Homocysteine Thiolactone Research Articles

Differentiation of myoblast cell lines and biological methylation: 3-deazaadenosine stimulates formation of multinucleated myofibers.

Treatment of myoblast cell lines with 3-deazaadenosine stimulates differentiation into myofibers. Myoblast clone L5/ 3B5 , which does not form myofibers after 6 days in fusion medium, was stimulated to form myofibers after 5 days of culture in fusion medium containing 50 microM 3-deazaadenosine. Myoblast clone L5/ 3C4 , which normally begins to form myofibers after 4 days in fusion medium, was stimulated by 50 microM 3-deazaadenosine to form myofibers after 3 days in culture and the extent of fusion was also increased. In the presence of 100 microM homocysteine thiolactone, the concentration of 3-deazaadenosine that stimulated maximal fusion was reduced by a factor of 10, from 50 microM to 5 microM 3-deazaadenosine. Stimulation of myofiber formation by 3-deazaadenosine suggests a requirement for one or more methylation reactions in myoblast differentiation and the potentiation by homocysteine thiolactone indicates that myofiber formation is specifically stimulated by an intracellular accumulation of 3-deazaadenosylhomocysteine.

Altered cell cycle distributions of cultured human lymphoblasts during cytotoxicity related to adenosine deaminase inhibition

Serial-flow cytometric analysis of DNA content of T lymphoblasts (MOLT-4) and B lymphoblasts (MGL-8) was performed to correlate the cytotoxic properties of adenosine deaminase inhibition with alterations of DNA synthesis and disruptions of the cell cycle. The addition of deoxyadenosine up to 50 μmol/L potently decreased the growth of T lymphoblasts, and these changes were enhanced with the addition of 100 μmol/L homocysteine thiolactone. These conditions caused a virtual absence of cells from S and G 2M phases after 24 hours. The DNA distribution was similar in cells cultured for 24 hours in 50 μmol/L deoxyguanosine or 2.5 μmol/L hydroxyurea. These observations suggested accumulation of cells in the G 1 phase. T lymphoblasts cultured with up to 50 μmol/L adenosine had a substantial decrease in growth, which was not modified by the addition of homocysteine thiolactone. Cell cycle distributions of T lymphoblasts cultured for 24 to 48 hours under these conditions showed mild decreases in the G 2M population. The addition of adenosine up to 50 μmol/L decreased the growth of B lymphoblasts, and these changes were enhanced by the addition of 100 μmol/L homocysteine thiolactone. These conditions induced mild decreases in the S-phase population in B lymphoblasts. The addition of deoxyadenosine, even with homocysteine thiolactone, did not modify growth in B lymphoblasts and the cell-cycle distributions were indistinguishable from distributions of control populations after 24 and 48 hours. The observations provide independent support for a reduction of DNA synthesis associated with cytotoxicity during adenosine-deaminase inhibition. These changes were more profound in T lymphoblasts than B lymphoblasts. Cytotoxicity during S and G 2M phases and/or a G 1 S transition block could account for the cell-cycle alterations observed.

Selection and characterization of a murine lymphoid cell line partially deficient in S-adenosylhomocysteine hydrolase

The exact role of S-adenosylhomocysteine hydrolase (EC 3.3.1.1) in mediating the toxic effects of adenosine toward mammalian cells has not been ascertained. The selection and characterization of S-adenosylhomocysteine hydrolase-deficient cell lines offers a biochemical genetic approach to this problem. In the present experiments, a mutant clone (Sahn 12) with 11–13% of wild-type S-adenosylhomocysteine hydrolase activity was selected from the murine T lymphoma cell line R 1.1 after mutagenesis and culture in adenosine, deoxycoformycin, uridine and homocysteine thiolactone-supplemented medium. In the presence of 0.5 mM homocysteine thiolactone and 10–200 μM adenosine, wild-type and mutant cells synthesized S-adenosylhomocysteine intracellularly at markedly different rates, and excreted the compound extracellularly. Thus, at time points up to 10 h, the S-adenosylhomocysteine hydrolase-deficient lymphoblasts required 5–10-fold higher concentrations of adenosine in the medium to achieve the same intracellular S-adenosylhomocysteine levels as wild-type cells. Similarly, the Sahn 12 lymphoblasts were 5–10-fold more resistant than R 1.1 cells to the toxic effects of adenosine plus homocysteine thiolactone. These results establish that (i) 11–13% of wild-type S-adenosylhomocysteine hydrolase activity is compatible with normal growth, (ii) in medium supplemented with both adenosine and homocysteine thiolactone, intracellular S-adenosylhomocysteine is synthesized by S-adenosylhomocysteine hydrolase, (iii) the net intracellular level of S-adenosylhomocysteine is determined by both the rate of S-adenosylhomocysteine synthesis and its rate of excretion, (iv) under such conditions the accumulation of S-adenosylhomocysteine is related to cytotoxicity, (v) in the absence of an exogenous homocysteine source, S-adenosylhomocysteine derives from endogenous sources, and the accumulation of S-adenosylhomocysteine is not the primary cause of adenosine induced cytotoxicity.

Hippocampal electrical activity and gamma-aminobutyrate metabolism in brain tissue following administration of homocysteine.

The concentration of gamma-aminobutyrate (GABA) and the activity of glutamate decarboxylase and GABA-transaminase were measured in extracts of mouse brain before the onset and during the course of generalized seizures induced by systemic administration of homocysteine thiolactone. The results indicate that whole brain GABA metabolism is unaffected by subconvulsive and convulsive doses of homocysteine at all stages of the generalized seizure. Electroencephalographic monitoring of rat brain electrical activity via hippocampal electrode implantation allowed the course homocysteine-induced seizures to be followed and afforded a means of quantifying such seizures.

Phospholipid methylation increases during capacitation of golden hamster sperm in vitro.

The present report describes in vitro experiments with golden hamster sperm designed to determine whether there is any relationship between sperm phospholipid methylation and capacitation and/or the acrosome reaction. Washed cauda epididymal hamster sperm were incubated in a capacitation medium containing [methyl-3H] methionine. After 0.5, 1.5, 2.5 and 3.5 h of incubation, sperm were extracted with a chloroform:methanol:2 N HCl mixture to extract total phospholipids. Liquid scintillation counting revealed that the methyl-3H-group was incorporated into phospholipids with maximum incorporation at 3.5 h and an increase of 50% between 2.5 and 3.5 h. Uptake of labeled methionine by sperm reached its plateau by 1.5 h of incubation. Some sperm were capacitated by 3.5 h because that is the time at which the rate of acrosome reactions began to increase and because at least 50% of them were able to undergo the acrosome reaction 10 min after the addition of the fusogen lysophophatidylcholine (LPC) at 3.5 h but not at 2.5 h. Homocysteine thiolactone and 3-deazadenosine, inhibitors of transmethylation, inhibited incorporation of methyl-3H into phospholipids at 3.5 h by approximately 90% and also inhibited LPC-induced acrosome reactions by 60%. Separation of methylated sperm phospholipid by thin-layer chromatography demonstrated the presence of 3H-labeled phosphatidyl-N-monomethylethanolamine, phosphatidyl-N,N-dimethylethanolamine, and to a lesser extent phosphatidylcholine. In addition, an unidentified lipid was also highly labeled. These results strongly suggest a positive correlation between phospholipid methylation and capacitation and/or the acrosome reaction of hamster sperm in vitro. Possible mechanisms for phospholipid methylation involvement in these events are discussed.

Possible role of increased brain methylation in methionine sulfoximine epileptogenesis: effects of administration of adenosine and homocysteine thiolactone.

An intraventricular pulse of [14COOH]L-methionine to mice pretreated with the convulsant L-methionine-dl-sulfoximine (MSO) resulted in significantly higher than control specific radioactivity values of cerebral [14COOH]L-methionine (Met), [14COOH]S-adenosyl-L-methionine (AdoMet) and [14COOH]S-adenosyl-L-homocysteine (AdoHcy). MSO administration (3 hr) also decreased brain steady-state levels of Met, AdoMet, and AdoHcy. Following an intraventricular pulse of [3H-methyl]L-methionine, the levels of [3H-methyl]phosphatidylmonomethylethanolamine and of membrane associated and soluble [3H-methyl]carboxylmethylated proteins were increased over corresponding saline-treated controls. The activity of cerebral histamine N-methyltransferase was also increased after MSO treatment. The administration of a combination of adenosine and homocysteine thiolactone to MSO-pretreated animals counteracted the MSO-induced decreases in brain Met, AdoMet, and AdoHcy as well as the increase in histamine N-methyltransferase activity. In addition, administration of adenosine together with homocysteine thiolactone decreased the incidence of, and increased the latency to MSO seizures, with the most effective anticonvulsant action occurring when cerebral AdoHcy levels were at their highest.

Mechanism of the cytostatic activity of 3-deazaaristeromycin, an inhibitor of adenosylhomocysteine hydrolase.

3-Deazaaristeromycin (100 microM) is cytostatic for the RAW264 mouse macrophage cell line. In cells incubated with 3-deazaaristeromycin, Cell replication is effectively stimulated by 0.5 microM homocysteine thiolactone and no further stimulation occurs at concentrations of homocysteine thiolactone greater than 50 microM. Cell replication occurs when homocysteine thiolactone is administered either simultaneously with, or 4 days after, the administration of 3-deazaaristeromycin. In the presence of 3-deazaaristeromycin, cell replication also occurs when either folate (10 to 100 microM) or hypoxanthine and thymidine are included in the medium. Incorporation of [14C]formate into trichloroacetic acid-precipitable material is inhibited when cells are incubated with 3-deazaaristeromycin, and this inhibition is partially reversed by inclusion of either homocysteine thiolactone or folate in the medium. Experiments showed that adenosylhomocysteine hydrolase is inhibited to the same extent after the cells have been incubated for 4 days with either 3-deazaaristeromycin or 3-deazaaristeromycin and homocysteine thiolactone. The findings suggest that in cells incubated with 3-deazaaristeromycin, the formation of homocysteine from S-adenosylhomocysteine is insufficient for the regeneration of tetrahydrofolate for the synthesis of purines and pyrimidines.

The preparation of [ 35S]homocysteine thiolactone free of [ 35S]methionine

The unambiguous study of homocysteine metabolism requires a source of [355]homocysteine completely free from contaminating [35S]methionine. Until now, such a reagent has not been readily available. Homocysteine is an important metabolite involved in methionine, cysteine and methylation metabolism. Homocysteine can be methylated to form methionine with either 5-methyltetrahydrofolate [1] or betaine [2] serving as the methyl donor, it can be condensed with serine to form cystathionine [3] which can be converted to cysteine, or it can be condensed with adenosine [4] to form S-adenosylhomocysteine, a strong inhibitor of all known cellular methylation reactions [5,6]. Additionally, there is evidence that in methionine-dependent tumor lines, methionine synthesized endogenously from homocysteine is used differently by the cell than exogenously supplied methionine [7]. In this report we describe a novel, simple and fast method of recovering methionine-free [35S]homocysteine thiolactone following its synthesis.

Absence of a direct role of phospholipid methylation in stimulus-secretion coupling and control of adenylate cyclase in guinea-pig and rat parotid gland.

The present study was undertaken to investigate a possible involvement of phospholipid methyltransferases in the coupling of receptor-mediated stimulation to secretion. Phospholipid methyltransferases were assayed in isolated parotid acini in the presence of carbamoylcholine or isoprenaline. Carbamoylcholine reduced the incorporation of methyl groups into phospholipids, whereas isoprenaline showed no effect. Amylase secretion stimulated either by carbamoylcholine or by isoprenaline could not be affected by inhibitors of methyltransferases (3-deaza-adenosine alone or plus homocysteine thiolactone) under conditions where phospholipid methylation was strongly inhibited. The activity of adenylate cyclase in isolated parotid microsomal membranes was not inhibited or stimulated by S-adenosyl-homocysteine or -methionine respectively. These results indicate that phospholipid methylation does not play an essential role in stimulus-secretion coupling in the parotid gland.

Role of transmethylation in the elicitation of an oxidative burst in macrophages

Elicited guinea pig peritoneal macrophages (MPs) respond by an oxidative burst (OB) to a variety of membrane stimulants. Evidence has recently accumulated, indicating that phospholipase A 2 activation resulting in unsaturated fatty acid liberation is a prerequisite for the induction of an OB by some stimulants. We examined the effect of inhibiting adenosylmethionine-dependent phospholipid methylation on the capacity of MPs to produce superoxide (O 2 −) in response to membrane stimulation. We found that preincubation of MPs with the transmethylation inhibitor, 3-deazaadenosine (DZAdo), totally eliminated the induction of an OB by concanavalin A, wheat germ agglutinin, and N-formyl- l-methionyl- l-leucyl- l-phenylalanine and partially blocked O 2 − production in response to NaF, phospholipase C, digitonin, the ionophore A23187, and phorbol myristate acetate (PMA). The PMA-elicited OB was the most resistant to inhibition by DZAdo. Homocysteine thiolactone enhanced the blocking effect of DZAdo. These findings suggest that stimulated O 2 − production by guinea pig peritoneal MPs requires enzymatic methylation of an unknown substrate, most likely a membrane phospholipid.

Possible metabolic basis for the different immunodeficient states associated with genetic deficiencies of adenosine deaminase and purine nucleoside phosphorylase.

An inherited deficiency of adenosine deaminase (Ado deaminase; adenosine aminohydrolase, EC 3.5.4.4) causes severe combined immunodeficiency disease in humans. A similar deficiency in purine nucleoside phosphorylase (Puo phosphorylase; purine-nucleoside:orthophosphate ribosyltransferase, EC 2.4.2.1) engenders a selective cellular immune deficit. To elucidate the possible metabolic basis for the contrasting immunologic phenotypes, we compared the toxicity toward mature resting human lymphocytes of the Ado deaminase substrates deoxyadenosine and adenosine and the Puo phosphorylase substrate deoxyguanosine. When Ado deaminase was inhibited, micromolar concentrations of deoxyadenosine progressively killed nondividing helper and suppressor-cytotoxic T cells, but not B cells. The toxicity required phosphorylation, with subsequent dATP formation. The deoxyadenosine analogs 2-chlorodeoxyadenosine, 2-fluorodeoxyadenosine, and adenine arabinonucleoside also killed resting T cells. Cell death was unrelated to inhibition of adenosylhomocysteinase (EC 3.3.1.1) but was preceded by a gradual decline in ATP levels. As much as 1 mM deoxyguanosine did not impair resting lymphocyte viability, despite the synthesis of dGTP. The combination of 200 microM adenosine plus 500 microM homocysteine thiolactone killed dividing lymphocytes but had no discernible toxic effect toward resting T cells, which accumulated adenosylhomocysteine over a 4-hr period but thereafter excreted the nucleoside into the culture medium. The different clinical syndromes associated with genetic deficiencies of Ado deaminase and Puo phosphorylase may be explained by the ability of dATP to kill mature resting T lymphocytes by depleting ATP levels.

Effect of ADA deficiency on cultured murine thymus transplantation

Nude mice were transplanted with thymus tissue cultured in media supported by adenosine deaminase (ADA)-deficient horse serum and containing various combinations of the ADA inhibitor 2-deoxycoformycin (2-dCF), deoxyadenosine, homocysteine thiolactone, and uridine. Reconstitution of the nude mice was not accomplished or only marginal when thymus was cultured in 2-dCF and deoxyadenosine or 2-dCF plus all of the above metabolites. 2-dCF alone or d-adenosine alone did not sufficiently affect the cultured tissue to impair restorative capability. Toxic metabolites known to accumulate in ADA deficiency gradually obliterate lymphocytes leading to combined immunodeficiency. These studies suggest that thymic epithelium may also suffer, making regeneration of T lymphocytes a difficult if not impossible task, should the metabolite accumulation be sufficiently severe.

Homocysteine thiolactone and experimental homocysteinemia

The usual cause of premature death in patients with homocystinuria due to cystathionine β-synthase deficiency (EC 4.2.1.22) is vascular disease which occurs with high plasma-homocysteine concentrations. In animal models, experimental homocysteinemia produced by chronic homocysteine thiolactone infusions has been associated with continuing endothelial damage which could be the initiating factor in the pathogenesis of vascular disease in homocystinuric patients. It is shown in the present experiments first that hydroiysis of 1 mol of homocysteine thiolactone at pH 7.4 in dilute phosphate buffer produces 1 mol of homocysteine and 1 equivalent of H + at 37°C ( t 1 2 > 30 hr ). Pig liver carboxylesterase strongly catalyzes this hydrolysis: for l-homo-cysteine thiolactone, V = 26.9 units A 280 , K m = 14.0 mm, and stereo-specificity favors the l-enantiomer over the d-enantiomer, but is weak. Thus in animals given homocysteine thiolactone infusions, most hydrolysis will occur in esterase-rich tissues, and the acid which this reaction produces may damage those tissues. Secondly, at neutral pH, the thiolactone covalently modifies proteins which in animal models may produce changes unrelated to homocysteinemia. Thirdly, concentrated solutions of the thiolactone at neutral pH readily form the diketopiperazine which may precipitate. We conclude the changes in vascular morphology reported after chronic homocysteine thiolactone infusions may not result solely from high plasma-homocysteine concentrations, and that this diminishes the relevance of these experiments to the pathogenesis of vascular disease in clinical homocystinuria. The infusion of homocysteine itself is preferable in animal model experiments.

Differential cyst(e)ine requirements in human T and B lymphoblastoid cell lines.

In an effort to find exploitable metabolic differences between human T and B lymphoblasts, we have compared the ability of lymphocytes of varying phenotype to grow in cystine-deficient medium. Only 6 of 12 human lymphoblastoid cell lines tested were able to utilize homocysteine thiolactone or cystathionine in place of cystine for growth. This difference in growth requirements was unrelated to the rate of cell division, the presence of Epstein-Barr viral genetic material, or whether or not the cell lines derived from benign or malignant tissues. Rather, all B lymphoblastoid cell lines grew in homocysteine thiolactone- or cystathionine-containing medium, while the T and non-T, non-B lymphoblastoid cell lines did not. Normal human peripheral blood T and B lymphoblasts did not respond to mitogens in the homocysteine thiolactone or cystathionine medium, but developed the ability to utilize these cysteine precursors after stimulation with concanavalin A, protein A, or Epstein-Barr virus. The differences in cysteine requirements among T and B cell lines may reflect a fundamental difference in de novo protein-synthesizing capacity of the two cell types.

IgE-mediated histamine release in rat basophilic leukemia cells: Receptor activation, phospholipid methylation, Ca 2+ flux, and release of arachidonic acid

Stimulation of IgE receptors on rat basophilic leukemia cells causes a transient rise and fall of methylated phopholipids, Ca 2+ influx, and release of arachidonic acid previously incorporated into phosphatidylcholine and liberation of histamine. Inhibition of phospholipid methylation by methyltransferase inhibitors, 3-deazaadenosine and homocysteine thiolactone, almost completely blocks the influx of Ca 2+, and release of arachidonic acid and histamine. Stimulation of immunoglobulin E receptors by antigen releases only [ 14C]arachidonic acid but not [ 14C]linoleic acid, [ 14C]oleic acid and [ 14C]stearic acid, all of which were previously incorporated into phospholipids. [ 14C]Arachidonate was found to be incorporated mainly into phosphatidylcholine. The phosphatidycholine rich in arachidonate appeared to be synthesized to a considerable extent by the transmethylation pathway. These findings suggest that in rat basophilic leukemia cells, immunoglobulin E receptors, phospholipid methyltransferases, Ca 2+ ion channel, and phospholipase(s) that cause release of arachidonic acid and the discharge of histamine are associated.

Stimulation of tubulin tyrosinolation in rabbit leukocytes evoked by the chemoattractant formyl-methionyl-leucyl-phenylalanine.

Cellular tubulin is subject to a posttranslational modification involving the reversible addition to tyrosine through peptide linkage to the C-terminal glutamate of the alpha-chain. The synthetic peptide chemoattractant, N-formyl-methionyl-leucyl-phenylalanine, causes a specific, dose-dependent stimulation of tubulin tyrosinolation in rabbit leukocytes. This stimulation is prevented by carbobenzoxy-phenylalanyl-methionine, benzoyl-tyrosine ethylester, and nordihydroguaiaretic acid, which are all inhibitors of chemotaxis presumed to act via membrane-associated events. The combination of 3-deazaadenosine and homocysteine thiolactone, which inhibits phospholipid methylation, and quinacrine, an inhibitor of phospholipase A2, also abolishes the response to the peptide. Colchicine, however, which causes a marked disassembly of cellular microtubules in these cells and also inhibits chemotaxis, does not have any inhibitory effect on the basal or peptide-stimulated rate of tubulin tyrosinolation. In contrast, taxol, a microtubule-stabilizing agent, has an inhibitory effect on both the basal and peptide-stimulated tyrosine incorporation. Taxol also inhibits chemotaxis in rabbit leukocytes. The results strongly suggest the role of closely linked membrane-cytoskeleton interactions in leukocyte chemotaxis, in which tyrosinolation of tubulin may be functionally involved.

Inhibition of the hamster sperm acrosome reaction by transmethylation inhibitors.

Two inhibitors of somatic cell transmethylation, homocysteine thiolactone (Hcy) and 3--deazadenosine (3--DZA), inhibited the acrosome reactions of hamster sperm in vitro. Hcy (250 microM) or 3--DZA (200 microM) inhibited acrosome reactions by approximately 30% after 4 hours of incubation with washed sperm in a capacitating medium containing epinephrine, taurine, and bovine serum albumin. Inhibition by 3--DZA decreased by 5 hours, but inhibition by Hcy did not. Incubation of sperm with the combination of both inhibitors inhibited acrosome reactions by nearly 60% at 4 hours, and did not appreciably decrease with time up to 5 hours. These results are the first to demonstrate that such inhibitors can interfere with the acrosome reaction and taken together with the effects of Hcy and 3--DZA on somatic cells, suggest that sperm protein and/or phospholipid transmethylation is/are involved in capacitation and/or the acrosome reaction.

Inhibition of phospholipid methyltransferase during zymosan induced secretion of platelet-activating factor in human polymorphonuclear leukocytes

Phagocytosis of zymosan particles coated with complement induces a time and dose dependent inhibition of the enzyme phospholipid methyltransferase in human polymorphonuclear cells. The extent of phospholipid methyltransferase inhibition induced by various concentrations of zymosan strongly correlates with the secretory process: liberation of platelet-activating factor (PAF) and β-glucuronidase. Zymosan also decreases the incorporation of 3H-methyl group into phospholipids in cells pre-labeled with ( 3H-methyl)-methionine. Finally, preincubation of cells with 3-deaza-adenosine and homocysteine thiolactone, inhibitors of phospholipid methyltransferase, decrease the incorporation of 3H-methyl group into phospholipids in cells pre-labeled with ( 3H-methyl)-methionine and modulate the release of PAF. These results suggest that phospholipid methylation plays an important role during the transduction of the secretory signal triggered by zymosan in human polymorphonuclear cells.

Homocysteine-induced seizures in the mouse cerebral cortex after propranolol pretreatment. Lack of correlation between cyclic adenosine monophosphate levels and phosphorylase activation

Cyclic AMP levels and phosphorylase a activity were investigated in the cerebral cortex of mice during seizures induced by homocysteine thiolactone, following pretreatment with various doses of propranolol and, also, after administration of subthreshold doses of homocysteine. The accumulation of cyclic AMP accompanying homocysteine-induced seizures was not influenced by theophylline, so that adenosine is not likely to be the mediator involved. Since it has been suggested that there exists a receptor for certain acidic amino acids which participates in cyclic AMP accumulation in brain cortex slices in vitro, it seemed of interest to ascertain whether homocysteine, or possibly its metabolic product, could not have a direct effect on the cyclic AMP-generating system. However, subthreshold doses of homocysteine thiolactone (which did not evoke seizures) did not influence the levels of cyclic AMP in the cerebral cortex for a period of up to 1 h. The accumulation of cyclic AMP was markedly reduced or completely prevented by pretreatment with DL-propranolol ; phentolamine exerted no effect. This suggests that, during homocysteine-induced seizures, catecholamines might interact with the cyclic AMP-generating system, possibly through β-adrenergic receptors. The increased activity of phosphorylase accompanied not only homocysteine-induced seizures, but was also present after pretreatment with subthreshold doses of homocysteine or propranolol. The pretreatment with subthreshold doses of homocysteine or 20 mg/kg propranolol enhances phosphorylase activity while cyclic AMP levels remain unchanged. This suggests that in the brain, as in some other tissues, phosphorylase activation may be independent of cyclic AMP in some situations.

Phospholipid methylation and phospholipase A2 activation in cytotoxicity by human natural killer cells.

The role of phospholipid methylation and phospholipase A2 (phosphatide 2-acylhydrolase, EC 3.1.1.4) in natural killer (NK) function by human peripheral blood mononuclear cells was studied. Pretreatment of effector cells with a methyltransferase inhibitor, 3-deazaadenosine, in the presence of homocysteine thiolactone, reduced cytotoxicity in a dose-dependent fashion. This effect was closely associated with inhibition of methylation of lipids but not of nucleic acids or proteins. The suggestion for a role of phospholipid methylation was supported by the observation that the interaction between NK-susceptible tumor targets and peripheral blood mononuclear cells caused increased phospholipid methylation only when susceptible target cells were used. Phospholipase A2 was also implicated in human NK activity. Inhibitors of the enzyme such as tetracaine, mepacrine, Rosenthal's inhibitor, and corticosteroids impaired NK function. Rosenthal's inhibitor was also shown to exert an inhibitory effect on a purified NK-cell population obtained by the isolation of large granular lymphocytes on Percoll gradients. Peripheral blood mononuclear cells were also directly shown to display phospholipase A2-like activity, as measured by the decrease in radioactive arachidonate from prelabeled phospholipids, specifically phosphatidylcholine, in effector cells. These data suggest that enhanced phospholipid methylation occurs during the recognition function of NK cells. Consequent activation of phospholipase A2 might be involved in the mechanisms leading to lytic events within the target cell.