Mesenchymal stem cells and immunomodulation: current status and future prospects.

The unique immunomodulatory properties of mesenchymal stem cells (MSCs) make them an invaluable cell type for the repair of tissue/ organ damage caused by chronic inflammation or autoimmune disorders. Although they hold great promise in the treatment of immune disorders such as graft versus host disease (GvHD) and allergic disorders, there remain many challenges to overcome before their widespread clinical application. An understanding of the biological properties of MSCs will clarify the mechanisms of MSC-based transplantation for immunomodulation. In this review, we summarize the preclinical and clinical studies of MSCs from different adult tissues, discuss the current hurdles to their use and propose the future development of pluripotent stem cell-derived MSCs as an approach to immunomodulation therapy.

Activation of NRG1-ERBB4 signaling potentiates mesenchymal stem cell-mediated myocardial repairs following myocardial infarction.

Mesenchymal stem cell (MSC) transplantation has achieved only modest success in the treatment of ischemic heart disease owing to poor cell viability in the diseased microenvironment. Activation of the NRG1 (neuregulin1)-ERBB4 (v-erb-b2 avian erythroblastic leukemia viral oncogene homolog 4) signaling pathway has been shown to stimulate mature cardiomyocyte cell cycle re-entry and cell division. In this connection, we aimed to determine whether overexpression of ERBB4 in MSCs can enhance their cardio-protective effects following myocardial infarction. NRG1, MSCs or MSC-ERBB4 (MSC with ERBB4 overexpression), were transplanted into mice following myocardial infarction. Superior to that of MSCs and solely NRG1, MSC-ERBB4 transplantation significantly preserved heart functions accompanied with reduced infarct size, enhanced cardiomyocyte division and less apoptosis during early phase of infarction. The transduction of ERBB4 into MSCs indeed increased cell mobility and apoptotic resistance under hypoxic and glucose-deprived conditions via a PI3K/Akt signaling pathway in the presence of NRG1. Unexpectedly, introduction of ERBB4 into MSC in turn potentiates NRG1 synthesis and secretion, thus forming a novel NRG1-ERBB4-NRG1 autocrine loop. Conditioned medium of MSC-ERBB4 containing elevated NRG1, promoted cardiomyocyte growth and division, whereas neutralization of NRG1 blunted this proliferation. These findings collectively suggest that ERBB4 overexpression potentiates MSC survival in the infarcted heart, enhances NRG1 generation to restore declining NRG1 in the infarcted region and stimulates cardiomyocyte division. ERBB4 has an important role in MSC-mediated myocardial repairs.

Metabolic syndrome, endothelial injury, and subclinical atherosclerosis in patients with systemic lupus erythematosus.

OBJECTIVES: To study the link between metabolic syndrome (MetS), endothelial injury, and atherosclerosis in patients with systemic lupus erythematosus (SLE). METHODS: Consecutive SLE patients without a history of arterial thrombosis were screened for atherosclerosis at the carotid and coronary arteries by B-mode ultrasound [intima-media thickness (IMT)] and multidetector computed tomography (MDCT) scan (Agatston calcium scores), respectively. Plasma levels of homocysteine, high-sensitivity C-reactive protein (hsCRP), soluble vascular cell adhesion molecule (sVCAM)-1, P-selectin, and soluble thrombomodulin (sTM) were assayed. Patients were stratified according to the National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATP III) criteria for MetS, using the Asian criteria for abdominal obesity. Risk factors for atherosclerosis were studied. RESULTS: Of the 123 SLE patients (93% women; age 47.9+/-11 years; SLE duration 10.9+/-7.0 years) studied, 20 (16.3%) had MetS. The prevalence of MetS in the SLE patients was significantly higher than in 492 age- and sex-matched healthy controls (9.6%; p=0.03). Coronary calcification and abnormal carotid IMT were detected in 38 (31%) and 72 (59%) of SLE patients, respectively. Patients with MetS had a significantly higher Agatston score (69.5+/-95 vs. 16.4+/-57; p=0.03) and a numerically higher carotid IMT (p=0.43) than those without. In a logistic regression model, the MetS [odds ratio (OR) 3.11, 95% confidence interval (CI) 1.01-9.59, p=0.049] was associated with coronary atherosclerosis after adjustment for age and other risk factors. In addition, patients with MetS had significantly higher levels of hsCRP (p=0.002), homocysteine (p=0.03), and sTM (p=0.01). CONCLUSIONS: The MetS is more prevalent in SLE patients than the general population and is associated with endothelial injury and coronary atherosclerosis. More aggressive control of risk factors is justified in these patients.

Photodynamic therapy-induced death of HCT 116 cells: Apoptosis with or without Bax expression.

Cell death following photodynamic therapy (PDT) with the photosensitizer Pc 4 involves the intrinsic pathway of apoptosis. To evaluate the importance of Bax in apoptosis after PDT, we compared the PDT responses of Bax-proficient (Bax(+/-)) and Bax knock-out (BaxKO) HCT116 human colon cancer cells. PDT induced a slow apoptotic process in HCT Bax(+/-) cells following a long delay in the activation of Bax and release of cytochrome c from mitochondria. Although cytochrome c was not released from mitochondria following PDT in BaxKO cells, an alternative mechanism of caspase-dependent apoptosis with extensive chromatin and DNA degradation was found in these cells. This alternative process was less efficient and slower than the normal apoptotic process observed in Bax(+/-) cells. Early events upon PDT, such as the loss of mitochondrial membrane potential, photodamage to Bcl-2, and activation of p38 MAP kinase, were observed in both HCT116 cell lines. In spite of differences in the efficiency and mode of apoptosis induced by PDT in the Bax(+/-) and BaxKO cells, they were found to be equally sensitive to killing by PDT, as determined by loss of clonogenicity. Thus, for Pc 4-PDT, the commitment to cell death occurs prior to and independent of Bax activation, but the process of cellular disassembly differs in Bax-expressing vs. non-expressing cells.

Bax is essential for mitochondrion-mediated apoptosis but not for cell death caused by photodynamic therapy.

The role of Bax in the release of cytochrome c from mitochondria and the induction of apoptosis has been demonstrated in many systems. Using immunocytochemical staining, we observed that photodynamic therapy (PDT) with the photosensitiser Pc 4 induced Bax translocation from the cytosol to mitochondria, and the release of cytochrome c from mitochondria as early signalling for the intrinsic pathway of apoptosis in human breast cancer MCF-7c3 cells. To test the role of Bax in apoptosis, MCF-7c3 cells were treated with Bax antisense oligonucleotides, which resulted in as much as a 50% inhibition of PDT-induced apoptosis. In the second approach, Bax-negative human prostate cancer DU-145 cells were studied. Following PDT, the hallmarks of apoptosis, including the release of cytochrome c from mitochondria, loss of mitochondrial membrane potential, caspase activation, and chromatin condensation and fragmentation, were completely blocked in these cells. Restoration of Bax expression in DU-145 cells restored apoptosis, indicating that the resistance of DU-145 cells to PDT-induced apoptosis is due to the lack of Bax rather than to another defect in the apoptotic machinery. However, despite the inhibition of apoptosis, the Bax-negative DU-145 cells were as photosensitive as Bax-replete MCF-7c3 cells, as determined by clonogenic assay. Thus, for Pc 4-PDT, the commitment to cell death occurs prior to Bax activation.

FADD null mouse embryonic fibroblasts undergo apoptosis after photosensitization with the silicon phthalocyanine Pc 4.

Oxidative stress, such as photodynamic therapy with the silicon phthalocyanine Pc 4 (Pc 4-PDT), can induce apoptosis and tumor necrosis factor alpha (TNF) production. TNF receptors, as well as other death receptors, have been implicated in stress-induced apoptosis. To assess directly the role of FADD, a death receptor-associated protein, in induction of apoptosis post-Pc 4-PDT, embryonic fibroblasts from FADD knock out (k/o) and wild-type (wt) mice were used. Pc 4-PDT induced casp-3 activation and apoptosis in both cell types. In the presence of zVAD, a pancaspase inhibitor, Pc 4-PDT-induced apoptosis was abrogated in both cell lines. Fumonisin B1 (FB), an inhibitor of ceramide synthase, had no effect on apoptosis after Pc 4-PDT in either cell line. Similar to Pc 4-PDT, exogenous C6-ceramide bypassed FADD deficiency and induced zVAD-sensitive apoptosis. In contrast to Pc 4 photosensitization, TNF did not induce either apoptosis or ceramide accumulation in FADD k/o cells. In the absence of FADD deficiency, TNF-induced apoptosis was zVAD-sensitive and FB-insensitive. Induced ceramide levels remained elevated after cotreatment with TNF and zVAD in FADD wt cells. Taken together, these data provide genetic evidence for a lack of FADD requirement in Pc 4-PDT- or C6-ceramide-induced apoptosis. FB-sensitive ceramide production accompanies, but does not suffice, for apoptosis after Pc 4 photosensitization or TNF.

Dissociation of mitochondrial depolarization from cytochrome c release during apoptosis induced by photodynamic therapy.

Photodynamic therapy (PDT) with the phthalocyanine photosensitizer Pc 4 induces rapid apoptosis in mouse lymphoma (LY-R) cells, initiating with the release of cytochrome c from mitochondria. It has been proposed that the opening of the mitochondrial membrane permeability transition pores, which results in the dissipation of the mitochondrial membrane potential (Deltapsi(m)), is essential for the escape of cytochrome c from mitochondria into the cytosol as well as for apoptotic cell death. Therefore, we have assessed the correlation between the loss of Deltapsi(m)and the release of cytochrome c following PDT. Treatment of LY-R cells with 300 nM Pc 4 and 60, 90 or 120 mJ/cm(2)of red light resulted in apoptosis of 80-90% of the cells, accompanied by >20-fold elevation in caspase-3-like activity within one h. At all 3 doses of PDT employed here, the majority of the cytochrome c was released from mitochondria at 15 min after irradiation, as determined by an immunohistochemical method. In contrast, the loss of Deltapsi(m)following PDT, as monitored by the uptake of JC-1 or Rh-123, depended on the PDT dose and the post-treatment time. In spite of the release of cytochrome c at 15 min after each of the 3 doses, a corresponding loss of Deltapsi(m)was observed only for those cells that received the highest dose of PDT. Virtually all cells that received one of the lower doses of PDT (300 nM Pc 4 plus 60 or 90 mJ/cm(2)) maintained normal Deltapsi(m). Hence, our results support the conclusion that the release of cytochrome c from mitochondria resulting from Pc 4-PDT-induced photodamage is independent of the loss of Deltapsi(m). Therefore, it is important to consider a range of doses of this or other apoptotic stimuli in deciphering the relationship of metabolic responses that contribute to apoptosis.

Recombinant human tumor necrosis factor alpha does not potentiate cell killing after photodynamic therapy with a silicon phthalocyanine in A431 human epidermoid carcinoma cells.

Photodynamic therapy (PDT) is a novel cancer treatment utilizing a photosensitizer, visible light and oxygen. PDT with the silicon phthalocyanine Pc 4, a new photosensitizer, is highly effective in cancer cell destruction and tumor ablation. The mechanisms underlying cancer cell killing by PDT are not fully understood. Tumor necrosis factor alpha (TNF) is a multifunctional cytokine that has been implicated in photocytotoxicity. We asked whether recombinant human TNF (rhTNF) affects Pc 4-PDT cytotoxicity in A431 human epidermoid carcinoma cells. Co-treatment of A431 cells with various doses of Pc 4-PDT and a sub-lethal rhTNF dose led to a sub-additive reduction in cell survival. In addition, in the presence of Pc 4-PDT or rhTNF, caspase-3 activity and apoptosis were induced. The combined treatment, however, did not potentiate either caspase-3 activity or apoptosis. Similar to previous findings we observed that Pc 4-PDT initiated a time-dependent extracellular TNF accumulation. The data suggest that: a) PDT and rhTNF induce cancer cell killing through different mechanisms; and b) Pc 4-PDT-induced TNF production is a stress response that may not directly affect photocytotoxicity.

Photodynamic therapy-induced death of MCF-7 human breast cancer cells: a role for caspase-3 in the late steps of apoptosis but not for the critical lethal event.

Photodynamic therapy (PDT) causes mitochondrial damage and induces apoptosis through release of cytochrome c and activation of caspase-3. To test whether caspase 3 is the sole executioner of apoptosis and its role in overall cell lethality, we compared the response of MCF-7c3 cells that express a stably transfected CASP-3 gene to that of parental MCF-7:SW8 cells transfected with vector alone (MCF-7v). Following photosensitization with the phthalocyanine Pc 4 and red light, cytochrome c was released from the mitochondria to equivalent extents in the two cell lines. However, the appearance of apoptotic indicators, such as active caspase-3 (DEVDase), cleavage of poly(ADP-ribose) polymerase, and oligonucleosomal DNA fragmentation, was observed only in MCF-7c3 cells during the first 6 h after photosensitization. Although production of 50-kb DNA fragments and chromatin condensation were found in PDT-treated MCF-7v cells by 20-24 h posttreatment, the rate and extent of apoptosis were much less than in MCF-7c3 cells. MCF-7c3 cells were more sensitive to photosensitization than were MCF-7v cells when assayed for loss of viability by reduction of a tetrazolium dye. However, the two cell lines were equally sensitive to photodynamic killing when evaluated by a clonogenic assay. These results show (a) the importance of assessing overall cell death by clonogenic assay; (b) that the critical lethal event is independent of caspase-3, perhaps at or near the release of cytochrome c from mitochondria; and (c) that the caspase-3-mediated events appear to be irrelevant in determining overall killing of cells.

Phthalocyanine 4-photodynamic therapy induces ceramide generation and apoptosis in acid sphingomyelinase-deficient mouse embryonic fibroblasts.

Photodynamic therapy (PDT), a novel cancer treatment using a photosensitizer and visible light, produces an oxidative stress in cells that can lead to apoptosis. PDT with the phthalocyanine photosensitizer Pc 4 (Pc 4-PDT), causes increased generation of ceramide, a lipid mediator, and subsequent induction of apoptosis in various cell types. Formation of ceramide by acid sphingomyelinase (ASMase) in response to stress has been implicated in apoptotic cell death. We assessed the role of ASMase in photocytotoxicity using mouse embryonic fibroblasts (MEFs) isolated from ASMase knockout (k/o) and wild-type (wt) mice. Exposure of wt or k/o MEFs to Pc 4-PDT led to increased caspase-3 activity and subsequent apoptosis. Similarly, ceramide levels were elevated in both cell types post-PDT. We suggest that in MEFs, ASMase is dispensable for ceramide accumulation and induction of apoptosis after Pc 4-PDT.

Fumonisin B1 does not prevent apoptosis in A431 human epidermoid carcinoma cells after photosensitization with a silicon phthalocyanine.

Photodynamic therapy with the phthalocyanine photosensitizer Pc 4 (Pc 4-PDT), an apoptosis inducer, is associated with accumulation of ceramide in various cell lines. The role of ceramide in Pc 4-PDT-induced apoptosis was investigated in A431 cells. Caspase-3 (casp-3) was activated and TUNEL positive cells began to appear 30 and 60 min post-Pc 4-PDT, respectively. A rapid increase (10 min) in cellular ceramide levels was observed after Pc 4-PDT. Induced ceramide accumulation was maintained over 60 min, Acid sphingomyelinase, a ceramide-generating enzyme, was inhibited after photosensitization with Pc 4, suggesting that the enzyme was not required for stimulated ceramide accumulation. Co-treatment of A431 cells with fumonisin B1, a ceramide synthase inhibitor, and Pc 4-PDT led to a decrease in ceramide levels without any effect on induced casp-3 activity or apoptosis. In the presence of zVAD, a pan-caspase inhibitor, apoptosis was abolished, while ceramide levels remained elevated after Pc 4-PDT. Exposure of A431 cells to exogenous C6-ceramide for 22 h, led to induction of apoptosis, and the process was abrogated by zVAD. In conclusion, C6-ceramide-, like Pc 4-PDT-induced apoptosis, is zVAD-sensitive. Furthermore, Pc 4 photosensitization can lead to apoptosis without FB-sensitive elevation in ceramide levels upstream of caspases.

Photodynamic therapy-induced apoptosis in lymphoma cells: translocation of cytochrome c causes inhibition of respiration as well as caspase activation.

L5178Y-R mouse lymphoma (LY-R) cells undergo rapid apoptosis when treated with photodynamic therapy (PDT) sensitized with the silicon phthalocyanine Pc 4. In this study we show that cytochrome c is released into the cytosol within 10 min of an LD99.9 dose of PDT. Cellular respiration is inhibited by 42% at 15 min, and 60% at 30 min after PDT treatment, and caspase 3-like protease activity is elevated by 15 min post-PDT. In digitonin-permeabilized cells addition of cytochrome c to the respiration buffer reverses PDT-induced inhibition of state 3 respiration via Complex I by 40-60%, and via Complex III by 50-90%. In contrast, extramitochondrial cytochrome c does not stimulate respiration in permeabilized control cells, and catalyzes only a low rate of oxygen consumption via electron transfer to cytochrome b5 on the outer mitochondrial membrane. These results demonstrate that PDT-induced inhibition of respiration is primarily due to leakage of cytochrome c into the cytosol rather than to damage to the major enzyme complexes of the electron transport chain. Whether or not inhibition of respiration influences the time course or extent of Pc 4-PDT-induced apoptosis in LY-R cells is not clear at the present time.

Differential dependence on chromatin structure for copper and iron ion induction of DNA double-strand breaks.

The induction of DNA DSB (double-strand breaks) in isolated nuclear chromatin by Cu(II) or Fe(II)-EDTA in the presence of H2O2 and ascorbate has been compared to DSB induction by gamma-radiation. V79 nuclei embedded in agarose plugs were treated with each agent on ice, and the resultant DNA fragments were analyzed by pulsed-field gel electrophoresis. In the absence of low molecular weight radical scavengers, both irradiation and treatment with iron ion induced random DSB, as judged by the size distribution of DNA fragments, and the yield of DSB in each case was enhanced by either the expansion of chromatin (approximately 5-fold) or the removal of histones (21-25-fold) before treatment. In contrast, treatment with Cu(II) produced small DNA fragments of uniform size (approximately 100-200 kbp), independent of the yield of DSB. In addition, neither the DNA fragment size nor the yield of DSB produced by Cu(II) was affected by the prior removal of histones from chromatin. Deproteinized DNA was degraded randomly by Cu(II) but at a slower rate than observed for chromatin. In the presence of ascorbate, H2O2 was found to be essential for DSB induction by Fe(II)-EDTA but not by Cu(II), possibly because H2O2 can be produced from ascorbate and Cu(II) in the presence of oxygen. Despite the above differences between the production of DSB by the two metal ions, DSB induction in native chromatin by either metal ion was blocked by 0.1 M EDTA or 0.25 M thiourea but was resistant to the hydroxyl radical scavengers 0.25 M DMSO and 0.25 M mannitol.(ABSTRACT TRUNCATED AT 250 WORDS)

Copper ion-mediated sensitization of nuclear matrix attachment sites to ionizing radiation.

Exposure of mammalian cells to ionizing radiation induces nuclear matrix proteins and their attached transcribing DNA sequences to form cross-links. To characterize the cellular and matrix components necessary for DNA-protein crosslink (DPC) formation, DPC yields have been examined in isolated nuclear matrices and in the intermediate steps during cell fractionation. It was found that, in both unirradiated and irradiated cells, all components of DPC are retained in isolated nuclei, and the formed DPC are retained as well during the cell fractionation procedure resulting in nuclear matrices. In contrast, nuclear matrices isolated from unirradiated cells are deficient in the ability to form DPC upon irradiation, indicating that elements necessary for DPC production have been disrupted or removed during the isolation procedure. When isolated nuclei were irradiated, the yield of radiation-induced DPC was about 2-fold higher than that for intact cells, presumably due to the removal of soluble cellular scavengers during the isolation procedure. Treatment of nuclei with Cu2+ to stabilize nuclear structural organization during the preparation of the nuclear matrix caused additional DNA, especially the matrix-associated newly replicated DNA, to become bound to protein. Such treatment also enhanced radiation-induced DPC production which was sensitive to OH radical scavengers. Moreover, radiation-induced DPC production in Cu(2+)-treated nuclei was more sensitive to EDTA and catalase than in untreated nuclei. It is therefore proposed that excess DPC induction in Cu(2+)-treated nuclei occurs preferentially at the sites of Cu2+ binding to chromatin where hydroxyl radicals are produced repeatedly through the Fenton reaction.

Radiation-induced binding of DNA from irradiated mammalian cells to hydroxyapatite columns.

In experiments designed to measure radiation-induced DNA damage using the DNA unwinding-hydroxyapatite chromatography technique, we observed that under some experimental conditions a significant proportion of the test DNA became tightly bound to the hydroxyapatite (HA) and could not be released even with a high concentration of phosphate buffer. Approximately 5-10% of DNA from unirradiated cells binds to the HA. With increasing radiation doses in air, the fraction of bound DNA increases, reaching about 30% at about 35 Gy. The binding exhibits many of the characteristics of a radiation-induced cell lesion: the proportion of DNA retained by the HA is less when cells are irradiated under hypoxic conditions or in the presence of the thiol radioprotector dithiothreitol; and the binding decreases when an incubation period is allowed between irradiation and harvest of the cells for assay. Studies to determine the nature of the lesion responsible for the binding demonstrated that lesion production requires a component found in cells since no binding was observed with irradiated isolated DNA or nuclear matrix; the binding is not a result of the production of DNA-protein crosslinks; and the bound DNA is single-stranded, based on its sensitivity to nuclease S1. Because of the dose dependence of the binding of DNA to HA, the slopes of the dose-response curves for DNA damage determined with this assay depend on the method used to calculate the fraction of double-stranded DNA. Our demonstration that the bound DNA is single-stranded guides the choice of the method for data analysis.

Comparison of DNA-protein cross-links induced by 4'-(9-acridinylamino)-methanesulfon-m-anisidide and by gamma-radiation.

The antitumor agent 4'-(9-acridinylamino)methanesulfon-m-anisidide (m-AMSA) inhibits topoisomerase II activity through the formation of a complex of DNA and covalently bound enzyme which, upon protein denaturation, yields DNA breaks (single strand breaks). In the present study, this complex served as a standard for analysis of radiation-induced DNA-protein cross-links (DPC). Following the treatment of exponentially growing mouse L929 cells with 0-100 ng/ml of m-AMSA for 1 h, a linear dose-dependent increase was found in the amount of DNA retained on nitrocellulose filters during subsequent analysis. This result indicates that the assay can detect DPC that have a single protein bound to each DNA fragment. The results of fractionation of nuclear DNA show that m-AMSA induces 20- to 45-fold more DPC in nuclear matrix-associated DNA than in the majority distal loop DNA, supporting the notion that topoisomerase II is located at the nuclear matrix. The frequency of single strand breaks induced by m-AMSA, which should be equal to the frequency of DPC, was determined by alkaline elution. Results of the alkaline elution assay could be correlated with the percentage of DNA retained on nitrocellulose filters; i.e., 1% DNA retention corresponded to 2560 DPC per log-phase L929 cell, which has been determined to have a DNA content of 22.25 pg. Using this standard curve, DPC induced by gamma-irradiation in air were estimated to be formed at a frequency of 133 DPC/cell/Gy, a frequency approximately 3% that of gamma-ray-induced single strand breaks. The radiation dose response for DPC production was unaffected by the high levels of DPC present in cells previously treated with m-AMSA. In addition, DPC induced by m-AMSA were rapidly reversed after the removal of the drug, in contrast to a slower removal of DPC induced by gamma-radiation. These observations suggest that although m-AMSA and gamma-radiation both preferentially induce DPC with matrix-attached DNA, they produce independent types of DPC.

Induction of DNA strand breaks in transcriptionally active DNA sequences of mouse cells by low doses of ionizing radiation.

The efficiency of DNA single-strand break induction was measured in transcriptionally active DNA, transcriptionally inert satellite DNA, and bulk DNA sequences of mouse L929 cells using the alkaline filter elution assay. The cells were exposed to increasing doses of X-radiation up to 1000 rad. DNA which either eluted from or was retained on polycarbonate filters during the assays was collected onto nitrocellulose filters and hybridized against radiolabeled poly(A+)RNA (to probe transcribing DNA sequences) or mouse satellite DNA. The increasing rate and extent of elution of bulk DNA or specific DNA sequences after increasing radiation doses was taken as a measure of the increased frequency of radiation-induced DNA strand breaks. The results indicate that a significant fraction of transcriptionally active DNA contains endogenous strand breaks. With increasing dose, the efficiency of radiation-induced DNA strand breakage in bulk, transcriptionally active and satellite DNA sequences was observed to be the same when the sum of all eluted DNA was considered. However, the early eluting fractions contained DNA which was enriched in active sequences. Since DNA elutes as a function of size, the early fractions contain smaller DNA than later fractions. Therefore, our results indicate that the fraction of active sequences which elutes early resides on smaller fragments on the average than the later eluting DNA, and that even low doses of radiation preferentially cause breaks in regions of DNA containing active sequences.

The formation, identification, and significance of DNA-protein cross-links in mammalian cells.

DNA-protein cross-links (DPC) are formed by a variety of radiations and chemicals which act via free radical formation. Covalency is inferred from the resistance of the cross-links to harsh treatments. In mammalian cells, a background of DPC (6000 per V79 cell) may result from normal associations of chromosomal loops with the nuclear protein matrix. After ionizing radiation, the elevated level of DPC (150 per Gy per V79 cell) are enriched in actively transcribing DNA and in a subset of proteins of the nuclear matrix. DPC formation is reduced by hydroxyl radical scavengers, by oxygen, and by hypertonic medium and is enhanced by hypotonic medium and by removal of intracellular glutathione. DPC are repaired more slowly than single-strand breaks and not at all when formed during metaphase. During the postirradiation period, changes in the sequence composition of the DNA of residual DPC are consistent with the preferential repair of DPC in actively expressed genes. Excision repair mechanisms have been proposed. Unrepaired DPC may block normal functions of the nuclear matrix, such as replication and transcription.

Yield of DNA-protein cross-links in gamma-irradiated Chinese hamster cells.

gamma-Irradiation of Chinese hamster V79 cells increases the percentage of nuclear DNA cross-linked to proteins. Studies were carried out to ascertain whether the radiation-induced increase in DNA-protein cross-links (DPC) is due to an increase in the number of DNA fragments which are cross-linked to protein or to an increase in the size of bound DNA fragments. Cells were prelabeled with [3H]thymidine and irradiated (10-600 Gy), and DPCs were collected on nitrocellulose filters. Native gel analyses of the DNA recovered from the filters indicate that the number average molecular weight of cross-linked DNA (1.22 X 10(7) Da) is the same in unirradiated cells and in cells given up to 100 Gy. Assuming 5 pg of DNA per V79 cell, it was possible to calculate that there are approximately 6 X 10(3) DPC per unirradiated cell and that 150 DPC are formed per gray of gamma-radiation for doses of 0-100 Gy. Thus, radiation increases the number of new linkages between DNA and protein. At radiation doses greater than 200 Gy the percentage of nuclear DNA cross-linked to protein approaches a plateau value. The number of DPC (greater than 6 X 10(4)) formed at higher doses is within the range of the estimated number of DNA attachment sites on the nuclear matrix.

Modification of DNA damage in transcriptionally active vs. bulk chromatin.

Our previous experiments have demonstrated that regions of nuclear chromatin, containing transcriptionally active DNA sequences and associated with the nuclear matrix, are hypersensitive to the production of both single-strand breaks and DNA-protein cross-links upon gamma-irradiation of exponentially growing mammalian cells. In this study, we have irradiated Chinese hamster V79 cells in buffered saline with or without DMSO to scavenge hydroxyl radicals and in buffered salines of various tonicities to expand or condense chromatin. The yield of DNA-protein cross-links was assayed by a nitrocellulose filter binding technique and the DNA recovered from the cross-links hybridized to 125I-poly(A+)RNA to determine the relative frequency of transcriptionally active sequences in the cross-links compared to the bulk DNA. In all cases, the data show that active DNA is affected to a greater extent than bulk, primarily inactive DNA. The more extensive alteration of the level of ionizing radiation-induced damage in active DNA by the diffusible agents tested suggests that other agents, such as chemical sensitizers and protectors, which need to diffuse to the nuclear DNA, may also be acting primarily on active, matrix-associated DNA.