RECQL5 cooperates with Topoisomerase II alpha in DNA decatenation and cell cycle progression

Abstract

DNA decatenation mediated by Topoisomerase II is required to separate the interlinked sister chromatids post-replication. SGS1, a yeast homolog of the human RecQ family of helicases interacts with Topoisomerase II and plays a role in chromosome segregation, but this functional interaction has yet to be identified in higher organisms. Here, we report a physical and functional interaction of Topoisomerase IIα with RECQL5, one of five mammalian RecQ helicases, during DNA replication. Direct interaction of RECQL5 with Topoisomerase IIα stimulates the decatenation activity of Topoisomerase IIα. Consistent with these observations, RECQL5 co-localizes with Topoisomerase IIα during S-phase of the cell cycle. Moreover, cells with stable depletions of RECQL5 display a slow proliferation rate, a G2/M cell cycle arrest and late S-phase cycling defects. Metaphase spreads generated from RECQL5-depleted cells exhibit undercondensed and entangled chromosomes. Further, RECQL5-depleted cells activate a G2/M checkpoint and undergo apoptosis. These phenotypes are similar to those observed when Topoisomerase II catalytic activity is inhibited. These results reveal an important role for RECQL5 in the maintenance of genomic stability and a new insight into the decatenation process.

Figure 1.

RECQL5 depletion is associated with slow proliferation rate, arrest in G2/M phase, and late S-phase cycling defects. (A) Quantitative RT–PCR analysis of the Recql5 mRNA levels normalized to endogenous GAPDH, in HeLa shScrambled, shRECQL5-1 (n = 3) P = 0.0006 and shRECQL5-2 (n = 3) P = 0.0002, Student's t-test. (B) Methylene blue stained colonies of HeLa shScrambled and HeLa shRECQL5-2. Growth assays were performed with (C) HeLa, (D) WI38 or (E) HCT116 using shScrambled, shRECQL5-1 and shRECQL5-2 as described in ‘Materials and Methods’ section. Error bars represent ±SD, n = 3. The non-linear fit was calculated and R square (Y = Y₀*e^k*^X₎ values are 0.989 for shScrambled, 0.982 for shRECQL5-1 and 0.7416 for shRECQL5-2 in HeLa, 0.983 for shScrambled, 0.927 for shRECQL5-1 and 0.25 for shRECQL5-2 for WI38 experiments and 0.995 for shScrambled, 0.993 for shRECQL5-1 and 0.976 for shRECQL5-2 in HCT 116 experiments. Graphical representation of the doubling time calculated as an average of two independent growth assays. Error bars represent ±SD, n = 6 for each sample, each day; ***P < 0.0005, **P < 0.005, *P < 0.05, Student's t-test. Western blot showing reduced expression of RECQL5 in (C) HeLa, (D) WI38 or (E) HCT116 whole-cell lysates 96 h following transduction with lentivirus harboring shRECQL5-1, shRECQL5-2 RNA compared to control shScrambled RNA. Equal loading was confirmed by probing with anti-Actin antibody. Quantification of PI staining of asynchronous populations in (F) HeLa and (G) WI38 shScrambled and RECQL5-depleted cells. Error bars represent ±SD, n = 2; *P = 0.0359 between HeLa shScrambled and sh RECQL5-1 and *P = 0.0262 between HeLa shScrambled and shRECQL5-2, Student's t-test.

Figure 2.

RECQL5 depletion is associated with late S-phase cycling defects. Representative flow cytometry images of (A and B) HeLa and (C and D) HCT116 scrambled and RECQL5 knockdown cells analyzed for the ability to incorporate bromodeoxyuridine (BrdU) and plotted against DNA content in (A and C) a 30 min BrdU pulse and (B and D) 24 h pulse. For panels A and C, the fraction of BrdU positive cells are denoted in the upper left. The late S-G2 of the BrdU positive cells subpopulation, boxed area, is also quantified in the upper right. For panels B and D, the first generation slow late S-G2 population is boxed and quantified.

Figure 3.

Stable RECQL5 depletion causes metaphase chromosome defects. (A) Metaphase spreads of U2OS cells after RECQL5 depletion. Two observed phenotypes, undercondensation and entanglement of chromosomes, are shown. (B) Quantification of metaphase spreads from shScrambled and shRECQL5-2 depleted U20S cells. Results are an average of two independent biological experiments. A minimum of 50 metaphase spreads per cell line were analyzed. Error bars represent ±SD, *P = 0.0349, ***P = 0.0006, Student's t-test (C) Western blot showing reduced expression of RECQL5 in U2OS whole-cell lysates 96 h following transduction with lentivirus harboring shRECQL5-2 RNA compared to shScrambled RNA. Equal loading was confirmed by probing with anti-Actin antibody. (D) Fluorescence in situ hybridization in metaphase spreads of U2OS shRECQL5-2 using a telomeric probe. Yellow arrows indicate ends of undercondensed chromatids.

Figure 4.

RECQL5 physically interacts with Topoisomerase IIα during S-phase. (A) Western blot of whole cell lysates from HeLa shScrambled, shRECQL5-1 and shRECQL5-2 cells, probed with indicated antibodies. Equal loading was confirmed by probing with anti-Actin antibody. (B) Immunoprecipitation of RECQL5 and (C) Topoisomerase IIα from HeLa whole cell extracts and probed with indicated antibodies. (D) Immunoprecipitation of RECQL5 from a mixture of RECQL5 and Topoisomerase IIα recombinant proteins and probed with indicated antibodies. (E) Representative flow cytometry histograms of HeLa cells analyzed after 2 mM hydroxyurea treatment and released for the time indicated. DNA content stained with PI was plotted against cell number. (F) Confocal images of representative HeLa cells fixed and stained for RECQL5 and Topoisomerase IIα after 2 mM hydroxyurea treatment and then released for the time indicated. Staining protocol is described in ‘Materials and Methods’ section. Images are pseudo colored: green-RECQL5, red-Topoisomerase IIα and blue-DAPI stain of nucleus, with yellow areas indicating the co-localization. The co-localization channel was generated by the Volocity software. Co-localization was determined by the following equation: (X_i −X_mean) (Y_i −Y_mean) where X_i is the intensity of the voxel for the Red Fluorescence Channel and Y_i is the intensity of the voxel for the Green Fluorescence channel (62). The fifth panel is the merge of all three channels.

Figure 5.

RECQL5 stimulates Topoisomerase IIα catalyzed DNA decatenation. (A) Decatenation assay carried out with nuclear extracts of HeLa shScrambled, shRECQL5-1 and shRECQL5-2 at indicated concentrations. (B) Quantification of the decatenation was performed by measuring the intensity of the decatenated bands. Error bars represent ±SD, n = 3. ***P < 0.0001, f-test (C) DNA decatenation assays were carried out as described in ‘Materials and Methods’ section using recombinant Topoisomerase IIα with recombinant RECQL5 added in increasing amounts as indicated. K-kinetoplast DNA and C-covalently closed minicircles. (D) Quantification of the decatenation was performed by measuring the intensity of the decatenated bands. Error bars represent ±SD, n = 3. *P < 0.01, Student's t-test. (E) Helicase assay using purified RECQL5 and increasing quantity of Topoisomerase IIα.

Figure 6.

RECQL5, but not WRN, can stimulate Topoisomerase IIα mediated DNA decatenation. (A) Western blot showing reduced expression of WRN in nuclear extracts of U2OS shWRN cells compared to control shScrambled cells. Equal loading was confirmed by probing with a Lamin A specific antibody. (B) Decatenation assay carried out with the nuclear extracts of U2OS shScrambled and shWRN at indicated concentrations. (C) Quantification of the decatenation was performed by measuring the intensity of the decatenated bands. Error bars represent ±SD, n = 3.

Figure 7.

Depletion of RECQL5 increases apoptosis and activates a G2/M checkpoint. (A) Dot plots of FITC-Annexin versus PI two parameter flow cytometry of HeLa shScrambled, shRECQL5-1 and shRECQL5-2 cells. (B) Quantification of Annexin positive and PI negative cells in HeLa. Error bars represent ±SD, n = 2; *P = 0.02, **P = 0.002, Student's t-test. (C and D) Western blot of whole cell lysates from HeLa shScrambled, shRECQL5-1 and shRECQL5-2 cells probed with indicated antibodies.

Figure 8.

Model to demonstrate the interaction of RECQL5 and Topoisomerase IIα. Model to demonstrate the interaction of RECQL5 and Topoisomerase IIα functioning to separate daughter chromatids at the end of replication. RECQL5 serves to unwind the parental DNA at the converging replication forks with Topoisomerase IIα behind functioning to decatenate newly formed daughter molecules. The helicase activity is then inhibited till Topoisomerase II performs the decatenation.