Phospholipid methylation by intact rat Leydig cells

Phospholipid methylation by intact Leydig cells was investigated by determining the incorporation of radioactivity from [3H-methyl] methionine into phospholipids. Leydig cells incorporated significantly more radioactivity into phospholipids than did unpurified testicular cells, non-Leydig testicular cells, or red blood cells. Approximately 40% of the radioactivity was found in phosphatidylcholine, indicating that the methyltransferase pathway for the synthesis of this phospholipid is highly active in rat Leydig cells. Addition of luteinizing hormone to cells preloaded with [3H-methyl] methionine did not alter the rate of phospholipid methylation. However, phospholipid methylation by Leydig cells desensitized by the injection of human chorionic gonadotropin 1 to 7 days previously was reduced by approximately 60%. Inhibition of phospholipid methylation to 75% of normal with homocysteine thiolactone did not affect luteinizing hormone-stimulated androgen production. Further inhibition of phospholipid (and protein) methylation by treatment with homocysteine thiolactone and 3-deazaadenosine significantly reduced luteinizing hormone-stimulated androgen production. The results of this study demonstrate that the methyltransferase pathway for the synthesis of phosphatidylcholine is highly active in intact Leydig cells but is reduced in desensitized Leydig cells. There does not appear to be a close association between the activity of this pathway and the ability of luteinizing hormone to acutely stimulate androgen production.     

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.     

Phorbol 12-myristate 13-acetate, A23187 and L-adrenaline inhibit phospholipid methylation in human monocytes and lymphocytes. Inhibition is independent of oxyradical production and phospholipid hydrolysis

The Ca++ ionophore A23187 and phorbol 12-myristate 13-acetate (PMA) caused dose-dependent inhibition of phospholipid (PL) methylation in unfractionated mononuclear cells (MNC), monocytes, and lymphocytes as measured by incorporation of 3H-methyl-groups from [3H-methyl]-L-methionine into phosphatidylcholine (PC), dimethyl phosphatidylethanolamine (PE), and monomethyl PE. This inhibitory effect did not correlate with monocyte superoxide release and was unaltered by the presence of either catalase and superoxide dismutase or the NADPH oxidase inhibitor, diphenylene iodonium (DPI), indicating that oxyradical-mediated oxidation of methionine was not the major cause of inhibition of PL methylation. Furthermore L-adrenaline, which elevates cAMP and does not stimulate superoxide release, also inhibited PL methylation. Inhibition by PMA was not due to reduction in intracellular levels of methionine or S-adenosyl methionine. A23187 caused reduction of S-adenosyl methionine levels only at 1 microM, and had no effect at lower concentrations. Inhibition of PL methylation was shown not to be due to phospholipase A2-dependent hydrolysis of newly methylated PL. Attempts to reverse the inhibitory effect of either A23187 or PMA with the putative protein kinase inhibitors W-7 and H-7 were inconclusive. The mechanism of inhibition of PL methylation by A23187 and PMA remains unclear, but does not appear to be due to oxidation of methionine or hydrolysis of newly methylated PL.     

Phospholipid methylation and taurine content of synaptosomes from cerebral cortex of developing rat.

Changes in taurine concentration and rate of methylation of phosphatidylethanolamine have been examined in rat brain synaptosomes over the course of development. At 7, 14, 21, 28 and 56 days of age, rats were injected i.p. with 300 microCi/kg [3H-methyl]methionine. Synaptosomes (P2B fraction) were isolated from the cerebral cortex 9 h later and incorporation of the methionine methyl group into phospholipid and protein was investigated. Synaptosomal taurine and methionine concentrations were determined at the same ages, as were the concentrations of the major classes of phospholipids (phosphatidylethanolamine, phosphatidylinositol, phosphatidylcholine and phosphatidylserine). Methionine concentration increased between day 7 and 14 and fell thereafter. Phospholipid methylation rates calculated from the specific activity of synaptosomal methionine were high from days 7 and 14 and then fell, whereas protein methionylation increased between day 7 and 28 and then decreased. A strong correlation was found between the taurine concentration of the synaptosome and phospholipid methylation rates during brain development. Protein methionylation rates, however, showed no correlation with taurine concentration.     

Phospholipid metabolism, calcium flux, and the receptor-mediated induction of chemotaxis in rabbit neutrophils

Rabbit neutrophils were stimulated with the chemotactic peptide fMet-Leu-Phe in the presence of the methyltransferase inhibitors homocysteine (HCYS) and 3-deazaadenosine (3-DZA). HCYS and 3-DZA inhibited chemotaxis, phospholipid methylation, and protein carboxymethylation in a dose-dependent manner. The chemotactic peptide-stimulated release of [14C]arachidonic acid previously incorporated into phospholipid was also partially blocked by the methyltransferase inhibitors. Stimulation by fMet-Leu-Phe or the calcium ionophore A23187 caused release of arachidonic acid but not of previously incorporated [14C]-labeled linoleic, oleic, or stearic acids. Unlike the arachidonic acid release caused by fMet-Leu-Phe, release stimulated by the ionophore could not be inhibited by HCYS and 3-DZA, suggesting that the release was caused by a different mechanism or by stimulating a step after methylation in the pathway from receptor activation to arachidonic acid release. Extracellular calcium was required for arachidonic acid release, and methyltransferase inhibitors were found to partially inhibit chemotactic peptide-stimulated calcium influx. These results suggest that methylation pathways may be associated with the chemotactic peptide receptor stimulation of calcium influx and activation of a phospholipase A2 specific for cleaving arachidonic acid from phospholipids.     

Fluorometric determination of phosphatidylcholine as a measure of phospholipid methylation.

The successive methylation of phosphatidylethanolamine to phosphatidylcholine (phospholipid methylation) has been measured by the incorporation of S-[methyl-3H]adenosylmethionine or colorimetric assay of phosphatidylcholine extracted from adipocyte plasma membranes. A fluorometric assay for phosphatidylcholine was developed to measure phospholipid methylation. This assay is 10 times more sensitive than the colorimetric assay and demonstrates no significant interference with other methylated phospholipids. The fluorometric assay was used to determine a biphasic insulin dose response in adipocyte plasma membranes. This fluorometric assay for phosphatidylcholine represents an alternative method for monitoring phospholipid methylation, especially when increased sensitivity is required.     

Plasma from uninephrectomized rats stimulates phospholipid methylation and arachidonic acid release in renal-cortical slices.

Phospholipid methylation and arachidonic acid release in renal-cortical slices was investigated in vitro after addition of plasma from uninephrectomized or sham-operated rats. Plasma from uninephrectomized rats ('uni-plasma') stimulated phospholipid methylation when obtained within the first 3 h after uninephrectomy. With different amounts of added plasma a graded response in phospholipid methylation was obtained. Addition of 50 nM-12-O-tetradecanoylphorbol 13-acetate for 10 min to intact slices also stimulated phospholipid methylation, whereas incubation of slices before addition of 'uni-plasma' with 100 microM-1-(5-isoquinolinylsulphonyl)-2-methylpiperazine prevented it, suggesting that protein kinase C stimulates phospholipid methylation in renal-cortical slices. Plasma from uninephrectomized rats also stimulates [3H]arachidonic acid release from phosphatidylcholine (PtdCho) and phosphatidylethanolamine (PtdEtn) via activation of phospholipase A2. Two mechanisms of phospholipase A2 activation are proposed: first, in which it is activated by protein kinase C and releases 3H radioactivity from PtdCho, and second, in which phospholipase A2 is stimulated by Ca2+ ions and releases 3H radioactivity from PtdEtn.     

Phospholipid methylation in pancreatic islets

Glucose provokes a transient stimulation of phospholipid methylation in rat pancreatic islets, possibly by increasing phospholipid methyltransferase activity. The association of DL-homocysteine and 3-deazaadenosine inhibits phospholipid methylation. The methylation of phospholipids may play a role in the stimulus-secretion coupling for glucose-induced insulin release.     

UPTAKE OF [3H-METHYL]CHOLINE BY MICROSOMAL, SYNAPTOSOMAL, MITOCHONDRIAL AND SYNAPTIC VESICLE FRACTIONS OF RAT BRAIN

Abstract— Microsomal, mitochondrial, synaptosomal and synaptic vesicle fractions of rat brain took up [³H-methyl]choline by a similar carrier-mediated transport system. The apparent K_m for the uptake of [³H-methyl]choline in these subcellular fractions was about 5 × 10^?5 M. Choline uptake was also observed in microsomal fractions prepared from liver and skeletal muscle. Virtually identical kinetic properties for [³H-methyl]choline transport were found in the synaptosomal fractions prepared from the whole brain, cerebellum or basal ganglia. Countertransport of [³H-methyl]choline from the synaptosomal fraction was demonstrated against a concentration gradient. HC-3 was a competitive inhibitor of the uptake of [³H-methyl]choline in brain microsomal, synaptosomal and mitochondria] fractions with respective values for K_i of 4.0, 2.1 and 2.3 × 10^?5 M. HC-15 was a competitive inhibitor of the transport of [³H-methyl]choline in the synaptosomal fraction, with a K_i of 1.7 × 10^?4 M. Upon entry into the microsomal fraction, 74 per cent of the radioactivity could be recovered as unaltered choline, 10 per cent as phosphorylcholine, 1.5 per cent as acetylcholine and 2.5 per cent as phospholipid. Choline acetyltransferase (EC 2.3.1.6) was assayed with [¹⁴C]acetylCoA in synaptosomal fractions prepared from basal ganglia and cerebellum, and in the 31,000 g supernatant fraction of a rat brain homogenate. Enzyme activity was 11-fold greater in the synaptosomal fraction from the basal ganglia than in that from the cerebellum. HC-3 did not inhibit choline acetyltransferase and there was no evidence for acetylation of HC-3. Our findings suggest that choline uptake is a ubiquitous property of membranes in the CNS and cannot serve to distinguish cholinergic nerve endings and their synaptic vesicles.     

Effects of insulin secretagogues and inhibitors on phospholipid metabolism in Langerhans' islets of rat pancreas

Glucose stimulation of [32P]-prelabelled pancreatic islets induced an increased incorporation of radioactivity into phosphatidylinositol. Glucagon and agents that increases cAMP in the islet did not affect phosphatidylinositol turnover, in spite of increased release of insulin. Furthermore, the potent inhibitor of insulin release, somatostatin did not alter phospholipid metabolism. Colchicine inhibited glucose-stimulated turnover of phosphatidylinositol as well as insulin release. These results may suggest that: (1) cAMP and phosphatidylinositol turnover are involved in different transmembrane control system for regulating insulin release; and (2) the function of microtubules modulate phospholipid metabolism.     

IgE-dependent activation of human lung mast cells is not associated with increased phospholipid methylation

Enhanced phospholipid methylation has been suggested to be an obligatory process in IgE-dependent stimulus-secretion coupling in human lung mast cells. Our studies with mast cell-enriched lung preparations do not support this hypothesis, demonstrating no increased 3H-methyl radiolabeling of chloroform/methanol-extracted lipids or chromatographically separated phospholipids accompanying anti-IgE-dependent histamine secretion. Inhibitors of transmethylation, 3-deazaadenosine, and homocysteine thiolactone inhibited histamine secretion by both anti-IgE and calcium ionophore A23187, reflecting a requirement of secretion for overall integrity of cellular transmethylation. These agents induced small increases in cAMP concentration which are considered to make at most a minor contribution to this inhibition. The inability of methylation inhibitors to diminish anti-IgE-dependent increases in lung mast cell cAMP levels would suggest that not only does phospholipid methylation have no role in histamine secretion but also it does not participate in the activation of adenylate cyclase by this stimulus.     

Studies on the mechanism by which melittin stimulates insulin secretion from isolated rat islets of Langerhans

Incubation of isolated rat islets of Langerhans with melittin resulted in a dose-dependent stimulation of insulin secretion with half the maximal response occurring at 4 micrograms/ml melittin. The effect of melittin on insulin secretion was dependent on extracellular calcium, was inhibited by the phospholipase A2 inhibitor quinacrine and by the lipoxygenase inhibitor nordihydroguaiaretic acid. Stimulation of insulin secretion by melittin was associated with a calcium-dependent loss of [3H]arachidonic acid from phospholipids in islet cells prelabelled with [3H]arachidonic acid. Analysis of the islet phospholipids involved in this response revealed that the [3H]arachidonic acid was released predominantly from phosphatidylcholine. These results suggest that melittin may stimulate insulin secretion by activating phospholipase A2 in islet cells, causing the release of arachidonic acid from membrane phospholipid. The results are consistent with suggestions that the subsequent metabolism of arachidonic acid via the lipoxygenase pathway may be involved in regulating the insulin secretory response.     

Calmodulin modulates phospholipid methylation in Dictyostelium discoideum

In the presence of (³H-methyl)-S-adenosyl-L-methionine $\text{Di}\text{ctyostelium discoideum}$ cell homogenates incorporate (³H-methyl) groups into mono- and dimethyl-phosphatidylethanolamine and phosphatidylcholine. The addition of bovine brain calmodulin enhances lipid transmethylation and an antiserum against rat brain calmodulin inhibits the reaction. EGTA and chlorpromazine, an inhibitor of calmodulin action, inhibit phospholipid methylation. Based on these results we propose that this reaction is modulated by calmodulin.     

Effect of S-adenosylhomocysteine on insulin-independent release of pyruvate dehydrogenase activator from rat adipocyte plasma membranes

The addition of insulin to adipocyte plasma membranes has been shown to release a low molecular weight, acid stable mediator which activates mitochondrial pyruvate dehydrogenase.The insulin-dependent release of this activator is dependent on the method used to prepare the plasma membranes. Adipocyte plasma membranes prepared in 0.25 M sucrose, 10 mM MOPS, pH 7.4 released an activator of pyruvate dehydrogenase in an insulin-independent manner. Insulin is required to stimulate phospholipid methylation in these membranes. The inhibition of insulin-stimulated phospholipid methylation with 1 mM S-adenosylhomocysteine resulted in a significant increase in amount and/or activity of the pyruvate dehydrogenase activator. The insulin-dependent dependent release of mediators of insulin action from adipocyte plasma may be regulated by phospholipid methylation.     

Inhibitory effects of GMCHA-OPhBut on phospholipid methylation and histamine release in mast cells activated by concanavalin A, anti-IgE, and antigen

[3H]Methyl group incorporation and histamine secretion in rat mast cells induced by anti-IgE and con A were strongly inhibited by trans-4-guanidinomethylcyclohexanecarboxylic acid 4-tert-butylphenyl ester (GMCHA-OPhBut), a strong and specific inhibitor for pH 7 tryptase (Muramatsu et al. (1988) Biol. Chem. Hoppe-Seyler 369, 617-625) which is present in rat mast cells. The IC50s for these events were of the order of 10(-6) M. Addition of GMCHA-OPhBut after the maximal increase in [3H]methyl group incorporation in rat mast cells activated by con A and anti-IgE induced rapid reduction of the methylated phospholipid, and the later histamine release was strongly suppressed. Mast cells were prepared with Mg2+-free Tyrode-HEPES solution, and challenged with anti-IgE with or without Mg2+. With Mg2+, [3H]methyl group incorporation was enhanced, and histamine was secreted time-dependently. Without Mg2+, [3H]methyl group incorporation fell to one-third, whereas histamine secretion was not affected. These results were incompatible with the above results. From these results it was strongly suggested that a trypsin-like protease, probably pH 7 tryptase, is involved not only in the early events, such as activation of phosphatidylethanolamine methyltransferase I and/or II, but also in the late events such as histamine release, and phospholipid methylation is not associated with histamine secretion.     

Structural requirements of the phospholipid substrate for phospholipid N-methylation in rat liver

The role of endogenous phospholipid substrates for phospholipid methylation was investigated in rat liver microsomes. The amount of phosphatidylethanolamine could be drastically reduced by treatment of microsomes with an amino group-blocking compound, methylacetimidate. Simultaneously, the formation of labelled phospholipids from S-adenosyl[Me-3H]methionine decreased, indicating that the amount of endogenous substrate influenced the reaction rate. Phosphatidylmonomethylethanolamine, phosphatidyldimethylethanolamine and phosphatidylmonoethylethanolamine added as dispersions to untreated or treated microsomes stimulated phospholipid methylation, whereas several other phospholipids were inactive. In other experiments the role of phospholipid substrates in intact cells was studied. Cultured rat hepatocytes were enriched in different phospholipids by preincubation with different amino alcohols, and the effects of phospholipid methylation was measured by incubation with [Me-14C]methionine. Phospholipid methylation was significantly stimulated after preincubation with ethanolamine, monomethylethanolamine, monoethylethanolamine and 2-aminobutanol. The results show that both the number and chain length of N-alkyl substituents on phosphatidylethanolamine, as well as other changes in the ethanolamine moiety, will affect the ability of different phospholipids to act as methyl acceptors.     

Enhanced carboxyl methylation of membrane-associated hemoglobin in human erythrocytes

The alpha- and beta-chains of hemoglobin (Hb) are methylated in intact erythrocytes and in cellular extracts by a protein D-aspartate methyltransferase (EC 2.1.1.77) specific for D-aspartyl and L-isoaspartyl residues. During an 18-h incubation of intact erythrocytes with L-[methyl-3H]methionine, the subfraction of Hb molecules associated with the membrane becomes progressively enriched with methyl esters, reaching a specific activity 10-fold that of cytosolic Hb. The enhanced methylation of membrane Hb in intact cells appears not to result from its methylation at sites with inherently greater stability, since salt-extracted membrane Hb 3H-methyl esters and cytosolic Hb 3H-methyl esters are hydrolyzed at similar rates at pH 8.4 in vitro. Oxidative treatment of column-purified Hb with acetylphenylhydrazine produces an immediate 4-fold increase in its specific methyl-accepting activity coincident with the production of hemichrome forms known to possess a higher affinity for membrane binding sites. Together, the results suggest that the methyltransferase preferentially recognizes partially denatured Hb molecules which possess a higher affinity for membrane binding sites, similar to Hb forms observed in senescent erythrocytes.     

Evidence for glucose-responsive and -unresponsive pools of phospholipid in pancreatic islets

The effect of glucose on the metabolism of phospholipids in pancreatic islets was studied with three radioactive phospholipid precursors, [32P]orthophosphate, [3H]myoinositol, and [3H]arachidonic acid, to determine the conditions necessary for studying the breakdown of prelabeled phospholipids. Islets were incubated in the presence of a radioactive precursor for 60 or 90 min and in the presence of either 3.3 or 16.7 mM glucose to prelabel phospholipids. To study the breakdown of prelabeled phospholipid, the unincorporated precursor was removed and the islets were reincubated for 15 or 20 min under conditions that either did or did not stimulate insulin release. Prelabeling in the presence of a noninsulinotropic concentration of glucose (3.3 mM) supported the incorporation of precursors into almost all islet phospholipids studied. Prelabeling in an insulinotropic concentration of glucose (16.7 mM) increased the incorporation of precursors into a number of phospholipids even more; and reincubation in 16.7 mM glucose caused a rapid loss of radioactivity from specific phospholipids (phosphatidylinositol and/or phosphatidylcholine, depending on the precursor). This breakdown was observed only when islets had been prelabeled in 16.7 mM glucose. The amount of radioactivity lost from phospholipid corresponded roughly to the additional amount incorporated during the prelabeling in the high concentration of glucose. Radioactivity in phospholipids in islets prelabeled in 3.3 mM glucose or in nonsecretagogue metabolic fuels, such as malate plus pyruvate, did not decrease when the islets were subsequently exposed to 16.7 mM glucose, nor did it decrease in 3.3 mM glucose when these islets had been prelabeled in 16.7 mM glucose. Glyceraldehyde, an insulin secretagogue, but not galactose or L-glucose which are not insulin secretagogues, stimulated phospholipid breakdown in islets that had been prelabeled in 16.7 mM glucose. Depriving islets of extracellular calcium, a condition that inhibits insulin release, inhibited phospholipid breakdown. The results suggest that pancreatic islets contain a glucose-responsive and a glucose-unresponsive phospholipid pool. The glucose-responsive pool becomes labeled and undergoes rapid turnover only under stimulatory conditions and may play a role in the stimulus-secretion coupling of insulin release.     

Release of histamine and arachidonate from mouse mast cells induced by glycosylation-enhancing factor and bradykinin

Stimulation of normal rat splenic T cells with pertussigen (lymphocytosis-promoting factor from Bordetella pertussis) resulted in the release of a soluble factor that enhanced the assembly of N-linked oligosaccharides to IgE-binding factors during their biosynthesis. The glycosylation-enhancing factor (GEF) is a kallikrein-like enzyme and is purified by absorption to p-aminobenzamidine-Agarose followed by elution with benzamidine. Incubation of normal mouse mast cells with affinity-purified GEF or bradykinin, a product of cleavage of kininogen by kallikrein, resulted in the release of histamine and arachidonate from the cells. Passive sensitization of mast cells with mouse IgE antibody, followed by pretreatment of the cells with a suboptimal concentration of GEF, resulted in an enhancement of antigen-induced histamine release. It was found that GEF and bradykinin induced the same biochemical events in mast cells as those induced by bridging of IgE receptors. Both GEF and bradykinin induced phospholipid methylation and an increase in intracellular cyclic AMP (cAMP). Incorporation of 3H-methyl groups into phospholipids and intracellular cAMP levels both reached a maximum 30 sec after challenge with GEF or bradykinin, and then declined to base-line levels within 2 to 3 min. These biochemical events were followed by 45Ca influx and histamine release; 45Ca uptake reached a plateau value at 2 min, and histamine release reached a maximum at 5 to 8 min. The initial rise in cAMP induced by GEF (or bradykinin) was not inhibited by indomethacin, indicating that the activation of adenylate cyclase is not the result of prostaglandin synthesis. In both IgE-mediated and GEF-induced histamine release, inhibitors of methyltransferases, such as 3-deaza adenosine and L-homocysteine thiolactone, inhibited not only phospholipid methylation but also the cAMP rise and subsequent Ca2+ uptake and histamine release. The results indicate that GEF induces activation of methyltransferases and that phospholipid methylation is involved in the cAMP rise, Ca2+ uptake, and histamine release. The induction of the same biochemical events in the same sequence by bridging of IgE receptors and by GEF (bradykinin) supports the hypothesis that receptor bridging induces the activation of serine protease(s) and cleavage products of this enzyme in turn activate methyltransferases in mast cells.     

Methylation reactions and the phytoalexin response in alfalfa suspension cultures

 In order to determine why the activated methyl cycle is up-regulated in plants undergoing defence responses to fungal pathogens we have monitored the utilisation of methyl groups derived from methionine in cell-suspension cultures of alfalfa (Medicago sativa L.) treated for various times with fungal elicitor, by carrying out a parallel labelling study with [³⁵S]methionine and [methyl-³H]methionine. The distribution of the two radiolabels among the medium, soluble cellular components and cell wall was then determined. In the absence of elicitor the utilisation of the two radiolabels was similar. However, in the presence of the elicitor the total incorporation of radioactivity from [methyl-³H]methionine into metabolites was far greater than from [³⁵S]methionine, indicating that the methyl label had been utilised in methylation reactions. Elicitor treatment resulted in up to a sixfold increase in the use of ³H-methyl groups in the methylation of hydrophobic metabolites. In the period 0–24 h after elicitor treatment, increased methylation was directed largely into the synthesis of the isoflavonoid phytoalexin medicarpin and related metabolites. Newly synthesized phytoalexins were exported into the medium, while a significant proportion of the medicarpin accumulating in the cell in the early stages of elicitation was derived from the hydrolysis of its respective conjugate. Elicitor treatment also modified the incorporation of ³H-methyl groups into the cell wall. Between 0 and 24 h after elicitor treatment the methylation of pectin in the cell wall declined. After 24 h, pectin methylation recovered and was associated with an increase in the methylation of other wall-bound polysaccharide components. Since no other major metabolic sink for the increased methylation was determined we conclude that the increased activity of the activated methyl cycle during defence interactions in alfalfa is required to support phytoalexin synthesis and cell wall modifications. Received: 1 August 1996 / Accepted: 24 October 1996