Purification and characterization of a Mg2+-dependent endonuclease (AN34) from etoposide-treated human leukemia HL-60 cells undergoing apoptosis

An important biochemical hallmark of apoptosis is the cleavage of chromatin into oligonucleosomal fragments. Here, we purified a Mg2+-dependent endonuclease from etoposide-treated HL-60 cells undergoing apoptosis. High levels of Mg2+-dependent endonuclease activity were detected in etoposide-treated HL-60 cells, and this activity increased in a time-dependent manner following etoposide treatment. Such an activity could not be detected in untreated cells or in cells treated with etoposide in the presence of the caspase inhibitor benzyloxycarbonyl-Val-Ala-Asp-(OMe)-fluoromethyl ketone (zVAD-fmk) or the serine protease inhibitor tosyl-L-phenylalanine chloromethyl ketone (TPCK). This Mg2+-dependent endonuclease was purified by a series of chromatographic procedures. The enzyme preparation showed a single major protein band with Mr 34,000, determined by SDS-PAGE. The presence of the Mr 34,000 Mg2+-dependent endonuclease was also confirmed by activity gel analysis. The enzyme required only Mg2+ for full activity. pH optimum was in the range of 6.5-7.5. This enzyme introduced single- and double-strand breaks into SV40 DNA and produced internucleosomal DNA cleavage in isolated nuclei from untreated cells. The DNA breaks were terminated with 3'-OH, consistent with characteristic products of apoptotic chromatin fragmentation. We propose to designate this Mr 34,000 Mg2+-dependent endonuclease AN34 (apoptotic nuclease Mr 34,000).     

Mechanism of UV-induced apoptosis in human leukemia cells: roles of Ca2+/Mg(2+)-dependent endonuclease, caspase-3, and stress-activated protein kinases

Ultraviolet light (UV) induced rapid apoptosis of U937 leukemia cells, concurrent with DNA fragmentation and cleavage of poly(ADP-ribose)polymerase (PARP) by activated caspase-3. The in vitro reconstitution of intact HeLa S3 nuclei and apoptotic U937 cytosolic extract (CE) revealed that (i) Ca2+/Mg(2+)-dependent, Zn(2+)-sensitive endonuclease activated in the apoptotic CE induced DNA ladder in HeLa nuclei at pH 6.8-7.4, (ii) activated caspase-3 cleaved PARP in HeLa nuclei, and (iii) when the apoptotic CE was treated with the caspase-3 inhibitor (1 microM Ac-DEVD-CHO) or the caspase-1 inhibitor (10 microM Ac-YVAD-CHO), the former, but not the latter, caused a 50% inhibition of DNA fragmentation and the complete inhibition of PARP cleavage in HeLa nuclei. Similarly, Ac-DEVD-CHO (100 microM) inhibited apoptosis and DNA ladder by 50% and PARP cleavage completely in UV-irradiated U937 cells, but Ac-YVAD-CHO (100 microM) did not. Thus, UV-induced apoptosis of U937 cells involves the Ca2+/Mg(2+)-dependent endonuclease pathway and the caspase-3-PARP cleavage-Ca2+/Mg(2+)-dependent endonuclease pathway. The former pathway produced directly 50% of apoptotic DNA ladder, and the latter involved activated caspase-3 and PARP cleavage, followed by formation of the remaining 50% DNA ladder by the activated endonuclease. In UV-irradiated B-cell lines, further, p53-dependent increase of Bax resulted in a greater caspase-3 activation compared to its absence. However, UV-induced activation of JNK1 and p38 was not affected by the caspase-1 and -3 inhibitors in U937 cells, so that caspases-1 and -3 do not function upstream of JNK1 and p38.     

Ca2+/Mg(2+)-dependent endonuclease in marrow CD34 positive and erythroid cells in myelodysplasia

Endonucleases capable of producing internucleosomal DNA cleavage are one of the key enzymes in apoptosis. We examined endonuclease activities contained in nuclei of CD34+ and erythroid cells in the bone marrow (BM) from 12 patients with the myelodysplastic syndromes. The levels of Mg(2+)-dependent and acidic endonucleases showed little changes as compared with those from normal BM. By contrast, the level of Ca2+/Mg(2+)-dependent endonuclease was appreciably higher in MDS erythroid cells than normal counterparts, although the activity varied markedly in CD34+ and erythroid cells. Our results suggested that Ca2+/Mg(2+)-dependent endonuclease is related to ineffective erythropoiesis in MDS.     

Detection of Mg(2+)-dependent endonuclease activity in myeloid leukemia cell nuclei capable of producing internucleosomal DNA cleavage

We detected Mg(2+)-dependent, Ca(2+)-independent endonuclease activity in non-apoptotic myeloid leukemia cell nuclei using autodigestion method which cleaved the chromatin of the autologous leukemia cells to an oligonucleosomal length pattern. Similar endonuclease activity could be successfully recovered in the protein extracts of the human leukemia cell nuclei. The extracts consistently elicited characteristic DNA cleavage of another leukemia cell (KG-1) nuclei as the target, the enzyme activity of which had been inactivated. We propose that this method is a useful tool for the study of endonucleases involved in apoptosis.     

DNA fragmentation into 200-250 and/or 30-50 kilobase pair fragments in rat liver nuclei is stimulated by Mg2+ alone and Ca2+/Mg2+ but not by Ca2+ alone

Internucleosomal cleavage of DNA has often been regarded as the biochemical hallmark of apoptosis. We now demonstrate in isolated rat liver nuclei that DNA is initially cleaved into > or = 700, 200-250 kbp and 30-50 kbp fragments via a multi-step process, which is activated by Mg2+ and Mg2+(+)Ca2+ but not by Ca2+ alone. The subsequent internucleosomal cleavage requires both cations. These findings demonstrate that a key event in the apoptotic process is the fragmentation of DNA into large kbp fragments by either a Mg(2+)-dependent process (which can be potentiated by Ca2+) and/or by a Ca2+/Mg2+ activated endonuclease(s).     

Apoptosis occurs in granulosa cells but not cumulus cells in the atretic antral follicles in pig ovaries

The porcine antral follicles, 3-6 mm in diameter, were dissected from the ovaries of mature pigs, and then granulosa and cumulus cells were isolated from each follicle. In atretic follicles, high activity of neutral Ca2+/Mg2+-dependent endonuclease and DNA ladder formation, estimated by electrophoresis, were noted in granulosa cells but not in cumulus cells. Extremely low activity of the endonuclease and no DNA ladder formation were observed in both types of cells obtained from healthy follicles. Moreover, apoptotic cells were observed histochemically among granulosa cells only. A good correlation (r = 0.987) between the endonuclease activity of granulosa cells and the progesterone/estradiol ratio of follicular fluid in each follicle was found. These results suggest that apoptosis occurs in granulosa cells but not cumulus cells in the atretic antral follicles in pigs.     

The nuclease activity of the cell nuclei in 2 lines of murine myeloma sp2/0 differing in their resistance to cytostatics in adriamycin-induced apoptosis

The endonuclease activity of two drug-sensitive and drug-resistant mouse myeloma cell lines during cytotoxic drug-induced apoptosis was studied. It was shown that internucleosomal fragmentation of DNA in drug-sensitive line sp2/0, undergoing apoptosis in the presence of adriamycin and colchicine, was not dependent on intracellular calcium content and was associated with activation of both Ca(2+)-Mg(2+)-dependent and acidic cation-independent endonucleases. In contrast, in multidrug resistant spEBR-5 cells, treated with the same drugs, only Ca(2+)-Mg(2+)-dependent endonuclease activity was detected. These data suggest that the differences in the pattern of endonuclease activity revealed in these cells are linked to drug-resistant phenotype and do not depend on the apoptosis-inducing agent used.     

Caspase-1 is not involved in CD95/Fas-induced apoptosis in Jurkat T cells

It is now well established that the caspases, a family of cysteine proteases, play a key role in apoptosis. Although overexpressing each of the caspases in cells triggered apoptosis, the precise role and contribution of individual caspases are still unclear. Caspase-1, the first caspase discovered, was initially implicated in mammalian apoptosis because of its similarity to the gene product ced-3. Using whole cells as well as an in vitro system to study apoptosis, the role of caspase-1 in Fas-mediated apoptosis in Jurkat T cells was examined in greater detail. Using various peptide-based caspase inhibitors, our results showed that N-acetyl-Tyr-Val-Ala-Asp chloromethyl ketone and benzyloxycarbonyl-Val-Ala-Asp (OMe) fluoromethyl ketone efficiently blocked Fas-mediated apoptosis in Jurkat T cells, whereas N-acetyl-Tyr-Val-Ala-Asp aldehyde, which is more specific for caspase-1, had little effect. Cell lysates derived from anti-Fas-stimulated cells, which readily induced apoptotic nuclei morphology and DNA fragmentation in isolated thymocyte nuclei, had no caspase-1 activity using proIL-1beta as a substrate. Time-course studies showed no caspase-1 activity during the activation of apoptosis in Jurkat cells by agonistic Fas antibodies. Furthermore, no pro-caspase-1 protein nor activated form of the protein was detected in normal or apoptotic Jurkat cells. In contrast, both caspase-2 and caspase-3 were readily detected as proenzymes in control cells and their activated forms were detected in apoptotic cells. Incubation of recombinant active caspase-1 with control cell lysates did not activate the apoptotic cascade as shown by the lack of detectable apoptotic nuclei promoting activity using isolated nuclei as substrate. However, under similar conditions proIL-1beta was readily processed into the mature cytokine, indicating that the recombinant caspase-1 remained active in the presence of control cell lysates. Taken together our results demonstrate that caspase-1 is not required for the induction of apoptosis in Jurkat T cells mediated by the Fas antigen.     

Role of proteolysis in apoptosis: involvement of serine proteases in internucleosomal DNA fragmentation in immature thymocytes

Three chemically distinct serine, but not cysteine, protease inhibitors (phenylmethylsulphonyl fluoride, N-tosyl-L-phenylalanylchloromethyl ketone and 3,4-dichloroisocoumarin) prevented, in a dose-dependent manner, the characteristic apoptotic internucleosomal DNA cleavage (DNA ladder) typically observed in thymocytes in response to dexamethasone and teniposide VM-26. This effect was not the result of a direct inhibition of the Ca2+,Mg(2+)-dependent endonuclease, since oligonucleosomal DNA cleavage occurred in the presence of these inhibitors in isolated nuclei. The proteolytic step occurred at a very early stage of apoptosis, and preincubation of thymocytes with the inhibitors before dexamethasone or teniposide VM-26 were added irreversibly suppressed ladder formation. This implied that the cellular effector(s) of these compounds preexisted and were not resynthesized in response to the inducers of apoptosis. Serine protease inhibitors also suppressed apoptotic cell shrinkage and complete nuclear collapse, suggesting that these morphological changes were directly related to internucleosomal fragmentation of DNA. However, the serine protease inhibitors did not prevent high molecular weight DNA cleavage (> 50 kilobases) that preceded the ladder formation and thymocytes still died by apoptosis. This supported the view that internucleosomal DNA cleavage, considered to be the biochemical marker of apoptosis, might in fact be a late and dispensable step and that the newly described high molecular weight DNA cleavage might be a better indicator of apoptosis.     

The neurochemical, molecular and ultrastructural mechanisms of the formation of latent postresuscitation encephalopathy

The content of total RNA and DNA, activity of Ca2+, Mg(2+)-dependent DNA endonuclease, and ultrastructural changes in nerve tissue cells were examined in the brain cortex of narcotized dogs 1 to 3 months after a 4-hour hemorrhagic shock (arterial pressure 40 mm Hg). A new variant of reconstruction of cell membranes and organelles formed by them was revealed, developing in the brain neurons in the course of adaptation during the first-third months of the postshock period. Evidently, the molecular base of development of an atypical variant of cell structure rearrangement in the remote period after shock is the internucleosomal fragmentation of a part of the DNA of nerve cells resultant from DNA endonucleolysis and subsequent information disintegration of a cell as a system. This distorts the process of biosynthesis of supramolecular ensembles, specifically, of nerve cell biomembranes.     

Spontaneously produced anti-DNA/DNase I autoantibodies modulate nuclear apoptosis in living cells

Apoptosis has received increased attention over the past decade, and it is established as an essential process in physiological and disease states. Much effort has been devoted to understanding the intracellular mechanisms culminating in apoptosis; intense investigation has also focused on its role during inflammation. Despite these efforts, these events remain incompletely understood. It has been suggested that the Ca(2+)- and Mg2+-dependent endonuclease that mediates DNA fragmentation is DNase I; however, the precise role of DNase I during apoptosis has been debated. Recent observations using anti-DNA antibodies derived from autoimmune mice (MRL-lpr/lpr) provided both the means and the reagents to approach these issues in a more direct manner. We previously discovered that many anti-DNA antibodies cross-react with DNase I, and a subset of these Ig inhibited DNase I enzymatic activity in vitro. Serendipitously, in separate studies, a subset of these antibodies were observed to enter and localize within the nuclei of living cells. The aim of the present investigation was to determine whether these nuclear-localizing anti-DNA antibodies could interact with DNase I in living cells. We found that, once internalized, these autoantibodies bound DNase I and inhibited activity of the enzyme. Furthermore, living cells containing the intracellular antibodies appeared resistant to apoptotic stimuli; both morphological features of nuclear apoptosis and DNA fragmentation were inhibited. These results support a pivotal role for DNase I in apoptosis, and they provide a novel paradigm for autoantibody-mediated inflammatory disease.     

Isolation, purification, and characterization of a neutral Mg(2+)-dependent deoxyribonuclease of the Colorado potato beetle Leptinotarsa decemlineata Say

A neutral Mg(2+)-dependent deoxyribonuclease from the Colorado potato beetle was isolated and characterized in physicochemical terms. An electrophoretically homogeneous preparation of the enzyme was obtained using salt fractionation, Sephadex G-100 gel filtration, and subsequent preparative isoelectrofocusing in an Ultrodex layer. The molecular weight of the purified DNase preparation (with a purification degree of 104) and its isoelectric point were 100 kD and 9.1, respectively. The enzyme activity was maximal at pH 7.2 and 46 degrees C in the presence of 10 mM Mg2+. The DNase of the Colorado beetle preferentially hydrolysed denatured DNA via the endonuclease pathway, degrading the substrate to oligonucleoside-3'-phosphates. As far as the physical and chemical properties are concerned, this Colorado beetle DNase seems different from previously investigated DNases of other insect species.     

Role of serine and ICE-like proteases in induction of apoptosis by etoposide in human leukemia HL-60 cells

This study was undertaken to examine the role of proteases in etoposide-induced apoptosis of human leukemia HL-60 cells. We found the potent activity to produce internucleosomal DNA fragmentation in a 150 000 g supernatant of cell lysate which was prepared from etoposide-treated HL-60 cells undergoing apoptosis. This nuclear-DNA fragmenting activity could be detected when the supernatant was incubated with isolated nuclei under Mg2+-dependent conditions. On the other hand, we could not detect such activity in the supernatant of cell lysate from non-treated HL-60 cells. Treatment of the supernatant with a serine protease inhibitor, N-tosyl-L-phenylala-nylchloromethyl ketone (TPCK), abolished the DNA fragmenting activity. An inhibitor of interleukin 1-beta-converting enzyme (ICE), Z-Val-Ala-Asp-fluoromethyl ketone (VAD-FMK), had no effect on this DNA fragmenting activity in vitro. However, when the cells were incubated with etoposide in the presence of VAD-FMK, the formation of TPCK-sensitive DNA fragmenting activity was blocked. Our data indicate that serine and ICE-like proteases may be involved in etoposide-induced apoptosis at the different stages, and especially a serine protease may be closely associated with the final step for induction of internucleosomal DNA fragmentation during apoptosis in HL-60 cells.     

Dexamethasone-induced apoptosis involves cleavage of DNA to large fragments prior to internucleosomal fragmentation

Apoptosis is a major form of cell death, characterized morphologically by chromatin condensation and biochemically by endonuclease cleavage of DNA into oligonucleosomal fragments. Recently, we reported that zinc arrested dexamethasone-induced apoptosis in thymocytes at an early stage, as characterized morphologically by condensation of heterochromatin into clumps abutting the nuclear membrane. In this study, we show that zinc completely inhibits endonuclease cleavage of DNA into oligonucleosomal fragments but does not prevent the cleavage of DNA into high molecular weight fragments. These results indicate that the formation of these high molecular weight fragments, which correlates with the very early morphological features of apoptosis, is a critical event in glucocorticoid-induced apoptosis. The formation of these high molecular weight fragments, despite the inhibition by zinc of the endonuclease cleavage of DNA, suggests that key enzyme(s), other than the Ca2+/Mg(2+)-dependent endonuclease, are involved at the earliest stages of induction of apoptosis.     

Frequency-shaped amplification changes the neural representation of speech with noise-induced hearing loss

Apoptosis is a mode of active cell death. We have examined whether 2-chloroethylethyl sulfide (CEES), a sulfur vesicating agent, triggers apoptosis as a cytotoxic mechanism. Incubation of thymocytes with CEES, resulted in an induction of apoptotic features of cell death. Treatment of cells with 100 microM CEES for 5 h increased DNA fragmentation to approximately 40% of control. The fragmentation of DNA was visualized by agarose gel electrophoresis. It showed ladder pattern of DNA fragmentation, which indicates internucleosomal cleavage of DNA. Further evidence of apoptosis was observed in morphological changes of nuclei by using the terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL) method. The percentage of TUNEL positive cells was dependent upon CEES concentrations. CEES induced the classical morphological features of apoptosis in nucleus. These features were accompanied by condensation of chromatin, which arranged in sharply declined clumps and fragmentation of nucleus. To study requirement for synthesis of new protein in CEES-induced apoptosis, we studied the effect of cycloheximide for apoptotic activity. This protein synthesis inhibitor did not suppress the CEES-induced apoptotic activity. Taken together, these results suggest that CEES-induced apoptosis as a cytotoxicmechanism and this process occurs independent of synthesis of new protein.     

Anatomical basis of a congenital hearing impairment: basilar papilla dysplasia in the Belgian Waterslager canary

We studied the role of proteases in apoptosis using a cell-free system prepared from a human leukemia cell line. HL60 cells are p53 null and extremely sensitive to a variety of apoptotic stimuli including DNA damage induced by the topoisomerase I inhibitor, camptothecin. We measured DNA fragmentation induced in isolated nuclei by cytosolic extracts using a filter elution assay. Cytosol from camptothecin-treated HL60 cells induced internucleosomal DNA fragmentation in nuclei from untreated cells. This fragmentation was suppressed by serine protease inhibitors. Serine proteases (trypsin, endoproteinase Glu-C, chymotrypsin A, and proteinase K) and papain by themselves induced DNA fragmentation in naive nuclei. This effect was enhanced in the presence of cytosol from untreated cells. Cysteine protease inhibitors (E-64, leupeptin, Ac-YVAD-CHO [ICE inhibitor]) did not affect camptothecin-induced DNA fragmentation. The apopain/Yama inhibitor, Ac-DEVD-CHO, and the proteasome inhibitor, MG-132, were also inactive both in the cell-free system and in whole cells. Interleukin-1 beta converting enzyme (ICE) or human immunodeficiency virus protease failed to induce DNA fragmentation in naive nuclei. Together, these results suggest that DNA damage activates serine protease(s) which in turn activate(s) nuclear endonuclease(s) during apoptosis in HL60 cells.     

IL-4 and interferon-gamma production in children with atopic disease

The mechanism by which 7,12-dimethylbenz[a]anthracene (DMBA) produces cytotoxicity in lymphocytes was investigated in these studies using the murine A20.1 B cell lymphoma. Results show that in vitro exposure of these cells to 10-30 microM DMBA for 4 hr produced an increase in intracellular Ca2+, DNA fragmentation, and subsequent cell death. Elevation of Ca2+ and DNA fragmentation induced by DMBA were greatly pronounced when the A20.1 cells were exposed at high cell density (10(7) cells/ml). DMBA-induced DNA fragmentation and cell death were inhibited by coexposure of A20.1 cells to a calcium chelator (EDTA), a general nuclease and polymerase inhibitor (aurintricarboxylic acid), and a protein synthesis inhibitor (cycloheximide). These agents have been previously shown to inhibit apoptosis in lymphocytes and other cells exposed to chemical agents. We also found that cyclosporin A, an inhibitor of Ca(2+)-dependent pathways of T and B cell activation, prevented apoptosis in the A20.1 cell line. These results demonstrate that DMBA induces programmed cell death (apoptosis) in the A20.1 murine B cell lymphoma by Ca(2+)-dependent pathways. The increased sensitivity of A20.1 at high cell density to Ca2+ elevation and DNA fragmentation suggests that cell to cell interactions may also be important in this process.     

Impairment of forearm vasodilatation to acetylcholine in hypercholesterolemia is reversed by aspirin

To study the intracellular apoptotic signaling pathways, we have established a cell-free system, in which DNA fragmentation of the isolated mouse liver nuclei was induced with lysates from the Fas-activated cells. We have found that the inactive nuclease present in the intact cell cytosol is activated by a caspase-3-like protease and the activated nuclease induces the nucleosomal DNA fragmentation. We attempted the purification of the inactive nuclease from bovine liver cytosol. The partially purified nuclease was activated by recombinant caspase-3, and to a lesser extent by caspase-6. The activated nuclease was able to digest plasmid DNA in addition to induce the DNA fragmentation of nuclei. DFF-45, which is a subunit of heterodimeric protein leading to DNA fragmentation upon its digestion by caspase-3, is found to inhibit the activity of the activated nuclease. These results suggest that the inactive nuclease in cytoplasm is converted to the active form by caspases, and the activated nuclease enters into nucleus to induce the DNA fragmentation. It is suggested that DFF-45 may function as an inhibitory factor of the caspase-sensitive nuclease in vivo.     

Absence of bcl-2 expression by activated CD45RO+ T lymphocytes in acute infectious mononucleosis supporting their susceptibility to programmed cell death

bcl-2 proto-oncogene encodes an inner mitochondrial membrane protein that blocks programmed cell death (apoptosis). There is now increasing evidence that regulation of bcl-2 expression is a determinant of life or death in normal lymphocytes. We have recently described that activated (CD45RO+) CD4+ and CD8+ T cells in acute infectious mononucleosis (IM) undergo apoptotic cell death on culturing, indicating an activation-driven cell death of mature T cells. In this work, we examine bcl-2 expression by activated T cells in acute IM using a flow-cytometric analysis with an anti-bcl-2 monoclonal antibody (MoAb). It was consistently observed that most T cells from acute IM patients displayed only much less bcl-2, while normal T cells expressed bcl-2 relatively strongly. Multicolor analysis showed that bcl-2-lacking T cells in acute IM were restricted to the CD45RO+ (activated) populations of CD4+, as well as CD8+ T cells. In contrast, the relatively intense levels of bcl-2 were expressed in both CD45RO+ and CD45RO- T-cell populations from normal subjects. This marked difference in bcl-2 expression of CD45RO+ T cells between acute IM and normal controls was also confirmed by Western blot analysis. Activated (CD45RO+) T cells with low bcl-2 expression, but not bcl-2-expressing CD45RO- T cells, in acute IM patients were found to die easily when cultured without added growth factors. However, in normal individuals, both CD45RO+ and CD45RO- T cells were relatively stable on culturing. These findings suggest that lack of bcl-2 expression by activated (CD45RO+) T cells in acute IM might be associated with their susceptibility to programmed cell death.     

Long-range fragmentation of the eukaryotic genome by exogenous and endogenous nucleases proceeds in a specific fashion via preferential DNA cleavage at matrix attachment sites

Small cell lung cancer cells (OC-NYH-VM) were permeabilized and treated with different nucleases. The long-range distribution of DNA cleavage sites in the amplified c-myc gene locus was then analyzed by pulsed field gel electrophoretic separation of the released 50-kilobase to 1-megabase DNA fragments followed by indirect end labeling. Exogenous DNase I and nucleases specific for the single-stranded DNA were found to generate similar nonrandom patterns of large DNA fragments. The cleavage sites were located close to or even colocalized with matrix attachment regions, which were mapped independently using a recently developed procedure for DNA loop excision by DNA topoisomerase II-mediated DNA cleavage. Endogenous acidic nuclease with the properties of DNase II also digested DNA preferentially in proximity to the matrix attachment regions, generating characteristic patterns of excised DNA loops and their oligomers. A similar, although less specific, pattern of DNA fragmentation was observed after incubation of permeabilized cells under conditions favoring the activity of endogenous neutral Ca(2+)- and Mg(2+)-dependent nucleases. These findings are discussed in the context of the current model of the spatial domain organization of eukaryotic genome.