened during five restreaks and then equilibrated at the same short length, which overlaps with the 577-bp band. Interestingly, although progressive telomere shortening was similar to that reported previously in strains deficient in both Tel1 and Mec1 function (10, 31, 39), eventual senescence of mec1Δ sml1Δ tel1Δ and mec1Δ sml1Δ tel1-ΔN40 cells was not observed in this strain background.

Immunoblotting of cell extracts was performed to assay expression of the extreme N-terminal mutant Tel1 proteins. As shown in Fig. 2B, bottom panel, each of these mutant proteins was expressed. The levels of 3xFLAG-Tel1-4A and 3xFLAG-Tel1-ΔN40 proteins were readily detectable as lower than the level of 3xFLAG-Tel1 (Fig. 2B, bottom panel). To address the possibility that reduced protein expression contributed to the telomere shortening observed for these two mutants, untagged Tel1, Tel1-4A, and Tel1-ΔN40 were each overexpressed using an inducible galactose promoter. Strains were initially grown on glucose, conditions under which the reduced levels of Tel1 expression from the PGAL1 promoter resulted in the expected telomere shortening. Strains were then transferred to plates containing galactose and grown for six streaks. As expected, levels of galactose promoter-expressed Tel1, Tel1-4A, and Tel1-ΔN40 were significantly higher than for Tel1 expressed from its endogenous promoter, which was below the detection threshold with the Tel1 antibody used (Fig. 2C, bottom panel). Despite this overexpression, telomeres from PGAL1-tel1-ΔN40 cells grown on galactose were no longer than telomeres in tel1Δ cells, and similarly, telomeres from cells overexpressing Tel1-4A were only slightly longer than telomeres in tel1Δ cells and were much shorter than telomeres in wild-type cells (Fig. 2C, top panel). In contrast, telomeres in the control PGAL1-TEL1 cells became longer than wild-type telomeres (Fig. 2C, top panel), consistent with a previous report of the effects of TEL1 overexpression (32). These results demonstrate that although the Tel1 TAN motif may affect Tel1 protein level, this function can be uncoupled from its essential role in telomere length maintenance by Tel1.

To further dissect the significance of the TAN motif, a TAN deletion mutation and point mutations were made in Tel1 proteins encoded by the endogenous chromosomal locus without an N-terminal 3xFLAG tag. Like 3xFLAG-Tel1-ΔN40 (Fig. 2B), untagged Tel1-ΔN40 caused telomeres to be as short as in the control tel1Δ strains (Fig. 2D). Alanine substitution of conserved TAN amino acids demonstrated that residues L13, K17, and R21 all contribute to telomere length maintenance (Fig. 2D). Furthermore, the functional significance of arginine 21, which is absolutely conserved in Tel1/ATM TAN-containing orthologs (Fig. 1A), was investigated by substituting it with another basic residue, lysine. This alteration resulted in slight telomere shortening (Fig. 2E), indicating that the specific chemical properties of the arginine, and not solely its basic property, are important for Tel1 function. These mutants define the functional significance of individual TAN residues in telomere length maintenance.

To determine whether the TAN motif is the only functionally significant portion of the Tel1 N terminus in telomere length maintenance, we investigated the functions of other sequences N-terminal to the kinase domain of Tel1 (see Fig. S2A, S2B, and S2C in the supplemental material). A set of deletion mutants spanning the entire region of Tel1 N terminal to the kinase domain revealed that multiple regions are also essential for telomere maintenance (see Fig. S2A, S2B, and S2C in the supplemental material). However, because of the sequence conservation of the TAN motif, further experiments were directed toward understanding its role in other functions of Tel1.

The TAN motif is required for Tel1 function in response to DNA damage.

We next examined the role of the TAN motif in responding to DNA damage. It was previously shown that MMS-treated mec1Δ sml1Δ sae2Δ cells display Rad53 phosphorylation, whereas mec1Δ sml1Δ sae2Δ tel1Δ cells do not, indicating that Tel1 is required for Rad53 phosphorylation in this background (41). Therefore, to allow ready detection of a Tel1-mediated DNA damage response, we used a mec1Δ sml1Δ sae2Δ background (41). Strains were grown in liquid culture in the presence or absence of 0.01% MMS for 105 min, and aliquots of the cultures were washed and then plated on YEPD to assess viability. In the absence of MMS treatment, the tel1-4A mutation caused mec1Δ sml1Δ sae2Δ cells to exhibit a modest growth defect, and the tel1-ΔN40 or tel1Δ mutations in the mec1Δ sml1Δ sae2Δ background caused a more severe growth defect (Fig. 4A). In mec1Δ sml1Δ sae2Δ cells treated with MMS, the 4A and ΔN40 TAN mutations each caused DNA damage sensitivity to a degree similar to that caused by the complete deletion of TEL1 (Fig. 4A).

FIG. 4.

The TAN motif is required for the Tel1 DNA damage response. (A) Mutations in the Tel1 TAN motif reduce viability following exposure to MMS. All strains were in the mec1Δ sml1Δ sae2Δ background. Cultures were grown at 30°C for 105 min with 0.01% MMS (+) or without MMS (−), washed in YEPD, and then plated on YEPD and grown for 2 days at 30°C. (B) Mutations in the TAN motif reduce Rad53 phosphorylation following exposure to MMS. The Western blot of cell extracts from cells in panel A was probed for Rad53. Reduced mobility of Rad53 indicates phosphorylation (Rad53-P).

The loss of viability caused by TAN mutation upon DNA damage could have been due to a defect in the sensing of DNA damage or its signal transduction; alternatively, DNA damage signaling could have been intact, but later steps in the repair reaction may have been impaired. To distinguish between these possibilities, aliquots of the MMS-treated or untreated cultures described above were examined for a marker of DNA damage signaling, Rad53 phosphorylation. Phosphorylation of Rad53 is indicated by its reduced mobility during electrophoresis, as observed for MMS-treated mec1Δ sml1Δ sae2Δ cells (Fig. 4B). As with the mec1Δ sml1Δ sae2Δ tel1Δ cells, no Rad53 mobility shift was detectable in the mec1Δ sml1Δ sae2Δ 4A or mec1Δ sml1Δ sae2Δ ΔN40 strains (Fig. 4B). Together, this loss of viability and failure to phosphorylate Rad53 caused by Tel1 TAN mutations after MMS treatment demonstrate that the requirement for the TAN motif in the role of Tel1 in responding to DNA damage occurs at a step in the DNA damage signaling cascade prior to Rad53 phosphorylation.

The TAN motif of Tel1 is not essential for nuclear localization or Xrs2 association.

We tested the effects of deleting the TAN motif on several different properties of Tel1. First, we examined whether the TAN motif is required for the nuclear import of Tel1. The Tel1 TAN motif contains several basic residues (K17, K19, and R21) in a pattern somewhat reminiscent of the cluster of basic residues found in the simian virus 40 large T-antigen nuclear localization signal (NLS). However, as measured by cell fractionation, deletion of the TAN motif did not appear to interfere with the nuclear localization of Tel1 (Fig. 5). Thus, the Tel1 TAN motif does not contain the sole NLS for Tel1, and mutation of the TAN motif does not impair the ability of the NLS(s) (not identified so far) in Tel1 to function. This result is consistent with the previously reported finding that in vertebrate ATM, basic residues, which the present study shows lie within the TAN motif, were not required for nuclear import of an ATM fragment in monkey kidney cells (43).

FIG. 5.

Deletion of the TAN motif does not alter the nuclear localization of Tel1. Western blot of whole-cell extracts (W), cytoplasmic fractions (C), and nuclear fractions (N) of strains containing 3xFLAG-Tel1 or 3xFLAG-Tel1-ΔN30. Fractionation was achieved as described in Materials and Methods. The same blot was probed consecutively for 3xFLAG-Tel1 proteins, a nuclear pore complex protein (NPC), and glucose-6-phosphate dehydrogenase (G6PD).

We have recently reported that the interaction of Tel1 with the protein Tel2, which is conserved in eukaryotes and acts in the Tel1 DNA damage response pathway in yeast, was unaffected by deletion of a large N-terminal region of Tel1 that included the TAN motif (2). Hence, the TAN motif is not critical for the Tel1-Tel2 interaction. We next tested whether the TAN motif is required for the interaction of Tel1 with Xrs2. When Xrs2 is mutated to prevent this association, there is also a severe reduction in Tel1 localization to a double-strand break (29) and to short telomeres (32). The interaction site for Xrs2 on budding yeast Tel1 has not been mapped, although the interaction of S. pombe Nbs1 maps to internal Tel1 sequences (42). Also, the interaction of mammalian NBS1 maps to internal ATM sequences, and the first 247 amino acids of ATM did not associate with NBS1 (11). In coimmunoprecipitation experiments, 3xFLAG-Tel1-ΔN30 associated as efficiently with Xrs2 as 3xFLAG-Tel1 did (Fig. 6). Although we cannot discern if the TAN mutation affected binding affinity, this result indicates that the TAN motif is not required for Tel1 to associate with Xrs2 in vivo and, furthermore, suggests that mutation of the TAN motif does not severely interfere with the structure of regions in Tel1 that interact with Xrs2.

FIG. 6.

The TAN motif is not required for association with Xrs2. (Top) Immunoblot of cell extracts used in coimmunoprecipitation experiments probed with anti-FLAG antibody. Asterisks denote background bands that serve as internal loading controls. FL, full-length Tel1. (Middle) Immunoblot of coimmunoprecipitation with anti-MYC. Blot was probed with anti-FLAG antibody. In different strains in the presence (+) of Xrs2-13MYC, the anti-MYC antibody pulled down 3xFLAG-Tel1 and 3xFLAG-Tel1-ΔN30. (Bottom) The blot in the middle panel was stripped and reprobed with anti-MYC for Xrs2-13MYC. The position of full-length Xrs2-13MYC is indicated to the right of the gel. The positions (in kilodaltons) of molecular mass markers in all three panels are indicated to the left of the gels.

We next tested whether deletion of the TAN motif impacts the ability of Tel1 to localize to a DSB. As for the Xrs2 interaction experiment, strains expressing untagged Tel1, 3xFLAG-Tel1, or 3xFLAG-Tel1-ΔN30 were used. Expression of 3xFLAG-Tel1-ΔN30 in asynchronous cells was estimated to be approximately 50% the level of 3xFLAG-Tel1 expression (Fig. 7A). Each strain had an HO endonuclease cleavage site cassette (13) located on chromosome VII, and the HO cleavage site in the MATa locus was mutated (2). These strains also harbored a plasmid expressing the HO endonuclease under the control of a galactose-inducible promoter. Cells were held in the G₁ phase of the cell cycle by treatment with α-factor, and a DSB was induced by the addition of galactose for 1 hour. Protein and DNA were cross-linked, and anti-FLAG antibody was used to immunoprecipitate 3xFLAG-Tel1 or 3xFLAG-Tel1-ΔN30. After reversing the protein-DNA cross-links, quantitative PCR using primers adjacent to the HO cleavage site was used to measure DNA near the HO cleavage site that had associated with Tel1 (2). As expected, when 3xFLAG-Tel1 was immunoprecipitated, there was a large (∼28-fold) increase in DNA retrieved from near the HO cleavage site relative to untagged controls (Fig. 7A). In contrast, DNA near the HO cleavage site was enriched only threefold when the deletion mutant 3xFLAG-Tel1-ΔN30 was pulled down (Fig. 7A). Because the level of 3xFLAG-Tel1-ΔN30 protein was lower than that of the full-length protein, we therefore tested the effect of boosting the level of Tel1 by expressing an extra copy of 3xFLAG-tel1-ΔN30 from a CEN plasmid. As shown in Fig. 7B, this expression system rendered the mutant protein at least as abundant as full-length 3xFLAG-Tel1. In this set of experiments, immunoprecipitation of full-length 3xFLAG-Tel1 resulted in a 68-fold enrichment of DNA near the break site (relative to the signal from control cells with no break induced), and this was reduced to 4.5-fold when 3xFLAG-Tel1-ΔN30 expressed from a single gene was pulled down. Interestingly, immunoprecipitation of 3xFLAG-Tel1-ΔN30 from the strain harboring the second copy of 3xFLAG-tel1-ΔN30 still resulted in only an 11.8-fold enrichment of DNA near the HO cleavage (Fig. 7B). This result indicates that the TAN motif has a critical role in efficient recruitment and/or retention of Tel1 near a DSB.

In summary, we identified a novel, evolutionarily conserved motif we term TAN in the N terminus of Tel1/ATM orthologs (Fig. 1A). The TAN motif is notable, given the overall low sequence homology among Tel1/ATM members in the N-terminal region and because the TAN motif appears to be specific to Tel1/ATM proteins. Functional analysis showed a critical role for the yeast Tel1 TAN both in telomere length maintenance (Fig. 2B, C, D, and E) and in responding to DNA damage (Fig. 4A and B). TAN mutations caused reduced protein levels, but overexpression of Tel1 TAN mutants did not rescue the telomere length defect (Fig. 2C).

The reduced localization of TAN-mutated Tel1 to DSBs we observed is likely to lead to the observed failure to initiate signaling through phosphorylation of Rad53 in MMS-treated TAN-mutated mec1Δ sml1Δ sae2Δ cells (Fig. 4B). The telomere shortening observed in Tel1-TAN mutants (Fig. 2B, C, D, and E) may also be due to a similar reduction in the ability of Tel1 to localize to short telomeres. How does deletion of the TAN motif lead to impaired localization of Tel1 to a DSB? Our experimental results suggest that the defect is not a result of reduced Tel1 expression levels, loss of Tel1 nuclear localization, or disruption of the Tel1-Tel2 or Tel1-Xrs2 interactions (2). Therefore, the TAN motif appears to regulate Tel1 localization to DSBs through a novel mechanism. The evolutionary conservation of the TAN motif indicates that it is also likely to be important for the function of ATM from other eukaryotes, including mammals.