Know thy neighbor: stromal cells can contribute oncogenic signals.

Although the stroma within carcinogenic lesions is known to be supportive and responsive to tumors, new data increasingly show that the stroma also has a more active, oncogenic role in tumorigenesis. Stromal cells and their products can transform adjacent tissues in the absence of pre-existing tumor cells by inciting phenotypic and genomic changes in the epithelial cells. The oncogenic action of distinctive stromal components has been demonstrated through a variety of approaches, which provide clues about the cellular pathways involved.

Combined haploinsufficiency for ATM and RAD9 as a factor in cell transformation, apoptosis, and DNA lesion repair dynamics.

Loss of function of oncogenes, tumor suppressor genes and DNA damage processing genes has been implicated in the development of many types of cancer, but for the vast majority of cases, there is no link to specific germ line mutations. In the last several years, heterozygosity leading to haploinsufficiency for proteins involved in DNA repair pathways was shown to play a role in genomic instability and carcinogenesis after DNA damage is induced. Because the effect of haploinsufficiency for one protein is relatively small, we hypothesize that predisposition to cancer could be a result of the additive effect of heterozygosity for two or more genes, critical for pathways that control DNA damage signaling, repair or apoptosis. To address this issue, primary mouse cells, haploinsufficient for one or two proteins, ATM and RAD9, related to the cellular response to DNA damage were examined. The results show that cells having low levels of both ATM and RAD9 proteins are more sensitive to transformation by radiation, have different DNA double-strand break repair dynamics and are less apoptotic when compared with wild-type controls or those cells haploinsufficient for only one of these proteins. Our conclusions are that under stress conditions, the efficiency and capacity for DNA repair mediated by the ATM/RAD9 cell signaling network depend on the abundance of both proteins and that, in general, DNA repair network efficiencies are genotype-dependent and can vary within a specific range.

NBS1 knockdown by small interfering RNA increases ionizing radiation mutagenesis and telomere association in human cells.

Hypomorphic mutations which lead to decreased function of the NBS1 gene are responsible for Nijmegen breakage syndrome, a rare autosomal recessive hereditary disorder that imparts an increased predisposition to development of malignancy. The NBS1 protein is a component of the MRE11/RAD50/NBS1 complex that plays a critical role in cellular responses to DNA damage and the maintenance of chromosomal integrity. Using small interfering RNA transfection, we have knocked down NBS1 protein levels and analyzed relevant phenotypes in two closely related human lymphoblastoid cell lines with different p53 status, namely wild-type TK6 and mutated WTK1. Both TK6 and WTK1 cells showed an increased level of ionizing radiation-induced mutation at the TK and HPRT loci, impaired phosphorylation of H2AX (gamma-H2AX), and impaired activation of the cell cycle checkpoint regulating kinase, Chk2. In TK6 cells, ionizing radiation-induced accumulation of p53/p21 and apoptosis were reduced. There was a differential response to ionizing radiation-induced cell killing between TK6 and WTK1 cells after NBS1 knockdown; TK6 cells were more resistant to killing, whereas WTK1 cells were more sensitive. NBS1 deficiency also resulted in a significant increase in telomere association that was independent of radiation exposure and p53 status. Our results provide the first experimental evidence that NBS1 deficiency in human cells leads to hypermutability and telomere associations, phenotypes that may contribute to the cancer predisposition seen among patients with this disease.

Alternative recombination pathways in UV-irradiated XP variant cells.

XP variant (XP-V) cells lack the damage-specific polymerase eta and exhibit prolonged replication arrest after UV irradiation due to impaired bypass of UV photoproducts. To analyse the outcome of the arrested replication forks, homologous recombination (HR, Rad51 events) and fork breakage (Rad50 events) were assayed by immunofluorescent detection of foci-positive cells. Within 1 h of irradiation, XP-V cells showed more Rad51-positive cells than normal cells, while neither cell type showed an increase in Rad50 foci. Beyond 1 h, the frequency of Rad51-positive cells reached similar levels in both cell types, then declined at higher UV doses. At these later times, Rad50-positive cells increased with dose and to a greater extent in XP-V cells. Few cells were simultaneously positive for both sets of foci, suggesting a mutually exclusive recruitment of recombination proteins, or that these pathways operate at different stages during S phase. Analysis of cells containing a vector of tandemly arranged enhanced green fluorescent protein genes also showed that UV-induced HR was higher in XP-V cells. These results suggest that cells make an early commitment to HR, and that at later times a subset of arrested forks degrade into double-strand breaks, two alternative pathways that are greater in XP-V cells.

Medium-mediated intercellular communication is involved in bystander responses of X-ray-irradiated normal human fibroblasts.

Although radiation-induced bystander effects have been demonstrated in a number of cell types, the studies have largely been performed using high linear energy transfer (LET) radiation, such as alpha-particles. The literature is contradictory on whether fibroblasts show bystander responses, especially after low LET radiation such as X- or gamma-rays and whether the same signal transmission pathways are involved. Herein, a novel transwell insert culture dish method is used to show that X-irradiation induces medium-mediated bystander effects in AGO1522 normal human fibroblasts. The frequency of micronuclei formation in unirradiated bystander cells increases from a background of about 6.5% to about 9-13% at all doses from 0.1 to 10 Gy to the irradiated cells. Induction of p21Waf1 protein and foci of gamma-H2AX in bystander cells is also independent of dose to the irradiated cells above 0.1 Gy. In addition, levels of reactive oxygen species (ROS) were increased persistently in directly irradiated cells up to 60 h after irradiation and in bystander cells for 30 h. Adding Cu-Zn superoxide dismutase (SOD) and catalase to the medium decreases the formation of micronuclei and induction of p21Waf1 and gamma-H2AX foci in bystander cells, suggesting oxidative metabolism plays a role in the signaling pathways in bystander cells. The results of clonogenic assay of bystander cells showed that survival of bystander cells decreases from 0 to 0.5 Gy, and then is independent of the dose to irradiated cells from 0.5 to 10 Gy. Unlike the response with p21Waf1 expression, gamma-H2AX foci and micronuclei, adding SOD and catalase has no effect on the survival of bystander cells. The data suggest that irradiated cells release toxic factors other than ROS into the medium.

Stress-specific signatures: expression profiling of p53 wild-type and -null human cells.

Gene expression responses of human cell lines exposed to a diverse set of stress agents were compared by cDNA microarray hybridization. The B-lymphoblastoid cell line TK6 (p53 wild-type) and its p53-null derivative, NH32, were treated in parallel to facilitate investigation of p53-dependent responses. RNA was extracted 4 h after the beginning of treatment when no notable decrease in cell viability was evident in the cultures. Gene expression signatures were defined that discriminated between four broad general mechanisms of stress agents: Non-DNA-damaging stresses (heat shock, osmotic shock, and 12-O-tetradecanoylphorbol 13-acetate), agents causing mainly oxidative stress (arsenite and hydrogen peroxide), ionizing radiations (neutron and gamma-ray exposures), and other DNA-damaging agents (ultraviolet radiation, methyl methanesulfonate, adriamycin, camptothecin, and cis-Platinum(II)diammine dichloride (cisplatin)). Within this data set, non-DNA-damaging stresses could be discriminated from all DNA-damaging stresses, and profiles for individual agents were also defined. While DNA-damaging stresses showed a strong p53-dependent element in their responses, no discernible p53-dependent responses were triggered by the non-DNA-damaging stresses. A set of 16 genes did exhibit a robust p53-dependent pattern of induction in response to all nine DNA-damaging agents, however.

Artemis deficiency confers a DNA double-strand break repair defect and Artemis phosphorylation status is altered by DNA damage and cell cycle progression.

Mutations in the Artemis gene are causative in a subset of human severe combined immunodeficiencies (SCIDs) and Artemis-deficient cells exhibit radiation sensitivity and defective V(D)J recombination, implicating Artemis function in non-homologous end joining (NHEJ). Here we show that Artemis-deficient cells from Athabascan-speaking Native American SCID patients (SCIDA) display significantly elevated sensitivity to ionizing radiation (IR) but only a very subtle defect in DNA double-strand (DSB) break repair in contrast to the severe DSB repair defect of NHEJ-deficient cells. Primary human SCIDA fibroblasts accumulate and exhibit persistent arrest at both the G1/S and G2/M boundaries in response to IR, consistent with the presence of persistent DNA damage. Artemis protein is phosphorylated in a PI3-like kinase-dependent manner after either IR or a number of other DNA damaging treatments including etoposide, but SCIDA cells are not hypersensitive to treatment with etoposide. Inhibitor studies with various DNA damaging agents establish multiple phosphorylation states and suggest multiple kinases function in Artemis phosphorylation. We observe that Artemis phosphorylation occurs rapidly after irradiation like that of histone H2AX. However, unlike H2AX, Artemis de-phosphorylation is uncoupled from overall DNA repair and correlates instead with cell cycle progression to or through mitosis. Our results implicate a direct and non-redundant function of Artemis in the repair of a small subset of DNA double-strand breaks, possibly those with hairpin termini, which may account for the pronounced radiation sensitivity observed in Artemis-deficient cells.

Interactions involving the Rad51 paralogs Rad51C and XRCC3 in human cells.

Homologous recombinational repair of DNA double-strand breaks and crosslinks in human cells is likely to require Rad51 and the five Rad51 paralogs (XRCC2, XRCC3, Rad51B/Rad51L1, Rad51C/Rad51L2 and Rad51D/Rad51L3), as has been shown in chicken and rodent cells. Previously, we reported on the interactions among these proteins using baculovirus and two- and three-hybrid yeast systems. To test for interactions involving XRCC3 and Rad51C, stable human cell lines have been isolated that express (His)6-tagged versions of XRCC3 or Rad51C. Ni2+-binding experiments demonstrate that XRCC3 and Rad51C interact in human cells. In addition, we find that Rad51C, but not XRCC3, interacts directly or indirectly with Rad51B, Rad51D and XRCC2. These results argue that there are at least two complexes of Rad51 paralogs in human cells (Rad51C-XRCC3 and Rad51B-Rad51C-Rad51D-XRCC2), both containing Rad51C. Moreover, Rad51 is not found in these complexes. X-ray treatment did not alter either the level of any Rad51 paralog or the observed interactions between paralogs. However, the endogenous level of Rad51C is moderately elevated in the XRCC3-overexpressing cell line, suggesting that dimerization between these proteins might help stabilize Rad51C.

Processing of bistranded abasic DNA clusters in gamma-irradiated human hematopoietic cells.

Clustered DNA damages--two or more lesions on opposing strands and within one or two helical turns--are formed in cells by ionizing radiation or radiomimetic antitumor drugs. They are hypothesized to be difficult to repair, and thus are critical biological damages. Since individual abasic sites can be cytotoxic or mutagenic, abasic DNA clusters are likely to have significant cellular impact. Using a novel approach for distinguishing abasic clusters that are very closely spaced (putrescine cleavage) or less closely spaced (Nfo protein cleavage), we measured induction and processing of abasic clusters in 28SC human monocytes that were exposed to ionizing radiation. gamma-rays induced approximately 1 double-strand break: 1.3 putrescine-detected abasic clusters: 0.8 Nfo-detected abasic clusters. After irradiation, the 28SC cells rejoined double-strand breaks efficiently within 24 h. In contrast, in these cells, the levels of abasic clusters decreased very slowly over 14 days to background levels. In vitro repair experiments that used 28SC cell extracts further support the idea of slow processing of specific, closely spaced abasic clusters. Although some clusters were removed by active cellular repair, a substantial number was apparently decreased by 'splitting' during DNA replication and subsequent cell division. The existence of abasic clusters in 28SC monocytes, several days after irradiation suggests that they constitute persistent damages that could lead to mutation or cell killing.

An ATM-independent S-phase checkpoint response involves CHK1 pathway.

After exposure to genotoxic stress, proliferating cells actively slow down the DNA replication through a S-phase checkpoint to provide time for repair. We report that in addition to the ataxia-telangiectasia mutated (ATM)-dependent pathway that controls the fast response, there is an ATM-independent pathway that controls the slow response to regulate the S-phase checkpoint after ionizing radiation in mammalian cells. The slow response of S-phase checkpoint, which is resistant to wortmannin, sensitive to caffeine and UCN-01, and related to cyclin-dependent kinase phosphorylation, is much stronger in CHK1 overexpressed cells, and it could be abolished by Chk1 antisense oligonucleotides. These results provide evidence that the ATM-independent slow response of S-phase checkpoint involves CHK1 pathway.

Gene conversion is strongly induced in human cells by double-strand breaks and is modulated by the expression of BCL-x(L).

Homology-directed repair (HDR) of DNA double-strand breaks (DSBs) contributes to the maintenance of genomic stability in rodent cells, and it has been assumed that HDR is of similar importance in DSB repair in human cells. However, some outcomes of homologous recombination can be deleterious, suggesting that factors exist to regulate HDR. We demonstrated previously that overexpression of BCL-2 or BCL-x(L) enhanced the frequency of X-ray-induced TK1 mutations, including loss of heterozygosity events presumed to arise by mitotic recombination. The present study was designed to test whether HDR is a prominent DSB repair pathway in human cells and to determine whether ectopic expression of BCL-x(L) affects HDR. Using TK6-neo cells, we find that a single DSB in an integrated HDR reporter stimulates gene conversion 40-50-fold, demonstrating efficient DSB repair by gene conversion in human cells. Significantly, DSB-induced gene conversion events are 3-4-fold more frequent in TK6 cells that stably overexpress the antiapoptotic protein BCL-X(L). Thus, HDR plays an important role in maintaining genomic integrity in human cells, and ectopic expression of BCL-x(L) enhances HDR of DSBs. This is the first study to highlight a function for BCL-x(L) in modulating DSB repair in human cells.

Silencing expression of the catalytic subunit of DNA-dependent protein kinase by small interfering RNA sensitizes human cells for radiation-induced chromosome damage, cell killing, and mutation.

Targeted gene silencing in mammalian cells by RNA interference (RNAi) using small interfering RNAs (siRNAs) was recently described by Elbashir et al. (S. M. Elbashir et al., Nature (Lond.), 411: 494-498, 2001). We have used this methodology in several human cell strains to reduce expression of the Prkdc (DNA-PKcs) gene coding for the catalytic subunit of the DNA-dependent protein kinase (DNA-PKcs) that is involved in the nonhomologous end joining of DNA double-strand breaks. We have also demonstrated a radiosensitization for several phenotypic endpoints of radiation damage. In low-passage normal human fibroblasts, siRNA knock-down of DNA-PKcs resulted in a reduced capacity for restitution of radiation-induced interphase chromosome breaks as measured by premature chromosome condensation, an increased yield of acentric chromosome fragments at the first postirradiation mitosis, and an increased radiosensitivity for cell killing. For three strains of related human lymphoblasts, DNA-PKcs-targeted siRNA transfection resulted in little or no increase in radiosensitivity with respect to cell killing, a 1.5-fold decrease in induced mutant yield in TK6- and p53-null NH32 cells, but about a 2-fold increase in induced mutant yield in p53-mutant WTK1 cells at both the hypoxanthine quanine phosphoribosyl transferase (hprt) and the thymidine kinase loci.

Transforming growth factor-beta1 mediates cellular response to DNA damage in situ.

Transforming growth factor (TGF)-beta1 is rapidly activated after ionizing radiation, but its specific role in cellular responses to DNA damage is not known. Here we use Tgfbeta1 knockout mice to show that radiation-induced apoptotic response is TGF-beta1 dependent in the mammary epithelium, and that both apoptosis and inhibition of proliferation in response to DNA damage decrease as a function of TGF-beta1 gene dose in embryonic epithelial tissues. Because apoptosis in these tissues has been shown previously to be p53 dependent, we then examined p53 protein activation. TGF-beta1 depletion, by either gene knockout or by using TGF-beta neutralizing antibodies, resulted in decreased p53 Ser-18 phosphorylation in irradiated mammary gland. These data indicate that TGF-beta1 is essential for rapid p53-mediated cellular responses that mediate cell fate decisions in situ.

A deficiency in DNA repair and DNA-PKcs expression in the radiosensitive BALB/c mouse.

We have studied the efficiency of DNA double strand break (DSB) rejoining in primary cells from mouse strains that show large differences in in vivo radiosensitivity and tumor susceptibility. Cells from radiosensitive, cancer-prone BALB/c mice showed inefficient end joining of gamma ray-induced DSBs as compared with cells from all of the other commonly used strains and F1 hybrids of C57BL/6 and BALB/c mice. The BALB/c repair phenotype was accompanied by a significantly reduced expression level of DNA-PKcs protein as well as a lowered DNA-PK activity level as compared with the other strains. In conjunction with published reports, these data suggest that natural genetic variation in nonhomologous end joining processes may have a significant impact on the in vivo radiation response of mice.

Different mechanisms of radiation-induced loss of heterozygosity in two human lymphoid cell lines from a single donor.

Allelic loss is an important mutational mechanism in human carcinogenesis. Loss of heterozygosity (LOH) at an autosomal locus is one outcome of the repair of DNA double-strand breaks (DSBs) and can occur by deletion or by mitotic recombination. We report that mitotic recombination between homologous chromosomes occurred in human lymphoid cells exposed to densely ionizing radiation. We used cells derived from the same donor that express either normal TP53 (TK6 cells) or homozygous mutant TP53 (WTK1 cells) to assess the influence of TP53 on radiation-induced mutagenesis. Expression of mutant TP53 (Met 237 Ile) was associated with a small increase in mutation frequencies at the hemizygous HPRT (hypoxanthine phosphoribosyl transferase) locus, but the mutation spectra were unaffected at this locus. In contrast, WTK1 cells (mutant TP53) were 30-fold more susceptible than TK6 cells (wild-type TP53) to radiation-induced mutagenesis at the TK1 (thymidine kinase) locus. Gene dosage analysis combined with microsatellite marker analysis showed that the increase in TK1 mutagenesis in WTK1 cells could be attributed, in part, to mitotic recombination. The microsatellite marker analysis over a 64-cM region on chromosome 17q indicated that the recombinational events could initiate at different positions between the TK1 locus and the centromere. Virtually all of the recombinational LOH events extended beyond the TK1 locus to the most telomeric marker. In general, longer LOH tracts were observed in mutants from WTK1 cells than in mutants from TK6 cells. Taken together, the results demonstrate that the incidence of radi-ation-induced mutations is dependent on the genetic background of the cell at risk, on the locus examined, and on the mechanisms for mutation available at the locus of interest.

Modest increased sensitivity to radiation oncogenesis in ATM heterozygous versus wild-type mammalian cells.

Subpopulations that are genetically predisposed to radiation-induced cancer could have significant public health consequences. Individuals homozygous for null mutations at the ataxia telangiectasia gene are indeed highly radiosensitive, but their numbers are very small. Ataxia Telangiectasia heterozygotes (1-2% of the population) have been associated with somewhat increased radiosensitivity for some end points, but none directly related to carcinogenesis. Here, intralitter comparisons between wild-type mouse embryo fibroblasts and mouse embryo fibroblasts carrying ataxia telangiectasia mutated (ATM) null mutation indicate that the heterozygous cells are more sensitive to radiation oncogenesis than their normal, litter-matched, counterparts. From these data we suggest that Ataxia Telangiectasia heterozygotes could indeed represent a societally-significant radiosensitive human subpopulation.

Elevated breast cancer risk in irradiated BALB/c mice associates with unique functional polymorphism of the Prkdc (DNA-dependent protein kinase catalytic subunit) gene.

Female BALB/c mice are unusually radiosensitive and more susceptible than C57BL/6 and other tested inbred mice to ionizing radiation (IR)-induced mammary tumors. This breast cancer susceptibility is correlated with elevated susceptibility for mammary cell transformation and genomic instability following irradiation. In this study, we report the identification of two BALB/c strain-specific polymorphisms in the coding region of Prkdc, the gene encoding the DNA-dependent protein kinase catalytic subunit, which is known to be involved in DNA double-stranded break repair and post-IR signal transduction. First, we identified an A --> G transition at base 11530 resulting in a Met --> Val conversion at codon 3844 (M3844V) in the phosphatidylinositol 3-kinase domain upstream of the scid mutation (Y4046X). Second, we identified a C --> T transition at base 6418 resulting in an Arg --> Cys conversion at codon 2140 (R2140C) downstream of the putative leucine zipper domain. This unique PrkdcBALB variant gene is shown to be associated with decreased DNA-dependent protein kinase catalytic subunit activity and with increased susceptibility to IR-induced genomic instability in primary mammary epithelial cells. The data provide the first evidence that naturally arising allelic variation in a mouse DNA damage response gene may associate with IR response and breast cancer risk.

Methylation of p16(INK4a) promoters occurs in vivo in histologically normal human mammary epithelia.

Cultures of human mammary epithelial cells (HMECs) contain a subpopulation of variant cells with the capacity to propagate beyond an in vitro proliferation barrier. These variant HMECs, which contain hypermethylated and silenced p16(INK4a) (p16) promoters, eventually accumulate multiple chromosomal changes, many of which are similar to those detected in premalignant and malignant lesions of breast cancer. To determine the origin of these variant HMECs in culture, we used Luria-Delbrück fluctuation analysis and found that variant HMECs exist within the population before the proliferation barrier, thereby raising the possibility that variant HMECs exist in vivo before cultivation. To test this hypothesis, we examined mammary tissue from normal women for evidence of p16 promoter hypermethylation. Here we show that epithelial cells with methylation of p16 promoter sequences occur in focal patches of histologically normal mammary tissue of a substantial fraction of healthy, cancer-free women.

Cyclooxygenase-2 expression is related to nuclear grade in ductal carcinoma in situ and is increased in its normal adjacent epithelium.

Cyclooxygenase-2 (COX-2) is emerging as an important cancer biomarker and is now an experimental target for solid tumor treatment.However, no study has exclusively focused on COX-2 expression in early lesions such as ductal carcinoma in situ (DCIS). We examined COX-2 expression by immunohistochemistry in 46 cases of women undergoing surgical resection for DCIS. We found that COX-2 expression was detected in 85% of all DCIS specimens, with increased COX-2 staining correlating with higher nuclear grade. Strikingly, COX-2 staining intensity in the normal adjacent epithelium was stronger than in the DCIS lesion itself. Our observations demonstrate that COX-2 is up-regulated in the normal adjacent epithelium and supports the hypothesis that the surrounding epithelial tissue is part of the disease process in DCIS.

Characterization of a novel epigenetic effect of ionizing radiation: the death-inducing effect.

The detrimental effects associated with exposure to ionizing radiation have long been thought to result from the direct targeting of the nucleus leading to DNA damage; however, the emergence of concepts such as radiation-induced genomic instability and bystander effects have challenged this dogma. After cellular exposure to ionizing radiation, we have isolated a number of clones of Chinese hamster-human hybrid GM10115 cells that demonstrate genomic instability as measured by chromosomal destabilization. These clones show dynamic and persistent generation of chromosomal rearrangements multiple generations after the original insult. We hypothesize that these unstable clones maintain this delayed instability phenotype by secreting factors into the culture medium. To test this hypothesis we transferred filtered medium from unstable cells to unirradiated GM10115 cells. No GM10115 cells were able to survive this medium. This phenomenon by which GM10115 cells die when cultured in medium from chromosomally unstable GM10115 clones is the death-inducing effect. Medium transfer experiments indicate that a factor or factors is/are secreted by unstable cells within 8 h of growth in fresh medium and result in cell killing within 24 h. These factors are stable at ambient temperature but do not survive heating or freezing, and are biologically active when diluted with fresh medium. We present the initial description and characterization of the death-inducing effect. This novel epigenetic effect of radiation has implications for radiation risk assessment and for health risks associated with radiation exposure.