The involvement of human RECQL4 in DNA double-strand break repair

Abstract

Rothmund-Thomson syndrome (RTS) is an autosomal recessive hereditary disorder associated with mutation in RECQL4 gene, a member of the human RecQ helicases. The disease is characterized by genomic instability, skeletal abnormalities and predisposition to malignant tumors, especially osteosarcomas. The precise role of RECQL4 in cellular pathways is largely unknown; however, recent evidence suggests its involvement in multiple DNA metabolic pathways. This study investigates the roles of RECQL4 in DNA double-strand break (DSB) repair. The results show that RECQL4-deficient fibroblasts are moderately sensitive to gamma-irradiation and accumulate more gammaH2AX and 53BP1 foci than control fibroblasts. This is suggestive of defects in efficient repair of DSB's in the RECQL4-deficient fibroblasts. Real time imaging of live cells using laser confocal microscopy shows that RECQL4 is recruited early to laser-induced DSBs and remains for a shorter duration than WRN and BLM, indicating its distinct role in repair of DSBs. Endogenous RECQL4 also colocalizes with gammaH2AX at the site of DSBs. The RECQL4 domain responsible for its DNA damage localization has been mapped to the unique N-terminus domain between amino acids 363-492, which shares no homology to recruitment domains of WRN and BLM to the DSBs. Further, the recruitment of RECQL4 to laser-induced DNA damage is independent of functional WRN, BLM or ATM proteins. These results suggest distinct cellular dynamics for RECQL4 protein at the site of laser-induced DSB and that it might play important roles in efficient repair of DSB's.

Fig. 1. RECQL4-deficient fibroblasts are sensitive to γ-irradiation compared to normal control fibroblasts

A, Normal (GM00323) and primary RTS fibroblasts (AG05013) cell lines were plated on a 60 mm petri dishes and treated with 0, 1, 2 and 3 Gy of γ-irradiation. Cells were allowed to grow for 14 days and fluorescent intensity was measured as described in Materials and Methods. B, Similar experiments were performed with normal fibroblasts (GM00969) and RTS cell line (AG17524). The experiments were performed in triplicate and the error bars represents the standard deviation (+/⁻). The degree of significance were calculated by Student’s t-Test method and the (*) represent the significant difference from the normal fibroblast with p value < 0.05.

Fig. 2. The processing of DNA double strand breaks is delayed in RECQL4 defective primary fibroblasts

After 5 Gy of γ-irradiation, the cells were allowed to recover under physiological conditions, fixed at indicated time points (0, 1, 3, 6 and 10 hrs) and immunostained with anti rabbit 53BP1 antibodies (as described in materials and methods). The images were captured with confocal microscope. Panel A and B show the persistence of more 53BP1 foci in RECQL4-deficient primary fibroblasts AG05013 and AG18371 (lower rows) compared to normal control fibroblasts GM00323 and GM01864 (upper rows), respectively at different time intervals as indicated. The first column represents the untreated cells (0 hour) as control. The total number of 53BP1 foci’s in both the RTS and normal control cell lines for each time point were counted and the average number of foci/ cell with respect to time are represented in panel C and D for two different cell lines, respectively. The repair kinetics of 53BP1 foci in both RTS and normal cell lines in terms of percentage of 53BP1 foci at each time point are represented in panel E and F for two different cell lines, respectively. In both panels, the total number of 53BP1 focis at 1 hr is taken as 100% foci for both normal and RTS cell lines.

Fig. 3. The formation of DNA double strand breaks at the laser microirradiation site requires higher doses of laser irradiaiton

A. Panel (a) shows the association of GFP tagged XRCC1, 53BP1 and RECQL4 in live HeLa cells at low (8%) and high (21%) doses of laser intensity. Transfected HeLa cells were microirradiated within the nucleus of the cell and images were captured for 2 min after photo bleaching. Panel (b) represents the association of endogenous XRCC1, 53BP1 and RECQL4 proteins at laser induced DNA damage site irradiated with low (8%) and high (21%) doses of laser intensity. Live HeLa cells were microirradiated and fixed after 5 min and immunostained with either XRCC1, 53BP1 or RECQL4 antibodies (see materials and methods). Arrow indicates the exact site of laser DNA damage. B. The live HeLa cells were microirradiated with the laser at 18% and 8% laser intensities. At 18% laser intensity recruitment of both endogenous γ-H2AX and XRCC1 was observed whereas at 8% laser intensity only XRCC1 is recruited and not the γ-H2AX. The arrow indicates the site of exact DNA damage. Asterisks (*) in the lower panel represents the DNA damage at 18% laser intensity.

Fig. 4. The recruitment kinetics of GFP tagged WRN, BLM and RECQL4 proteins at the site of laser induced DNA damage

The HeLa cells were transfected with GFP tagged WRN, BLM or RECQL4 plasmids. After 24 hr post transfections, the cells were laser microirradiated within the nucleus with 435 nm laser light at 21% laser intensity. The images were captured for 60 seconds through FRAP channel using confocal microscope. The association of GFP-tagged WRN, BLM and RECQL4 proteins is shown at different time intervals after microirradiation in panel A, B, and C, respectively. Arrow represents the site of laser irradiation. The normalized intensity values of the kinetic of association of different proteins have been represented in panel D.

Fig. 5. Endogenous γ-H2AX and RECQL4 colocalizes at the laser induced DNA damage site

HeLa cells were laser microirradiated at 21% laser intensity followed by fixation after 5 min and immunostaining with anti rabbit γ-H2AX and anti goat RECQL4 antibodies. The untreated cell is shown in top left panel, γ-H2AX is shown in red (top middle), RECQL4 is depicted in green (top right) and the merge images of the two showing co-localization is shown in yellow (lower panels).

Fig. 6. Analysis of association of GFP tagged RECQL4 mutants at the site of laser DNA damage

The different GFP-RECQL4 mutant constructs were transiently transfected in the HeLa cell and association kinetics were monitored at 21% laser intensity. Panel A, shows the schematics of different mutant and their association at DNA damage site (represented with + sign). Two Nuclear targeting sequences (NTS1 & NTS2) and the helicase domain are shown. Panel B show one representative images of each mutant association at the DNA damaged site.

Fig. 7. The recruitment of GFP tagged RECQL4 is independent of functional WRN, BLM and ATM proteins

The association kinetics of transiently transfected GFP-RECQL4 protein at the site of laser induced DNA damage in SV40 transformed Normal fibroblasts (GM00637), BLM (GM08505), WRN (AG07066), and ATM (GM05849) mutant fibroblasts cells are shown in panel A, B, C and D, respectively. The normalized intensity value of the kinetic of association of RECQL4 in normal fibroblasts, BLM, WRN and ATM mutant cells have been represented in panel E.

Fig. 8. Retention of RECQL4 at the DNA damaged site is shorter compared to WRN and BLM proteins

The HeLa cells were transiently transfected with GFP tagged WRN, BLM and RECQL4 plasmids and after 24 hr post transfections, the cells were microirradiated at 21% laser intensity and the dissociation kinetics were monitored by live cell imaging. The cells were placed in an environmental chamber and the dissociation kinetics of GFP tagged WRN, BLM and RECQL4 at the site of laser induced DNA damage is shown in panel A, B and C, respectively. The time for image acquisition is shown for each panel. D. The percent maximum intensity values with respect to time of the dissociation of GFP tagged WRN, BLM and RECQL4 in HeLa cells have been represented below.