Skip to main content
NCBI home page
Nucleic Acids Research logoLink to Nucleic Acids Research
. 2003 Oct 15;31(20):5848–5857. doi: 10.1093/nar/gkg775

Modification of the ionizing radiation response in living cells by an scFv against the DNA-dependent protein kinase

Shuyi Li, Yoshihiko Takeda, Stéphanie Wragg, John Barrett 1, Andrew Phillips, William S Dynan *
PMCID: PMC219464  PMID: 14530433

Abstract

The non-homologous end joining pathway uses pre-existing proteins to repair DNA double-strand breaks induced by ionizing radiation. Here we describe manipulation of this pathway in living cells using a newly developed tool. We generated a single chain antibody variable fragment (scFv) that binds to the DNA-dependent protein kinase catalytic subunit (DNA-PKcs), a key enzyme in the pathway. In contrast to existing pharmacologic inhibitors, the scFv binds a newly defined regulatory site outside the kinase catalytic domain. Although the scFv inhibits kinase activity only modestly, it completely blocks DNA end joining in a cell-free system. Microinjection of the scFv sensitizes human cells to radiation, as measured by a reduction in efficiency of colony formation and induction of apoptosis at an otherwise sublethal dose of 1.5 Gy. The scFv blocks non-homologous end joining in situ at a step subsequent to histone γ-H2AX focus formation but preceding γ-H2AX dephosphorylation. Blockage occurs in cells exposed to as little as 0.1 Gy, indicating that DNA-PKcs is essential for double-strand break repair even at low radiation doses. The ability to modify the radiation response in situ in living cells provides a link between biochemical, genetic and cytologic approaches to the study of double-strand break repair intermediates.

INTRODUCTION

Human exposure to ionizing radiation (IR) comes from cosmic, terrestrial, occupational and medical sources. Interest in the IR response derives from a desire to understand and mitigate the risks of environmental exposure. Interest also comes from a desire to increase the therapeutic gain from radiation therapy, which is the most common non-surgical treatment for a variety of human tumors, including lung, prostate, colon and breast cancer.

The biological effects of IR exposure arise largely from its unique ability to induce DNA double-strand breaks (DSBs) (1). Even a single DSB per cell, if unrepaired, can lead to irreversible growth arrest or cell death (2). Eukaryotic cells have evolved several DSB repair mechanisms to reduce the severity of IR damage (3). In humans, the non-homologous end joining (NHEJ) pathway repairs most breaks within minutes of their occurrence by direct, DNA ligase-mediated end joining. An alternative repair mechanism, homologous recombination, uses an intact copy of the gene as a template for synthesis of new DNA spanning the DSB. In higher eukaryotes, homologous recombination occurs predominantly in the G2 phase of the cell cycle, when sister chromatids are available as template (4,5).

Although not all components of the NHEJ system have been identified, the DNA-dependent protein kinase is crucial. This enzyme is composed of a regulatory component, Ku protein, and the DNA-dependent protein kinase catalytic subunit (DNA-PKcs), which bind cooperatively to free DNA ends to form an active protein kinase complex (6,7). DNA-PKcs phosphorylates itself, other repair proteins and p53 (8). In rodents, DNA-PKcs mutants show greatly increased sensitivity to IR (9,10) and in human tumors, there is an inverse correlation between the level of DNA-PKcs and radiation sensitivity (11). The radiosensitive phenotype of mutant cells can be rescued by introduction of a functional DNA-PKcs cDNA, but this is not seen when using a DNA-PKcs point mutant that lacks kinase activity (12). Thus, kinase activity itself is essential for DSB repair. The avid binding of DNA-PKcs to DNA ends, together with its ability to phosphorylate a variety of nuclear targets, suggests that it could act as a decision maker, determining whether a break is repaired by NHEJ, redirected for repair by an alternative pathway or allowed to remain unrepaired, leading to irreversible growth arrest or cell death.

DSB repair takes place in vivo within cytologically defined foci characterized by the presence of a modified histone (γ-H2AX), autophosphorylated DNA-PKcs and a number of other signaling and repair proteins (13–20). Two general approaches have been taken to investigate the role of DNA-PKcs within these foci, including its interaction with cellular DNA damage signaling pathways. In one of these, the expression of DNA-PKcs has been attenuated or eliminated through the use of antisense RNA, siRNA or targeted gene disruption (9,10,21,22). To a limited extent, the function of individual residues within DNA-PKcs has been investigated by reintroduction of mutant alleles. The very large size of the coding region (>12 000 nt) complicates the use of this approach. Pharmacological inhibitors provide a more facile approach for investigating the consequences when DNA-PKcs is present but not active. The most widely used of these compounds, wortmannin and LY294002, effectively inhibit DNA-PKcs in vivo and in vitro, but lack specificity for DNA-PKcs over related phosphatidylinositol 3-kinase family members such as ATM and ATR (23,24). This again limits the utility of the approach.

Here we describe the use of a novel modifier of DNA-PKcs activity, which consists of a single chain antibody variable fragment (scFv) derived from a previously described monoclonal antibody (mAb) directed against DNA-PKcs, mAb 18-2 (25). We show here that scFv recognizes a 25 residue linear peptide unique to DNA-PKcs, outside the conserved protein kinase catalytic domain. We show that the scFv sensitizes cells to radiation by altering DSB repair in situ in living cells. These studies provide direct evidence that the NHEJ pathway is involved in repair of DSBs not only at previously studied, cytotoxic doses of IR, but also at a low dose that induce only a few DSBs per cell. The ability to block the DSB repair pathway at a defined point in vivo allows us to order individual events in the radiation response, i.e. to define which events occur before and which after the step that is blocked. Exploitation of the novel regulatory site in DNA-PKcs defined by these studies may also facilitate new, mechanism-based approaches for increasing therapeutic gain from radiation therapy.

MATERIALS AND METHODS

scFv cloning and expression

mRNA was purified from hybridoma cells expressing mAb 18-2 (25) and used to generate an scFv-encoding cDNA as described (26). Products were subcloned in pCANTAB 5 E (Amersham Biosciences, Piscataway, NJ). A 1 l culture of Escherichia coli containing the plasmid was grown in 2× YT medium with 100 µg/ml ampicillin and 2% glucose at 30°C to an A600 of 0.8–1.0. Cells were collected, resuspended in the same medium lacking glucose and containing 1 mM isopropyl-β-d-thiogalactopyranoside and cultured overnight. A periplasmic extract was prepared by sequential extraction of the cell pellet with 20 ml of ice-cold TES buffer (0.2 M Tris–HCl, pH 8.0, 0.5 mM EDTA, 0.5 M sucrose) and 30 ml of 0.2× TES buffer. After 1 h, centrifugation was performed and scFv was purified from the supernatant using a 5 ml HiTrap anti-E tag column (Amersham Biosciences). In some experiments, scFv was prepared by an alternative procedure, where protein was precipitated from the periplasmic extract with (NH4)2SO4 (75% saturation) and subjected to Superdex 75 gel filtration chromatography (Amersham Biosciences) in buffer containing 50 mM Tris–HCl pH 7.9, 12.5 mM MgCl2, 1 mM EDTA, 5% glycerol, 1 mM dithiothreitol and 0.1 M KCl.

Clones encoding DNA-PKcs fragments

Clones were obtained by reverse transcription–PCR amplification of Jurkat cell mRNA using the following primer sets: residues 411–780, d(CCGGGATCCCCAAGCTTCCTCCA GTCTGTTGCAAG) and d(CAAGCGGCCGCCAATATAA ATTGACCATTCTTCTAG); residues 765–1276, d(CGG GGATCCTTGGCAGAAGTAGGCCTGAATGCTC) and d(GAAGCGGCCGCCTACAGTTCTCTCGCCAATGAACG); residues 1247–1761, d(CGGGGATCCCCATTCAGCCTGC AGGCCACGCTA) and d(GGGGCGGCCGCTCAATTC CAACAACATAGGGCTT); residues 1734–2228, d(GGG GGATCCCCGCGGTTCAATAATTATGTGGACTGC) and d(CAAGCGGCCGCCTCTTTTTGGATGAAAGACATGT TTC); residues 2204–2714, d(GGGGGATCCGGGGTCC CTAAAGATGAAGTGTTAGC) and d(GAAGCGGCCGC AGTTATCCACCTCGTCCCCTGGAAG). The cDNAs were subcloned in pCITE4a(+) (Novagen, Madison, WI) and expressed using the TNT Coupled Rabbit Reticulocyte In Vitro Transcription/Translation kit (Promega, Madison, WI) with [35S]methionine. Immunoprecipitation was as described (27).

Binding parameters for scFv–DNA-PKcs interaction

Surface plasmon resonance measurements were made using a Biacore X instrument (Biacore, Piscataway, NJ). Interaction between scFv 18-2 and purified DNA-PKcs was measured by amine coupling of scFv to one channel of a Biosensor chip CM-5. The other channel was used as a reference. Analyte, consisting of purified DNA-PKcs diluted in HBS-EP (10 mM HEPES, pH 7.4, 150 mM NaCl, 3 mM EDTA, 0.005% Surfactant P20) was flowed over the chip at 30 µl/min. Interaction between scFv and peptides was measured by immobilizing the specific peptide, biotin-KKKYIEIRK EAREAANGDSDGPSYM, in one channel of a Biosensor chip SA and a non-specific peptide, representing a nearby non-binding sequence, LADSTLSEEMSQFDFSTGVQSYSYS, in the other channel. Analyte, consisting of scFv 18-2 diluted in HBS-EP, was flowed over the chip as above. For both experiments, regeneration between runs was with HBS-EP supplemented with 4 mM MgCl2, 100 mM glycine, pH 2.3, and 1 M NaCl. Duplicate measurements were made at 25°C. Data were additionally double referenced and evaluated using the 1:1 interaction with the mass transfer limitation model of the BioEvaluation 3.1 software.

Functional assays

Cell-free DNA end joining assays were performed as described (28). Peptide phosphorylation assays (25 µl) contained 12.5 µM ATP, 0.5 µCi [γ-32P]ATP, 0.4 mM biotinylated p53 peptide, 2 µg bovine serum albumin and 3 ng DNA fragment (400 bp). Reactions were preincubated at 30°C for 5 min and then DNA-PK (1 µl) and scFv or mAb were added. Reactions were incubated at 30°C for 15 min, 12.5 µl of termination buffer (7.5 M guanidine hydrochloride) was added and 10 µl from each reaction was spotted on a biotin capture membrane (SignaTECT DNA-PK Assay System; Promega). After washing, binding of phosphopeptide was determined by liquid scintillation counting.

Microinjection

Retinal pigment epithelial cells were immortalized with telomerase. An expression construct for the oncoprotein, adenovirus E1A, was introduced using a retroviral vector (29). Cells were grown in a 50:50 (v/v) mixture of MEM and F12 media, supplemented with 10% fetal bovine serum and antibiotics. The SK-MEL-28 human skin melanoma cell line (30) was grown in DMEM supplemented with 10% fetal bovine serum and antibiotics. Cells were seeded on 175 µm CELLocate coverslips (Eppendorf AG, Hamburg, Germany) and microinjected using sterile microcapillaries (Femtotips II; Eppendorf AG) mounted on an automated microinjection system (FemtoJet and InjectMan; Eppendorf AG) attached to a Zeiss Axiovert microscope. The injection mixture consisted of 1 mg/ml scFv, 15 µg/ml pEGFP-N1 DNA (Clontech, Palo Alto, CA), 10 mM KH2PO4, pH 7.4, and 75 mM KCl. Each injection was performed at a pressure of 50 hPa for 0.2 s. Unless otherwise indicated in the figure legend, cells were allowed to recover for 2–3 h at 37°C and irradiated using a 137Cs source (GammaCell 40 Exactor; MDS Nordion, Ottawa, Canada) at a rate of 1 Gy/min. Successfully injected cells were identified by GFP fluorescence after 12–24 h.

Immunofluorescence staining

Cells were fixed in 2% formaldehyde for 10 min. They were permeabilized and blocked by incubation for 1 h in phosphate-buffered saline containing 0.5% Triton X-100, 15% goat serum, 0.2% fish skin gelatin and 0.03% NaN3. Samples were incubated with one or more of the following antibodies: 1:250 dilution of anti-E-tag (detect scFv; Amersham Pharmacia Biotech, Piscataway NJ), 1:1000 anti-DNA-PKcs (human serum FT) (31), 1:500 anti-green fluorescent protein (Novus Biologicals, Littleton, CO or Molecular Probes, Eugene, OR), 1:200 anti-activated caspase 3 (Promega) or 1:500 anti-γ-H2AX (mAb JBW301; Upstate Cell Signaling Solutions, Waltham, MA). Staining was visualized using secondary antibodies of appropriate specificity conjugated to Alexa Fluor 488 or Alexa Fluor 594 (Molecular Probes).

RESULTS

scFv 18-2 recognizes a novel site in DNA-PKcs

A reverse transcription–PCR strategy was used to amplify the rearranged heavy and light chain variable region genes from mAb 18-2-expressing cells. Amplified genes were assembled into a scFv-encoding cDNA, which was subcloned for overexpression in the E.coli periplasm. Purified scFv preparations, obtained by affinity chromatography as described in Materials and Methods, contained a prominent 30 kDa band (Fig. 1A). This was identified as the scFv based on its size and anti-epitope tag immunoblotting. The presence of authentic heavy and light chain variable fragment sequences was verified by comparison with the Kabat immunoglobulin sequence database using the AbCheck tool (32) and by molecular modeling using the WAM tool (33).

Figure 1.

Figure 1

Open in a new tab

(A) Appearance of scFv in crude periplasmic extract and after affinity purification. The results of SDS–PAGE with Coomassie blue staining are shown. Migration of molecular weight markers is indicated in kDa. (B) Purified DNA-PKcs binds to scFv 18-2 with a Kd of ∼1.4 × 10–9 M. Analyte containing DNA-PKcs was flowed over a surface containing immobilized scFv 18-2 ligand and interaction was measured by surface plasmon resonance. The measured kassoc was 1.9 × 105 per M per s and kdissoc was 2.7 × 10–4 per s. (C) scFv 18-2 binds selectively to the N-terminal caspase 3 fragment of DNA-PKcs. Jurkat cells were incubated for 3 h in the presence or absence of 100 ng/ml anti-Fas antibody and lysates were analyzed by immunoblotting with the indicated antibodies. (D and E) The scFv 182 epitope maps within residues 1901–2081. Segments of DNA-PKcs were subcloned and expressed by coupled transcription–translation in the presence of [35S]methionine. Products were immunoprecipitated with scFv 18-2 and immune complexes were analyzed by SDS–PAGE. Input represents 20% of material used for immunoprecipitation. (F) The scFv 18-2 epitope maps within residues 2001–2025. The indicated biotinylated peptides were immobilized on the surface of a streptavidin-coated multiwell plate and ELISA was performed to detect scFv binding. (G) scFv 18-2 binds to peptide A (residues 2001–2025) with a Kd of ∼2 × 10–9 M. Analyte containing scFv was flowed over a surface containing immobilized peptide ligand and interaction was measured by surface plasmon resonance. In the experiment shown, the measured kassoc was 8.2 × 104 per M per s and kdissoc was 1.7 × 10–4 per s. (H) Diagram showing location of fragments and peptides used for epitope mapping relative to kinase catalytic domain.

scFv binding parameters were evaluated by surface plasmon resonance (Fig. 1B). The Kd was ∼1.4 nM, which is typical for antibody–antigen interactions (34). Immunoblotting showed the ability of scFv 18-2 to selectively recognize DNA-PKcs in a mixture of total cellular proteins (Fig. 1C). Immunoblotting also showed that the scFv 18-2 epitope lies within a caspase cleavage fragment spanning residues 1–2713 (35), which was produced by treating Jurkat cells with anti-Fas antibody (36). This is consistent with results of prior studies using the parental mAb (35,36). The pattern of binding differs from that of mAb 42-27, which recognizes a different epitope C-terminal to the caspase site (37).

The epitope was further delineated using overlapping cDNAs providing full coverage of the 1–2713 fragment. Proteins were expressed using a coupled in vitro transcription–translation system and scFv 18-2 binding was tested by immunoprecipitation with anti-epitope tag antibody. The epitope mapped to a fragment spanning residues 1734–2228 (Fig. 1D). This sequence was further subdivided by the same approach. Residues 1901–2228 were required for recognition by the scFv (Fig. 1E). Further subcloning (not shown) and studies with synthetic peptides (Fig. 1F) identified a 25 residue sequence, 2001–2025, as necessary and sufficient for epitope formation. Surface plasmon resonance showed that binding parameters for interaction between scFv and the peptide were comparable to those for interaction with whole DNA-PKcs (Fig. 1G). The epitope mapping is summarized in Figure 1H. The epitope is located outside the kinase catalytic domain, within sequences unique to DNA-PKcs and not shared with ATM or ATR.

Inhibition of DNA-PK activity in cell-free assays

To test the effect of scFv 18-2 on DNA end joining, reactions were performed in a cell-free system containing linearized plasmid substrate, HeLa cell nuclear extract and recombinant DNA ligase IV (DNL IV)/XRCC4 complex (28). Consistent with previous results, nuclear extract and purified DNL IV/XRCC4 each had little activity when tested alone, but catalyzed efficient conversion of linear substrate to dimers and higher oligomeric products when tested as a mixture (Fig. 2A, lanes 1–4). scFv 18-2 strongly inhibited end joining, whereas an unrelated control scFv had little effect at an equal concentration (lanes 5 and 6). The parental mAb 18-2 inhibited end joining, although the inhibition was incomplete, even at the highest concentration tested (lanes 7 and 8). Control mouse IgG did not inhibit (lane 9). LY 294002, a relatively non-specific phosphatidylinositol 3-kinase inhibitor, blocked end joining completely under the conditions used (lane 10).

Figure 2.

Figure 2

Open in a new tab

scFv 18-2 inhibits the end joining and protein kinase activities of DNA-PKcs in cell-free assays. (A) Linearized radiolabeled plasmid DNA was incubated with HeLa cell nuclear extract, recombinant DNL IV/XRCC4 and other components as indicated and products were analyzed by SDS–agarose electrophoresis. (Upper) PhosphorImager analysis; (bottom) quantitation. End joining was expressed as the ratio of ligated products to total DNA, normalized to the value obtained in lane 4. All lanes are from the same gel. (B) Peptide phosphorylation reactions contained scFv 18-2 mixed with unrelated scFv to give a constant total concentration of 1 µM. Control reactions were performed where mAb 18-2 (2.1 µM) or LY 294002 (1.4 µM) were added or where DNA-PK was omitted, as indicated. Activities have been normalized to the value obtained in the presence of control scFv. Reactions were performed in duplicate, and error bars denote range.

Separate assays were performed to test the effect of scFv 18-2 on kinase activity. Both the scFv and the parental mAb inhibited p53 peptide phosphorylation activity by ∼50% at the highest concentrations tested (Fig. 2B), similar to results obtained in the initial characterization of mAb 18-2 (25). The complete inhibition of end joining activity under conditions that give only partial inhibition of kinase activity is consistent with the epitope mapping results showing that the scFv 18-2 recognition sequence is outside the kinase domain.

Intracellular binding of scFv 18-2 to DNA-PKcs

Microinjection of antibodies is a well-established method to study intracellular protein function (38,39). Although scFvs can, in principle, be expressed intracellularly by gene transfer, microinjection was chosen for the present study because it allows introduction of native, folded antibody directly into the nucleus. This eliminates concerns over disulfide bond formation and folding in the intracellular environment, which are common obstacles to use of scFvs for intracellular applications (40). Initial experiments were performed using telomerase-immortalized human retinal pigment epithelial (RPE) cells expressing the adenovirus E1A oncoprotein (29). These cells are sensitive to p53-mediated apoptotic signaling and are expected to respond strongly to any increase in unrepaired DSBs.

A preliminary experiment was performed to determine whether microinjected scFv 18-2 was stable and associated with DNA-PKcs in vivo. scFv was injected into the nucleus and cells were immunostained after 6 h to allow simultaneous visualization of endogenous DNA-PKcs and microinjected scFv 18-2. (Fig. 3A). In a non-injected control cell (a–e), DNA-PKcs has a punctate nuclear distribution, consistent with previous results (41), and scFv staining is not seen. In injected cells, two patterns were seen. In the more common (f–j), DNA-PKcs retains its punctate nuclear appearance and microinjected scFv 18–2 adopts a coincident nuclear distribution. In an estimated 20% of the cells, scFv 18-2 was primarily cytoplasmic (k–o). This may reflect variability in the microinjection technique. In these cells, a portion of the DNA-PKcs appears to be drawn into the cytoplasm, where it assumes a focal distribution coincident with the scFv. Results demonstrate that scFv 18-2 is stable inside cells for at least 6 h post-injection and are consistent with binding to endogenous DNA-PKcs. In separate experiments, microinjected scFv 18-2 was detected 18 h post-injection, albeit at lower levels (not shown).

Figure 3.

Figure 3

Open in a new tab

(A) Microinjected scFv 18-2 is stable in RPE E1A cells. Cells were fixed at 6 h after injection and stained with DAPI, 1:1000 anti-DNA-PKcs human autoimmune serum and 1:250 anti-E tag mouse monoclonal antibody to detect scFv18-2. (a–e) Non-injected cell. (f–j) Microinjected cell showing nuclear pattern of scFv distribution. (k–o) Microinjected cell showing cytoplasmic pattern of scFv distribution. (B) A combination of microinjected scFv 18-2 and irradiation blocks colony formation. (a–e) scFv 18-2 was co-injected with pEGFP N1 vector into RPE E1A cells. Living cells were observed after 16 h with a combination of phase contrast and epifluorescence illumination. Green fluorescence indicates EGFP expression. (f–j) Cells were fixed after 88 h and stained with 1:500 anti-GFP antibody (Novus Biologicals, Littleton, CO) and 1:500 Alexa-Fluor 488-conjugated goat anti-rabbit antibody (Molecular Probes, Eugene, OR). The same field as in (a–e) was observed with DIC and fluorescence. (C) A combination of microinjected scFv 18-2 and irradiation induces apoptosis. scFv 18-2-injected RPE E1A cells were fixed at 60 h after injection and stained with DAPI (blue), anti-GFP antibody (green) and anti-active caspase 3 antibody (red). Scale bar indicates 10 µm in all panels.

Combination of scFv 18-2 and IR inhibits microcolony formation

We next tested the effect of microinjected scFv 18-2 on colony forming ability. Cells were co-injected with scFv and a plasmid encoding enhanced green fluorescent protein (EGFP), which serves as a tracer, allowing the fate of injected cells to be tracked in real time. Cells received either scFv 18-2 or an unrelated control antibody, scFv 147 (42) and received 0 or 1.5 Gy IR at 6 h post-injection. The dose was chosen on the basis of preliminary experiments indicating that 1.5 Gy was somewhat below the threshold required to reduce growth or induce apoptosis in non-injected cells.

The cell growth substrate was marked with a grid pattern, permitting the same field to be observed repeatedly. At 16 h post-injection, individual microinjected cells could be recognized by intrinsic EGFP fluorescence against a background of non-injected cells, which was visualized by phase contrast illumination (Fig 3B, a–e). There was no difference between treatment groups at this early time. However, when cells were re-observed 88 h post-injection, an estimated 60–80% of the cells that received a combination of scFv 18-2 and 1.5 Gy had disappeared from the plate (f–g) and the few remaining cells had failed to divide (see below). Surrounding non-injected cells, visualized by DIC optics, proliferated normally. In contrast to cells in the treatment group, almost all of the cells in the three control groups, which had received a combination of control scFv and 1.5 Gy, scFv 18-2 and 0 Gy, scFv 147 and 0 Gy, respectively, divided to form microcolonies of 4–8 cells (h–j).

scFv 18-2 sensitizes cells to IR-induced apoptosis

Non-dividing cells that remained on the coverslip after combination treatment with the scFv and 1.5 Gy IR were immunostained with antibodies against active caspase 3 (Fig. 3C). EGFP was used as a tracer to allow visualization of microinjected cells. At 60 h post-injection, the majority of cells remaining after treatment with scFv 18-2 and 1.5 Gy IR stained brightly for activated caspase 3 (Fig. 3C, k–l), whereas surrounding non-injected cells were negative. Microinjected cells in the control groups were also negative (Fig. 3C, m–o). Table 1 provides a quantitation of the results. Differences in the frequency of apoptotic cells in different groups were statistically significant (P < 0.001).

Table 1. Induction of apoptosis.

Irradiation scFv 18-2 scFv 147
  1.5 Gy 0 Gy 1.5 Gy 0 Gy
Injected cells 20/21 (95.2%)a,b,c 3/23 (13.0%)a 3/30 (10.0%)b 4/29 (13.8%)
Non-injected cells (same coverslip) 23/203 (11.3%)c 22/223 (9.9%) 24/224 (10.7%) 20/220 (9.1%)
Open in a new tab

a,b,cPair-wise comparisons that show significant differences (P < 0.001) by one-sided Fisher’s exact test.

scFv 18-2 prevents repair of DNA damage

To determine whether scFv 18-2 directly inhibited repair in vivo, we monitored the fate of individual DSBs using histone γ-H2AX as a marker. Phosphorylation of the H2A variant, H2AX, which creates the γ-H2AX form, occurs in situ within a megabase domain of chromatin flanking each DSB (13). Recent studies have validated the use of γ-H2AX foci as a surrogate marker for unrepaired DSBs over a wide range of doses (13,43). The appearance and disappearance of γ-H2AX foci closely tracks the kinetics of DSB repair and disappearance of foci is impaired in cells that are deficient in DNL IV, an essential enzyme in the NHEJ pathway (43). Because we had difficulty identifying well-defined γ-H2AX foci in RPE cells, we switched to SK-MEL-28 cells, which are derived from a radioresistant melanoma (30). In preliminary experiments (not shown), prominent foci were induced in non-injected SK-MEL-28 cells in response to IR. Their appearance was dose dependent, they formed within 30 min and most were resolved within 90 min.

The effect of scFv 18-2 on γ-H2AX foci is shown in Figure 4. Irradiation was performed at two doses, 1.5 and 0.1 Gy, which are calculated to induce ∼50 and ∼3 DSBs per cell, respectively, assuming a diploid genome content (44,45). As in previous experiments, EGFP vector was co-injected to allow tracking of injected cells. At 30 min following exposure to 1.5 Gy, bright nuclear staining for γ-H2AX was seen, independently of whether cells were microinjected with scFv 18-2, with control scFv or were non-injected bystanders (Fig. 4A, a and e). No staining was seen in non-irradiated control cells (d and g). At 90 min post-irradiation, the γ-H2AX persisted at high levels in cells receiving scFv 18-2 (b and c), but disappeared from non-injected cells in the same field and from cells receiving control scFv (f). Figure 4B shows the same experiment at 0.1 Gy. Again, induction of γ-H2AX foci was similar in both irradiated groups (a and e). The foci persisted in cells receiving scFv 18-2 (b and c), but were quickly resolved in non-injected cells (not shown) and in cells receiving control scFv (f). Together, these results suggest that scFv 18-2 blocks or delays repair of DSBs in vivo.

Figure 4.

Figure 4

Open in a new tab

scFv 18-2 prolongs the lifetime of γ-H2AX foci. (A) SK-MEL-28 cells were injected with scFv 18-2 or with scFv 147 and irradiated at 1.5 Gy. Cells were fixed at 30 or 90 min after irradiation and stained with DAPI (blue), anti-γ-H2AX antibody (red) and anti-GFP antibody as described in Materials and Methods. (B) As in (A), except with 0.1 Gy of ionizing radiation. Scale bar indicates 10 µm in both panels. There were 50–100 successfully injected cells in each experimental group, and results shown are typical.

DISCUSSION

Here we have demonstrated that an scFv directed against DNA-PKcs blocks the NHEJ pathway of DSB repair. The scFv blocks end joining, both in an in vitro system based on a cell-free extract and in situ in living cells. Unlike known pharmacologic inhibitors of DNA-PKcs, scFv 18-2 does not target the conserved kinase domain, but rather a unique site near the middle of the primary amino acid sequence, within a region of previously undefined function. This sequence is unique to DNA-PKcs and is not present in related DNA damage-responsive kinases, ATM and ATR.

Because the scFv is directed against a specific molecular target required for NHEJ, it provides a unique tool for dissecting the radiation response in living cells. Consistent with the binding target being outside the kinase domain, the scFv produces only modest inhibition of kinase activity, although it appears to produce a complete inhibition of end joining. Thus, blockage of DSB repair is not attributable to loss of kinase activity per se, but rather to some other mechanism. In vitro studies suggest that the antibody does not block DNA interaction, as there is no change in crosslinking of DNA-PKcs to photoreactive DNAs in vitro (S.Wragg and W.S.Dynan, unpublished results). Consistent with this, residual kinase activity seen in the presence of mAb 18-2 remains DNA-dependent (25). Presumably, the mechanism of action involves either steric hindrance of an essential protein–protein interaction surface on DNA-PKcs or blockage of a conformational change required for progression of the end joining reaction.

The scFv appears to be at least as effective at inhibiting the end joining reaction as the parental mAb, and possibly more effective, based on the comparison in Figure 2. It could be that the smaller size of the scFv renders it better able to interact with its epitope in native DNA-PKcs. We cannot rule out the possibility, however, that the apparently greater potency of the scFv reflects differences in the methods of preparation. The scFv is from a recombinant source and is therefore homogeneous, whereas the mAb is purified from cell culture supernatant and could be contaminated with other IgGs from the growth medium. In our experience, the mAb 18-2-producing line is not amenable to growth in serum-free medium.

Cytologic studies suggest that DSB end joining occurs within the context of repair foci characterized by a distinctive histone phosphorylation event, by the accumulation of autophosphorylated DNA-PKcs and by the recruitment of repair and signaling proteins to the break site, including 53BP1, NFBD1/MDC1 and the chromatin-bound form of the Mre11/Rad50/NBS1 complex. The order of assembly of the proteins, the number of copies that are present and their interactions within the foci remain largely unknown. The ability to block the repair process in situ at a specific stage, using an scFv directed against an epitope in an individual repair protein, provides an opportunity to dissect the process of assembly of repair foci. For example, in the present study we show that appearance of histone γ-H2AX foci precedes the step blocked by scFv 18-2, whereas dephosphorylation of γ-H2AX occurs subsequent to this step. Investigation of the effects of scFv 18-2 on trafficking of other proteins involved in the assembly of repair foci is in progress.

The effect of scFv 18-2 on histone γ-H2AX foci provides direct evidence for a role of DNA-PKcs in the low dose radiation response. Prior work demonstrating a requirement for DNA-PKcs has used high, cytotoxic doses of radiation. Most human exposure, however, is to low doses, and the relevance of DNA-PKcs to the low dose response has not been established. One recent study showed that human cells with a mutant form of DNL IV were impaired in the ability to resolve histone γ-H2AX foci, implicating NHEJ as a significant mechanism of repair at low doses (43). Our present work on DNA-PKcs confirms and extends this result.

A limitation of the present studies is that they rely on microinjection. We have not yet explored methods for introduction of scFv 18-2 into cells on a bulk scale. scFvs vary widely in their intracellular stability, apparently because of differences in the ability to fold in the intracellular environment (40). Strategies have been described for re-engineering of scFvs for more efficient intracellular expression (46). Intracellular expression of scFv 18-2 in active and stable form will permit more extensive studies of its effect on the radiation response.

Blockade of the NHEJ pathway in the RPE cells was effectively irreversible, as the cells that received a combination of scFv 18-2 and radiation underwent apoptosis. The long-term effects of the scFv in other cell types, including the SK-MEL-28 melanoma line, are unknown. However, in vitro studies suggest that the scFv blocks progression of the end joining reaction without interfering with the DNA-binding capability of DNA-PKcs. The induction of an arrested DNA-PKcs complex at DNA termini is potentially a more effective strategy for tumor cell radiosensitization than reduction of the expression of DNA-PKcs itself. The presence of a non-functional repair complex may block access by proteins associated with other pathways of DSB repair, creating a persistent unrepaired DSB. Whereas persistent DSBs may be tolerated in quiescent normal tissues, even a small number of unrepaired DSBs in proliferating tumor cells would be expected to lead to generation of acentric chromosomal fragments, loss of essential genes and cell death. Thus, the scFv itself, or pharmacologic agents that target the new regulatory site defined by the scFv, may prove useful for increasing the therapeutic gain from radiotherapy.

Acknowledgments

ACKNOWLEDGEMENTS

We thank Dr Deborah Lewis for instruction in the microinjection technique, Dr Alex Chiu for the use of the microinjection apparatus, Dr Miyake Katsuya for assistance with image collection, Dr David Munn for suggesting the use of the SK-MEL-28 cell line, Dr Raymond Mernaugh for advice on scFv expression and for provision of a control scFv preparation, Dr Andrew Hayhurst for scFv 147 and Dr David Bickel for statistical advice. We acknowledge the Medical College of Georgia Imaging Core Facility, Medical Illustration and Photography, and Office of Biostatistics and Bioinformatics for their services. This work was supported by an award from the US Department of Energy Low Dose Radiation Research Program (DE-FG07-99ER62875) and by a US Public Health Service grant (GM 35866).

REFERENCES

  • 1.Ward J.F. (1998) Nature of lesions formed by ionizing radiation. In Nickoloff,J.A. and Hoekstra,M.F. (eds), DNA Damage and Repair. Humana Press, Totowa, NJ, Vol. II, pp. 65–84. [Google Scholar]
  • 2.Di Leonardo A., Linke,S.P., Clarkin,K. and Wahl,G.M. (1994) DNA damage triggers a prolonged p53-dependent G1 arrest and long-term induction of Cip1 in normal human fibroblasts. Genes Dev., 8, 2540–2451. [DOI] [PubMed] [Google Scholar]
  • 3.Pastink A., Eeken,J.C. and Lohman,P.H. (2001) Genomic integrity and the repair of double-strand DNA breaks. Mutat. Res., 480/481, 37–50. [DOI] [PubMed] [Google Scholar]
  • 4.Sonoda E., Takata,M., Yamashita,Y.M., Morrison,C. and Takeda,S. (2001) Homologous DNA recombination in vertebrate cells. Proc. Natl Acad. Sci. USA, 98, 8388–8394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Lee S.E., Mitchell,R.A., Cheng,A. and Hendrickson,E.A. (1997) Evidence for DNA-PK-dependent and -independent DNA double-strand break repair pathways in mammalian cells as a function of the cell cycle. Mol. Cell. Biol., 17, 1425–1433. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Dvir A., Peterson,S.R., Knuth,M.W., Lu,H. and Dynan,W.S. (1992) Ku autoantigen is the regulatory component of a template-associated protein kinase that phosphorylates RNA polymerase II. Proc. Natl Acad. Sci. USA, 89, 11920–11924. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Gottlieb T.M. and Jackson,S.P. (1993) The DNA-dependent protein kinase: requirement for DNA ends and association with Ku antigen. Cell, 72, 131–142. [DOI] [PubMed] [Google Scholar]
  • 8.Smith G.C. and Jackson,S.P. (1999) The DNA-dependent protein kinase. Genes Dev., 13, 916–934. [DOI] [PubMed] [Google Scholar]
  • 9.Taccioli G.E., Amatucci,A.G., Beamish,H.J., Gell,D., Xiang,X.H., Torres Arzayus,M.I., Priestley,A., Jackson,S.P., Marshak Rothstein,A., Jeggo,P.A. et al. (1998) Targeted disruption of the catalytic subunit of the DNA-PK gene in mice confers severe combined immunodeficiency and radiosensitivity. Immunity, 9, 355–366. [DOI] [PubMed] [Google Scholar]
  • 10.Gao Y., Chaudhuri,J., Zhu,C., Davidson,L., Weaver,D.T. and Alt,F.W. (1998) A targeted DNA-PKcs-null mutation reveals DNA-PK-independent functions for KU in V(D)J recombination. Immunity, 9, 367–376. [DOI] [PubMed] [Google Scholar]
  • 11.Vaganay-Juery S., Muller,C., Marangoni,E., Abdulkarim,B., Deutsch,E., Lambin,P., Calsou,P., Eschwege,F., Salles,B., Joiner,M. et al. (2000) Decreased DNA-PK activity in human cancer cells exhibiting hypersensitivity to low-dose irradiation. Br. J. Cancer, 83, 514–518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Kurimasa A., Kumano,S., Boubnov,N.V., Story,M.D., Tung,C.S., Peterson,S.R. and Chen,D.J. (1999) Requirement for the kinase activity of human DNA-dependent protein kinase catalytic subunit in DNA strand break rejoining. Mol. Cell. Biol., 19, 3877–3884. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Rogakou E.P., Boon,C., Redon,C. and Bonner,W.M. (1999) Megabase chromatin domains involved in DNA double-strand breaks in vivo. J. Cell Biol., 146, 905–916. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Paull T.T., Rogakou,E.P., Yamazaki,V., Kirchgessner,C.U., Gellert,M. and Bonner,W.M. (2000) A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr. Biol., 10, 886–895. [DOI] [PubMed] [Google Scholar]
  • 15.Schultz L.B., Chehab,N.H., Malikzay,A. and Halazonetis,T.D. (2000) p53 binding protein 1 (53BP1) is an early participant in the cellular response to DNA double-strand breaks. J. Cell Biol., 151, 1381–1390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Chan D.W., Chen,B.P., Prithivirajsingh,S., Kurimasa,A., Story,M.D., Qin,J. and Chen,D.J. (2002) Autophosphorylation of the DNA-dependent protein kinase catalytic subunit is required for rejoining of DNA double-strand breaks. Genes Dev., 16, 2333–2338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Mirzoeva O.K. and Petrini,J.H. (2001) DNA damage-dependent nuclear dynamics of the Mre11 complex. Mol. Cell. Biol., 21, 281–288. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Shang Y.L., Bodero,A.J. and Chen,P.L. (2003) NFBD1, a novel nuclear protein with signature motifs of FHA and BRCT and an internal 41-amino acid repeat sequence, is an early participant in DNA damage response. J. Biol. Chem., 278, 6323–6329. [DOI] [PubMed] [Google Scholar]
  • 19.Xu X. and Stern,D.F. (2003) NFBD1/KIAA0170 is a chromatin-associated protein involved in DNA damage signaling pathways. J. Biol. Chem., 278, 8795–8803. [DOI] [PubMed] [Google Scholar]
  • 20.Lou Z., Chini,C.C., Minter-Dykhouse,K. and Chen,J. (2003) Mediator of DNA damage checkpoint protein 1 regulates BRCA1 localization and phosphorylation in DNA damage checkpoint control. J. Biol. Chem., 278, 13599–13602. [DOI] [PubMed] [Google Scholar]
  • 21.Sak A., Stuschke,M., Wurm,R., Schroeder,G., Sinn,B., Wolf,G. and Budach,V. (2002) Selective inactivation of DNA-dependent protein kinase with antisense oligodeoxynucleotides: consequences for the rejoining of radiation-induced DNA double-strand breaks and radiosensitivity of human cancer cell lines. Cancer Res., 62, 6621–6224. [PubMed] [Google Scholar]
  • 22.Peng Y., Zhang,Q., Nagasawa,H., Okayasu,R., Liber,H. and Bedford,J. (2002) Silencing expression of the catalytic subunit of DNA-dependent protein kinase by small interfering RNA sensitizes human cells for radiation-induced chromosome damage, cell killing and mutation. Cancer Res., 62, 6400–6404. [PubMed] [Google Scholar]
  • 23.Izzard R.A., Jackson,S.P. and Smith,G.C. (1999) Competitive and noncompetitive inhibition of the DNA-dependent protein kinase. Cancer Res., 59, 2581–2586. [PubMed] [Google Scholar]
  • 24.Sarkaria J.N., Tibbetts,R.S., Busby,E.C., Kennedy,A.P., Hill,D.E. and Abraham,R.T. (1998) Inhibition of phosphoinositide 3-kinase related kinases by the radiosensitizing agent wortmannin. Cancer Res., 58, 4375–4382. [PubMed] [Google Scholar]
  • 25.Carter T., Vancurova,I., Sun,I., Lou,W. and DeLeon,S. (1990) A DNA-activated protein kinase from HeLa cell nuclei. Mol. Cell. Biol., 10, 6460–6471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Yuan Q., Clarke,J.R., Zhou,H.R., Linz,J.E., Pestka,J.J. and Hart,L.P. (1997) Molecular cloning, expression and characterization of a functional single-chain Fv antibody to the mycotoxin zearalenone. Appl. Environ. Microbiol., 63, 263–269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Takeda Y., Caudell,P., Grady,G., Wang,G., Suwa,A., Sharp,G.C., Dynan,W.S. and Hardin,J.A. (1999) Human RNA helicase A is a lupus autoantigen that is cleaved during apoptosis. J. Immunol., 163, 6269–6274. [PubMed] [Google Scholar]
  • 28.Huang J. and Dynan,W.S. (2002) Reconstitution of the mammalian DNA double-strand break end-joining reaction reveals a requirement for an Mre11/Rad50/NBS1-containing fraction. Nucleic Acids Res., 30, 667–674. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Blint E., Phillips,A.C., Kozlov,S., Stewart,C.L. and Vousden,K.H. (2002) Induction of p57(KIP2) expression by p73beta. Proc. Natl Acad. Sci. USA, 99, 3529–3534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Carey T.E., Takahashi,T., Resnick,L.A., Oettgen,H. and Old,L.J. (1976) Cell surface antigens of human malignant melanoma: mixed hemadsorption assays for humoral immunity to cultured autologous melanoma cells. Proc. Natl Acad. Sci. USA, 73, 3278–3282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Jafri F., Hardin,J.A. and Dynan,W.S. (2001) A method to detect particle-specific antibodies against Ku and the DNA-dependent protein kinase catalytic subunit in autoimmune sera. J. Immunol. Methods, 251, 53–61. [DOI] [PubMed] [Google Scholar]
  • 32.Martin A.C. (1996) Accessing the Kabat antibody sequence database by computer. Proteins, 25, 130–133. [DOI] [PubMed] [Google Scholar]
  • 33.Whitelegg N.R. and Rees,A.R. (2000) WAM: an improved algorithm for modelling antibodies on the WEB. Protein Eng., 13, 819–824. [DOI] [PubMed] [Google Scholar]
  • 34.Siegel R.W., Allen,B., Pavlik,P., Marks,J.D. and Bradbury,A. (2000) Mass spectral analysis of a protein complex using single-chain antibodies selected on a peptide target: applications to functional genomics. J. Mol. Biol., 302, 285–293. [DOI] [PubMed] [Google Scholar]
  • 35.Casciola-Rosen L., Nicholson,D.W., Chong,T., Rowan,K.R., Thornberry,N.A., Miller,D.K. and Rosen,A. (1996) Apopain/CPP33 cleaves proteins that are essential for cellular repair: a fundamental principle of apoptotic death. J. Exp. Med., 183, 1957–1964. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.McConnell K.R., Dynan,W.S. and Hardin,J. (1997) The DNA-dependent protein kinase catalytic subunit (p460) is cleaved during Fas-mediated apoptosis in Jurkat cells. J. Immunol., 158, 2083–2089. [PubMed] [Google Scholar]
  • 37.Song Q., Lees-Miller,S.P., Kumar,S., Zhang,Z., Chan,D.W., Smith,G.C., Jackson,S.P., Alnemri,E.S., Litwack,G., Khanna,K.K. et al. (1996) DNA-dependent protein kinase catalytic subunit: a target for an ICE-like protease in apoptosis. EMBO J., 15, 3238–3246. [PMC free article] [PubMed] [Google Scholar]
  • 38.Morgan D.O. and Roth,R.A. (1988) Analysis of intracellular protein function by antibody injection. Immunol. Today, 9, 84–88. [DOI] [PubMed] [Google Scholar]
  • 39.McNeil P.L. (1989) Incorporation of macromolecules into living cells. Methods Cell Biol., 29, 153–173. [DOI] [PubMed] [Google Scholar]
  • 40.Cattaneo A. and Biocca,S. (1999) The selection of intracellular antibodies. Trends Biotechnol., 17, 115–121. [DOI] [PubMed] [Google Scholar]
  • 41.Mo X. and Dynan,W.S. (2002) Subnuclear localization of Ku protein: functional association with RNA polymerase II elongation sites. Mol. Cell. Biol., 22, 8088–8099. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Hayhurst A. and Harris,W.J. (1999) Escherichia coli skp chaperone coexpression improves solubility and phage display of single-chain antibody fragments. Protein Expr. Purif., 15, 336–343. [DOI] [PubMed] [Google Scholar]
  • 43.Rothmann K. and Löbrich,M. (2003) Evidence for a lack of DNA double-strand break repair in human cells exposed to very low X-ray doses. Proc. Natl. Acad. Sci. USA, 100, 5057–5062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Ward J.F. (1988) DNA damage produced by ionizing radiation in mammalian cells: identities, mechanisms of formation and repairability. Prog. Nucleic Acid Res. Mol. Biol., 35, 95–125. [DOI] [PubMed] [Google Scholar]
  • 45.Metzger L. and Iliakis,G. (1991) Kinetics of DNA double-strand break repair throughout the cell cycle as assayed by pulsed field gel electrophoresis in CHO cells. Int. J. Radiat. Biol., 59, 1325–1339. [DOI] [PubMed] [Google Scholar]
  • 46.Jermutus L., Honegger,A., Schwesinger,F., Hanes,J. and Pluckthun,A. (2001) Tailoring in vitro evolution for protein affinity or stability. Proc. Natl Acad. Sci. USA, 98, 75–80. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press

RESOURCES