Apoptosis and necrosis in toxicology: a continuum or distinct modes of cell death?

Aponecrosis: morphological and biochemical exploration of a syncretic process of cell death sharing apoptosis and necrosis

A rat fibroblastic cell line (rat-1/myc-ERtrade mark) was treated with different concentration of Antimycin A, a metabolic poison that affects mitochondrial respiratory chain complex III. The modes of cell death were analyzed by time-lapse videomicroscopy, in situ end-labeling (ISEL) technique, and ultrastructural analysis. Intracellular ATP levels were also measured in order to detect whether the energetic stores were determinant for the type of cell death. It was found that while apoptosis was the prevalent cell death in the fibroblasts treated with low doses, 100 or 200 microM Antimycin A, a new type of cell demise that shared dynamic, molecular, and morphological features with both apoptosis and necrosis represents the most common cell death when the cells were exposed to high doses, 300 or 400 microM, of the hypoxic stimulus. This new type of cell death has been chimerically termed aponecrosis. The inhibition of caspase 3, an enzyme critical for the apoptotic DNA degradation, caused a clear shift from aponecrosis to necrosis in the cell culture, suggesting that this new type of cell death could account for an incomplete execution of the apoptotic program and the following degeneration in necrosis. After being treated with higher doses, i.e., 1000 microM Antimycin A, almost all of the cells died by true necrosis. The analysis of the cellular energetic stores showed that the levels of ATP were a primary determinant in directing toward active cell death (apoptosis), aponecrosis, or necrosis. We conclude that chemically induced hypoxia produces different types of cell death depending on the intensity of the insult and on the ATP availability of the cell, and that the classic apoptosis and necrosis may represent only two extremes of a continuum of intermediate forms of cell demise.

Parvovirus H-1-induced cell death: influence of intracellular NAD consumption on the regulation of necrosis and apoptosis

The autonomous parvovirus H-1 exerts tumor-suppressive effects in living organisms and has been shown to specifically interfere with the survival of transformed cells in culture. The mechanism(s) by which H-1 virus induces death of transformed cells is not yet well understood. It has recently been reported that H-1 virus induces apoptotic cell death in the human monocytic U937 cell line, as assessed by biochemical and morphological changes of infected cells (Rayet, B., Lopez-Guerrero, J.-A., Rommelaere, J., Dinsart, C., 1998. Induction of programmed cell death by parvovirus H-1 in U937 cells: connection with the TNFalpha signalling pathway. J. Virol. 72, 8893-8903). Here we show that parvovirus H-1 infection induced early biochemical changes pointing to apoptotic events also in the transformed human keratinocyte cell line, HeLa, and the transformed rat fibroblast cell line, P1. Morphologic changes, however, and in particular the early breakdown of plasma membrane integrity, suggested that apoptosis did not go to completion, leading to necrotic cell death as the major result of parvovirus infection of HeLa and P1 cells. Parvovirus infection of these, and to a significantly lesser extent of U937 cells, was accompanied by rapid depletion of intracellular NAD stores. Inhibition of NAD-consuming enzymes interfered with parvovirus-induced NAD depletion and increased the proportion of H-1 virus-infected cells displaying apoptotic features of cell death. In contrast, a similar prevention of NAD depletion through stimulation of NAD production had little influence on the cell death pathway, suggesting that NAD-consuming enzymes may promote necrosis in a direct way rather than through inducing the overall drop of intracellular NAD.

Tumor cell viability in clear cell sarcoma requires DNA binding activity of the EWS/ATF1 fusion protein

Chimeric proteins resulting from characteristic chromosomal translocations are believed to play a key role in the development of neoplasia. The consistent chromosomal translocation t(12;22) found in Clear Cell sarcoma (CCS) fuses the genes for Ewing's sarcoma protein (EWS) and activating transcription factor 1 (ATF1). Contribution of the chimeric EWS/ATF1 protein to maintenance of the tumor phenotype was investigated using intracellular expression of an inhibitory anti-ATF1 single chain antibody fragment (scFv4). Transfection of scFv4 into a cell line (SU-CCS-1) derived from CCS resulted in a 90% reduction in cyclic AMP response element-driven reporter activity. The delivery of scFv4 into SU-CCS-1 cells by a Moloney sarcoma retroviral vector (SRalpha-Fv4) significantly reduced viability and induced apoptosis as measured by terminal deoxynucleotidetransferase-mediated dUTP-biotin nick end labeling and flow cytometry. Conversely, scFv4 had no effect on viability of HeLa cells. The level of EWS/ATF1 expression was found to be significantly higher in primary tumor tissue than in SU-CCS-1 cells or in 293T cells following introduction of an EWS/ATF1 expression vector. These studies demonstrate a direct role for the EWS/ATF1 fusion protein in maintaining tumor cell viability of Clear Cell sarcoma and indicate that intracellular antibodies may be used to achieve a phenotypic knockout of tumor-related proteins as a method to explore their function.

Structure-activity relationship studies of novel heteroretinoids: induction of apoptosis in the HL-60 cell line by a novel isoxazole-containing heteroretinoid

In a search for retinoic acid receptor (RAR and RXR)-selective ligands, a series of isoxazole retinoids was synthesized and evaluated in vitro in transcriptional activation and competition binding assays for RARs and RXRs. In addition, these compounds were evaluated for their differentiating, cytotoxic, and apoptotic activities. In general, these derivatives showed scarcely any binding affinity and were not active in the transcriptional assay. However, among these isoxazole derivatives, the cis-isomer 14b was identified as a potent inducer of apoptosis, and its activity was found to be 6.5 and 4 times superior than that of 13-cis- and 9-cis-retinoic acids, respectively. On the other hand, compound 13b, which has the trans stereochemistry at the double bond, was found not to be active in the apoptotic assay, but it was endowed with appreciable differentiating activity. Therefore, it seems that the different stereochemistry of the double bond may be associated with a different biological activity: potent apoptotic activity for the cis-isomer but differentiating activity for the trans structure. This biological behavior was found, at least in part, for the 9-cis- and 13-cis-retinoic acids with respect to the all-trans-retinoic acid. Thus, structure 14b could offer an interesting model for the design of new compounds endowed with apoptotic activity.

Hyperoxia inhibits proliferation of Mv1Lu epithelial cells independent of TGF-beta signaling

High concentrations of O(2) inhibit epithelial cell proliferation that resumes on recovery in room air. To determine whether growth arrest is mediated by transforming growth factor-beta (TGF-beta), changes in cell proliferation during exposure to hyperoxia were assessed in the mink lung epithelial cell line Mv1Lu and the clonal variant R1B, which is deficient for the type I TGF-beta receptor. Mv1Lu cells treated with TGF-beta accumulated in the G(1) phase of the cell cycle as determined by propidium iodide staining, whereas proliferation of R1B cells was unaffected by TGF-beta. In contrast, hyperoxia inhibited proliferation of both cell lines within 24 h of exposure through an accumulation in the S phase. Mv1Lu cells treated with TGF-beta and exposed to hyperoxia accumulated in the G(1) phase, suggesting that TGF-beta can inhibit the S phase accumulation observed with hyperoxia alone. Cyclin A was detected in cultures exposed to room air or growth arrested by hyperoxia while decreasing in cells growth arrested in the G(1) phase by TGF-beta. Finally, hyperoxia failed to activate a TGF-beta-dependent transcriptional reporter in both Mv1Lu and R1B cells. These findings reveal that simple growth arrest by hyperoxia involves a defect in S phase progression that is independent of TGF-beta signaling.

Neuronal apoptosis in acute necrotic insults: why is this subject such a mess?

It is now recognized that necrotic neurological insults often trigger apoptosis in a subset of neurons. It is also now apparent that such apoptosis rarely matches the 'classical' apoptosis seen during development or the physiological turnover of cells outside the nervous system. As a result, the view has emerged that the 'apoptosis-like' changes that follow necrotic insults represent a different phenomenon, which is on a vague continuum with the necrotic features of cell death. We suggest that apoptosis following neurological insults is, in actuality, mechanistically identical to classical apoptosis. However, the atypical apoptotic endpoints that are observed are inevitable, given the way in which insult-triggered apoptosis is likely to have evolved.

Cocaine cytotoxicity in hepatocyte cultures from phenobarbital-induced rats: involvement of reactive oxygen species and expression of antioxidant defense systems

The present study was designed to investigate whether cocaine modifies the production of reactive oxygen species, affects cellular enzyme-mediated antioxidant defense systems and, subsequently, promotes apoptosis and/or necrosis of hepatocytes. Primary cultures of hepatocytes isolated from phenobarbital-induced rats were exposed to cocaine (0-1000 microM) for 24 hr, and cell death (apoptosis or necrosis), antioxidant enzyme activities and mRNA levels, and peroxide generation were determined. Cocaine cytotoxicity by apoptosis was observed by detecting apoptotic nuclei using optic microscopy and by measurement of the hypodiploid peak (<2C) in DNA histograms obtained by flow cytometry. Necrosis was evidenced by lactate dehydrogenase (LDH) leakage, and peroxide production was quantified with 2',7'-dichlorodihydrofluorescein diacetate. Low concentrations of cocaine (less than 100 microM) resulted in an increase in dichlorofluorescein fluorescence, associated with an enhancement in apoptotic cell death and sharp decreases in the enzyme activities and RNAs of catalase and manganese-superoxide dismutase (Mn-SOD). The progressive decrease in peroxide production in cell cultures detected in the range of 250-1000 microM cocaine was associated with increases in LDH leakage and decreases in the percentage of apoptotic cells, accompanied by low levels in catalase and Mn-SOD enzyme activities and mRNAs, without apparent changes in apoptosis. These data indicate that oxygen radicals may contribute directly or indirectly to cocaine-induced apoptosis in cultured hepatocytes. We conclude that, in primary hepatocyte cultures, cocaine-induced cell death by necrosis was dependent on cocaine concentration, while cell death by apoptosis was parallel to peroxide concentration. The down-regulation of the gene expression of antioxidant enzyme systems should be one of the mechanisms involved in cocaine toxicity.

Inhibition of Fas receptor (CD95)-induced hepatic caspase activation and apoptosis by acetaminophen in mice

The mechanism of liver cell injury induced by an overdose of the analgesic acetaminophen (AAP) remains controversial. Recently, it was hypothesized that a significant number of hepatocytes die by apoptosis. Since caspases have been implicated as critical signal and effector proteases in apoptosis, we investigated their potential role in the pathophysiology of AAP-induced liver injury. Male C3Heb/FeJ mice were fasted overnight and then treated with 500 mg/kg AAP. Liver injury became apparent at 4 h and was more severe at 6 h (plasma ALT activities: 4110 +/- 320 U/liter; centrilobular necrosis). DNA fragmentation increased parallel to the increase of plasma ALT values. At 6 h there was a 420% increase of DNA fragmentation and a 74-fold increase of terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL)-positive cells located predominantly around central veins. However, the activity of the proapoptotic caspase-3 was not increased at any time after AAP. In contrast, injection of the anti-Fas antibody Jo-2 (positive control) caused a 28-fold increase of caspase-3 activity and severe DNA fragmentation before significant ALT release. Treatment with the caspase inhibitor ZVAD-CHF2 had no effect on AAP toxicity but completely prevented Jo-mediated apoptosis. In contrast, Jo-induced caspase activation and apoptosis could be inhibited by AAP treatment in a time- and dose-dependent manner. We conclude that AAP-induced DNA fragmentation does not involve caspases, suggesting a direct activation of endonucleases through elevated Ca2+ levels. In addition, electrophilic metabolites of AAP may inactivate caspases or their activation pathway. This indicates that AAP metabolism has the potential to inhibit signal transduction mechanisms of receptor-mediated apoptosis.

The potential of neuroprotection in glaucoma treatment

Visual field loss in glaucoma is due to death of retinal ganglion cells. Reducing or slowing down the loss of ganglion cells in glaucoma, a concept known as neuroprotection, would appear to be the only way forward. This does not imply that treatment of risk factors, such as elevated intraocular pressure, must not be continuously implemented. In this paper we point out that very little is known about the mechanisms of ganglion cell death in glaucoma and that data derived from studies on the "ideal animal model for glaucoma" must not be overemphasized. We also propose that the death processes of neurones in various diseases are fundamentally the same but vary in cause. Experimental data show that the death rate of neuronal populations is dependent on the impact of the insult and that neuroprotectants are more likely to benefit a patient in diseases in which the neurones die slowly, as in glaucoma, than in a disease in which the death of a set of neurones is rapid. We conclude that if a putative neuroprotectant can be administered in such a way that it reaches the retina in appropriate amounts and has insignificant side effects, it is likely to attenuate ganglion cell death and thus benefit the glaucoma patient.

The perturbation of apoptosis and mitosis by drugs and xenobiotics

Drugs such as the barbiturate phenobarbitone and fibrate hypolipidaemic agents, in addition to a range of chemicals of environmental and industrial significance, are able to perturb rodent tissue homeostasis, leading to tissue enlargement. Many of these xenobiotics are rodent nongenotoxic carcinogens since they do not damage DNA, yet cause tumours in the rat and mouse. These nongenotoxic carcinogens display both species and tissue specificity; for example, rat and mouse hepatocytes display S-phase induction and a suppression of apoptosis in response to drugs such as phenobarbitone or the hypolipidaemic peroxisome proliferators (PPs). In contrast, human hepatocytes or other types of rodent cells are refractory to these effects. However, in the absence of a discrete mechanism of action, the clear species differences preclude extrapolation of rodent data to provide an accurate human risk assessment. Recent data have demonstrated that PPs activate the PP-activated receptor alpha in rodent liver, leading to enzyme induction, stimulation of S-phase, and a suppression of apoptosis. How these acute effects may lead to hepatocarcinogenesis and the relevance of this for humans will be discussed.

Differential sensitivity of transcription factors to mustard-damaged DNA

Nitrogen mustard (bis(2-chloroethyl) methylamine, HN2) inhibited the binding of upstream factors Sp1 and AP2 to their consensus sequences. At concentrations where 50% of the consensus sequence DNA contained at least one lesion, HN2 inhibited formation of the Sp1 complex by 37% (40 microM HN2) and the AP2 complex by 40% (50 microM HN2). The binding of the TATA binding protein (TBP) to the TATA element was also inhibited by HN2, whereas sulphur mustard and the monofunctional sulphur mustard 2-chloroethyl ethyl sulphide (CEES) resulted in a disproportional extent of inhibition with respect to the level of alkylation. The level of alkylation of the TBP oligonucleotide varied significantly at 100 microM drug, with 80, 42 and 15% of HN2, sulphur mustard and CEES, respectively. However, this level of alkylation inhibited formation of the TBP-DNA complex by 70, 70 and 45%, respectively. This differential sensitivity of transcription factors to mustard-induced DNA damage therefore appears to reside dominantly in the stereochemical differences between the specific mustard lesions.

Cellular disposition of sulphamethoxazole and its metabolites: implications for hypersensitivity

1. Bioactivation of sulphamethoxazole (SMX) to chemically-reactive metabolites and subsequent protein conjugation is thought to be involved in SMX hypersensitivity. We have therefore examined the cellular metabolism, disposition and conjugation of SMX and its metabolites in vitro. 2. Flow cytometry revealed binding of N-hydroxy (SMX-NHOH) and nitroso (SMX-NO) metabolites of SMX, but not of SMX itself, to the surface of viable white blood cells. Cellular haptenation by SMX-NO was reduced by exogenous glutathione (GSH). 3. SMX-NHOH and SMX-NO were rapidly reduced back to the parent compound by cysteine (CYS), GSH, human peripheral blood cells and plasma, suggesting that this is an important and ubiquitous bioinactivation mechanism. 4. Fluorescence HPLC showed that SMX-NHOH and SMX-NO depleted CYS and GSH in buffer, and to a lesser extent, in cells and plasma. 5. Neutrophil apoptosis and inhibition of neutrophil function were induced at lower concentrations of SMX-NHOH and SMX-NO than those inducing loss of membrane viability, with SMX having no effect. Lymphocytes were significantly (P<0.05) more sensitive to the direct cytotoxic effects of SMX-NO than neutrophils. 6. Partitioning of SMX-NHOH into red blood cells was significantly (P<0.05) lower than with the hydroxylamine of dapsone. 7. Our results suggest that the balance between oxidation of SMX to its toxic metabolites and their reduction is an important protective cellular mechanism. If an imbalance exists, haptenation of the toxic metabolites to bodily proteins including the surface of viable cells can occur, and may result in drug hypersensitivity.

Protection against necrosis but not apoptosis by heat-stress proteins in vascular smooth muscle cells: evidence for distinct modes of cell death

We have reported previously that cultured vascular smooth muscle cells (VSMC) isolated from spontaneously hypertensive rats (SHR) show higher proliferation and cell death than normotensive controls. In addition to protecting cells against death, heat stress proteins (HSPs) appear to play a role in cell proliferation. This investigation examines the involvement of HSP72 and HSP27 in altered SHR VSMC proliferation and death. We have performed detailed discriminatory analysis to characterize which type of VSMC death is induced by heat stress (HS) and serum deprivation. Serum deprivation induced apoptosis (caspase-3 cleavage and DNA laddering) and secondary necrosis, the 2 processes being a continuum of each other. In contrast, acute HS (46 degrees C, 30 minutes), which inhibited BN. lx and SHR VSMC proliferation by 2-fold, increased necrosis (by 5-fold and 2-fold, respectively) but not apoptosis. HSP72 and HSP27 expression evoked in VSMC by mild HS (44 degrees C, 15 minutes) 6 hours before acute HS prevented the inhibition of proliferation and induction of necrosis with no effect on serum deprivation-induced or staurosporine-induced apoptosis. This induced expression of HSP72 and HSP27 did not eliminate the higher basal proliferation, apoptosis, and necrosis of SHR VSMC compared with BN.lx VSMC, suggesting that these HSPs are not involved in altered SHR VSMC proliferation and death. Also, although apoptosis and necrosis may be a continuum, in VSMC the 2 processes may be distinguished by HS, in which only necrosis is prevented by prior HSP accumulation. This observation may be of use in designing strategies for cellular protection.

Apoptosis and cell death in neuronal cells: where does Ca2+ fit in?

Mounting evidence shows that neuronal death is an important and essential component of brain tissue homeostasis, with major forms of cell death occurring: necrosis and apoptosis. No general consensus exists as to whether these two forms of neuronal death represent separate cellular processes or just two different forms of a common 'death pathway'. One difference between them is the role played by intracellular Ca2+: central and obligatory, in necrosis and possible, but not always necessary in triggering apoptosis. Furthermore, the same assessment of the involvement of Ca2+ signalling could also distinguish between two possible apoptotic states in the nervous system: one, the 'developmental apoptosis', involving immature and developing neurons, in which Ca2+ plays mainly an apoprotector role, and another one, associated mainly with pathological instances and involving fully matured and established neurons, in which Ca2+ plays an apo-inducing role.

Role of Hsp70 in regulation of stress-kinase JNK: implications in apoptosis and aging

Cell protection from stresses by the major heat shock protein Hsp72 was previously attributed to its ability to prevent aggregation and to accelerate refolding of damaged proteins. This repair function of Hsp72 may play an important role in cell survival after extremely harsh protein damaging treatments leading to necrotic cell death. On the other hand, protein repair function of Hsp72 cannot explain how it protects cells from stresses which do not cause direct protein damage, e.g. some genotoxic agents. These stresses kill cells through activation of apoptosis, and Hsp72 increases cell survival by interfering with the apoptotic program. Recently it has been found that Hsp72 mediates suppression of a stress-activated protein kinase, JNK, an early component of stress-induced apoptotic signalling pathway. This finding provides the basis for the anti-apoptotic activity of Hsp72. These observations can explain increased stress sensitivity of aged cells in which compromised inducibility of Hsp72 leads to a loss of control of JNK activation by stresses and subsequently to a higher rate of apoptotic death.