Epigenetics and the plasticity of differentiation in normal and cancer stem cells.

Embryonic stem cells are characterized by their differentiation to all cell types during embryogenesis. In adult life, different tissues also have somatic stem cells, called adult stem cells, which in specific niches can undergo multipotent differentiation. The use of these adult stem cells has considerable therapeutic potential for the regeneration of damaged tissues. In both embryonic and adult stem cells, differentiation is controlled by epigenetic mechanisms, and the plasticity of differentiation in these cells is associated with transcription accessibility for genes expressed in different normal tissues. Abnormalities in genetic and/or epigenetic controls can lead to development of cancer, which is maintained by self-renewing cancer stem cells. Although the genetic abnormalities produce defects in growth and differentiation in cancer stem cells, these cells have not always lost the ability to undergo differentiation through epigenetic changes that by-pass the genomic abnormalities, thus creating the basis for differentiation therapy. Like normal stem cells, cancer stem cells can show plasticity for differentiation. This plasticity of cancer stem cells is also associated with transcription accessibility for genes that are normally expressed in different tissues, including tissues other than those from which the cancers originated. This broad transcription accessibility can also contribute to the behavior of cancer cells by overexpressing genes that promote cell viability, growth and metastasis.

The RUNX3 gene--sequence, structure and regulated expression.

The RUNX3 gene belongs to the runt domain family of transcription factors that act as master regulators of gene expression in major developmental pathways. In mammals the family includes three genes, RUNX1, RUNX2 and RUNX3. Here, we describe a comparative analysis of the human chromosome 1p36.1 encoded RUNX3 and mouse chromosome 4 encoded Runx3 genomic regions. The analysis revealed high similarities between the two genes in the overall size and organization and showed that RUNX3/Runx3 is the smallest in the family, but nevertheless exhibits all the structural elements characterizing the RUNX family. It also revealed that RUNX3/Runx3 bears a high content of the ancient mammalian repeat MIR. Together, these data delineate RUNX3/Runx3 as the evolutionary founder of the mammalian RUNX family. Detailed sequence analysis placed the two genes at a GC-rich H3 isochore with a sharp transition of GC content between the gene sequence and the downstream intergenic region. Two large conserved CpG islands were found within both genes, one around exon 2 and the other at the beginning of exon 6. RUNX1, RUNX2 and RUNX3 gene products bind to the same DNA motif, hence their temporal and spatial expression during development should be tightly regulated. Structure/function analysis showed that two promoter regions, designated P1 and P2, regulate RUNX3 expression in a cell type-specific manner. Transfection experiments demonstrated that both promoters were highly active in the GM1500 B-cell line, which endogenously expresses RUNX3, but were inactive in the K562 myeloid cell line, which does not express RUNX3.

Spatial and temporal expression pattern of Runx3 (Aml2) and Runx1 (Aml1) indicates non-redundant functions during mouse embryogenesis.

The human RUNX3/AML2 gene belongs to the 'runt domain' family of transcription factors that act as gene expression regulators in major developmental pathways. Here, we describe the expression pattern of Runx3 during mouse embryogenesis compared to the expression pattern of Runx1. E10.5 and E14.5-E16.5 embryos were analyzed using both immunohistochemistry and beta-galactosidase activity of targeted Runx3 and Runx1 loci. We found that Runx3 expression overlapped with that of Runx1 in the hematopoietic system, whereas in sensory ganglia, epidermal appendages, and developing skeletal elements, their expression was confined to different compartments. These data provide new insights into the function of Runx3 and Runx1 in organogenesis and support the possibility that cross-regulation between them plays a role in embryogenesis.

Elevated Cu/Zn-SOD exacerbates radiation sensitivity and hematopoietic abnormalities of Atm-deficient mice.

Patients with the genetic disorder ataxia-telangiectasia (A-T) display a pleiotropic phenotype that includes neurodegeneration, immunodeficiency, cancer predisposition and hypersensitivity to ionizing radiation. The gene responsible is ATM, and ATM:-knockout mice recapitulate most features of A-T. In order to study the involvement of oxidative stress in the A-T phenotype, we examined mice deficient for Atm and overexpressing human Cu/Zn superoxide dismutase (SOD1). We report that elevated levels of SOD1 exacerbate specific features of the murine Atm- deficient phenotype, including abnormalities in hematopoiesis and radiosensitivity. The data are consistent with the possibility that oxidative stress contributes to some of the clinical features associated with the A-T phenotype.

Regulation of p53 stability and p53-dependent apoptosis by NADH quinone oxidoreductase 1.

The tumor suppressor gene wild-type p53 encodes a labile protein that accumulates in cells after different stress signals and can cause either growth arrest or apoptosis. One of the p53 target genes, p53-inducible gene 3 (PIG3), encodes a protein with significant homology to oxidoreductases, enzymes involved in cellular responses to oxidative stress and irradiation. This fact raised the possibility that cellular oxidation-reduction events controlled by such enzymes also may regulate the level of p53. Here we show that NADH quinone oxidoreductase 1 (NQO1) regulates p53 stability. The NQO1 inhibitor dicoumarol caused a reduction in the level of both endogenous and gamma-irradiation-induced p53 in HCT116 human colon carcinoma cells. This reduction was prevented by the proteasome inhibitors MG132 and lactacystin, suggesting enhanced p53 degradation in the presence of dicoumarol. Dicoumarol-induced degradation of p53 also was prevented in the presence of simian virus 40 large T antigen, which is known to bind and to stabilize p53. Cells overexpressing NQO1 were resistant to dicoumarol, and this finding indicates the direct involvement of NQO1 in p53 stabilization. NQO1 inhibition induced p53 degradation and blocked wild-type p53-mediated apoptosis in gamma-irradiated normal thymocytes and in M1 myeloid leukemic cells that overexpress wild-type p53. Dicoumarol also reduced the level of p53 in its mutant form in M1 cells. The results indicate that NQO1 plays an important role in regulating p53 functions by inhibiting its degradation.

Suppression or induction of apoptosis by opposing pathways downstream from calcium-activated calcineurin.

Ca(2+)-mobilizing compounds such as the Ca(2+) ionophore A23187 or the endoplasmic reticulum Ca(2+) ATPase inhibitor thapsigargin can suppress or induce apoptosis in the same cells. The use of different calcineurin inhibitors has shown that both suppression and induction of apoptosis by the Ca(2+)-mobilizing compounds were mediated by calcineurin activation. Ca(2+)-mobilizing compounds activated p38 and p44/42 mitogen-activated protein kinases (MAPKs). Induction of apoptosis by the Ca(2+)-mobilizing compounds was suppressed by an inhibitor of p38 MAPK but not by an inhibitor of p44/42 MAPK. These MAPK inhibitors did not suppress apoptosis induction by wild-type p53 or by withdrawal of IL-6 from IL-6-dependent cells that are mediated by calcineurin-independent pathways. These MAPK inhibitors also did not affect the ability of Ca(2+)-mobilizing compounds to suppress apoptosis. The results indicate that (i) Ca(2+)- mobilizing compounds activate different and opposing pathways that diverge downstream from calcineurin activation that can either suppress or induce apoptosis in the same cells; (ii) p38 MAPK but not p44/42 MAPK is involved in induction of apoptosis but not in its suppression by the Ca(2+)-mobilizing compounds; and (iii) neither p38 nor p44/42 MAPKs mediate induction of apoptosis by some calcineurin-independent pathways.

Cytokines as suppressors of apoptosis.

Many cytokines have been isolated by their ability to induce growth and have been called growth factors. But these cytokines are also essential to induce cell viability, and cell viability and growth can be separately regulated. Using as examples myeloid hematopoietic cells, lymphocytes and neuronal cells, in vitro and in vivo studies have shown the role of cytokines in inducing viability of different cell types during development to mature cells. Some cytokines can act on more than one cell type. Cytokines induce viability of normal and cancer cells by suppressing the apoptotic machinery activated by wild-type p53, or by cytotoxic agents including irradiation and compounds used in cancer chemotherapy. Cytokines can be used to decrease apoptosis in normal cells and inhibition of cytokine activity may improve cancer therapy by enhancing apoptosis in cancer cells. The apoptosis suppressing function of cytokines is mediated by changing the balance in the activity of apoptosis inducing and suppressing genes. Apoptosis suppression is upstream of caspase activation in the apoptotic process. Cytokines can suppress multiple pathways leading to apoptosis, only some of which were suppressed by other agents such as some antioxidants, Ca(2+)-mobilizing compounds and protease inhibitors.

Different mechanisms for suppression of apoptosis by cytokines and calcium mobilizing compounds.

Overexpression of wild-type p53 in M1 myeloid leukemia cells induces apoptotic cell death that was suppressed by the calcium ionophore A23187 and the calcium ATPase inhibitor thapsigargin (TG). This suppression of apoptosis by A23187 or TG was associated with suppression of caspase activation but not with suppression of wild-type-p53-induced expression of WAF-1, mdm-2, or FAS. In contrast to suppression of apoptosis by the cytokines interleukin 6 (IL-6) and interferon gamma, a protease inhibitor, or an antioxidant, suppression of apoptosis by A23187 or TG required extracellular Ca2+ and was specifically abolished by the calcineurin inhibitor cyclosporin A. IL-6 induced immediate early activation of junB and zif/268 (Egr-1) but A23187 and TG did not. A23187 and TG also suppressed induction of apoptosis by doxorubicin or vincristine in M1 cells that did not express p53 by a cyclosporin A-sensitive mechanism. Suppression of apoptosis by A23187 or TG was not associated with autocrine production of IL-6. Apoptosis induced in IL-6-primed M1 cells after IL-6 withdrawal was not suppressed by A23187 or TG but was suppressed by the cytokines IL-6, IL-3, or interferon gamma. The results indicate that these Ca2+-mobilizing compounds can suppress some pathways of apoptosis suppressed by cytokines but do so by a different mechanism.

Expression of AML1-d, a short human AML1 isoform, in embryonic stem cells suppresses in vivo tumor growth and differentiation.

The human AML1 gene encodes a heterodimeric transcription factor which plays an important role in mammalian hematopoiesis. Several alternatively spliced AML1 mRNA species were identified, some of which encode short protein products that lack the transactivation domain. When transfected into cells these short isoforms dominantly suppress transactivation mediated by the full length AML1 protein. However, their biological function remains obscure. To investigate the role of these short species in cell proliferation and differentiation we generated embryonic stem (ES) cells overexpressing one of the short isoforms, AML1-d, as well as cells expressing the full length isoforms AML1-b and AML2. The in vitro growth rate and differentiation of the transfected ES cells were unchanged. However, overexpression of AML1-d significantly affected the ES cells' ability to form teratocarcinomas in vivo in syngeneic mice, while a similar overexpression of AML1-b and AML2 had no effect on tumor formation. Histological analysis revealed that the AML1-d derived tumors were poorly differentiated and contained numerous apoptotic cells. These data highlight the pleiotropic effects of AML1 gene products and demonstrate for the first time an in vivo growth regulation function for the short isoform AML1-d.

Cytokine suppression of protease activation in wild-type p53-dependent and p53-independent apoptosis.

M1 myeloid leukemic cells overexpressing wild-type p53 undergo apoptosis. This apoptosis can be suppressed by some cytokines, protease inhibitors, and antioxidants. We now show that induction of apoptosis by overexpressing wild-type p53 is associated with activation of interleukin-1beta-converting enzyme (ICE)-like proteases, resulting in cleavage of poly(ADP- ribose) polymerase and the proenzyme of the ICE-like protease Nedd-2. Activation of these proteases and apoptosis were suppressed by the cytokine interleukin 6 or by a combination of the cytokine interferon gamma and the antioxidant butylated hydroxyanisole, and activation of poly(ADP-ribose) polymerase and apoptosis were suppressed by some protease inhibitors. In a clone of M1 cells that did not express p53, vincristine or doxorubicin induced protease activation and apoptosis that were not suppressed by protease inhibitors, but were suppressed by interleukin 6. In another myeloid leukemia (7-M12) doxorubicin also induced protease activation and apoptosis that were not suppressed by protease inhibitors, but were suppressed by granulocyte-macrophage colony-stimulating factor. The results indicate that (i) overexpression of wild-type p53 by itself or treatment with cytotoxic compounds in wild-type p53-expressing or p53-nonexpressing myeloid leukemic cells is associated with activation of ICE-like proteases; (ii) cytokines exert apoptosis-suppressing functions upstream of protease activation; (iii) the cytotoxic compounds induce additional pathways in apoptosis; and (iv) cytokines can also suppress these other components of the apoptotic machinery.

Oxidative stress mediates impairment of muscle function in transgenic mice with elevated level of wild-type Cu/Zn superoxide dismutase.

Cases of familial amyotrophic lateral sclerosis (fALS; a neurodegenerative disorder) have been reported in which the gene for Cu/Zn superoxide dismutase (CuZnSOD) was mutated. Several studies with the fALS mutant CuZnSOD in transgenic mice and cells showed that the fALS mutations act through an as yet undefined dominant gain-of-function mechanism. Wild-type CuZnSOD catalyzes the dismutation of superoxide (O(2)(-).) but also produces hydroxyl radicals (.OH) with H(2)O(2) as substrate. Two laboratories have recently demonstrated that the .OH production ability was preferentially enhanced by the fALS mutant CuZnSOD, suggesting that this might be the function gained in fALS. In this study, we used transgenic CuZnSOD (Tg-CuZnSOD) mice with elevated levels of CuZnSOD to determine whether overexpression of wild-type CuZnSOD was also associated with increased .OH production and impaired muscle function. Enhanced formation of .OH was detected, by spin trapping, in brain and muscle extracts of the Tg-CuZnSOD mice. Three independently derived Tg-CuZnSOD lines showed muscle abnormalities, reflected by altered electromyography (EMG) and diminished performance in the rope grip test. After treatment with paraquat (PQ), a widely used herbicide and O(2)(-).-generating compound, muscle disability significantly deteriorated in Tg-CuZnSOD mice but not in control mice. The results indicate that elevated levels of CuZnSOD cause indigenous long-term oxidative stress leading to impairment of muscle function. These findings may provide valuable clues about the concurred role of indigenous oxidative stress and exogenous agents in the etiology of sporadic ALS and several other neurodegenerative diseases in which a specific subset of neurons is affected.

Differential suppression by protease inhibitors and cytokines of apoptosis induced by wild-type p53 and cytotoxic agents.

Apoptosis induced in myeloid leukemic cells by wild-type p53 was suppressed by different cleavage-site directed protease inhibitors, which inhibit interleukin-1 beta-converting enzyme-like, granzyme B and cathepsins B and L proteases. Apoptosis was also suppressed by the serine and cysteine protease inhibitor N-tosyl-L-phenylalanine chloromethylketone (TPCK) [corrected], but not by other serine or cysteine protease inhibitors including N alpha-p-tosyl-L-lysine chloromethylketone (TLCK), E64, pepstatin A, or chymostatin. Protease inhibitors suppressed induction of apoptosis by gamma-irradiation and cycloheximide but not by doxorubicin, vincristine, or withdrawal of interleukin 3 from interleukin 3-dependent 32D non-malignant myeloid cells. Induction of apoptosis in normal thymocytes by gamma-irradiation or dexamethasone was also suppressed by the cleavage-site directed protease inhibitors, but in contrast to the myeloid leukemic cells apoptosis in thymocytes was suppressed by TLCK but not by TPCK. The results indicate that (i) inhibitors of interleukin-1 beta-converting enzyme-like proteases and some other protease inhibitors suppressed induction of apoptosis by wild-type p53 and certain p53-independent pathways of apoptosis; (ii) the protease inhibitors together with the cytokines interleukin 6 and interferon-gamma or the antioxidant butylated hydroxyanisole gave a cooperative protection against apoptosis; (iii) these protease inhibitors did not suppress induction of apoptosis by some cytotoxic agents or by viability-factor withdrawal from 32D cells, whereas these pathways of apoptosis were suppressed by cytokines; (iv) there are cell type differences in the proteases involved in apoptosis; and (v) there are multiple pathways leading to apoptosis that can be selectively induced and suppressed by different agents.

Cellular oxidative stress and the control of apoptosis by wild-type p53, cytotoxic compounds, and cytokines.

Apoptosis induced by wild-type p53 or cytotoxic compounds in myeloid leukemic cells can be inhibited by the cytokines interleukin 6, interleukin 3, granulocyte-macrophage colony-stimulating factor, and interferon gamma and by antioxidants. The antioxidants and cytokines showed a cooperative protective effect against induction of apoptosis. Cells with a higher intrinsic level of peroxide production showed a higher sensitivity to induction of apoptosis and required a higher cytokine concentration to inhibit apoptosis. Decreasing the intrinsic oxidative stress in cells by antioxidants thus inhibited apoptosis, whereas increasing this intrinsic stress by adding H2O2 enhanced apoptosis. Induction of apoptosis by wild-type p53 was not preceded by increased peroxide production or lipid peroxidation and the protective effect of cytokines was not associated with a decrease in these properties. The results indicate that the intrinsic degree of oxidative stress can regulate cell susceptibility to wild-type p53-dependent and p53-independent induction of apoptosis and the ability of cytokines to protect cells against apoptosis.

Control of apoptosis in hematopoiesis and leukemia by cytokines, tumor suppressor and oncogenes.

Hematopoietic cells require certain cytokines including colony-stimulating factors and interleukins to maintain viability. Without these cytokines the program of apoptotic cell death is activated. Cells from many myeloid leukemias require cytokines for viability, and apoptosis is also activated in these leukemic cells after cytokine withdrawal resulting in reduced leukemogenicity. The same cytokines protect normal and leukemic cells from induction of apoptosis by irradiation and cytotoxic chemotherapeutic compounds. This suggests that decreasing the levels of viability inducing cytokines may increase the effectiveness of cytotoxic anti-cancer therapy. The susceptibility of normal and cancer cells to induction of apoptosis is also regulated by the balance between apoptosis-inducing genes such as the tumor suppressor wild-type p53, and c-myc and bax, and apoptosis-suppressing genes such as the oncogene mutant p53, and bcl-2 and bcl-XL. Cell susceptibility to induction of apoptosis in leukemic cells could be enhanced by increased expression of apoptosis-inducing genes and/or decreased expression of apoptosis-suppressing genes. Modulation of expression of apoptosis-regulating genes should thus also be useful for improvement of anti-cancer therapy.

A large variety of alternatively spliced and differentially expressed mRNAs are encoded by the human acute myeloid leukemia gene AML1.

The human chromosome 21 acute myeloid leukemia gene AML1 is frequently rearranged in the leukemia-associated translocations t(8;21) and t(3;21), generating fused proteins containing the amino-terminal part of AML1. In normal blood cells, five size classes (2-8 kb) of AML1 mRNAs have been previously observed. We isolated seven cDNAs corresponding to various AML1 mRNAs. Sequencing revealed that their size differences were mainly due to alternatively spliced 5' and 3' untranslated regions, some of which were vast, exceeding 1.5 kb (5') and 4.3 kb (3'). These untranslated regions contain sequences known to control mRNA translation and stability and seem to modulate AML1 mRNA stability. Further heterogeneity was found in the coding region due to the presence of alternatively spliced stop codon-containing exons. The latter led to production of polypeptides that were smaller than the full-length AML1 protein; they lacked the trans-activation domains but maintained DNA binding and heterodimerization ability. The size of these truncated products was similar to the AML1 segment in the fused t(8;21) and t(3;21) proteins. In thymus, only one mRNA species of 6 kb was detected. Using in situ hybridization, we showed that its expression was confined to the cortical region of the organ. The 6-kb mRNA was also prominent in cultured peripheral blood T cells, and its expression was markedly reduced upon mitogenic activation by phorbol myristate acetate (TPA) plus concanavalin A (ConA). These results and the presence of multiple coding regions flanked by long complex untranslated regions, suggest that AML1 expression is regulated at different levels by several control mechanisms generating the large variety of mRNAs and protein products.

Protection of human myeloid leukemic cells against doxorubicin-induced apoptosis by granulocyte-macrophage colony-stimulating factor and interleukin 3.

Hematopoietic cells require certain cytokines to maintain viability by preventing apoptotic cell death. These cytokines can also protect leukemic cell lines against induction of apoptosis by cytotoxic anticancer compounds. We now show that the cytokines granulocyte-macrophage colony-stimulating factor and interleukin 3 can protect primary human myeloid leukemic cells against doxorubicin-induced apoptosis. Protection was detected in cells from 72% of the myeloid leukemic patients tested. The results indicate that these, and perhaps other, hematopoietic cytokines can decrease the effectiveness of cytotoxic anticancer therapy in some human myeloid leukemias. Leukemic cell sensitization to cytotoxic therapy may, therefore, require decreasing the availability of certain cytokines.

A mutant p53 antagonizes the deregulated c-myc-mediated enhancement of apoptosis and decrease in leukemogenicity.

Myeloid leukemic M1 cells that do not express p53 and transfected M1 clones that constitutively express the [Val135]p53 mutant or deregulated c-myc or coexpressing both genes grew autonomously in culture with a similar growth rate and cloning efficiency. Expression of deregulated c-myc in M1 leukemic cells enhanced susceptibility to induction of apoptotic cell death and resulted in a reduced leukemogenicity when injected into isologous mice. Expression of the [Val135]p53 mutant did not change cell susceptibility to induction of apoptosis or leukemogenicity, but expression of this mutant p53 suppressed the effects of deregulated c-myc on these properties. The results indicate that the [Val135]p53 mutant can show a gain of function for susceptibility to apoptosis and leukemogenicity in leukemic cells with deregulated c-myc and, thus, enhance tumor development.

Thymic abnormalities and enhanced apoptosis of thymocytes and bone marrow cells in transgenic mice overexpressing Cu/Zn-superoxide dismutase: implications for Down syndrome.

The copper-zinc superoxide dismutase (CuZnSOD) gene resides on chromosome 21 and is overexpressed in Down syndrome (DS) patients. Transgenic CuZnSOD mice with elevated levels of CuZnSOD were used to determine whether, as in DS, overexpression of CuZnSOD was also associated with thymus and bone marrow abnormalities. Three independently derived transgenic CuZnSOD strains had abnormal thymi showing diminution of the cortex and loss of corticomedullary demarcation, resembling thymic defects in children with DS. Transgenic CuZnSOD mice were also more sensitive than control mice to in vivo injection of lipopolysaccharide (LPS), reflected by an earlier onset and enhanced apoptotic cell death in the thymus. This higher susceptibility to LPS-induced apoptosis was associated with an increased production of hydrogen peroxide and a higher degree of lipid peroxidation. When cultured under suboptimal concentrations of interleukin 3 or in the presence of tumour necrosis factor, bone marrow cells from transgenic CuZnSOD mice produced 2- to 3-fold less granulocyte and macrophage colonies than control. The results indicate that transgenic CuZnSOD mice have certain thymus and bone marrow abnormalities which are similar to those found in DS patients, and that the defects are presumably due to an increased oxidative damage resulting in enhanced cell death by apoptosis.

Regulation of bcl-2, bcl-XL and bax in the control of apoptosis by hematopoietic cytokines and dexamethasone.

Treatment of M1 myeloid leukemic cells with interleukin 6 (IL-6) or dexamethasone (DEX), both of which induce differentiation in these cells, down-regulated expression of the apoptosis-suppressing gene bcl-2 and the apoptosis-promoting gene bax but up-regulated expression of the apoptosis-suppressing gene bcl-XL. There was a higher expression of bcl-XL in cells treated with DEX or DEX plus IL-6 compared to cells treated with IL-6 alone. The alternatively spliced bcl-X gene, bcl-Xs, which interferes with the action of bcl-2, was not expressed. Treatment with IL-6 increased the susceptibility of these cells to induction of apoptosis by Adriamycin or cycloheximide, but treatment with DEX or with IL-6 and DEX did not. Withdrawal of DEX after up-regulation of bcl-XL resulted in a decrease in bcl-XL expression and a concomitant increase in cell susceptibility to induction of apoptosis. Another myeloid leukemia that shows barely detectable expression of bcl-2 also showed up-regulated expression of bcl-XL but no change in bax after induction of differentiation with granulocyte-macrophage colony-stimulating factor, and this reduced cell susceptibility to induction of apoptosis by Adriamycin or cycloheximide. The results indicate that the related apoptosis-regulating genes bcl-2, bcl-XL, and bax are differently regulated and that up-regulation of bcl-XL expression may compensate for down-regulation of bcl-2 in the balance between genes that control apoptosis.