WO2011161457A1

WO2011161457A1 - Cancer treatment

Info

Publication number: WO2011161457A1
Application number: PCT/GB2011/051180
Authority: WO
Inventors: Boris Kynsela
Original assignee: The University Of Birmingham
Priority date: 2010-06-23
Filing date: 2011-06-23
Publication date: 2011-12-29
Also published as: GB2497216A; GB201010588D0; GB201301168D0; GB2497216B

Abstract

The invention provides a method of increasing the sensitivity of a cell to a double strand break (dsb) damage, comprising decreasing levels of functional histone H1x in a cell. This may be carried out using, for example, RNAi or anti-H1x antibodies. The double strand breaks may be introduced using ionising radiation (such as γ or x-rays or protons) or chemically. Inhibitors of H1x function are also provided.

Description

Cancer Treatment

The invention relates to methods of increasing the sensitivity of cells to DNA double strand break (dsb) damage, to cells with increased sensitivity to dsb damage, to agents capable of increasing sensitivity to dsb damage, and to the use of such agents in the treatment of cancer.

DNA double-strand breaks (DSB) are the most lethal form of DNA damage, since the continuity of genetic information is disrupted in both strands. The defects in DSB repair lead to accumulation of mutations resulting in developmental defects and cancer ^l'².

In all cellular organisms DSB are repaired by evolutionary conserved pathways of nonhomologous end-joining (NHEJ) and homologous recombination (HR) that require co- ordinated action of many proteins acting on the broken DNA molecule . The basic mechanistic aspects of these processes have been well documented, however, much less is known about how these pathways operate within the nuclear environment and which components of chromatin are critical for the successful outcome of the DNA repair.

Linker histones are the family of related proteins that have been hypothesised to act as structural components of chromatin stabilising the DNA entry and exit points of nucleosomal core particles and maintaining the higher order chromatin structure. The actual biological roles of individual linker histones in the chromatin mediated processes are poorly understood⁴.

Ionizing radiation and many anti-cancer drugs such as bleomycin cause DNA double-strand breaks (DSB). In human cells, the majority of DSB are repaired by the process of NHEJ that is complemented by HR when the second copy of the genome becomes available in S and G₂ phases of the cell cycle. The nuclear environment represents a significant challenge for any process involving DNA molecules and only limited information is available on the contribution of the chromatin components to the successful outcome of DSB repair. In our previous work using NHEJ reconstituted in vitro, we have shown that linker histones mixture purified from somatic cells exhibited general inhibitory effects on the efficiency of the process⁵. Linker histones are the family of structurally related proteins with distinct species, tissue and developmental specificities⁶. In higher eukaryotes all eleven linker histone variants adhere to the same general structure plan with a highly conserved central globular domain and much less conserved N-teraiinal and C-terminal tails. The moderate level of conservation of tail domains between the species is suggestive of a specific function, nevertheless the biological roles of individual linker variants are not well understood. Historically, the linker histones have been considered to have a structural role in establishing and maintaining the higher-order chromatin structure, acting as a general repressor of transcription. More recently, some HI histones were implicated in chromatin mediated processes such as gene expression⁷,

8 9

DNA replication and DNA repair . The linker histone Hlx is ubiquitously present in all somatic cells⁶. Hl (but not other linker variants) is over-expressed in neuroendocrine cells and tumours¹⁰. Interestingly, the cancer stem cells that can be particularly resistant to cell death induced by radiation or chemotherapy¹¹ have recently been identified to be of neuroendocrine origin in some tumours .

A first aspect of the invention provides a method of increasing the sensitivity of a cell to DNA double strand break (dsb) damage comprising decreasing the level of functional histone Hlx in a cell.

A further aspect of the invention provides a cell having decreased levels of functioning histone Hlx compared to normal cells. That is compared to, for example, a cell of the same type, but without the level of functional histone Hlx inhibited or otherwise decreased, such as by using the inhibitors of Hlx function described herein.

Inhibitors of Hlx function are also provided, as is their use to treat cancer.

The inventors have identified that inhibiting levels of functional Hlx, by, for example, using inhibitors of Hlx function in the cell, increases sensitivity of the cell to dsb. This results in increased sensitivity to dsb-inducing agents such as ionising radiation or dsb-inducing agents such as bleomycin, doxorubicin, camptothecin (which interferes with DNA toposisomerases resulting in increased dsb levels) and other DNA replication poisons, many of which have been found to be associated with increased dsb.

The ionising radiation may be, for example γ-radiation, X-radiation or particulate radiation such as a- or β-radiation or accelerated particle beams such as electrons or proton beams generally known in the art. Conversely in a further aspect, radioresistance may be increased by increasing Hlx levels in cells, by for example, over expressing Hlx in the cell. For example, Hlx may be provided via a vector, such as an adenoviral vector, limited to a constitutive or inducible promoter.

The cell may be in vitro, for example in tissue culture or in vivo. The cell may be a cancer cell. It may be part of a tumour.

Preferred tumours are those typically treated with radiotherapy, such as breast, neck and head (especially brain) cancers.

The levels of functional Hlx in the cells may be reduced with inhibitors of functional Hlx. These may decrease the concentration of Hlx found in the cell or inhibit the function of Hlx in the cell.

Levels of Hlx may be reduced for example using RNA interference (RNAi) technology.

Hlx levels may be determined by looking at mRNA concentrations and protein levels (e.g. via Western Blotting). It may be compared to a levels in a normal cell, for example one of the same cell type.

In vitro stimulation of the DNA ligaseIV/xrcc4 activity may be used to assess functional activity

RNA interference (RNAi) technology has been known for several years. It utilises the use of a complementary sequence to the mRNA encoding a protein of interest. The complementary sequence binds to the mRNA strand and generates a double-stranded molecule. This inhibits production of protein from the mRNA sequence.

RNAi can be carried out using several methods using for example small interfering RNA (siRNA), DNA-directed RNA (ddRNA), microRNA (miRNA), and short hairpin (shRNAs). The use of RNAi is summarised in, for example, Advances in RNAi, Technical Insights (2009), Frost & Sullivan, San Antonio, Tx, USA. According to the invention siRNA having a preferred length of 19-21 nucleotides, is typically used to regulate Hlx production within the cell.

The Hlx siRNA used typically comprises the sequence:

5 ' -CCAAGAAGGUUCCGUGGUUTT-3 ' siRNA allows efficient and specific down regulation of gene expression to be achieved, utilising relatively low concentrations of siRNA. Typical concentrations of siRNA used have been approximately 10 nM. This latter concentration has been achieved with the preferred use of interferin as a transfection agent, instead of the standard oligofectamine methods generally used in the art. The interferin based methods are described in detail below.

RNAi may be introduced into cells by a number of methods generally known in the art including:

Transfection, using, for example, a cationic carrier to facilitate entry into a cell or, for example, cyclodextrin containing polymers (Calando Pharmaceuticals Inc, marketed under the trademark RONDEL).

Electroporation uses an electric pulse to introduce nucleic acids into cells. This is less preferred as it potentially damages the cell.

Viral delivery has been used to introduce RNAi into cells. Examples generally known in the art include viral vectors such as retroviruses, adenoviruses and lentiviruses. Tkachenko et al (JACS (2003) 125, 4700) shows targeting of the nucleus for RNAi.

RXi Pharmaceuticals Corporation have developed a number of approaches to delivery of RNAi by oral, systemic and local delivery approaches. These include the use of beta-glucan shells to encapsulate RNAi embedded within networks of polyethylenimine (PEI). siRNAs have been successfully targeted against tumour cells using siRNA linked to a RNA aptamer. The RNA aptamer binds to cell surface antigens associated with the target cancer, see for example McNamara J.O. et al (Nature Biotechnology (2006) 24, 1005-1015. The siR A may therefore be targeted by aptamers capable of binding cell surface antigens.

Receptors which are over expressed in cancers which may be targeted include Folate (Leamon, Drug Discov. Today (2001) 6, 44 and Parker et al (2005) 338, 284) and Shiga toxin (Johannes (online 21 Dec 2009), Nat. Rev. Microbiol).

The lifespan of siRNAs have been previously shown to be extended within the body by binding to polyethylene glycol (PEG) a process called PEGylation. PEGylated siRNA according to the invention is provided.

Hlx may be also inhibited by using anti-Hl antibodies. The invention provides anti-Hlx antibodies or fragments thereof capable of binding Hlx for use according to any aspect of the invention. The antibody may be a polyclonal antibody, but may be a monoclonal antibody. Antibodies may be raised by conventional methods in the art, such as that used by Kohler & Milstein. Polyclonal antibodies against Hlx are commercially available from, for example, Abeam, Cambridge, UK

The fragments may include Fab, F(ab)₂ or scFv fragments.

The antibodies specifically bind to and reduce the activity of Hlx.

Antibodies have been successfully targeted at the cell nucleus before. Weisbart et al (J. Autoimm. (1998), JT, 539-546) for example, shows modified antibodies to target the cell nucleus (see also WO 97/32602). Xiang H et al (Protein Exp. Purif (2009) 66, 172-80) have recently demonstrated the production of scFv proteins with nuclear location signals. Such a system may be used to target anti-Hlx antibody fragments to the nucleus.

Preferably the siRNA or anti-Hlx antibodies or fragments of the invention, are targeted against a cancer cell, such as those discussed above, for example by the methods or constructs described herein.

The siRNAs and antibodies or fragments may be provided in combination with pharmaceutically acceptable carriers. Pharmaceutical compositions comprising Hlx inhibiting agents according to the invention are provided.

The use of the Hlx inhibiting agents according to the invention in the methods of the invention is also provided.

Inhibitors of Hlx function, for example as described herein, for use in the treatment of cancer is provided. Methods of treating cancer are also provided using such inhibitors.

The invention also provides inhibitors of Hlx as defined herein, in the manufacture of a medicament to treat cancer.

The inhibitors may be used in combination with dsb-inducing agents such as ionising radiation or chemicals as discussed above.

The dosage used is typically tumour/patient dependent - for radiation treatment a typical fraction is between 1.8 to 2Gy and total dose between 45-80 Gy

They may be used with other radiosensitisers, such as misonidazole or metronidazole.

Many tumours are have relatively low amounts of vasculature and therefore relatively hypoxic. Hypoxic cytotoxins for use in the treatment of tumours are generally known in the art. The inhibitors may be used in combination with such cytotoxins, such as tirapazamine.

The invention will now be described by way of example only with reference to the following figures:

Figure 1. siRNA mediated depletion of Hlx protein renders cells sensitive to ionising radiation. In the transiently transfected cells, the depletion reached the maximum efficiency between 48 and 96 hours after transfection and no "off target" effects had been observed for the core histone H2A and the linker histone HI .2 used as controls (Panel a). Vice versa, no "off-target" effects on t he expression of Hlx protein had been detected in the controls transfected with Luciferase and HI .2 siRNA constructs (Panel b). Hlx depleted cells show sensitivity to ionizing radiation in clonogenic survival assay (Panel c). On the contrary, depletion of HI .2 linker histone leads to increased resistance to the effects of radiation at higher doses (Panel d). All quantitative results are average ± SD of at least 3 independent experiments.

Figure 2. The analysis of the cell cycle responses in the Hlx depleted cells after exposure to γ-rays. No differences in the overall cell cycle dynamics had been detected by dual parameter Propidium Iodide/Bromodeoxiuridine (PI/BrdU) cell cycle analysis in the siRNA Hlx cells compared to the controls without radiation (Panel a). In γ-irradiated (5 Gy) cell populations, the Hlx depleted cells exhibited prolonged stay in the G₂M phase of the cell cycle, as demonstrated by delayed transition of BrdU positive cells to Gj phase (Panel b, black rectangle areas). The quantitative analysis of the G₂M cell cycle block in irradiated Hlx depleted cells compared to controls (Panel c). No defects have been found in the activation of the key G₂ checkpoint kinases in response to ionizing radiation, suggesting the increased amount of unrepaired DNA damage in Hlx depleted cells caused the prolonged G₂M cell cycle block (Panel d).

Figure 3. The analysis of DNA repair foci formation and activation of key DSB repair proteins in the Hlx depleted cells in response to radiation damage. Hlx depleted cells exhibited increased numbers of γ-Η2ΑΧ unrepaired foci 24h after irradiation compared to the controls (Panel a) detected by immunofluorescence. Persistent unrepaired co-localizing foci were also detected for MDC1 , p53BPl, BRCAland Rad51 proteins, suggesting increased amounts of unrepaired DNA damage in Hlx depleted cells (Panel b). Quantitative analysis of DNA repair foci in Hlx depleted cells and controls at 24 h after radiation (Panel c). No defects in the activation of key DSB repair proteins have been detected in Hlx depleted cells compared to controls (Panel d).

Figure 4. Hlx interacts with LX complex in vivo and specifically stimulates its ligation activity. Stably overexpressed GFP-tagged Hlx fusion protein interacts with XRCC4 protein as detected by immunoprecipitation using GFP specific antibody (Panel a). The interaction of Hlx with LX complex was further confirmed using co-immunoprecipitation of native proteins (Panel b). The functional impact of Hlx linker histone on ligation activity of LX or T4 DNA ligase had been analyzed using reconstituted ligation assay. Up to 60 fold stimulation of ligation was observed for LX complex, but no stimulation had been detected for T4 DNA ligase (Panels c and d). No ligation was observed in the absence of the ligase complex (Panel d, lanes 10 and 11). Reaction efficiencies were quantified using ImageQuant software (Molecular Dynamics) and expressed as the percentage of the radioactively labelled reaction substrate converted into higher order concatemers. Each data point represents an average ± SD of at least three independent experiments. In the absence of magnesium ions the presence of both Hlx and LX results in the formation of two distinct DNA-protein complexes, suggesting Hlx stabilizes the interaction of LX with DNA, increasing its ligation efficiency (Panel e, lanes 15 - 18). No DNA-protein complex formation had been observed for Hl or LX alone (Panel e lanes 2 - 10).

Figure 5. Generation of stable overexpressing eGFP-Hlx MRC5VA clones. Human Hlx was cloned into the eGFP expression vector and stably transfected into MRC5VA cells. Stable clones were analysed by fluorescent microscopy and two clones were selected for further studies, labelled HD4 (Panel A) and HD11 (Panel B). Empty eGFP vector was also transfected into MRC5VA cells as a control (Panel C). GFP-tagged overexpressed Hlx protein shows the same characteristic nuclear distribution profile with increased nucleolar staining as the wild type protein (ref. 21).

Figure 6. Purified Recombinant Human Hlx Protein. Human Hlx was cloned into the pET-28b expression vector, purified by metal chelated affinity chromatography and dialysed into water. Proteins were visualised on an SDS-PAGE gel and stained by Coomassie. Lane 1 shows BL21 (DE3) E. coli whole cell lysate expressing his-Hlx fusion protein. Lanes 2-5 show the affinity elution profile of his-Hlx from TALON affinity resin beads and lane 6 shows final dialysed recombinant his-Hlx used for further experiments.

Figure 7. Co-immuno-precipitation of nucleolin with GFP-H1.X

MRC5VA cells were stably transfected with GFP tagged Hl .X. Co-immuno-precipitation experiments show that GFP-H1.X interacts with endogenous nucleolin in response to exposure to 30Gy IR. Co-immuno-precipitation experiments using non-specific antibodies do not co-immuno-precipitate any endogenous nucleolin protein. IP = Co-immuno-precipitation; IB = Western Blot Figure 8. DNA-Pk phosphorylation sites by in vitro kinase assay showing Hlx is phosphorylated at S49 by DNA-Pk.

Methods Cell Culture

MRC5VA normal human epithelial lung fibroblasts were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% foetal calf serum (FCS), 3.0 mM L- glutamine, non-essential amino acids and 1% sodium pyruvate without antibiotics.

RNAi

MRC5VA cells were seeded at a density of 1.5xl0⁴ (6 well dish) or lxlO⁶ (T75) and transfected with ΙΟηΜ siRNA duplex to Hlx: (5'-CCAAGAAGGUUCCGUGGUUTT-3'), HI .2: 5'-AAGAGCGUAGCGGAGUUUCUCTT-3', or a control siRNA duplex (CGUACGCGGAAUACUUCGA) using Interferin (AutogenBioclear). Cells were incubated for 48 h before further experiments were undertaken.

Clonogenic Survival Assay

Clonogenic survival assays were performed essentially as described . MRC5VA cells were treated with siRNA constructs as described before being trypsinised and irradiated using a ¹³⁷Cs source at a dose rate of 2.5Gy min^"1. Following irradiation 50-500,000 cells were seeded into six well plates and incubated at 37°C in a humidified 5% CO² atmosphere for 14 days. Colonies were stained with 0.25% crystal violet in 50% methanol. Colonies of >50 cells were counted.

Bivariate Cell Cycle Analysis

MRC5VA cells were treated with siRNA constructs as described above before they were incubated with 20μΜ bromodeoxyuridine (BrdU) for 20mins. Cells were trypsinised, counted and divided into two aliquots, one of which was irradiated with 5Gy of Cs γ-radiation, the other was left unirradiated and used as a control. Aliquots of 3x10⁵ cells were seeded to allow a time point to be collected every 2 hours over a 24 hour period or every 4 hours over a 48 hour period time course. Cells were incubated for the indicated period of time before being trypsinised, washed and fixed with ice cold 70% ethanol. Following fixation, cell suspensions were washed in PBS and treated with 2M HC1 containing 0.1% pepsin for 20 mins at room temperature. Cells were washed and incubated with 1 :50 dilution of anti-BrdU antibody (DAKO) in PNT buffer (PBS, 0.5% normal goat serum, 0.5% Tween-20) for 60mins. After this time, cells were washed before being incubated with 1 :50 dilution of Alexa-Fluor 488 conjugated secondary antibody (Invitrogen) in PNT for 60mins. Cells were washed before being treated with 20μg/ml propidium iodide and analysed by FACS (Beckman Coulter Epics XL).

Immunoblot analysis

Cells were sonicated in UTB buffer (8M urea, 150mM β-mercaptoethanol, 50mM Tris/HCl pH 7.5) or lysed in RIPA buffer (50mM Tris-HCl pH 8.0, 1% NP40, 0.5% deoxycholic acid, 0.1% SDS, 150mM NaCl and complete protease inhibitor mix (ROCHE)) and cellular debris removed by centrifugation. Protein concentration was determined by using the BioRad Bradford Protein determination reagent. Proteins were fractionated by SDS-Page. Proteins were transferred to nitrocellulose, and immunoblots were performed using the appropriate antibodies. The antibodies were used according to manufacturer's instructions: anti-phospho- H2AX, phospho-ATM and H2A antibodies (Upstate Biotechnology), anti-Chkl antibody (Santa Cruz), DNA-P _Cs, and BRCA1 antibodies (Merck), anti-phospho-53BPl and phospho-Chkl antibodies (Cell Signaling), anti-53BPl and Nbsl antibodies (Novus), anti- phospho-SMCl and phospho-DNA-PKcs antibodies (Bethyl), anti-phospho-Nbsl and phospho-Chk2 (R&D biosystems), Hlx antibody and HI .2 antibody (Abeam) and a-tubulin antibody (SIGMA). Antibodies to Chk2 were a gift from Dr. Stephen Elledge.

H2AX Foci Formation and DNA Repair Protein Foci Formation

MRC5VA cells were grown on coverslips under standard growing conditions and treated with siRNA constructs and irradiated as described above. Following irradiation, the cells were allowed to recover for the indicated period of time (see results section) before being permeabilised with lOmM PIPES buffer, pH 6.8, (300mM Sucrose, 20mM NaCl, 3mM MgC12, 0.5% Triton X-100). Next, cells were fixed in 4% paraformaldehyde before being blocked with 10% Foetal Calf Serum in Phosphate Buffered Saline (PBS). Cells were incubated with mouse monoclonal anti-y-H2AX antibody, (Millipore Corporation, Billerica, MA), MDC1 antibody (Stewart et al., 2003), 53BP1 antibody (Novus), BRCA1 antibody (Santa Cruz), or Rad51 antibody (Abeam) for 60mins at room temperature according to the manufacturer's instructions. Cells were washed in PBS before being incubated with 1 :500 dilution of Alexa Fluor 488-conjugated secondary antibody (Invitrogen) for 60mins at room temperature. Cells were washed before being mounted in Vectashield (VectorLabs, UK) containing 4',6-diamidino-2-phenylindole (DAPI). Images were captured using a Zeiss Axioplan 1 microscope with a lOOx objective. A minimum of 75 nuclei per experimental group were scored in multiple independent experiments.

Co-Immunoprecipitation

MRC5VA cells stably expressing pEGFP-Hlx were grown to confluency and lysed (50mM Tris-HCl pH7.5, lmM EGTA, ImM EDTA, 50mM sodium fluoride, 5mM sodium pyrophosphate, ImM sodium orthovanadate, 0.27M sucrose, 1% Triton X-100, 0.1% β- mercaptoethanol, and complete protease inhibitor cocktail (Roche)). One milligram of total protein was pre-cleared with protein G beads before the protein lysate was incubated with 4μg of primary antibody overnight at 4°C. The antibodies were used according to the manufacturer's instructions: GFP and Hlx antibodies (ABC AM), Ligase IV antibody (Santa Cruz) and in-house made rabbit polyclonal XRCC4 antibody. Complexes were isolated by incubating with protein G-beads for 4 hours at 4°C before proteins were eluted in lx SDS buffer without reducing agents at 40°C for 20mins.

Purification of Recombinant Hlx protein

Amino-terminal histidine-tagged recombinant human Hlx was expressed from pET28b expression vector (Invitrogen) in BL21 (DE3) E.coli cells and purified by affinity chromatography and dialysis (see Supplementary Information for more details). Purified human proteins were quantified using a BCA protein assay (Pierce).

In Vitro Ligation Assay

The ligation reactions were performed using modified procedure, described previously²³'²⁴. Briefly, the indicated amounts of proteins were incubated for 10 minutes at 37°C in 30 μΐ reaction mixture (66 mM Tris-HCl, 5 mM MgCl_2, 1 mM DTT, ImM ATP, 0.5mg/ml BSA, pH 7.5) with 70fmol of the Afllll-Pstl fragment of BLUESCRIPT plasmid as a substrate (γ- ³²P-ATP labelled on the 5 '-end). Reactions with Hlx were pre-incubated for 10 minutes on ice with indicated amounts of protein, and ligation was started by adding the LX complex or T4 DNA ligase and transfer to 37°C. After incubation, the reactions were deproteinized in 0.6% SDS and 0.6 mg/ml proteinase K (SIGMA) for 10 minutes at 37°C, phenol/chloroform extracted and precipitated with Pellet-Paint co-precipitant (Novagen). Aliquots of the reactions were run on 0.8% agarose gels. Dried gels were analyzed and quantified using a STORM Phosphorimager (Molecular Dynamics).

Electrophoresis Mobility Shift Assays

Indicated amounts of proteins were incubated with 70fmol of the Afllll-Pstl fragment of BLUESCRIPT plasmid as a substrate (γ-³²Ρ-ΑΤΡ labelled on the 5' -end) in the reaction buffer identical to the ligation buffer without magnesium ions for 10 minutes on ice. The DNA protein complexes were resolved on 0.8% agarose gels. Dried gels were visualized and analyzed using STORM Phosphorimager (Molecular Dynamics).

Results

To investigate a possible role for the linker histone Hlx in DNA damage responses in human cells, the inventors have targeted the expression of Hlx protein in human MRC5VA fibroblasts using siRNA technology. The transient transfection of the siRNA had been optimised using interferin as transfection agent allowing the use of 15 fold lower molar concentrations of targeting oligonucleotides (ΙΟηΜ) compared to the standard oligofectamine-mediated method (150nM) to minimize non-specific "off-target" effects. It had been previously shown that the specificity of siRNA constructs is concentration dependent. At concentrations of —100 nM, siRNA nonspecifically induces a significant number of genes, many of which are known to be involved in apoptosis, cell cycle regulation

1 ^

and the stress response .

However, reduction of the siRNA concentration to 20 nM eliminated nonspecific responses. Using optimized procedure with low concentration of targeting oligonucleotides, very substantial reduction in the expression levels of Hlx protein had been achieved and this reduction was maintained for up to 5 days post-transfection (Figure la). No nonspecific changes were observed on the expression levels of the core histone H2A and another linker histone variant HI.2, used as controls. Similarly, no changes in expression levels of Hlx and H2A proteins were observed in the control cells transfected with Luciferase and HI .2 targeting constructs when compared to the untransfected cells (Figure lb). The long-term maintenance of the targeting effect allowed us to investigate the impact of depletion of Hlx protein on radiation responses using standard clonogenic survival assay. The Hlx targeted cells showed significant reduction in the cell survival after exposure to ⁶⁰Co γ-rays when compared to non-transfected and mock-transfected cells (Figure lc).

Interestingly, the cells targeted for linker histone HI .2 used as a control showed statistically significant (p<0.01) increased survival compared to the non-transfected and mock-transfected cells at higher doses of ionizing radiation (Figure Id). Similar enhanced resistance to the effects of ionizing radiation, hydroxyurea and methyl-methane-sulfonate were previously reported for the mouse embryonic stem cells depleted for 3 linker histone variants including H1.2¹⁴. This observation in the context of the increased sensitivity of Hlx depleted cells suggest the intriguing possibility of a complex interplay between individual linker histone variants in DNA damage responses. The recent report attempting to characterize the biological activity of Hlx suggested the protein was required for the mitotic progression and proper chromosome segregation in undamaged HeLa cells¹⁵. This observation opened the possibility that defective cell cycle responses to DNA damage could be the underlying cause of increased sensitivity of Hlx cells to radiation. To investigate this possibility, we have analyzed the cell cycle progression and the checkpoint function in both control and irradiated cell populations using two-dimensional (Propidium Iodide/Bromodeoxyuridine - PI) flow cytometry analysis. Surprisingly, we have not found any differences in the overall cell cycle time and duration of individual cell cycle phases in Hlx depleted undamaged cells compared to the controls (Figure 2a), or any evidence of prolonged mitosis. One possible explanation for this discrepancy could be that the cell cycle effects previously reported might be a nonspecific effect due to the high concentration of siRNA used in the published study (120nM) that can induce significant changes in the expression of non-targeted genes involved in the cellular stress responses as described above . On the contrary, the Hlx depleted cells exhibited a significantly increased G₂M cell cycle delay after exposure to 5Gy of IR (Figure 2b and 2c).

The increased G₂M delay can be symptomatic of either increased amount of unrepaired persistent DNA damage, or defects in the checkpoint activation and maintenance. In eukaryotes, the cell cycle responses to DNA damage are coordinated by phosphorylation- based transduction cascades mediated by activation of two critical checkpoint kinases Chkl and Chk2¹⁶' ¹⁷. To this extent, we analyzed the dynamics of Chkl and Chk2 phosphorylation induced by DNA damage. Consistent with the flow cytometry data, the levels of Chkl^S345P and Chk2^T68p in response to IR were found to be unchanged in Hlx depleted cells compared to controls, suggesting normal activation of both kinases in DNA damage responses (Figure 2d). This result suggested that the most likely explanation for the prolonged cell cycle delay would be the higher level of unrepaired DNA damage in the Hlx depleted cells. Phosphorylated histone H2AX (γ-Η2ΑΧ) is an exceptionally sensitive indicator of DSB that acts as a surrogate measure of lethal damage after exposure to ionizing radiation and drags. Induction of DSB causes phosphorylation of histone H2AX at serine 139, and the resulting megabase size phosphorylated regions flanking each DSB appear as spots or foci when examined microscopically after antibody staining. For ionizing radiation, the number of phosphorylated γ-Η2ΑΧ foci correlates well with the expected number of double-strand breaks produced per genome and the γ-Η2ΑΧ foci resolve with progression of DNA repair.

In Hlx depleted cells a high proportion of IR induced γ-Η2ΑΧ (5 Gy) remained unresolved even after 24h repair incubation interval compared to the control cells (Figure 3a and 3d). The unresolved foci represent sites of incomplete DSB repair and are invariably associated with radiation sensitivity phenotype. This observation clearly indicates that inhibition of Hlx protein expression significantly compromises DSB repair in response to radiation damage. The elevated level of γ-Η2ΑΧ could also indicate the requirement for the functional Hlx protein in the activation and recruitment of other known DNA repair proteins to the sites of DNA damage. A similar requirement, dependent on the previously unknown ubiquitin-ligase.

RNF168, has recently been described for the P53BP1 and BRCA1 proteins in radiosensitive RIDDLE syndrome²⁰. No defects in the recruitment of MDC1, p53BPl, BRCA1 or RAD51 to the repair foci had been found in the Hlx depleted cells and in all cases the elevated levels of these foci correlated closely with elevated levels of γ-Η2ΑΧ foci (Figure 3b and Figure 3c). Similarly, no defects in the activation of DNA-P _CS, NBS1 , SMC1 and p53BPl were found in Hlx depleted cells using phosphospecific antibodies against critical phosphorylated residues required for the activation of DNA damage responses mediated by these proteins (Figure 3d). Collectively, these results implicated the possibility of the direct involvement of Hlx protein in the repair of DSB. To test whether Hlx protein physically interacts with the components of NHEJ, GFP-tagged Hlx fusion protein was stably over-expressed in MRC5VA cells (Figure 5). Significant interaction with XRCC4 protein had been revealed following immunoprecipitation using GFP specific antibody (Figure 4a). The interaction of Hlx with LX complex was further substantiated by co-immunoprecipitation of native proteins (Figure 4b). These interactions were not affected by the presence of EtBr in the reaction mix, suggesting they are not mediated by DNA itself (data not shown). To test for functional significance of these interactions, recombinant his-tagged-Hlx protein had been over-expressed and purified to homogeneity using affinity liquid column chromatography (Figure 6).

The effect of Hlx protein on the ultimate ligation step of the NHEJ had been analyzed using the in vitro system reconstituted on radioactively-labelled 445bp DNA substrate. Our results show that the pre-incubation of the Hlx protein with the DNA substrate significantly stimulated the reaction catalyzed by DNA LigaseIV/Xrcc4 complex (LX) (up to 60-fold, Figure 4c and Figure 4d compare the lanes 2 and 7). This stimulation was specific for LX, since no stimulation could be detected for the T4 DNA ligase in the presence of Hlx (Figure 4c and data not shown) and Hlx protein did not exhibit any intrinsic ligation activity on its own (Figure 4d, lanes 10 and 11). To analyze the possible mechanism for the LX ligation stimulation, identical reactions have been reconstituted in the absence of Mg²⁺ to assess the impact of Hlx protein on the interaction of LX complex with the DNA. No stable DNA- protein complexes have been detected for either LX or Hlx proteins alone (Figure 4e lanes 2- 9 and lane 10), but two distinctive DNA-protein complexes have been identified with both proteins present (Figure 4e lanes 16-18).

In summary, the inventors have identified the novel functional role for Hlx in the DNA damage responses in human cells. Before this study, very little had been known about the biological function of this least-conserved linker histone family member, ubiquitously present

15 21

in all somatic cells ' . Their data demonstrate that low levels of Hlx protein result in the increased sensitivity of affected cells to DNA damage, protracted DNA damage checkpoint signalling and deficiency in DSB repair. The analysis of the mechanistic basis for the requirement of Hlx in DSB repair has revealed that Hlx physically interacts with the LX effector complex of NHEJ machinery, specifically and efficiently stimulates the ligation step of NHEJ and stabilizes the interaction of LX complex with DNA. These findings establish Hlx in DNA damage responses in human cells and suggest that the high level of expression reported in some types of tumours might affect responses towards radio and chemotherapy treatment.

Proteomic analysis of additional Hlx molecular partners Summary of the approach:

The whole cell lysates from irradiated (30 Gy) and un-irradiated cells (stably overexpressing GFP tagged Hlx cell line) were used for co-immunoprecipitation using anti-GFP antibody. This method provides the ability to isolate the overexpressed protein of interest, Hlx, and any proteins which may be interacting with it. The resulting precipitates were resolved on polyacrylamide gels and Coomassie Blue stained to detect the presence of novel protein bands in the irradiated samples compared with the un-irradiated samples and the control GFP expressing cell line (GFP not tagged to a protein). Novel bands were excised from the gel and analysed by mass spectrometry (MS) to identify interacting proteins. Ethidium bromide co- immunoprecipitations were also carried out to eliminate any DNA mediated protein-protein interactions, allowing the determination of those interactions which are either direct or indirectly mediated via another protein. Potential interesting interactions resulting from MS analysis were confirmed by Western blotting using co-immunoprecipitation samples and specific antibodies to these proteins.

Results revealed that Hlx interacts with a number of proteins functioning in DNA damage responses in human cells. These include nucleolin, DNA-PKcs, Ku70, interleukin enhancer binding factors 2 and 3, heat shock protein 70, stress-70 protein and DNA topoisomerase I and nuclease-sensitive element binding protein 1. The identification of these proteins as Hlx interactors further supports the possibility that Hlx (apart from its direct role in NHEJ) may have a major role in co-ordination of DNA damage responses.

Summary of the results:

Coverage Peptides Score Protein Identification

38.31 58 1239.1 Nucleolin

56.48 24 537.38 Nuclease-sensitive element-binding protein 1

25.27 12 284.88 Heat shock 70kDa protein (HSP70.1)

32.25 16 229.67 Stress-70 protein (mortalin)

17.30 8 233.55 Polyadenylate binding protein 1 1.81 60.70 Ku70

1.16 3 55.76 DNA-dependent protein kinase catalytic subunit

8.24 2 45.13 14-3-3 protein epsilon

3.66 2 32.85 DNA topoisomerase 1

Hlx specifically interacts with nucleolin following exposure to DNA damage

The highest scoring interacting protein was that of nucleolin. To validate this interaction, GFP-H1.X interacting proteins were co-immuno-precipitated, in the presence of ethidium bromide, and, using Western blotting, directly probed for nucleolin. Interestingly, the interaction was only seen following ionizing radiation (IR) exposure, suggesting that the interaction occurs following DNA damage:

Nucleolin was identified as an interacting partner for Hl .X (see Figure 7), however, unlike the LX complex, this interaction was only detected following DNA damage induced by IR. Nucleolin is predominantly a nucleolar protein, although it can be found in the nucleus, the cytoplasm and the cell membrane (Borer et al., 1989, Hovanessian et al., 2000). Interestingly, nucleolin has already been linked with other types of repair. Nucleolin can accelerate oligonucleotide binding, including oligonucleotides which contain mismatches (Hanakahi et al., 2000). Nucleolin has also been shown to interact with PCNA, and following exposure to UV, the interaction is enhanced having an inhibitory effect on nucleotide excision repair (C. Yang et al., 2009). Nucleolin has also been proposed to bind to RAD51 and regulate HR (De et al., 2006). Also, down-regulation of nucleolin following DNA damage increases the translation of p53, helping to initiate the Gl phase checkpoint (De et al., 2006).

Interactions with the heat-shock family members of proteins

The highly conserved heat shock protein (Hsp) 70 family of proteins is a group of chaperones involved in protein folding, stabilization, and shuttling functions throughout the cell. Members of this protein family can be induced by various cellular stresses, including sublethal heat stress, radiation, heavy metals, ischemia, nitric oxide radicals, certain chemotherapeutics. The human Hsp70 family contains 11 distinct genes located on several chromosomes, including both constitutive and inducible isoforms. The 70 kDa heat shock proteins (HSP70s) were initially identified by their elevated expression following hyperthermic cell stress, however, these highly conserved proteins also protect critical cellular functions from a wider range of important environmental and physiological stresses. HSP70 have been reported upregulated in tumor cells, selective inhibition of such proteins might be valuable approach to treat cancer. It has been demonstrated recently that mice lacking Hsp70.1 gene are radiosensitive and the absence of function results in genomic instability (Hunt et al. 2004). Interestingly, it had also been shown that HSP70 might play a crucial role in disassembling RAG/DNA complexes stably formed during V(D)J recombination, the process that is intrinsically linked to DSB repair in mammalian cells (Son et al 2008). More recently a further study demonstrated that, upon DNA damage, Hsp70 translocates to the cell nucleus and nucleoli, binds to PARP-1 and XRCCl proteins and protects cells form DNA breaks (Kotoglou et al. 2009).

Interactions with DNA-PKcs and interleukin enhancer binding factors 2 and 3

DNA-PKcs has also been found in relatively high-levels in the co-immunoprecipitate. This is a substantial finding since DNA-PK is a core complex of the NHEJ machinery and its interaction with Hlx has lead to further insight into the regulation of the role of HI x in the process (see section on Hlx phosphorylation), also considering that Hlx has already been unequivocally identified as a functional partner of LX complex, the major effector complex of NHEJ. From this point of view it is also interesting that interleukin enhancer binding factors 2 and 3 had been found to interact with Hlx. Both of these proteins have been found to promote the formation of a stable DNA-PK complex on the DNA (Ting and Kao 1998).

Interactions with Polyadenylate binding protein 1 and 14-3-3 protein epsilon

These two proteins together link apoptosis, DNA repair and nucleolin , suggesting a possible higher levels of co-ordination of processes that could ultimately lead to an more complex role of Hlx as one of the central controllers of repair and apoptosis interplay. 14-3-3 protein epsilon is a regulatory protein known to regulate ABL1, a tyrosine protein kinase that initiates the apopototic response when DNA damage exceeds the capacity of repair machinery, and BAD, another protein involved in regulation of apoptosis (Won et al. 2003). The BAD 14-3-3 complex is cleaved by caspase 3 and the released BAD moves to mitochondria and releases cytochrome C to initiate apoptosis and interacts with Bcl-xL, reducing its anti-apoptotic effect (Won et al. 2003). Interestingly, Bcl-xL expression is promoted by nucleolin via binding to Bcl-xL mRNA that is dependent polyadenylate tail and polyadeylate binding protein 1 (Zhang et al. 2008).

Mapping DNA-PK phosphorylation sites in Hlx

In the previous published experiments, we have determined that the phosphorylation of linker histone variants has a functional impact on NHEJ (Kysela et al., 2005). The exact composition of the linker histone mix used in these studies was not known, therefore the exact target for DNA-PK phosphorylation could be any one or a combination of the 7 somatic linker histone variants. The direct interaction of Hlx protein with DNA-PK complex suggested the possibility that Hlx function may be regulated by the phosphorylation mediated by DNA-PK. DNA-PK has also been shown to phosphorylate several members of the NHEJ pathway including the KU complex, XRCC4, XLF and Artemis in vitro and in vivo (Cao et al, 1994, Douglas et al., 2005, Ma et al., 2002, Yu et al., 2008, Yu et al., 2003c). Using site directed mutagenesis, putative phosphorylation sites in XRCC4 and XLF were removed, however, there was no loss-of-NHEJ-function (Yu et al., 2008, Yu et al., 2003c). Interestingly however, the phosphorylation activity of DNA-PK has been shown to be crucial for NHEJ activity, highlighting the importance in discovering phosphorylation targets in vivo (Kurimasa et al., 1999). The ATM/DNA-PK phosphorylation consensus sequence has been identified as S/T-Q (S. T. Kim et al., 1999). All somatic linker histone variants were examined by bioinformatics sequence analysis (not shown) to determine if any of these contained putative S/T-Q sites. Only two of the linker histone variants, H1.0 and Hl .X, contained an SQ motif. Interestingly, the serine residue of the SQ motif in H1.0 is conserved throughout all the linker histone variants with the exception of Hl .X, whereas, the serine residue of the SQ motif in Hl .X is conserved in all linker histones variants. In both cases the glutamine residue following the serine residue is not conserved at all.

Conservation of these SQ sites in Hl .X and H1.0 were then studied across different species. The SQ site in Hl.X is conserved from Homo sapiens through to Gallas gallas, along with a surrounding 15 amino acid region (not shown). However, the site is not conserved in Danio rerio with the glutamine residue being replaced by a lysine residue. The full length Hl .X protein is conserved 94% and 91% with Pan troglodytes and Bos taurus respectively, however it is much less conserved with other species. To identify potential DNA-PK phosphorylation sites on Hlx, a recombinant purified linker histone Hlx was used as a substrate for DNA-PK in an in vitro kinase assay. Protein has been resolved on an SDS-PAGE gel and stained with PageBlue. The corresponding band was excised from the SDS-PAGE gel and phospho-peptide mapped by Functional Genomics (University of Birmingham) using the collision induced dissociation technique. Whilst this technique confirms the presence of a modification, it does not necessarily determine which residue is modified (Sweet et al., 2006). Therefore all phosphorylatable residues within a peptide are listed as potential sites of phosphorylation. Three novel potential phosphorylation sites had been identified in Hlx: pY-48 (tyrosine), pS-49 (serine) pT-55 (threonine). In order to determine the exact phosphorylation site of DNA-PK, the serine 49 had been changed to alanine by site-directed mutagenesis and the mutation carrying recombinant protein subjected to in vitro kinase assay (Figure 4). The results show clearly that Hlx is exclusively phosphorylated by DNA-PK at the S49.

References:

References

1. Risinger, M. A., Groden, J. Crosslinks and crosstalk: human cancer syndromes and DNA repair defects. Cancer Cell, 6, 539-545, (2004).

2. Girard, P. M, Kysela B. Harer, C. J. Doherty A. J. Jeggo P. A. Analysis of DNA ligase IV mutations found in LIGIV syndrome patients: the impact of two linked polymorhisms. Hum Mol Genet. 13, 2369-2376, (2004).

3. Weller, G. R., Kysela, B., Roy, R., Tonkin, L. M., Scanlan, E., Delia, M, Devine, S.

K., Day, J. P., Wilkinson, A., di Fagagna, F., Devine, K. M., Bowater, R. P., Jeggo, P. A., Jackson, S. P., Doherty A. J. Identification of a DNA nonhomologous end-joining complex in bacteria. Science, 297, 1686-1689, (2002). Godde, J. S., Ura, K. Cracking the enigmatic linker histone code. J. Biochem. 143, 287-293, (2008). Proc. Natl. Acad. Sci. USA, 102, 1877-1882, (2005). Kysela, B., Chovanec, M, Jeggo, P. A. Phosphorylation of linker histones by DNA- dependent protein kinase is required for DNA ligase IV-dependent ligation in the presence of histone HI. Ausio, J. Histone variants - the structure behind the function. Brief Funct Genomic Proteomic, 5, 228-243, (2006). Brown, D.T. Histone HI and the dynamic regulation of chromatin function.

Biochem Cell Biol, 81, 221-227, (2008). Alexandrow, M.G., Hamlin, J. L. Chromatin decondensation in S-phase involves recruitment of Cdk2 by Cdc45 and histone HI phosphorylation. J Cell Biol, 168, 875- 886, (2005). Downs, J. A., Kosmidou, E., Morgan, A., Jackson, S. P. Suppression of homologous recombination by the Saccharomyces cerevisiae linker histone. Mol Cell, 11,1685- 1692 (2003). Warneboldt, J., Haller, F., Horstmann, O., Danner, B.C., Fiizesi, L., Doenecke, D., Happel, N. Histone Hlx is highly expressed in human neuroendocrine cells and tumours. BMC Cancer, 8, 388, (2008). Baumann, M., Krause, M., Hill, R. Exploring the role of cancer stem cells in radioresistance. Nat Rev Cancer,7, 45-554, (2008).

Palapattu, G.S., Wu, C, Silvers, C.R., Martin, H.B., Williams, K., Salamone, L., Bushnell, T., Huang, L.S., Yang, Q., Huang, J. Selective expression of a CD44, a putative prostate cancer stem cell marker, in neuroendocrine tumour cells of human prostate cancer. Prostate, 69, 787-798, (2009). Semizarov, D., Frost, L., Sarthy, A., Kroeger, P., Halbert, D. N., Fesik, S. W. Specificity of short interfering RNA determined through gene expression signatures. Proc. Natl. Acad. Sci. USA 100, 6347-6352, (2003). Murga, M., Jaco, I., Fan, Y., Soria, R., Martinez-Pastor, B., Cuadrado, M., Yang, S.M, Blasco, M.A., Skoultchi, A.I., Fernandez-Capetillo, O. Global chromatin compaction limits the strength of the DNA damage response. J Cell Biol, 178, 1101-1108, (2007). Takata, H., Matsunaga, S., Morimoto, A., Ono-Maniwa, R., Uchiyama S, Fukui, K. Hl .X with different properties from other linker histones is required for mitotic progression. FEBS Lett, 581, 3783-3788, (2007). Zhao, FL, Watkins, J. L., Piwnica- Worms, H. Disruption of the checkpoint kinase 1/cell division cycle 25 A pathway abrogates ionizing radiation-induced S and G2 checkpoints. Proc. Natl. Acad. Sci. USA. 99, 14795-14800, (2002). Rainey, M. D., Black, E. J., Zachos, G., Gillespie, D. A. Chk2 is required for optimal mitotic delay in response to irradiation-induced DNA damage incurred in G2 phase. Oncogene, 27, 896-906, (2008). Rogakou, E. P., Boon, C, Redon, C, Bonner, W. M. Megabase chromatin domains involved in DNA double-strand breaks in vivo. J Cell Biol, 146, 905-16, (1999). Rothkamm, K., Lobrich, M., Evidence for a lack of DNA double-strand break repair in human cells exposed to very low x-ray doses. Proc Natl Acad Sci U S A, 100, 5057-5062, (2003). Stewart, G. S., et al. The RIDDLE syndrom protein mediates a ubiquitin-dependent signalling cascade at sites of DNA damage. Cell, 136, 420-434, (2009). Stold, S., Wenzel, D., Schulze E., Doenecke D., Happel, N. Gl phase-dependent nucleolar accumulation of human histone Hlx. Biol Cell, 99, 541-552, (2007). Franken, N. A., Rodermond, H. M., Stap, J., Haveman, J. vn Bree, C. Clonogenic assays of cells in vitro. Nat Protoc, 1,2315-2319 (2006) Riballo, E., Doherty A. J., Stiff, T., Oettinger, M. A., Jeggo P. A., Kysela, B. Cellular and biochemical impact of a mutation in DNA ligase IV conferring clinical radiosensitivity. J Biol Chem, 276, 31 124-31132 (2001). Henges, P., Ahnesorg, P., Pitcher, R. S., Bruce, C. K., Kysela, B., Green, A. J., Bianchi, J., Wilson, T. E., Jackson, S. P., Doherty, A. J. Evolutionary and functional conservation of the DNA non-homologous end-joining protein, XLF/Cernunnos. J Biol Chem, 281, 37517-37526 (2006). 25. Borer, R.A., Lehner, C.F., Eppenberger, H.M., et al. (1989) Major nucleolar proteins shuttle between nucleus and cytoplasm. Cell, 56: (3): 379-390. Cao, Q.P., Pitt, S., Leszyk, J., et al. (1994) DNA-dependent ATPase from HeLa cells is related to human Ku autoantigen. Biochemistry, 33: (28): 8548-8557. De, A., Donahue, S.L., Tabah, A., et al. (2006) A novel interaction of nucleolin with Rad51. Biochem Biophys Res Commun, 344: (1): 206-213. Hanakahi, L.A., Bu, Z. and Maizels, N. (2000) The C-terminal domain of nucleolin accelerates nucleic acid annealing. Biochemistry, 39: (50): 15493-15499. Hovanessian, A.G., Puvion-Dutilleul, F., Nisole, S., et al. (2000) The cell-surface- expressed nucleolin is associated with the actin cytoskeleton. Exp Cell Res, 261 : (2): 312-328. Hunt CR, Dix DJ, Sharma GG, Pandita RK, Gupta A, Funk M, Pandita TK (2004) Genomic instability and enhanced radiosensitivity in Hsp70.1- and Hsp70.3-deficient mice. Mol Cell Biol. 24: (2): 899-911. Kim, S.T., Lim, D.S., Canman, C.E., et al. (1999) Substrate specificities and identification of putative substrates of ATM kinase family members. J Biol Chem, 274: (53): 37538-37543. Kotoglou P, Kalaitzakis A, Vezyraki P, Tzavaras T, Michalis LK, Dantzer F, Jung JU, Angelidis C (2009) Hsp70 translocates to the nuclei and nucleoli, binds to XRCC1 and PARP-1, and protects HeLa cells from single-strand DNA breaks. Cell Stress Chaperones, 14: (4): 391-406 Kurimasa, A., Kumano, S., Boubnov, N.V., et al. (1999) Requirement for the kinase activity of human DNA-dependent protein kinase catalytic subunit in DNA strand break rejoining. Mol Cell Biol, 19: (5): 3877-3884. Kysela, B., Chovanec, M. and Jeggo, P.A. (2005) Phosphorylation of linker histones by DNA-dependent protein kinase is required for DNA ligase IV-dependent ligation in the presence of histone HI . Proc Natl Acad Sci U S A, 102: (6): 1877-1882. Ma, Y., Pannicke, U., Lu, H., et al. (2005) The DNA-dependent protein kinase catalytic subunit phosphorylation sites in human Artemis. J Biol Chem, 280: (40): 33839-33846. Son YM, Lee JH, Kim DR (2008) Modulation of RAG7DNA complex by HSP70 in V(D)J recombination. Biochem Biophys Res Commun. 365: (1): 113-1 17. Sweet, S.M., Creese, A.J. and Cooper, H.J. (2006) Strategy for the identification of sites of phosphorylation in proteins: neutral loss triggered electron capture dissociation. Anal Chem, 78: (21): 7563-7569. Ting NSY, Kao PN (1998) DNA dependent protein kinase interacts with antigen response element binding proteins NF90 and NF45. J Biol Chem, 273: (4): 2136-2141. Won J, Kim DY (2003) Cleavage of 14-3-3 protein by Caspase-3 facilitates BAD interaction with Bcl-xL during apoptosis. J Biol Chem, 278: (21): 19347. Yang, C, Kim, M.S., Chakravarty, D., et al. (2009) Nucleolin Binds to the Proliferating Cell Nuclear Antigen and Inhibits Nucleotide Excision Repair. Mol Cell Pharmacol, 1 : (3): 130-137. Yu, Y., Mahaney, B.L., Yano, K., et al. (2008) DNA-PK and ATM phosphorylation sites in XLF/Cernunnos are not required for repair of DNA double strand breaks. DNA Repair (Amst), 7: (10): 1680-1692. Yu, Y., Wang, W., Ding, Q., et al. (2003c) DNA-PK phosphorylation sites in XRCC4 are not required for survival after radiation or for V(D)J recombination. DNA Repair (Amst), 2: (1 1): 1239-1252. Zhang J, Tsaprailis G (2008) Nucleolin stabilizes Bcl-xL messanger RNA in response to UVA radiation. Cancer Res, 68: (4): 1046.

Claims

1. A method of increasing the sensitivity of a cell to DNA double strand break (dsb) damage, comprising decreasing levels of functional histone Hlx in the cell.

2. A method according to claim 1, wherein the Hlx production is inhibited in the cell with using RNAi.

3. A method according to claim 2, wherein the RNAi is siRNA.

4. A method according to claim 1, comprising binding functional Hlx to Hlx- specific antibodies, or fragments thereof.

5. A method according to claim 1, wherein the dsb damage is induced into the DNA of the cell by ionising radiation or chemically induced into the DNA of the cell.

6. A method according to claim 1, wherein the ionising radiation is selected from γ- radiation, x-rays and protons from proton therapy.

7. A method according to claim 1, additionally comprising adding to the cell a radiosensitiser and/or a hypoxia cytotoxin.

8. A radiosensitive cell having decreased levels of functional histone Hlx compared to normal cells.

9. A cell according to claim 9, comprising therein RNAi, such as siRNA, or anti-Hlx antibody or fragment thereof.

10. A cell according to claim 8 or claim 9, additionally comprising a radiosensitiser or hypoxic cytotoxin.

11. An siR A comprising a nucleotide sequence complementary to a portion of a nucleotide sequence encoded by Hlx mR A and a moiety targeting for targeting a cancer cell.

12. An siR A according to claim 11, comprising a moiety capable of binding a receptor on a cancer cell.

13. An siRNA according to claims 11 to 12, in combination with a pharmaceutically acceptable carrier.

14. An anti-Hlx antibody, or fragment thereof, capable of specifically binding to Hlx, for use in a method according to claims 1 to 8.

15. A pharmaceutical formulation comprising a siRNA or anti-Hlx antibody or fragment, according to claims 11 to 14.

16. An inhibitor of Hlx function for use in the treatment of cancer in combination with a DNA double strand breaking agent.

17. An inhibitor according to claim 16 which is an siRNA or antibody or fragment thereof, according to claims 11 to 15.

18. An inhibitor according to claim 16, wherein the DNA double strand breaking agent is ionising radiation.

19. An inhibitor according to claims 16 to 18, in combination with a radiosensitiser or hypoxic cytotoxin.