The radioresistance kinase TLK1B protects the cells by promoting repair of double strand breaks

Abstract

Background: The mammalian protein kinase TLK1 is a homologue of Tousled, a gene involved in flower development in Arabidopsis thaliana. The function of TLK1 is not well known, although knockout of the gene in Drosophila or expression of a dominant negative mutant in mouse cells causes loss of nuclear divisions and missegregation of chromosomes probably, due to alterations in chromatin remodeling capacity. Overexpression of TLK1B, a spliced variant of the TLK1 mRNA, in a model mouse cell line increases it's resistance to ionizing radiation (IR) or the radiomimetic drug doxorubicin, also likely due to changes in chromatin remodeling. TLK1B is translationally regulated by the availability of the translation factor eIF4E, and its synthesis is activated by IR. The reason for this mechanism of regulation is likely to provide a rapid means of promoting repair of DSBs. TLK1B specifically phosphorylates histone H3 and Asf1, likely resulting in changes in chromatin structure, particularly at double strand breaks (DSB) sites.

Results: In this work, we provide several lines of evidence that TLK1B protects the cells from IR by facilitating the repair of DSBs. First, the pattern of phosphorylation and dephosphorylation of H2AX and H3 indicated that cells overexpressing TLK1B return to pre-IR steady state much more rapidly than controls. Second, the repair of episomes damaged with DSBs was much more rapid in cells overexpressing TLK1B. This was also true for repair of genomic damage. Lastly, we demonstrate with an in vitro repair system that the addition of recombinant TLK1B promotes repair of a linearized plasmid incubated with nuclear extract. In addition, TLK1B in this in vitro system promotes the assembly of chromatin as shown by the formation of more highly supercoiled topomers of the plasmid.

Conclusion: In this work, we provide evidence that TLK1B promotes the repair of DSBs, likely as a consequence of a change in chromatin remodeling capacity that must precede the assembly of repair complexes at the sites of damage.

Figure 1

Pattern of γ-radiation sensitivity by clonogenic assays. 10⁴ untransfected MM3MG and cells overexpressing TLK1B were irradiated with the indicated doses and plated in triplicates of 100, 500, 1000, or 5000 cells in multiwell plates. The colonies were counted 10 days later. The average from 2 independent experiments is shown. S.F. = surviving fraction.

Figure 2

Phosphorylation of H3 and H2AX after irradiation. MM3MG and MM3MG over-expressing TLK1B cells were grown to 80% confluence prior to gamma-radiation (10 Gy). Cells were harvested at different times (0 h, 30 min, 1 h, 2 h, 4 h, 8 h and 16 h) after irradiation and lysed in RIPA buffer. Equal amount of protein of each sample was loaded on a 15% SDS-PAGE gel and electrophoresed. Blots were probed with phospho-Histone H3 (Ser-10) antibody, or phospho-Histone H2AX (Ser-139) antibody (Upstate Cell Signaling). Equal loading of proteins was confirmed by staining the blots with Ponceau S prior to processing. These blots are representative of two separate experiments.

Figure 3

Phosphorylation of H3 and H2AX after irradiation in the presence of wortmannin. The experiment was carried out as detailed in the legend to Fig. 2, but the cells were pre-treated with 30 μM wortmannin to inhibit ATM.

Figure 4

Analysis of episomes. 2 × 10⁷ cells transformed with BK-Shuttle or TLK1B were irradiated with the indicated dose of γ-radiation. The cells were returned to the incubator and the plasmids were isolated 1 hr later by the Hirt's protocol and separated on a 1% agarose/TAE gel. The mobility of the forms (circular, linear, and supercoiled) is indicated. The structure of the BK-Shuttle episomal vector is shown on the right. The bands were quantified with ImageQuant vs. 5 (Molecular Dynamics).

Figure 5

Analysis of episomes during a time course of recovery from IR. 2 × 10⁷ cells transformed with BK-Shuttle or TLK1B were irradiated (or not, C) with 20 Gy of γ-radiation. The cells were returned to the incubator and the plasmids were isolated immediately after radiation (IR), or after the indicated times of recovery, 2 to 8 hr (R2-R8). The episomes were recovered by alkaline lysis, which removes plasmids with DSBs by strand separation followed by the rapid renaturation. This is because it is not always easy to separate on agarose gels the linearized from the supercoiled form of the plasmid (these plasmids are 11–14 kb in size, without and with insert). The bands were quantified with ImageQuant. The intensities of the bands relative to control are indicated underneath each lane. Size markers (Novagen 1 kb ladder) are shown in the first lane.

Figure 6

Assay of repair of genomic damage. Examples of non-irradiated cells (A, D) or cells irradiated and allowed to recover for 0 hr (B, E) or 2 hr (C, F) are shown. MM3MG cells are shown in A-C; TLK1B cells are shown in D-F. An arrow points to one of the DAB spots.

Figure 7

A. Ligation reactions and supercoiling. Reactions were prepared as described Experimental Procedures. The position of linearized plasmid (lane 1, input), bacterial supercoiled (lane 2) and dimers is indicated. The various topomeric forms of supercoiled plasmid (the result of deposition of nucleosomes on the template) are not resolved well in this gel without chloroquine, and appear as a slight smear (lanes 6, 9, 10). B. Assay of chromatin assembly. Nucloesomes assembly was carried out on 2 μg of Bluescript plasmid as described in Methods. Each band corresponds to the addition of one nucleosome, which decreases the linking number. These gels are representative of two different experiments.