The structure of the plasmids used in this study. The pNCT-tel plasmid is located adjacent to the telomere on chromosome 16p in clone B3 of the EJ-30 tumor cell line, and was previously used to demonstrate the sensitivity of subtelomeric regions to DSBs. The pNCT-tel plasmid contains a β-lactamase gene for resistance to ampicillin and bacterial origin of replication (Amp/ori), a Neo gene for resistance to G418, and a gene for Herpes simplex virus thymidine kinase (HSV-tk). The pEJ5-GFP-tel and pDR-GFP-tel plasmids were generated from the pEJ5-GFP and pDR-GFP plasmids by insertion of telomeric repeat sequences. The telomeric repeat sequences were added to seed the formation of a new telomere following targeted integration through the shared homology in the Amp/ori sequences. The pDR-GFP-tel plasmid was used to monitor the frequency of HRR and contains a GFP gene that is defective due to an I-SceI site in the coding sequence, a puro gene for selection with puromycin, and a complimentary fragment of the GFP gene for repair of the I-SceI-induced DSB. The pEJ5-GFP-tel plasmid was used to monitor the frequency of NHEJ and contains a puro gene flanked by I-SceI sites, which is located between the GFP coding sequence and promoter. NHEJ between the two I-SceI-induced DSBs results in activation of the GFP gene. The pGFP-ISceI-tel plasmid was used to determine the frequency of large deletions (>50 bp) and GCRs. The cell clones containing the pGFP-ISceI-tel plasmid were generated from the clones containing the pEJ5-GFP-tel plasmid by I-SceI-mediated deletion of the puro gene, and contain an active GFP gene with an I-SceI site located between the coding sequence and promoter.

near telomeres and interstitial sites containing telomeric repeat sequences. The percentage of GFP-positive cells with (+) or without (-) constitutive expression of I-SceI endonuclease was used to monitor large deletions and GCRs in cell clones containing the pGFP-ISceI plasmid adjacent to a telomere (6D1 and 6D6), at an interstitial site (7F1 and 7F3), and an interstitial site containing telomeric repeat sequences (7C1 and 7C2). Experiments were performed in triplicate and standard deviations are shown.

No difference in small deletions due to DSBs at interstitial and subtelomeric sites. (A) Cell clones with the pEJ5-GFP plasmid integrated at telomeric (EJ5-6D, EJ5-6F, EJ5-6J) or interstitial (EJ5-7F, EJ5-AB, EJ5-AE) sites were used to determine the percentage of cells in the population that contain small deletions resulting from I-SceI-induced DSBs. Genomic DNA from pooled populations of cells infected with the pQCXIH-ISceI retrovirus and selected with hygromycin for 10 days was amplified by PCR using oligonucleotide primers that span one of the I-SceI sites in pEJ5-GFP (small black arrows). (B) The PCR products were then digested with I-SceI endonuclease, and the percentage of cut and uncut DNA was determined from the intensity of the bands using Image J. (C) The percentage of cells in the population containing small deletions was then determined by correcting the fraction of uncut DNA for the number of cells in the population that contained large deletions or GCRs (see Fig. 1), since the DNA from these cells would not amplify and would therefore cause an overestimation of the percentage of cells containing small deletions (see Materials and Methods). Experiments were performed in triplicate and standard deviations are shown.

Decreased NHEJ in repair of DSBs near telomeres and interstitial sites containing telomeric repeat sequences. The appearance of GFP-positive cells following infection with the pQCXIH-ISceI retrovirus and selection with hygromycin for 10 days was used to monitor the frequency of NHEJ in cell clones containing the pEJ5-GFP plasmid at telomeric sites (EJ5-6D, EJ5-6F, EJ5-6J), interstitial sites (EJ5-7F, EJ5-AB, EJ5-AD, EJ5-AE, EJ5-AI, EJ5-AL) or interstitial sites containing telomeric repeat sequences (EJ5-6B, EJ5-7C, EJ5-7G, EJ5-AC, EJ5-AF, EJ5-AJ). Experiments were performed in triplicate and standard deviations are shown.

No difference in HRR in repair of DSBs at interstitial and subtelomeric sites. The appearance of GFP-positive cells following infection with the pQCXIH-ISceI retrovirus and selection with hygromycin for 10 days was used to monitor the frequency of HRR in cell clones containing the pDR-GFP plasmid at telomeric (DR-1, DR-2, DR-7) or interstitial (DR-3, DR-4, DR-8) sites. Experiments were performed in triplicate and standard deviations are shown.

Increased loss of the I-SceI site during NHEJ of DSBs near telomeres. The frequency of loss of the I-SceI site during NHEJ between the two I-SceI sites in the pEJ5-GFP plasmid was compared for cell clones containing interstitial and telomeric integration sites. (A) The pEJ5-GFP plasmid integrated at telomeric (EJ5-6D, EJ5-6F, EJ5-6J) or interstitial (EJ5-7F, EJ5-AB, EJ5-AE) sites were used to determine the relative percentage of loss and restoration of the I-SceI site. Genomic DNA from pooled populations of cells infected with the pQCXIH-ISceI retrovirus and selected with hygromycin for 10 days was amplified by PCR using oligonucleotide primers distal to the two I-SceI sites in pEJ5-GFP (small black arrows). (B) The PCR products were then digested with I-SceI endonuclease. (C) The percentage of cut and uncut DNA was determined from the intensity of the bands using Image J. Experiments were performed in triplicate and standard deviations are shown.

A model to explain the deficiency in C-NHEJ near telomeres. (A) The processing of the leading strand of the telomere following DNA replication is regulated by the telomeric protein TRF2, which inhibits the 3′-5′ nuclease activity of MRE11, and stimulates the 5′-3′ nuclease activity of MRE11. Further resection is prevented by TPP1 and POT1, which bind to the single-stranded telomeric repeat sequences and promote T-loop formation. (B) The processing of DSBs near telomeres is also regulated by TRF2 and involves the resection of the 5′ strand by MRE11 to generate the 3′ single strand overhang. However, because the DNA is not composed of telomeric repeat sequence, TPP1 and POT1 cannot bind and prevent further resection, resulting in large single stranded regions, as are found in cells deficient in TPP1 and POT1 [62]. Further resection then occurs through EXO1 and DNA2, similar to HRR. However, cells that have not undergone DNA replication, and therefore do not contain a template for HRR, cannot repair the break cannot repair the break and experience large deletions, telomere loss, and GCRs.