Thiopurine treatment of human leukemia cells deficient in components of the mismatch repair system (Nalm6) initiated apoptosis after incorporation into DNA, as revealed by caspase activation and terminal deoxynucleotidyl transferase-mediated nick end labeling assay. To elucidate the cellular sensor(s) responsible for recognition of DNA damage in cells with an inactive mismatch repair system, we isolated a multiprotein nuclear complex that preferentially binds DNA with thioguanine incorporated. The components of this nuclear multiprotein complex, as identified by protein mass spectroscopy, included high mobility group proteins 1 and 2 (HMGB1, HMGB2), heat shock protein HSC70, protein disulfide isomerase ERp60, and glyceraldehyde 3-phosphate dehydrogenase. The same complex was also shown to bind synthetic oligodeoxyribonucleotide duplexes containing the nonnatural nucleosides 1-β-d-arabinofuranosylcytosine or 5-fluoro-2′-deoxyuridine. Fibroblast cell line derived from Hmgb1−/− murine embryos had decreased sensitivity to thiopurines, with an IC50 10-fold greater than Hmgb1-proficient cells (P < 0.0001) and exhibited comparable sensitivity to vincristine, a cytotoxic drug that is not incorporated into DNA. These findings indicate that the HMGB1-HMGB2-HSC70-ERp60-glyceraldehyde 3-phosphate dehydrogenase complex detects changes in DNA structure caused by incorporation of nonnatural nucleosides and is a determinant of cell sensitivity to such DNA modifying chemotherapy.

Induction of apoptosis in malignant cells by damaging DNA is a common mechanism of cancer chemotherapy (1). DNA lesions caused by irradiation or chemical modification of DNA, trigger the apoptotic machinery and thereby eliminate cells with competent mechanisms for recognizing DNA damage and carrying out apoptosis.

Treatment of cells with DNA-reactive compounds (e.g., cisplatin) or structural analogs of DNA precursors (e.g., cytarabine, fludarabine, and thiopurines) results in growth arrest and apoptosis (2, 3, 4). Although major components of the apoptotic machinery have been elucidated, the molecular interaction between damaged DNA and cellular processes that initiate apoptosis remain obscure. Recent studies demonstrated that distinct structural elements of damaged DNA provide the sites for binding of DNA damage-recognizing proteins and subsequent assembly of multiprotein complexes, which normally maintain the integrity and fidelity of genetic information (4, 5, 6). The postreplicative MMR3 system is another example of protein complexes involved in recognition of DNA damage after anticancer therapy, and the binding of MMR proteins to modified DNA has been shown in in vitro experiments (7, 8).

Cytotoxic effect of thiopurines, a class of widely used antileukemic agents (9), is achieved through a common mechanism that exploits the conversion of these drugs into deoxythioguanosine triphosphate, with subsequent incorporation into DNA (8, 10, 11). In contrast to γ-irradiation, such incorporation induces local structural changes in DNA (12) rather than destroy its integrity and affects DNA-protein interactions, including topoisomerase II and RNase H (13, 14). This mechanism may also contribute to the association of thiopurine treatment with an increased risk of therapy-related leukemia or brain tumors, when thiopurines are used concurrently with topoisomerase II inhibitors or irradiation (15, 16).

MMR proteins have been demonstrated to modulate thiopurine cytotoxicity, and the loss of MMR activity can confer a resistance phenotype to thiopurine treatment (17). However, several human and murine cells deficient in MSH2, a key component of MMR, demonstrated high sensitivity to anticancer drugs, including thiopurines (13, 18). These paradoxical observations indicate the existence of complementary biochemical pathways that activate cell death in response to DNA damage in MMR-deficient cells.

To further understand the mechanisms by which thiopurines exert their cytotoxic effects, we investigated how incorporation of thiopurines into DNA is recognized by cellular machinery in human MSH2-deficient acute lymphoblast leukemia cells. Characterization of this process and identification of sensor proteins in these cells will provide new insights into the determinants of drug sensitivity in cancer cells and may facilitate the development of more effective and less toxic anticancer regimens. Here we present new findings indicating that DNA surveillance mechanisms identify chemical modification of DNA by detecting nonnatural nucleosides with an HMGB1-containing multiprotein complex. Furthermore, we report the isolation and identification of components of a nuclear multiprotein complex that has high affinity to duplex DNA modified by incorporation of dGS, ara-C, or 5FdU and show that murine fibroblasts deficient in one component of the complex, Hmgb1, are 10-fold more resistant to thiopurine treatment than their wild-type counterparts.

Cell Culture and Preparation of Nuclear Extracts.

The human B-lineage leukemia cell line Nalm6 was obtained from the DSMZ-German Collection of Microorganisms and Cell Cultures (Braunschweig, Germany). Cells were grown in complete RPMI 1640 (BioWhittaker, Walkersville, MD), which contained 10% fetal bovine serum and 2 mml-glutamine, at 37°C in an atmosphere of 5% CO2. The doubling time of Nalm6 cells under these conditions was ∼19 h. The number and viability of the cells were determined in duplicate by trypan-blue exclusion, using a Burker-Turk chamber. In all experiments, the initial concentration of cells was 0.25 × 106/ml. A single dose of 6-MP, which was dissolved in medium, was added to the culture medium so that the final concentration was 10 μm (determined by measurement of the absorbance of the medium; wavelength, 320 nm; extinction coefficient, 19,600). Cytotoxicity was expressed as the ratio of the number of viable cells incubated with MP (i.e., cell count × percentage of viable cells) to the number of viable cells in the control medium, which lacked MP (i.e., cell count × percentage of viable cells). Nuclear extracts were prepared according to the method of Dignam et al.(19) Protein concentration was determined by Bradford dye-binding procedure using Bio-Rad Protein assay (Bio-Rad, Hercules, CA).

Cell Cycle Synchronization.

Nalm6 cells were seeded at a concentration of 1 × 106/ml in complete RPMI 1640 that contained 2 mm thymidine (Sigma, St. Louis, MO). The culture was then incubated at 37°C for 16 h. Cells were washed twice with PBS and centrifuged at 500 × g for 10 min at room temperature. Cells were resuspended in 200 ml of RPMI 1640, and 2 h after the removal of thymidine, MP was added to 20 ml of the cell suspension so that the final concentration of the drug was 10 μm.

DNA Modification.

The level of dGS incorporation into DNA after MP treatment was determined by high-performance liquid chromatography analysis by methods we have previously described in detail (13).

Cell Cycle Analysis, TUNEL Assay, and Caspase Activation Assay.

For cell cycle analysis, cells were centrifuged and resuspended at a concentration of 1 × 106/ml in a PI staining solution (0.05 mg/ml PI, 0.1% sodium citrate, 0.1% Triton X-100). Each sample was treated with DNase-free RNase (5 ng/ml; Calbiochem, San Diego, CA) at room temperature for 30 min, filtered through 40-μm mesh, and analyzed by a FACScan flow cytometer (Becton Dickinson, San Jose, CA) collecting the fluorescence (wavelength range, 563–607 nm) from PI-bound DNA (∼15,000 cells). The percentages of cells within each phase of the cell cycle were computed by using ModFit software (Verity Software House, Topsham, ME).

For TUNEL analysis of apoptosis, 1 × 106-3 × 106 fixed cells were end labeled with brominated dUTP (BD Biosciences, San Jose, CA) by using TdT (Roche Diagnostics, Indianapolis, IN). The samples were then incubated with FITC-conjugated anti-BrdUrd monoclonal antibody (BD Biosciences) for 1 h at room temperature. After cellular DNA was counterstained with PI, the cells were treated with RNase and analyzed by a FACS Calibur flow cytometer (Becton Dickinson). Matched aliquots of each sample were treated identically, except one aliquot of each pair was not incubated with TdT during the end-labeling procedure. Thus, aliquots that lacked TdT served as an internal negative control for FITC fluorescence of each sample. Cells exhibiting FITC fluorescence greater than background levels were classified as apoptotic.

For assays of caspase activation, ∼3 × 105 cells were incubated with the caspase inhibitor FAM-VAD-FMK according to the manufacturer’s protocol (CaspaTag Fluorescein Caspase Activity Kit; Intergen, Purchase, NY) to stain activated caspases 1 through 9. After the cells were counterstained with PI, they were analyzed by a FACS Calibur instrument (Becton Dickinson).

Synthesis of Duplex DNA.

The synthesis of modified oligodeoxyribonucleotides containing dGS was accomplished by standard phosphoramidite chemical methods with S6-DNP-dG-CE phosphoramidite (13). Oligodeoxyribonucleotides containing ara-C and 5FdU were synthesized and purified by TriLink Biotechnologies (San Diego, CA). The 5′ ends of the single-stranded oligodeoxyribonucleotides were radiolabeled by using the RTS T4 Kinase Labeling System (Life Technologies, Inc., Gaithersburg, MD). Sequences of the oligodeoxyribonucleotide duplexes are shown in Table 1. Oligodeoxyribonucleotide duplexes were prepared by annealing complementary single strands and were purified by nondenaturing gel electrophoresis (12% polyacrylamide gel) at 4°C as previously described (13) or by Fast Protein Liquid Chromatography using a Superdex 75 column (Amersham Pharmacia Biotech, Piscataway, NJ).

Electrophoretic Mobility Shift Assay and Isolation of a Nucleoprotein Complex.

DNA binding assays were performed in a 20–30-μl solution containing buffer [20 mm Tris-HCl (pH 8.0); 50 mm NaCl; 3 mm MgCl2; 1 mm DTT; 10% glycerol; 10 ng/μl poly(deoxyinosinic-deoxycytidylic acid)] as previously described (8). For the competition assay, 0–225 nm of unlabeled competitor were preincubated with nuclear extract (0–0.8 μg/μl total protein) in the presence of 2.5 mm ATP for 5 min at room temperature. Then, P32-labeled oligo duplex was added to final concentration 10 nm, and the reaction mixture was incubated for 15 min at room temperature. After addition of loading buffer containing 40% sucrose and 0.04% bromphenol blue, the samples were loaded onto 6% polyacrylamide gel and run for 2.5 h at 330 V (+4°C). In a supershift assay, reaction mixture was prepared as described above. After 10 min of incubation at room temperature, 5 μl (2.5 μg) of anti-HMGB1 polyclonal antibody was added, and the reaction mixture was incubated for additional 5 min before loading on the gel. For quantitative measurement of bound oligonucleotide, the gels were dried and autoradiographed in a PhosphoroImager cassette overnight. The signal intensities were normalized versus total radioactivity in the lane.

Radioactive fractions corresponding to the DNA-protein complex were electroeluted from the gel and analyzed in a 7.5% SDS-polyacrylamide gel. The bands containing the proteins of interest were excised from the gel, and the proteins were digested with trypsin and subsequently identified by mass spectroscopy, which was performed by Protana (Odense, Denmark).

Immunological Methods.

Polyclonal rabbit antibodies against human HMGB1 and HMGB2 (PharMingen, San Diego, CA) were used at a dilution of 1:5000 and 1:200, respectively, for Western blot analysis. Polyclonal rabbit antibody against the keyhole limpet hemocyanin-coupled peptide 491–505 of ERp60 (SwissProt no. P30101) was raised and affinity-purified by Rockland, Inc. (Gilbertsville, PA), and used at a 1:5000 dilution. Polyclonal goat anti-HSC70 antibody (Santa Cruz Biotechnology, Santa Cruz, CA) was used at a 1:100 dilution. Monoclonal antibody to GAPDH (MAB 374; Chemicon, Temecula, CA) was used at a dilution of 1:500. Secondary horseradish peroxidase-conjugated antibodies were used at dilutions ranging from 1:5000 to 1:10,000. The ECL-Plus Western blotting detection system (Amersham Pharmacia Biotech) was used to detect antibodies bound to the Western blots.

EMSA-Western Blot Analysis.

After completion of EMSA electrophoresis in nondenaturing conditions as described in a preceding section, the proteins were transferred to a Hybond-P membrane (Amersham Pharmacia Biotech) under which a sheet of DE-81 paper (Whatman, Maidstone, England) was placed, as described previously (20). After electrotransfer, the membrane was incubated with the antibodies to the proteins that comprised the dGS-DNA-indexing complex. Bound antibodies were detected by using secondary antibodies and the ECL-Plus kit (Amersham Pharmacia Biotech). The radiolabeled DNA bound to the DE-8l paper was detected by exposing the dried paper in a PhosphorImager cassette. For Western analysis of proteins in supershift experiments, radioactive bands were excised, incubated at 70°C for 30 min in NuPage LDS Sample buffer (Invitrogen, Carlsbad, CA), and the supernatant was loaded on the SDS-containing gel.

Immunoprecipitation.

The anti-HMGB1 antibody (300 μg) was cross-linked to protein G conjugated to Sepharose (Pierce, Rockford, IL). This complex was used with the Seize X Mammalian Immunoprecipitation kit (Pierce) to immunoprecipitate cellular or nuclear proteins from Nalm6 cell (50 mg of cells). Immobilization of antibody, loading of the cell lysate, and washing was performed according to manufacturer’s instructions. Eluted proteins were further separated using 4–12% NuPage Bis-Tris gel in MES buffer (Invitrogen). After electrophoresis, the proteins were transferred to a PVDF membrane for Western blot analysis.

MTT Cytotoxicity Assays.

Fibroblast cell lines deficient and proficient in Hmgb1 expression were generated as described previously (21). Thiopurine cytotoxicity was determined using the MTT assay after incubation of C1 (Hmgb1−/−) and VA1 (Hmgb1+/+) cells with MP or vincristine for 3–6 days (22). Thiopurines were dissolved in DMEM without fetal bovine serum, and the drug concentration was determined spectrophotometrically, as described earlier (8). The IC50s were obtained by fitting a sigmoid Emax model to the cell viability versus drug concentration data, determined in triplicate.

S-Phase Arrest and Apoptosis of MP-treated Synchronized Cells.

Untreated (control) synchronized Nalm6 cells progressed through the first and second cell cycles after release from thymidine block (Fig. 1,A) and were no longer synchronized by the end of the second cell cycle. When Nalm6 cells were treated with 10 μm MP (IC50 = 3.7 ± 0.8 μm after 48 h), 2 h after the block was released, cells progressed through the cell cycle during the first 18 h after the addition of MP (Fig. 1,A, 18 h). After beginning the second cell cycle, additional progression of treated cells ceased, and arrest occurred during S phase (Fig. 1 A, 34 h).

TUNEL assay revealed that up to 12% of synchronized MP-treated cells accumulated DNA strand breaks after 34 h of incubation (i.e., in the second cycle) with MP (Fig. 1,B). Consistent with this finding, FACScan analysis demonstrated activation of caspases in ∼20% of MP-treated cells, compared with 4% in untreated cells (Fig. 1,C). Caspase activation occurred simultaneously with accumulation of cells in S phase and the increase of sub-G1 fraction (up to 14–17%) in synchronized Nalm6 cells treated with MP. Interestingly, both accumulation of DNA strand breaks and caspase activation took place only after treated cells entered S phase of the second cycle (Figs. 1, A–C and 2 A).

Continuation of DNA Synthesis in Synchronized Nalm6 Cells Treated with MP.

Pulse-chase experiments showed that the level of BrdUrd incorporation into DNA was similar in untreated and MP-treated cells, indicating that DNA synthesis was not abrogated by MP treatment (Fig. 2, A and B). dGS incorporation (0.06%, about one dGS/1700 dG) was detected in MP-treated cells after completion of the first cycle and increased to 0.4% during the second cycle (one dGS/250 dG; Fig. 2,C), consistent with our previous findings in unsynchronized cells (13). The level of dGS incorporated into DNA increased during the second cycle of cell division in parallel with the relative number of cells in S-phase, also showing that DNA synthesis was not abrogated (Fig. 2, A–C).

High Affinity of Nuclear Proteins to dGS-DNA Duplex.

High affinity of nuclear extract proteins from mismatch-deficient lymphoblastic cells to dGS-DNA was demonstrated by EMSA experiments with 34-mer oligo duplexes. Titration of the complex with a competitor oligo duplex that had the identical sequence, except contained only natural nucleosides, demonstrated stability of the complex in the presence of 4–5-fold excess of competitor (Fig. 3, A and B). Further addition of competitor resulted in a decrease of the radiolabeled complex because of displacement of the labeled dGS-DNA duplex from the complex with nonlabeled competitor. Therefore, subsequent EMSA experiments were performed in the presence of 5-fold excess of nonradioactive competitor. An increase of the protein concentration in the reaction mix led to a linear increase of band intensity, further confirming the specificity of complex formation (Fig. 3, C and D).

Identification and Characterization of a dGS-DNA-Protein Complex.

A dGS-DNA-protein complex, formed in the presence of radiolabeled 34-mer DNA duplexes (see Table 1) containing one modified base, was isolated from Nalm6 nuclear extract by EMSA electrophoresis (Fig. 4,A). Analysis by SDS-PAGE revealed that the complex contained four or five major proteins (Fig. 4 B). Mass spectroscopy established that the protein with a molecular mass of 70 kDa was human HSC70 (SwissProt no. P11142; total coverage: 389/646 amino acids, 60%; average mass, 71,126); the protein with a molecular mass of 58 kDa was the putative ERp60 (SwissProt no. P30101; total coverage: 225/505 amino acids, 45%; average mass, 57,182); the protein with molecular mass of 25 kDa was human HMG protein 1 HMGB1 (SwissProt no. P09429; total coverage: 109/214 amino acids, 51%; average mass, 24,933); the protein with a molecular mass of 23 kDa was human HMGB2 (SwissProt no. P26583; total coverage: 145/208 amino acids, 70%; average mass 24,073); and the protein with a molecular mass of 37 kDa was GAPDH (SwissProt no. P04406; total coverage: 77/334 amino acids, 23%; average mass, 36,093). Identification of GAPDH as a component of the complex corroborates our earlier findings obtained by NH2-terminal sequencing (8).

Additional confirmation of the identity of the proteins was provided by the results of EMSA in the presence of antibodies specific to identified proteins. Two of the antibodies gave a notable effect: anti-HMGB1 antibody resulted in a supershift of the dGS-DNA-protein complex (Fig. 4 A, Lanes 1 and 2); and anti-GAPDH antibody abrogated formation of the complex, as reported earlier (8). Treatment with antibodies to HMGB2, ERp60, and HSC70 did not change mobility of the complex in EMSA.

In a separate set of EMSA experiments, the radioactive band formed after the treatment of the reaction mixture with anti-HMGB1 antibody (Fig. 4,A, Lane 2, top band) was excised and subjected to Western blot analysis. The presence of HMGB2, GAPDH, and HSC70 proteins in the excised (supershift) band was confirmed based on immunoreactivity to the antibodies and molecular weights of the proteins, as depicted in Fig. 4 C. We were unable to unambiguously identify ERp60 in this band because of cross-reactivity of the secondary antibody to anti-HMGB1 antibody and anti-ERp60 antibody. Identities of the major components of the dGS-DNA-protein complex (HMGB1, HMGB2, HSC70, ERp60, GAPDH) were also confirmed by Western analysis after the transfer of the proteins from the EMSA gel onto a PVDF membrane, with the radiolabeled oligonucleotide duplex retained on the ion-exchange paper underlying the PVDF membrane (data not shown).

Coimmunoprecipitation experiments were performed using anti-HMGB1 antibody. HMGB1 was isolated from the nuclear extract of untreated Nalm6 cells by affinity chromatography using anti-HMGB1 antibody immobilized on protein G-conjugated Sepharose. Western blot analysis of the bound proteins showed that three components of the complex (i.e., HMGB2, ERp60, HSC70) coimmunoprecipitated with HMGB1 (Fig. 5, Lanes 2–5). We did not detect GAPDH in the coimmunoprecipitated fraction.

Incubation of the nuclear extract with synthetic duplexes containing the modified nucleosides ara-C or 5FdU (Table 1) resulted in protein binding similar to that observed when the extract was incubated with dGS-containing DNA duplexes (Fig. 6,A, Lanes 7 and 9). Again, addition of anti-HMGB1 antibody to the reaction mixture resulted in supershift of the complex, suggesting that the same complex was involved in the binding of dGS-, araC-, and 5FdU-containing oligonucleotide duplexes (Fig. 6,A, Lanes 8 and 10). Comparison of band intensities indicated that the protein complex has ∼4–5-fold greater affinity for chemically modified oligonucleotide duplexes than for duplex DNA containing only natural nucleosides (Fig. 6 B).

Thiopurine Resistance of Hmgb1-deficient Murine Fibroblasts.

Immortalized fibroblast cell lines established from Hmgb1−/− and wild-type (Hmgb1+/+) murine embryos (21) were used to assess thiopurine sensitivity by MTT assay. Estimation of cell viability in C1 (Hmgb1−/−) and VA1 (Hmgb1+/+) cells after MP treatment revealed that after 5 days of incubation, Hmgb1−/− cells were ∼10-fold more resistant to MP treatment, compared with Hmgb1+/+ fibroblasts (IC50C1 = 8.9 ± 0.7 μm and IC50VA1 = 0.78 ± 0.11 μm, P = 0.0001; Fig. 7). In contrast, C1 (Hmgb1−/−) and VA1 (Hmgb1+/+) cells demonstrated similar viability after 5 days of incubation with vincristine (IC50C1 = 0.28 ± 0.032 μm; IC50VA1 = 0.39 ± 0.094 μm), a cytotoxic antileukemic agent that is not incorporated into DNA.

The thiopurine drugs MP, thioguanine, α-2′-deoxythioguanosine, and β-2′-deoxythioguanosine form a tight cluster when grouped according to their activity or gene expression profiles in 60 sensitive and resistant human cancer cell lines, implying that thiopurines have a similar molecular mechanism of cytotoxicity (11). Although the MMR system was shown to be an important determinant of thiopurine sensitivity in several types of cells (17), our prior studies have shown that MMR-deficient human leukemia cells can be highly sensitive to thiopurine treatment (8, 13). This observation led to our current studies to elucidate alternative mechanisms by which cells respond to the DNA modification after thiopurine treatment. To this end, we used nuclear extracts from human leukemia cells (i.e., Nalm6) that are highly sensitive to thiopurines (IC50 = 3.7 ± 0.8 μm after 48 h exposure to MP) but do not have detectable levels of MSH2 or MSH6, essential components of the MMR system (8).

Treated cells progressed successfully through G2-M of the first cell cycle and G1 phase of the second cell cycle, without cycle arrest (Fig. 1). During the second cell cycle, additional dGS was incorporated into DNA, a result consistent with formation and accumulation of deoxythioguanosine triphosphate within cells (23) and with continued DNA synthesis. As Fig. 1 and 2, A and B, depict, thiopurine metabolites neither inhibit DNA replication nor activate checkpoints that determine entrance into S phase. Moreover, DNA synthesis was not abrogated, as evidenced by the incorporation of BrdUrd and dGS into DNA (Fig. 2, B and C). However, thiopurine-treated cells did not progress beyond the second S phase and did not accomplish duplication of DNA (Fig. 1, 34 h +MP).

Interestingly, apoptotic changes in MP-treated cells became prominent only during the second S phase, after incorporation of dGS into DNA. Both activation of caspases and accumulation of DNA breaks occurred, indicating activation of the apoptotic cascade in the second cell cycle (Fig. 1, B and C). These results are consistent with earlier findings in nonhuman CHO and L1210 cell lines, which demonstrated accumulation of DNA breaks in the second cycle after thiopurine treatment (24, 25). The fact that initiation of apoptosis occurs in the second S phase after the incorporation of dGS into DNA suggests existence of cycle-specific mechanism(s) that are sensitive to such DNA modification.

On the basis of the above findings in MMR-deficient cells, we hypothesized that chemically modified DNA with dGS-inserts must be a substrate for a protein(s) that recognizes DNA distortion rather than formation of specific mispairs. Formation of DNA-protein complex with such altered DNA may be considered as the initial step toward activation of DNA repair or apoptosis (26, 27, 28). Structural analysis of dGS-containing duplex DNA has demonstrated local changes in geometry and dynamics of the double helix around the site of dGS incorporation (12), without loss of structural integrity of the DNA duplex. Thus, the protein(s) recognizing dGS-modified DNA is detecting relatively modest and localized changes in DNA structure.

Synthetic DNA duplexes containing nonnatural nucleosides provide a useful tool to characterize cellular components sensitive to DNA modification (29). We used a similar approach to identify cellular components sensitive to chemical modifications that do not affect integrity of DNA. In contrast to incorporation of 2′, 2′-difluorodeoxycitidine into DNA (29), dGS does not cause cessation of DNA strand elongation. Therefore, to characterize and isolate the protein complex interacting with dGS-modified DNA, we performed EMSA using synthetic oligonucleotide duplexes containing thioguanine base-paired with cytosine or thymine in the central position of the duplex (Table 1). Titration experiments demonstrated that in the presence of 4–5-fold excess of the nonmodified oligo duplex, dGS-DNA remained bound with the identified nuclear complex, demonstrating higher affinity of these nuclear proteins to modified DNA (Fig. 3, A and B). Selectivity for modified DNA is also demonstrated by preferential complex formation with oligo duplexes containing nonnatural nucleoside inserts (Fig. 6).

In this study, we demonstrated that the nuclear complex, which recognizes nonnatural nucleosides incorporated into DNA, contains HMGB1, HMGB2, HSC70, ERp60, and GAPDH proteins and binds dGS-modified DNA with substantially (4–5-fold) higher affinity than DNA containing only natural nucleotides. Coimmunoprecipitation of HMGB1 with three components of the complex from nuclear extracts of untreated cells (Fig. 5) indicates that these proteins interact in the absence of drug challenge, rather than assemble subsequent to DNA modification. Moreover, a similar DNA-protein complex is formed in the presence of oligonucleotide duplexes modified by incorporation of two other chemically distinct nonnatural nucleosides, araC and 5FdU (Table 1), as evidenced by supershift experiments in the presence of polyclonal rabbit anti-HMGB1 antibody (Fig. 6). Thus, three nonnatural nucleoside analogs (dGS, araC, and 5FdU) create DNA lesions that are recognized by this complex, indicating that this multiprotein complex may be a general sensor of DNA that has been modified by incorporation of structurally distinct nonnatural nucleosides.

HMGB1 and HMGB2 are abundant proteins localized in the nucleus (30, 31), and ERp60, a member of the large protein disulfide isomerase family, is localized in the nucleus and cytosol (32). Two components of the complex, a multifunctional protein GAPDH, (33) and HSC70, which is associated with the nuclear matrix (34), shuttle into the nucleus after stress, e.g., thiopurine treatment (8). All of the identified proteins in the complex have protein-protein binding properties and participate in the formation of homo- or heterocomplexes, and four of five proteins (HMGB1/2, ERp60, and GAPDH) have known DNA-binding properties (30, 32, 33, 35). Among the components of this complex, it is most likely that HMGB1 and HMGB2 are involved in direct binding with modified DNA. Although these proteins do not reveal sequence specificity, they have increased affinity to cruciform DNA, B-Z DNA junction (30) and DNA modified with cisplatin, an anticancer agent that significantly changes DNA geometry (36, 37). HMGB1/2 proteins contain two similar tandem HMG-box domains A and B, organized in a characteristic l-shape consisting of three α-helices and are capable of introducing bends into DNA (30). It is plausible that dGS incorporation facilitates the ability of DNA to acquire a bent conformation when bound to HMGB1/2 proteins. Because destabilizing alterations within the structure decrease DNA rigidity (38), we speculate that HMGB1/2 binding to flexible regions in DNA could provide a common mechanism for active scanning of DNA damage. HMG proteins are considered architectural regulators of transcription; therefore, DNA modifications that increase DNA binding of HMGB1/2 may initiate a stress response through transcriptional activation. Also, HMG functional motifs are hypothesized to play a major role in facilitating the orderly progression of many DNA-related activities, including replication, recombination, and repair (31).

To substantiate the hypothesis about involvement of HMGB1 into DNA stress response, we determined thiopurine sensitivity in immortalized fibroblast cell lines derived from Hmgb1-deficient and -proficient murine embryos (21). Fibroblasts lacking Hmgb1 were ∼10-fold more resistant (higher IC50) to MP (Fig. 7) but demonstrated similar sensitivity to vincristine, an antileukemic agent that is cytotoxic by arresting cell mitosis rather than incorporation into DNA (39). The direct interaction of HMGB1-containing protein complex with dGS-DNA, coupled with the lower sensitivity in Hmgb1−/− cells suggests that the HMGB1-containing complex plays an important role in cellular response to chemical modification of DNA. Recently, HMGB1 was shown to associate with the nuclei in apoptotic cells, indicating that during apoptosis, chromatin undergoes some chemical or structural transition that makes it susceptible to HMGB1 binding (40). Elucidation of the details of HMGB1 interactions with its various partners has considerable therapeutic potential (41). Investigation of the interactions between HMGB1, HMGB2, and DNA containing different nonnucleoside analogs, as well as the interactions between HMGB1 and its protein partners (HMGB2, HSC70, Erp60 and GAPDH), is ongoing in our lab.

Identification of a new multiprotein nuclear complex and its role in cell sensitivity to thiopurines in MMR-deficient cells opens new insights to a heretofore unrecognized cellular response to incorporation of nonnatural nucleosides into DNA. These findings provide the foundation for future studies to elucidate the function of individual components of this complex and their role in determining the sensitivity of cells to anticancer agents that exert their effects via incorporation into DNA. An important future direction will be to determine whether this multiprotein complex is the initial sensor of DNA modification that provides the primary pro-apoptotic signal in response to incorporation of nonnatural nucleosides in DNA.

Fig. 1.

A, FACS analysis of DNA content in Nalm6 cells synchronized by a thymidine block and released by medium change. Two h after the change, MP was added to a final concentration of 10 μm. The time of MP addition corresponds to the zero time point. Cells are shown without MP treatment (−MP, control) and after MP treatment (+MP) at 0, 4, 18, and at 34 h. B, fraction of cells containing DNA strand breaks (apoptotic cells), as detected by the TUNEL assay. □ depicts controls (no MP treatment); ▪, after MP treatment. C, fraction of synchronized cells that contain activated caspases after treatment with MP, as determined by flow cytometry using FAM-VAD-FMK as the substrate (see “Materials and Methods”). □, control (no MP treatment); ▪, after MP treatment. Data were collected from three independent experiments (mean ± SE).

Fig. 1.

A, FACS analysis of DNA content in Nalm6 cells synchronized by a thymidine block and released by medium change. Two h after the change, MP was added to a final concentration of 10 μm. The time of MP addition corresponds to the zero time point. Cells are shown without MP treatment (−MP, control) and after MP treatment (+MP) at 0, 4, 18, and at 34 h. B, fraction of cells containing DNA strand breaks (apoptotic cells), as detected by the TUNEL assay. □ depicts controls (no MP treatment); ▪, after MP treatment. C, fraction of synchronized cells that contain activated caspases after treatment with MP, as determined by flow cytometry using FAM-VAD-FMK as the substrate (see “Materials and Methods”). □, control (no MP treatment); ▪, after MP treatment. Data were collected from three independent experiments (mean ± SE).

Close modal
Fig. 2.

Response of synchronized Nalm6 cells to MP treatment. Two h after the block release, MP was added to a final concentration of 10 μm. The time at which MP was added corresponds to the zero time point on the chart. ○ represents untreated (control) synchronized cells; □ and ▪ represent synchronized cells treated with 10 μm MP. A, fraction of cells in the S phase of the cell cycle at each time point. B, relative number of cells that incorporated BrdUrd into their DNA after the thymidine block was removed. Symbols are same as in A. C, accumulation of dGS (expressed as dGS/100 dG residues) in DNA during the first and the second cycles of division of synchronized Nalm6 cells. Data were collected from three independent experiments (mean ± SE).

Fig. 2.

Response of synchronized Nalm6 cells to MP treatment. Two h after the block release, MP was added to a final concentration of 10 μm. The time at which MP was added corresponds to the zero time point on the chart. ○ represents untreated (control) synchronized cells; □ and ▪ represent synchronized cells treated with 10 μm MP. A, fraction of cells in the S phase of the cell cycle at each time point. B, relative number of cells that incorporated BrdUrd into their DNA after the thymidine block was removed. Symbols are same as in A. C, accumulation of dGS (expressed as dGS/100 dG residues) in DNA during the first and the second cycles of division of synchronized Nalm6 cells. Data were collected from three independent experiments (mean ± SE).

Close modal
Fig. 3.

Preferential binding of Nalm6 nuclear extract proteins with dGS-DNA oligo duplex. A, EMSA experiments were performed with 10 nm duplex 2 (see Table 1) with increasing concentration of duplex 1 (0–225 nm). B, quantification of DNA-protein binding shown on A. The results present three independent experiments (mean ± SE). C, EMSA experiments were performed with 10 nm duplex 2 in the presence of 50 nm duplex 1 with increasing concentrations of nuclear proteins (0–0.8 μg/μl). D, quantification of DNA-protein binding shown on C. The results present three independent experiments (mean ± SE).

Fig. 3.

Preferential binding of Nalm6 nuclear extract proteins with dGS-DNA oligo duplex. A, EMSA experiments were performed with 10 nm duplex 2 (see Table 1) with increasing concentration of duplex 1 (0–225 nm). B, quantification of DNA-protein binding shown on A. The results present three independent experiments (mean ± SE). C, EMSA experiments were performed with 10 nm duplex 2 in the presence of 50 nm duplex 1 with increasing concentrations of nuclear proteins (0–0.8 μg/μl). D, quantification of DNA-protein binding shown on C. The results present three independent experiments (mean ± SE).

Close modal
Fig. 4.

Identification of the nuclear proteins from Nalm6 cells involved in binding of dGS-DNA. A, the DNA-protein complex binding dGS-modified DNA (indicated by arrow) was isolated by EMSA (Lane 1). Incubation of the complex with anti-HMGB1 antibody resulted in formation of the supershift band (Lane 2). B, SDS-PAGE analysis of proteins eluted from the DNA-protein complex; proteins were visualized by silver staining (42). The most intense bands were excised, and the proteins were identified by mass spectroscopy. C, Western blot analysis of proteins extracted from the DNA-protein complex band (A, Lane 1) and the supershift band (A, Lane 2, top band). Extracted proteins were loaded on three separate membranes and developed with anti-GAPDH Ab, anti-HSC70 Ab, anti-HMGB2 Ab, as indicated on the membrane images. Proteins extracted from DNA-protein complex (A, Lane 1) are in Lane 1, and proteins from the supershift band (A, Lane 2, top band) are in Lane 2.

Fig. 4.

Identification of the nuclear proteins from Nalm6 cells involved in binding of dGS-DNA. A, the DNA-protein complex binding dGS-modified DNA (indicated by arrow) was isolated by EMSA (Lane 1). Incubation of the complex with anti-HMGB1 antibody resulted in formation of the supershift band (Lane 2). B, SDS-PAGE analysis of proteins eluted from the DNA-protein complex; proteins were visualized by silver staining (42). The most intense bands were excised, and the proteins were identified by mass spectroscopy. C, Western blot analysis of proteins extracted from the DNA-protein complex band (A, Lane 1) and the supershift band (A, Lane 2, top band). Extracted proteins were loaded on three separate membranes and developed with anti-GAPDH Ab, anti-HSC70 Ab, anti-HMGB2 Ab, as indicated on the membrane images. Proteins extracted from DNA-protein complex (A, Lane 1) are in Lane 1, and proteins from the supershift band (A, Lane 2, top band) are in Lane 2.

Close modal
Fig. 5.

HMGB1-associated proteins coimmunoprecipitated by the anti-HMGB1 antibody from nuclear extracts of untreated Nalm6 cells. Lane 1 depicts proteins after silver staining. Lanes 2–5 depict results of Western analysis using the anti-HMGB1 antibody, anti-HMGB2 antibody, anti-ERp60 antibody, and anti-HSC70 antibody, as indicated.

Fig. 5.

HMGB1-associated proteins coimmunoprecipitated by the anti-HMGB1 antibody from nuclear extracts of untreated Nalm6 cells. Lane 1 depicts proteins after silver staining. Lanes 2–5 depict results of Western analysis using the anti-HMGB1 antibody, anti-HMGB2 antibody, anti-ERp60 antibody, and anti-HSC70 antibody, as indicated.

Close modal
Fig. 6.

DNA-protein complex formed in the presence of DNA containing nonnatural nucleosides. A, EMSA experiment was performed in the presence of 10 nm 5′-[32P]-labeled synthetic DNA duplexes 2–4 (see Table 1) that differed by the central nucleosides involved in bp formation: GSC in duplex 2 (Lanes 1 and 2), GST in duplex 3 (Lanes 3 and 4), GC in duplex 1 (Lanes 5 and 6), GaraC in duplex 4 (Lanes 7 and 8), or A5FU in duplex 5 (Lanes 9 and 10). Five-fold excess of nonradiolabelled duplex 1 was added to each reaction mixture. Anti-HMGB1 antibody was added to one reaction in each pair (Lanes 2, 4, 6, 8, and 10), as indicated. B, quantification of the DNA-protein complexes (indicated by arrow) formed with chemically modified oligonucleotide duplexes 1–5 (see Table 1).

Fig. 6.

DNA-protein complex formed in the presence of DNA containing nonnatural nucleosides. A, EMSA experiment was performed in the presence of 10 nm 5′-[32P]-labeled synthetic DNA duplexes 2–4 (see Table 1) that differed by the central nucleosides involved in bp formation: GSC in duplex 2 (Lanes 1 and 2), GST in duplex 3 (Lanes 3 and 4), GC in duplex 1 (Lanes 5 and 6), GaraC in duplex 4 (Lanes 7 and 8), or A5FU in duplex 5 (Lanes 9 and 10). Five-fold excess of nonradiolabelled duplex 1 was added to each reaction mixture. Anti-HMGB1 antibody was added to one reaction in each pair (Lanes 2, 4, 6, 8, and 10), as indicated. B, quantification of the DNA-protein complexes (indicated by arrow) formed with chemically modified oligonucleotide duplexes 1–5 (see Table 1).

Close modal
Fig. 7.

The absence of Hmgb1 alters sensitivity to MP. Viability of Hmgb1−/− (C1) and Hmgb1+/+ (VA1) cells (MEFs) after 5 days of MP treatment, determined by MTT assay. ⋄, C1 cells; •, VA1 cells. Each point represents the results of three parallel experiments (mean ± SE). In contrast, Hmgb1−/− cells were not more resistant to vincristine (see “Results”).

Fig. 7.

The absence of Hmgb1 alters sensitivity to MP. Viability of Hmgb1−/− (C1) and Hmgb1+/+ (VA1) cells (MEFs) after 5 days of MP treatment, determined by MTT assay. ⋄, C1 cells; •, VA1 cells. Each point represents the results of three parallel experiments (mean ± SE). In contrast, Hmgb1−/− cells were not more resistant to vincristine (see “Results”).

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported, in part, by Grant R37 CA36401 from the National Institutes of Health, Cancer Center Support Grant CA 21765, and by the American Lebanese Syrian Associated Charities.

3

The abbreviations used are as follows: MMR, mismatch repair; MP, mercaptopurine; dGS, 2′-deoxy-6-thioguanosine; PI, propidium iodide; TdT, terminal deoxynucleotidyl transferase; TUNEL, Tdt-mediated nick end labeling; BrdUrd, bromodeoxyuridine; ara-C, 1-β-d-arabinofuranosylcytosine; 5FdU, 5-fluoro-2′-deoxyuridine; EMSA, electrophoretic mobility shift assay; HSC70, heat shock cognate protein 70; ERp60, protein disulfide isomerase; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; PVDF, polyvinylidene difluoride; HMGB1, high mobility group protein B1; HMGB2, high mobility group protein B2; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

Table 1

Sequences of 34-mer synthetic DNA duplexes containing a single modified nucleoside

Oligonucleotide duplexAbbreviationSequence
GC duplexa 5′-ACTCTTGCCTTTAAGGAAAGTATCTAAATGCTTC 
  3′-TGAGAACGGAAATTCCTTTCATAGATTTACGAAG 
GSC duplex 5′-ACTCTTGCCTTTAAGGSAAAGTATCTAAATGCTTC 
  3′-TGAGAACGGAAATTCCTTTCATAGATTTACGAAG 
GST duplex 5′-ACTCTTGCCTTTAAGGSAAAGTATCTAAATGCTTC 
  3′-TGAGAACGGAAATTCT TTTCATAGATTTACGAAG 
GaraC duplex 5′-ACTCTTGCCTTTAAG GAAAGTATCTAAATGCTTC 
  3′-TGAGAACGGAAATTCaraCTTTCATAGATTTACGAAG 
AfU duplex 5′-ACTCTTGCCTTTAAGG AAAGTATCTAAATGCTTC 
  3′-TGAGAACGGAAATTCCfUTTCATAGATTTACGAAG 
Oligonucleotide duplexAbbreviationSequence
GC duplexa 5′-ACTCTTGCCTTTAAGGAAAGTATCTAAATGCTTC 
  3′-TGAGAACGGAAATTCCTTTCATAGATTTACGAAG 
GSC duplex 5′-ACTCTTGCCTTTAAGGSAAAGTATCTAAATGCTTC 
  3′-TGAGAACGGAAATTCCTTTCATAGATTTACGAAG 
GST duplex 5′-ACTCTTGCCTTTAAGGSAAAGTATCTAAATGCTTC 
  3′-TGAGAACGGAAATTCT TTTCATAGATTTACGAAG 
GaraC duplex 5′-ACTCTTGCCTTTAAG GAAAGTATCTAAATGCTTC 
  3′-TGAGAACGGAAATTCaraCTTTCATAGATTTACGAAG 
AfU duplex 5′-ACTCTTGCCTTTAAGG AAAGTATCTAAATGCTTC 
  3′-TGAGAACGGAAATTCCfUTTCATAGATTTACGAAG 
a

Control nonmodified DNA duplex.

We thank Dr. R. A. Ashmun and the staff of the Flow Cytometry and Cell Sorting Shared Resource at St. Jude Children’s Research Hospital for their invaluable help and expertise, Dr. J. C. Panetta for assistance with statistical analysis of the results, X. S. Jiang for help with experiments involving synchronized cells, and M. L. Hankins, M. V. Mane, and Y. Su for their excellent technical assistance.

1
Johnstone R. W., Ruefli A. A., Lowe S. W. Apoptosis: a link between cancer genetics and chemotherapy.
Cell
,
108
:
153
-164,  
2002
.
2
Huang P., Plunkett W. Fludarabine- and gemcitabine-induced apoptosis: incorporation of analogs into DNA is a critical event.
Cancer Chemother. Pharmacol.
,
36
:
181
-188,  
1995
.
3
Miyashita T., Reed J. C. Bcl-2 oncoprotein blocks chemotherapy-induced apoptosis in a human leukemia cell line.
Blood
,
81
:
151
-157,  
1993
.
4
Kaufmann S. H., Earnshaw W. C. Induction of apoptosis by cancer chemotherapy.
Exp. Cell Res.
,
256
:
42
-49,  
2000
.
5
Nelms B. E., Maser R. S., MacKay J. F., Lagally M. G., Petrini J. H. In situ visualization of DNA double-strand break repair in human fibroblasts.
Science (Wash. DC)
,
280
:
590
-592,  
1998
.
6
Paull T. T., Cortez D., Bowers B., Elledge S. J., Gellert M. From the cover: direct DNA binding by Brca1.
Proc. Natl. Acad. Sci. USA
,
98
:
6086
-6091,  
2001
.
7
Griffin S., Branch P., Xu Y. Z., Karran P. DNA mismatch binding and incision at modified guanine bases by extracts of mammalian cells: implications for tolerance to DNA methylation damage.
Biochemistry
,
33
:
4787
-4793,  
1994
.
8
Krynetski E. Y., Krynetskaia N. F., Gallo A. E., Murti K. G., Evans W. E. A novel protein complex distinct from mismatch repair binds thioguanylated DNA.
Mol. Pharmacol.
,
59
:
367
-374,  
2001
.
9
Elion G. B. The purine path to chemotherapy.
Science (Wash. DC)
,
244
:
41
-47,  
1989
.
10
LePage G. A. Basic biochemical effects and mechanism of action of 6-thioguanine.
Cancer Res.
,
23
:
1202
-1206,  
1963
.
11
Scherf U., Ross D. T., Waltham M., Smith L. H., Lee J. K., Tanabe L., Kohn K. W., Reinhold W. C., Myers T. G., Andrews D. T., Scudiero D. A., Eisen M. B., Sausville E. A., Pommier Y., Botstein D., Brown P. O., Weinstein J. N. A gene expression database for the molecular pharmacology of cancer[see comments].
Nat. Genet.
,
24
:
236
-244,  
2000
.
12
Somerville L., Krynetski E. Y., Krynetskaia N. F., Beger R. D., Zhang W., Marhefka C. A., Evans W. E., Kriwacki R. W. Structure and dynamics of thioguanine-modified duplex DNA.
J. Biol. Chem.
,
27B
:
2003
.
13
Krynetskaia N. F., Krynetski E. Y., Evans W. E. Human RNase H-mediated RNA cleavage from DNA-RNA duplexes is inhibited by 6-deoxythioguanosine incorporation into DNA.
Mol. Pharmacol.
,
56
:
841
-848,  
1999
.
14
Krynetskaia N. F., Cai X., Nitiss J. L., Krynetski E. Y., Relling M. V. Thioguanine substitution alters DNA cleavage mediated by topoisomerase II.
FASEB J.
,
14
:
2339
-2344,  
2000
.
15
Relling M. V., Yanishevski Y., Nemec J., Evans W. E., Boyett J. M., Behm F. G., Pui C. H. Etoposide and antimetabolite pharmacology in patients who develop secondary acute myeloid leukemia.
Leukemia (Baltimore)
,
12
:
346
-352,  
1998
.
16
Relling M. V., Rubnitz J. E., Rivera G. K., Boyett J. M., Hancock M. L., Felix C. A., Kun L. E., Walter A. W., Evans W. E., Pui C. H. High incidence of secondary brain tumours after radiotherapy and antimetabolites[see comments].
Lancet
,
354
:
34
-39,  
1999
.
17
Karran P. Mechanisms of tolerance to DNA damaging therapeutic drugs.
Carcinogenesis (Lond.)
,
22
:
1931
-1937,  
2001
.
18
Pichierri P., Franchitto A., Piergentili R., Colussi C., Palitti F. Hypersensitivity to camptothecin in MSH2 deficient cells is correlated with a role for MSH2 protein in recombinational repair.
Carcinogenesis (Lond.)
,
22
:
1781
-1787,  
2001
.
19
Dignam J. D., Lebovitz R. M., Roeder R. G. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei.
Nucleic Acids Res.
,
11
:
1475
-1489,  
1983
.
20
Bianchi M. E., Beltrame M., Paonessa G. Specific recognition of cruciform DNA by nuclear protein HMG1.
Science (Wash. DC)
,
243
:
1056
-1059,  
1989
.
21
Calogero S., Grassi F., Aguzzi A., Voigtlander T., Ferrier P., Ferrari S., Bianchi M. E. The lack of chromosomal protein Hmg1 does not disrupt cell growth but causes lethal hypoglycaemia in newborn mice.
Nat. Genet.
,
22
:
276
-280,  
1999
.
22
Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays.
J. Immunol. Methods
,
65
:
55
-63,  
1983
.
23
Krynetski E. Y., Evans W. E. Mercaptopurine and related compounds Lipp H. P. eds. .
Anticancer Drug Toxicity
,
392
-429, Marcel Dekker, Inc. New York  
1999
.
24
Fairchild C. R., Maybaum J., Kennedy K. A. Concurrent unilateral chromatid damage and DNA strand breakage in response to 6-thioguanine treatment.
Biochem. Pharmacol.
,
35
:
3533
-3541,  
1986
.
25
Christie N. T., Drake S., Meyn R., Nelson J. A. 6-thioguanine-induced DNA damage as a determinant of cytotoxicity in cultured Chinese hamster ovary cells.
Cancer Res.
,
44
:
3665
-3671,  
1984
.
26
da Silva C. P., de Oliveira C. R., da Conceicao M., de Lima P. Apoptosis as a mechanism of cell death induced by different chemotherapeutic drugs in human leukemic T-lymphocytes.
Biochem. Pharmacol.
,
51
:
1331
-1340,  
1996
.
27
Hortelano S., Bosca L. 6-Mercaptopurine decreases the Bcl-2/Bax ratio and induces apoptosis in activated splenic B lymphocytes.
Mol. Pharmacol.
,
51
:
414
-421,  
1997
.
28
Tsurusawa M., Saeki K., Fujimoto T. Differential induction of apoptosis on human lymphoblastic leukemia Nalm-6 and Molt-4 cells by various antitumor drugs.
Int. J. Hematol.
,
66
:
79
-88,  
1997
.
29
Achanta G., Pelicano H., Feng L., Plunkett W., Huang P. Interaction of p53 and DNA-PK in response to nucleoside analogues: potential role as a sensor complex for DNA damage.
Cancer Res.
,
61
:
8723
-8729,  
2001
.
30
Bustin M., Reeves R. High-mobility-group chromosomal proteins: architectural components that facilitate chromatin function.
Prog. Nucleic Acid Res. Mol. Biol.
,
54
:
35
-100,  
1996
.
31
Bustin M. Regulation of DNA-dependent activities by the functional motifs of the high-mobility-group chromosomal proteins.
Mol. Cell. Biol.
,
19
:
5237
-5246,  
1999
.
32
Ferraro A., Altieri F., Coppari S., Eufemi M., Chichiarelli S., Turano C. Binding of the protein disulfide isomerase isoform ERp60 to the nuclear matrix-associated regions of DNA.
J. Cell Biochem.
,
72
:
528
-539,  
1999
.
33
Sirover M. A. New insights into an old protein: the functional diversity of mammalian glyceraldehyde-3-phosphate dehydrogenase.
Biochim. Biophys. Acta
,
1432
:
159
-184,  
1999
.
34
Jin M. L., Zhang P., Ding M. X., Yun J. P., Chen P. F., Chen Y. H., Chew Y. Q. Altered expression of nuclear matrix proteins in etoposide-induced apoptosis in HL-60 cells.
Cell Res.
,
11
:
125
-134,  
2001
.
35
Xanthoudakis S., Nicholson D. W. Heat-shock proteins as death determinants.
Nat. Cell Biol.
,
2
:
E163
-E165,  
2000
.
36
Cohen S. M., Lippard S. J. Cisplatin: from DNA damage to cancer chemotherapy.
Prog. Nucleic Acid Res. Mol. Biol.
,
67
:
93
-130,  
2001
.
37
Pasheva E. A., Ugrinova I., Spassovska N. C., Pashev I. G. The binding affinity of HMG1 protein to DNA modified by cisplatin and its analogs correlates with their antitumor activity.
Int. J. Biochem. Cell Biol.
,
34
:
87
-92,  
2002
.
38
Marathias V. M., Jerkovic B., Bolton P. H. Damage increases the flexibility of duplex DNA.
Nucleic Acids Res.
,
27
:
1854
-1858,  
1999
.
39
Lipp H. P., Bokemeyer C. Clinical pharmacokinetics of cytostatic drugs: efficacy and toxicity Lipp H. P. eds. .
Anticancer Drug Toxicity
,
11
-201, Marcel Dekker, Inc. New York  
1999
.
40
Scaffidi P., Misteli T., Bianchi M. E. Release of chromatin protein HMGB1 by necrotic cells triggers inflammation.
Nature (Lond.)
,
418
:
191
-195,  
2002
.
41
Bustin M. At the crossroads of necrosis and apoptosis: signaling to multiple cellular targets by HMGB1.
Sci STKE.
,
2002
:
PE39
2002
.
42
Shevchenko A., Wilm M., Vorm O., Mann M. Mass spectrometric sequencing of proteins from silver-stained polyacrylamide gels.
Anal. Chem.
,
68
:
850
-858,  
1996
.