An ordered assembly of MYH glycosylase, SIRT6 protein deacetylase, and Rad9-Rad1-Hus1 checkpoint clamp at oxidatively damaged telomeres

MYH, SIRT6, Hus1, and Rad9 are associated with oxidatively damaged telomeric chromatin

Although MYH, SIRT6, and 9-1-1 are all telomeric associated proteins [22, 24, 31, 32, 40], their responses to DNA damage are not clear. To investigate how MYH, 9-1-1, and SIRT6 respond to oxidatively damaged telomeres, we employed our newly developed inducible-ROS systems to confine DNA damage to telomeric chromatin [52]. In this method, local oxidative DNA damage within telomeric DNA is induced by activating KillerRed (KR) protein fused to telomeric repeat binding protein TRF1. Upon visible light illumination (550–580 nm), the photosensitizer KillerRed releases superoxide which may produce clustered oxidative base lesions leading to the production of single-stranded and double-stranded DNA breaks [52, 57]. Thus, after light induction, KR-TRF1 fusion protein bound to the telomeric TTAGGG repeat sequence can induce localized DNA damage. This novel approach allows us to study, on fine scale, exactly how the BER complex is assembled at lesion sites in live mammalian cells.

We analyzed ectopically expressed GFP-tagged MYH, SIRT6, and Hus1 in MEF cells expressing either KR-TRF1 or a non-phototoxic red fluorescent protein (DsRed)-tagged TRF1 (DsRed-TRF1). First, we examined whether GFP-tagged proteins were biologically active. GFP-tagged hMYH and hSIRT6 proteins were active because they could reduce 8-oxoG levels in the respective knockout cells following peroxide treatment (see Supplementary Figures 1A and 1B). Expression of GFP-hHus1 protein could reduce the sensitivity of hus1 KO cells to hydroxyurea (a DNA replication blocker) (Supplementary Figure 1C). In addition, MYH, sirt6, and hus1 KO cells expressing GFP-MYH, GFP-SIRT6, GFP-Hus1, respectively, contained fewer apoptotic cells than vector-transfected KO cells (Supplementary Figure 1D–1F). This is the first demonstration that SIRT6 is important in preventing 8-oxoG accumulation and apoptosis following oxidative stress. In cells expressing DsRed-TRF1, both GFP-MYH and GFP-SIRT6 appeared granulated in faint spots throughout the nucleoplasm of MEF cells (Figure 1A and 1D). Interestingly, about 15% of granulated GFP-MYH and GFP-SIRT6 spots were localized to DsRed-TRF1 at telomeres, as seen in the merged images of 20 cells from each group (Figure 1C and 1F). In undamaged cells, GFP-tagged Hus1 was localized mainly in cytoplasm and only about 3% of Hus1 foci colocalized with telomeres (Figure 1G and 1I). In cells transfected with KR-TRF1 without light induction (Supplementary Figure 2), the response and distribution of MYH, SIRT6, and Hus1 were essentially the same as cells with DsRed-TRF1 (Figure 1). Before light activation of KR, 11-14% of GFP-MYH, GFP-SIRT6, or endogenous mMyh granules were localized to telomeres (Supplementary Figure 2A, 2B, 2E, and 2G), while only about 3-7% of GFP-Hus1, FLAG-Rad9, or endogenous mHus1 granules were colocalized with telomeres (Supplementary Figure 2C, 2D, 2F, and 2G).

Figure 1. Expressed MYH, SIRT6, Hus1, and Rad9 are recruited to oxidatively damaged telomeric sites in mouse embryonic fibroblast (MEF) cells. (A), (D), (G), and (J), Distribution of GFP-hMYH, GFP-hSIRT6, GFP-hHus1, and FLAG-Rad9, respectively, in undamaged MEF cells containing DsRed-TRF1 or RFP-TRF1. Some GFP-hMYH and GFP-hSIRT6 granules are localized to DsRed-TRF1 at telomeres as shown in enlarged merged images. (B), (E), (H), and (K), GFP-hMYH, GFP-hSIRT6, GFP-hHus1, and FLAG-Rad9, respectively, form foci at KR-TRF1 damaged telomeric sites after light activation. Images were captured 30 min after light activation with an Olympus FV1000 confocal microscopy system. (C), (F), (I), and (L), Quantitative analyses of 20 cells in each undamaged and KR-induced damaged group. About 15% of granulated GFP-MYH and GFP-SIRT6 spots were localized to DsRed-TRF1 at telomeres (C, F). Over 50% of GFP-MYH, GFP-SIRT6, or GFP-Hus1 foci are colocalized with KR-TRF1 after KR activation. Error bars indicate SD; n ≥ 20. The P-value is calculated by Student’s t-test using Stat Plus software; P < 0.01 is shown as ***. (M), Enlarged merged images from (H) showing segregation of green GFP-hHus1 and red KR-TRF1 foci. (N), Enlarged merged images from (K) showing segregation of green FLAG-hRad9 and red KR-TRF1 foci.

We then performed the response kinetics of GFP-proteins to DNA damage at telomeres. As shown in Supplementary Figure 3, 30 min was the maximum time point for the damage response to occur. After ROS induction by activating the KR protein, GFP-tagged MYH, SIRT6, and Hus1 all formed discrete nuclear foci which colocalized with telomeres (Figure 1B, 1E, and 1H, yellow foci in the merged images). Because MYH removes adenine misinserted opposite 8-oxoguanine following DNA replication, we reason that some unsynchronized MEF cell populations are at G1- or S-phase. Therefore, some cells analyzed at 30 min after KR-induction are at S-phase and are replicating their DNA. In addition, MYH also binds to cytosine paired with 8-oxoG with high affinity [58, 59]. In this case, MYH can form foci at telomeres at any cell cycle stage. By analyzing 20 cells from each group, over 50% of GFP-MYH, GFP-SIRT6, or GFP-Hus1 foci exhibited colocalization with KR-TRF1 after KR activation (Figures 1C, 1F, and 1I). The colocalization of GFP-MYH, GFP-SIRT6 or GFP-Hus1 foci at sites of telomeres after light activation of KR was significantly increased compared to before light activation (Supplementary Figure 2). Using immunofluorescence staining, we also showed that FLAG-tagged Rad9 (one subunit of 9-1-1) acted similarly to Hus1 in the damage response (Figure 1J, 1K and 1L). However, in undamaged cells, about 8% of Rad9 foci colocalized with telomeres (Figure 1J and 1L). The FLAG-Rad9 protein has been shown to be active by rescuing the homologous recombination defect caused by Rad9 knockdown [60]. Interestingly, about 40% of Hus1 and Rad9 foci at telomeres expanded unidirectionally from KR-TRF1 sites as seen in the enlarged images (Figure 1M and 1N). Since nuclear Hus1 is likely to form a complex with Rad9, and Rad1 [61], and is the partner of MYH and SIRT6 [40, 55], Hus1 staining is used to represent the entire 9-1-1 complex in the nucleus.

We have shown that the endogenous SIRT6 and tagged SIRT6 proteins are recruited similarly to telomeres [39]. To confirm that endogenous mMyh and mHus1 proteins behave in a similar manner as tagged proteins in live cells, we proceeded with immunofluorescence staining to determine their response to DNA damage at telomeres. As shown in Supplementary Figure 4, the endogenous mMyh and mHus1 behaved in a similar manner as GFP-tagged proteins in regard to their colocalization with telomeres. About 20% of granulated mMyh spots were localized to RFP-TRF1 at telomeres, as seen in the merged images of 20 cells from each group (Supplementary Figure 4C). In undamaged cells, only 7% of mHus1 foci colocalized with telomeres (Supplementary Figure 4D and 4F). After ROS induction by activating the KR protein, mMyh and mHus1 formed discrete nuclear foci (Supplementary Figure 4B and 4E, yellow foci in the merged images). By analyzing 20 cells from each group, over 80% of mMyh or mHus1 foci showed colocalization with KR-TRF1 after KR activation (Supplementary Figure 4C and 4F). In summary, our data suggest that MYH, SIRT6, and 9-1-1 are recruited to telomeric chromatin to repair oxidative DNA damages.

MYH foci induced at oxidatively damaged telomeres are dependent on Hus1 while Hus1 foci formation at damaged telomeres is independent of MYH

To examine the mutual dependence of MYH and 9-1-1 association at DNA damage sites, we used the KR-TRF1 system in conjunction with cell knockout approaches. We transfected both CT2 (hus1^+/+) and CT7 (hus1^-/-p21^-/-) MEF cells with both KR-TRF1 (or DsRed-TRF1 as controls) and GFP-hMYH. As observed in Figure 2A, undamaged CT2 and CT7 cells contained granulated GFP-MYH spots and some of spots were localized with telomeres (Figure 2A and 2C). After ROS induction by activating the KR protein, GFP-hMYH formed discrete nuclear foci, of which 50% colocalized with KR-TRF1 marked telomeres in CT2 cells (Figure 2B and 2E). In contrast, GFP-MYH did not form foci in CT7 cells. Thus, the formation of MYH foci at damaged telomeres is dependent on Hus1.

Figure 2. The formation of MYH foci induced at oxidatively damaged telomeres is dependent on Hus1. (A and C) GFP-MYH does not form foci at sites with DsRed-TRF1 in undamaged CT2 (hus1^+/+) and CT7 (hus1^-/-p21^-/-) MEF cells, respectively. These cells contain granulated GFP-MYH spots and some of spots were localized with telomeres. (B and D) Damage response of GFP-MYH to the sites of KR-TRF1 after light activation in CT2 and CT7 MEF cells, respectively. GFP-MYH foci are not found at the sites of KR-TRF1 in CT7 cells. (E), Analyses of about 20 cells in each (A–D) group indicated that approximately 50% of GFP-MYH foci colocalized with telomeres in CT2 cells, in contrast, less than 5% of GFP-MYH foci colocalized with telomeres in CT7 cells. (F–J), GFP-Hus1 foci formation at telomeres in control HEK-293T and MYH KO HEK-293T human cells. Experiments were performed similarly to (A–E) except using GFP-hHus1 and different cells. After ROS induction by activating the KR protein, over 80% of GFP-Hus1 foci are colocalized with KR-TRF1 in both HEK-293T and MYH KO HEK-293T cells (J). ND indicates no difference.

Next, we examined GFP-Hus1 foci formation at telomeres in KR-TRF1-transfected MYH KO cells. For this purpose, the hMYH gene in human HEK-293T cells was knocked out by CRISPR-Cas9 method [62]. After transfection, with MYH crRNA 2 plasmid colonies were selected with 1.25 μg/ml puromycin and single colonies were screened, expanded, and confirmed by DNA sequencing and Western blot analysis. MYH KO clone 7a did not contain hMYH proteins in Western blotting (Supplementary Figure 5A). PCR and DNA sequencing analyses indicated that one allele of the hMYH gene contained an adenine insertion and the other allele contained a 229-bp deletion in MYH KO clone 7a. The majority of Hus1 were in cytoplasm (Figure 2F and 2H) and very few granulated GFP-Hus1 spots were colocalized with telomeres (Figure 2J) in undamaged control HEK-293T and MYH KO cells. After ROS induction by activating the KR protein, GFP-Hus1 formed discrete nuclear foci of which about 90% colocalized with KR-TRF1 marked telomeres in both HEK-293T and MYH KO HEK-293T cells (Figure 2G, 2I, and 2J). Thus, Hus1 foci formation at damaged telomeres is independent of MYH.

MYH foci induced at oxidatively damaged telomeres are dependent on SIRT6 while SIRT6 foci formation at damaged telomeres is independent of MYH

Next, we compared MYH response to KR-TRF1 induced telomeric damage in wild-type (WT) and sirt6 KO (sirt6^-/-) MEF cells. Undamaged WT and sirt6^-/- MEF cells contained GFP-MYH granulated spots, some of which were localized to telomeres (Figure 3A and 3C). After ROS induction by activating the KR protein, about 70% of GFP-hMYH foci colocalized with telomeres in WT MEF cells (Figure 3B and 3E). Importantly, GFP-MYH foci formation was substantially reduced in damaged sirt6^-/- cells (Figure 3D and 3E).

Figure 3. The formation of MYH foci induced at oxidatively damaged telomeres is dependent on SIRT6. (A and C), GFP-MYH does not form foci at sites with DsRed-TRF1 in undamaged control MEF and sirt6 KO MEF cells, respectively. (B and D), Damage response of GFP-MYH to the sites of KR-TRF1 after light activation in control and sirt6 KO MEF cells, respectively. GFP-MYH foci were not found at the sites of KR-TRF1 in sirt6 KO cells. (E), Analyses of about 20 cells in each (A–D) group indicated that approximately 70% of GFP-MYH foci colocalized with telomeres in control cells, in contrast, less than 10% of GFP-MYH foci colocalized with telomeres in sirt6 KO cells. (F–J), GFP-hSIRT6 foci formation at telomeres in control HEK-293T and MYH KO HEK-293T human cells. Experiments were performed similarly to (A–E) except using GFP-hSIRT6 and different cells. After ROS induction by activating the KR protein, about 70% of GFP-SIRT6 foci are colocalized with KR-TRF1 in both HEK-293T and MYH KO cells (J).

We then examined SIRT6 foci formation at KR-TRF1-induced damaged telomeres in HEK-293T and MYH KO HEK-293T cells. There were some granulated GFP-SIRT6 spots in both undamaged cells (Figure 3F and 3H). It appeared that more granulated GFP-SIRT6 spots were in undamaged MYH KO cells than control HEK-293T cells, although this was not statistically significant (Figure 3J). After ROS induction by activating the KR protein, GFP-SIRT6 formed discrete nuclear foci of which about 70% colocalized with telomeres in both HEK-293T and MYH KO cells (Figure 3G, 3I, and 3J). In summary, MYH foci formation at damaged telomeres is dependent on SIRT6, but SIRT6 recruitment to damaged telomeres is independent of MYH.

The recruitment of Hus1 to damaged telomeres is partially dependent on SIRT6 while SIRT6 recruitment to damaged telomeres is independent of Hus1

We also determined the order of Hus1 and SIRT6 association at damaged telomeres. As expected, about 80% of GFP-Hus1 foci colocalized with telomeres in WT MEF cells after ROS induction (Figure 4B and 4E). In undamaged sirt6^-/- MEF cells, Hus1 was present in both the cytoplasm and nucleus (Figure 4C). GFP-Hus1 did form foci at damaged telomeres in damaged sirt6^-/- MEF cells, however, only about 30% of Hus1 foci colocalized with KR-TRF1 marked telomeres (Figure 4D and 4E, yellow foci in the merged image) and some telomeres did not contain Hus1 foci (Figure 4D, red foci in the merged image). Therefore, Hus1 foci formation at damaged telomeres is partially dependent on SIRT6. We then examined SIRT6 foci formation at KR-TRF1-induced damaged telomeres in CT2 (hus1^+/+) and CT7 (hus1^-/-) cells. The responses of SIRT6 to damaged telomeres were similar in both cell lines (Figure 4F–4I). As shown in Figure 4J, about 60% of GFP-SIRT6 foci colocalized with telomeres in both WT and hus1 KO MEF cells after ROS induction. Thus, SIRT6 foci formation at damaged telomeres is independent of Hus1.

Figure 4. The formation of Hus1 foci induced at oxidatively damaged telomeres is partially dependent on SIRT6. (A and C), GFP-Hus1 does not form foci at sites with DsRed-TRF1 in undamaged control MEF and sirt6 KO MEF cells, respectively. (B and D), Damage response of GFP-Hus1 to the sites of KR-TRF1 after light activation in control and sirt6 KO MEF cells, respectively. GFP-hHus1 foci are significantly reduced at the sites of KR-TRF1 in sirt6 KO cells. (E), Analyses of about 20 cells in each (A–D) group indicated that approximately 80% of GFP-Hus1 foci colocalized with telomeres in control cells, in contrast, only 30% of GFP-Hus1 foci colocalized with telomeres in sirt6 KO cells. (F–J), GFP-hSIRT6 foci formation at telomeres in control CT2 and Hus1 KO CT7 MEF cells. Experiments were performed similarly to (A–E) except using GFP-hSIRT6 and different cells. After ROS induction by activating the KR protein, about 60% of GFP-SIRT6 foci are colocalized with KR-TRF1 in both CT2 and CT7 cells (J).

SIRT6 enzymatic activities are not important for SIRT6 response but are necessary for the recruitment of MYH to damaged telomeres

It has been shown that both protein deacetylase and ADP-ribosylase activities of SIRT6 facilitate DNA repair [63]. Since SIRT6 is required for MYH recruitment to damaged telomeres, we examined whether its catalytic activities were necessary for this function. We tested whether SIRT6^H133Y mutation could affect the recruitment of MYH to damaged telomeres. SIRT6^H133Y mutant has been shown to lack both NAD⁺-dependent protein deacetylation and ADP-ribosylation [38, 41] while also having very weak defatty-acylase activity [64]. MEF sirt6 KO cells were transfected with vector or vector containing wild-type (WT) or H133Y mutant hSIRT6 gene along with KR-TRF1 plasmid. As shown in Figure 5B–5D, direct imaging showed that similar percentages of hSIRT6^WT and hSIRT6^H133Y were colocalized with damaged telomeres. This indicates that the catalytic activities of SIRT6 are not necessary for SIRT6 response to damaged sites. Because the transfection frequencies with three plasmids into MEF cells are very low, we selected to examine endogenous mMyh in MEF cells transfected with two plasmids. We first tried to detect endogenous mMyh by immunofluorescence staining with blue emission secondary antibody in MEF cells containing GFP-SIRT6 and KR-TRF1. However, we were unable to clearly define blue mMyh foci because their signal was obscure and also interfered by the green and red fluorescence. We modified the method by stripping all fluorescence by HCl treatment after KR activation. mMyh, GFP-SIRT6, and Myc-KR-TRF1 were then detected by MYH, GFP, and Myc antibodies which were reacted with green, blue, and red emission secondary antibodies, respectively. Cells containing blue-colored GFP-SIRT6 were selected for analyses of the colocalization of mMyh green foci and red-colored telomeres (Figure 5E–5G). Approximately 9% of mMyh foci were observed in damaged sirt6^-/- cells transfected with vector alone (Figure 5E and 5H). This result is similar to that with GFP-MYH in sirt6^-/- cells (Figure 3E). After ROS induction by activating the KR protein, approximately 60% of mMyh foci were colocalized with KR-TRF1 in MEF sirt6 KO cells expressing hSIRT6^WT protein (Figure 5F and 5H). Interestingly, the telomere colocalizations of mMyh was absent in sirt6 KO cells expressing hSIRT6^H133Y mutant (Figure 5G and 5H). This result indicates that the activities of SIRT6 are essential for MYH recruitment to damaged telomeres.

Figure 5. The recruitment of MYH at oxidatively damaged telomeres is dependent on the catalytic activities of SIRT6. MEF sirt6 KO cells were transfected with pEGFP-C1 vector or vector containing human wild-type (WT) or H133Y mutant hSIRT6 gene along with KR-TRF1 plasmid. (A–D), Both GFP-tagged hSIRT6^WT and hSIRT6^H133Y respond similarly to damaged telomeres. GFP-SIRT6 and Myc-tagged KR-TRF1 were detected by direct imaging. Images were captured 30 min after light activation with an Olympus FV1000 confocal microscopy system. (E–H), Response of endogenous mMyh to the damage sites in sirt6 KO cells expressing hSIRT6^WT or hSIRT6^H133Y mutant. Cell fluorescence was stripped by HCl treatment. GFP-SIRT6, MYH, and Myc-KR-TRF1 were them detected by immunofluorescence staining with GFP, MYH, and Myc antibodies (Ab), respectively, and reacted with blue, green, and red emission secondary antibodies, respectively. Cells containing blue-colored GFP-SIRT6 were selected for analyses to detect mMyh green foci and red-colored telomeres. (D) and (H), Quantitative analyses of 20 cells as in (A–C) and (E–F) groups, respectively. White, green, and black bars represent protein colocalization with telomeres in MEF sirt6 KO cells containing GFP alone, GFP-hSIRT6^WT, or GFP-hSIRT6^H133Y, respectively.

Mutations in the interdomain connector of MYH influence MYH association with damaged telomeres

The IDC of MYH has a unique architecture [56] that serves as a hub for interactions with Hus1, SIRT6, and APE1 [40, 55, 65]. V315 and Q324 of hMYH are important for interaction with Hus1 [55, 66] (red stars in Figure 6A). However, residue Q324 of hMYH is dispensable for interaction with SIRT6 [40]. To determine how these residues control MYH association with damaged telomeres, we expressed GFP-MYH^WT, GFP-MYH^V315A, or GFP-MYH^Q324H along with KR-TRF1 in CT2 MEF cells. The results (Figure 6B and 6C) demonstrated that the association of GFP-MYH^V315A (but not GFP-MYH^Q324H) with damaged telomeres was substantially attenuated. Taken together with the results, and the deficiency of MYH mutants in protein-protein interactions, SIRT6 is an essential partner for MYH association with damaged telomeres.

Figure 6. Human V315A mutant MYH protein expressed in mouse cells cannot associate with damaged telomeres and interrupting the MYH interactions with its partners can increase cell’s sensitivity to H₂O₂ and/or elevate cellular 8-oxoG levels. (A) Sequence alignment of the IDC regions of eukaryotic MYH proteins. V315 and Q324 of hMYH are important for interaction with Hus1 [55, 66] (red stars). However, residue Q324 of hMYH is dispensable for interacting with SIRT6 [40]. (B) and (C) GFP-MYH^WT, GFP-MYH^V315A, and GFP-MYH^Q324H along with KR-TRF1 were expressed in MEF cells to determine their association with damaged telomeres. The association of GFP-MYH^V315A, but not GFP-MYH^Q324H, with damaged telomeres is substantially attenuated. (D) HEK-293T cells transformed with pEGFP-N1 vector or vector with IDC sequences were treated with 700 mM H₂O₂ for 1 h and recovered in fresh medium for two days. Cell viability was measured as described in Materials and Methods. (E) HEK-293T cells transformed with pEGFP-N1 vector or vector with IDC sequences were treated with 700 mM H₂O₂ for 1 h and recovered in fresh medium for 1 h. 8-oxoG levels were measured. Each average value was obtained by subtracting the value of cells containing pEGFP-N1 vector without H₂O₂ treatment from H₂O₂-treated cells with listed plasmids. *, **, and *** represent P <0.1, P <0.05, and P < 0.01, respectively.

Human cells expressing GFP-IDC peptides have increased sensitivity to H₂O₂ and/or elevated 8-oxoG levels

The IDC of MYH interacts with Hus1, SIRT6, and APE1 [40, 55, 65]. We hypothesized that IDC overproduction might inhibit MYH interactions with its partners and diminish BER. We expressed GFP-tagged hMYH-IDC peptides containing WT, V315A mutant, and Q324H mutant sequences in HEK-293T cells. IDC-WT expression is expected to impede MYH interaction with Hus1, SIRT6, and APE1; IDC-V315A expression is expected to interfere with MYH interaction with APE1, but not as much with Hus1 and SIRT6; IDC-Q324H expression is expected to block MYH interaction with APE1 and SIRT6, but not with Hus1. After H₂O₂ treatment, cell viability and 8-oxoG levels were measured. As indicated in Figure 6D and 6E, cells expressing IDC-WT peptide had increased sensitivity to H₂O₂ and elevated 8-oxoG levels as compared to cells transfected with vector alone. Cells expressing IDC-V315A peptide had slightly increased sensitivity to H₂O₂, but the 8-oxoG levels were the same as cells transfected with vector. Cells expressing IDC-Q324H peptide had significantly increased sensitivity to H₂O₂, while the 8-oxoG levels were slightly increased when compared to cells transfected with vector alone, but this increase was statistically insignificant. Thus, interrupting the MYH interactions with its partners can increase cell’s sensitivity to H₂O₂ and/or elevate cellular 8-oxoG levels after H₂O₂ treatment.

Cell culture

HEK-293T cells [from American Type Culture Collection (ATCC)] were maintained in DMEM (Thermo Fisher Scientific) supplemented with 4.5 mg/ml D-glucose, 2 mM L-glutamine, 1 mM sodium pyruvate, and 10% fetal bovine serum. Sirt6^+/+ (wild-type, WT) and sirt6^-/- (KO) mouse embryonic fibroblast (MEF) cells (obtained from Dr. Raul Mostoslavsky at Harvard Medical School) were maintained in DMEM supplemented with 15% fetal bovine serum and 1% Penicillin-Streptomycin. CT2 (hus1^+/+) and CT7 (hus1^−/−p21^−/−) MEF cells (obtained from Dr. Robert Weiss at Cornell University) were maintained in media as described for Sirt6^+/+ and sirt6^-/- MEF cells.

The hMYH gene in HEK-293T cells was knocked out by CRISPR-Cas9 method [62]. gRNA (MUTYH CRISPR Guide RNA 2 or crRNA 2) with sequence ACTGTGATCAACTACTATAC (located in Exon 5 of hMYH gene) in PX459 plasmid [81] was purchased from GenScript. MYH crRNA 2 was transfected into HEK-293T cells followed by 1.25 μg/ml puromycin selection and single colonies were screened, expanded, and confirmed by DNA sequencing and Western Blot analysis.

Plasmids

Plasmids pEGFP-hMYH [40], pEGFP-hSIRT6 [40], and pEGFP-hSIRT6 encoding catalytically inactive SIRT6^H133Y mutant [39] have been described. pCMV KR-TRF1, pCMV DsRed-TRF1, and pCMV RFP-TRF1 plasmids have been described [52]. pEGFP-hMYH^V315A and pEGFP-hMYH^Q324H were generated by site-directed mutagenesis with primers listed in Supplementary Table 1. hHus1 were subcloned by PCR amplification using template pET21a-hHus1 [55] and primers listed in Supplementary Table 1. The PCR product of hHus1 was digested with BamHI and SalI and ligated into the BamHI-SalI digested pEGFP-C1 vector (Clonetech Laboratories). FLAG-hRad9 [60] was obtained from Dr. Mihoko Kai at Johns Hopkins University. GFP-tagged IDC peptides (residues 295-350) with wild-type, V315A, or Q324H sequences were cloned using plasmids containing the full length wild-type and mutant hMYH genes [40] as templates, utilizing the primers listed in the table, into pEGFP-N1 (Clonetech Laboratories). A nuclear localization signal sequence was added at the C-terminus of the IDC peptide.

Transfections

For confocal microscopy studies, plasmids were transfected with Lipofectamine2000 (Thermo Fisher Scientific) using a standard protocol. For cell viability and 8-oxoG assays, plasmids were transfected into mammalian cells using X’tremeGENE HP transfection reagents (Millipore-Sigma, 6366244001). Transiently transfected cells were used for all analyses.

Confocal microscopy and image quantification

Twenty four or 36 hours after transfection with plasmids, cells were exposed to 15-W SYLVANIA cool white fluorescent bulbs for 30 minutes in a stage UVP (Uvland) to induce DNA damage as described [52, 82]. Cells were cultured for various times after white light exposure. All cells were then fixed with 4% (vol/vol) formaldehyde for 15 min at room temperature (RT) and subsequently washed 3 times with PBS. The images were acquired using the Olympus FV1000 confocal microscopy system (Cat. F10PRDMYR-1, Olympus) and FV1000 software. For calculation of the percentage of GFP-fusion proteins colocalized with KR-TRF1, 20 cells were counted manually. To avoid the interference of background signal, foci with fold increase of mean intensity of the protein foci at telomere/non-telomere background larger than 1.3 are defined as positive foci. Three independent experiments were performed, and representative data are shown. Fluoview Soft (Olympus) was used for data analysis.

Immunofluorescence staining

After cells were transfected with plasmids and exposed to light and recovered as described above, they were fixed with 3.7% (v/v) formaldehyde for 15 min at RT, followed by three washes with PBS. Cell were then permeabilized with PBS containing 0.2% Triton X-100 for 5 min at RT and washed with PBS twice. After being blocked in PBS containing 15% fetal bovine serum for 15 min at 37°C, the cells were reacted with polyclonal antibodies against an hMYH peptide (α344) [83] or hHus1 monoclonal antibody (Novus, NBP1-89445) overnight at 4°C. Next, the cells were washed three times for 15 min each in PBS and incubated with Alexa Fluor® 594 goat anti-rabbit secondary antibodies (Invitrogen) diluted in DMEM + Azide for 30 min at RT. Cell samples were then mounted in drops of PermaFluor (Immunon). Image quantification was analyzed as described in section 2.4.

To analyze endogenous mMyh foci in sirt6 KO MEF cells expressing GFP-SIRT6 and Myc-tagged KR-TRF1, immunofluorescence staining procedures were modified as follows. The sirt6 KO MEF cells were transfected with pEGFP-C1 vector (Clonetech Laboratories) (or pEGFP-C1-SIRT6) and pCMV KR-TRF1. After fixing with formaldehyde, cells were incubated with 2.5M HCl for 1 min to strip the green and red fluorescence. Cells were then permeabilized with PBS containing 0.2% Triton X-100 for 10 min followed by incubating with 5% bovine serum albumin for one hour. Next, the cells were reacted with primary antibodies of Myc (mouse monoclonal, ab32, Abcam), MYH (Rabbit polyclonal, α344), GFP (chicken polyclonal, ab13970, Abcam) overnight. After being washed three times for 5 min each with PBS, the cells were incubated with secondary antibodies (goat anti-mouse, Fluor®594, ab150116; goat anti-rabbit, Fluor®488, ab150077; goat anti-chicken Fluor®405, ab175674) for one hour and then followed by three washes times for 5 min each with PBS.

Cell viability analysis

Cell viability was measured using the neutral red uptake assay as described [72]. Cells were seeded in 12-well flat bottom tissue culture plates. One day post-seeding, the cells were treated with 700 mM H₂O₂ for 1 h. After recovery in fresh medium for two days, the plates were incubated for 2 h in regular medium containing 40 μg/ml of neutral red (3-amino-7-dimethylamino-2-methyl-phenazine hydrochloride, Sigma). The cells were then washed with PBS twice; the dye was extracted from each well with acidified ethanol solution. Plates were incubated at room temperature with gentle shaking for 15 minutes and the absorbance at 540 nm was read in a Multiskan Spectrum microplate spectrometer (Thermo Scientific).

Quantification of 8-oxo-G

Cells were seeded on a 2-chamber culture slide, treated with 700 mM H₂O₂ for 1 h and recovered for 1 h in fresh media. Cells were fixed in 1:1 methanol:acetone for 20 min at −20°C followed by 0.05 N HCl treatment for 5 min. RNA was removed by 100 μg/ml RNase A in 150 mM NaCl and 15 mM sodium citrate for 1 h at 37°C and DNA was denatured in situ with 0.15 N NaOH in 70% ethanol for 4 min. Cells were then treated with 5 μg/ml proteinase K in 20 mM Tris–HCl (pH 7.5) and 1 mM EDTA for 10 min at 37°C and blocked with 5% normal goat serum in PBS at room temperature for 1 h. Cells were then incubated with anti-8-oxo-dG antibody (Trevigen, 4354-MC-050) in PBS containing 1% BSA at 4°C overnight. Next, the cells were washed three times for 5 min each in PBS and incubated with Alexa Fluro 592 donkey anti-mouse antibodies (Molecular Probes) at a 1:250 dilution in PBS containing 1% BSA for 1 h at room temperature at dark. The slides were then washed three times in PBS with 0.05% Tween-20. Nuclear DNA was counterstained with 5 μg/ml DAPI. Slides were mounted with cover slip using mounting medium (Leica micromount) and images were captured with DMi8b fluorescent microscope (Leica).