Synthesis of N-alkyl substituted indolocarbazoles as potent inhibitors of human cytomegalovirus replication

The synthesis and antiviral evaluation of unsymmetrical indolocarbazole derivatives of Arcyriaflavin A, substituted with a range of alkyl groups at the indole nitrogen, is described. Structure-activity relationships in this series against human cytomegalovirus (HCMV) replication in cell culture are reported. Compound 4b was identified as potent inhibitor of HCMV (IC(50)=19 nM), which retained activity against a range of HCMV strains including ganciclovir resistant isolates.

Indolocarbazoles: potent, selective inhibitors of human cytomegalovirus replication

In our search for new, safer anti-HCMV agents, we discovered that the natural product Arcyriaflavin A (la) was a potent inhibitor of HCMV replication in cell culture. A series of analogues (symmetrical indolocarbazoles) was synthesised to investigate structure activity relationships in this series against a range of herpes viruses (HCMV, VZV, HSV1, and 2). This identified a number of novel, selective and potent inhibitors of HCMV, 12,13-dihydro-2,10-difluoro-5H-indolo[2,3-a]pyrrolo[3,4-c]carbazol e-5,7-(6H)-dione (1d) being the best example (IC50=40 nM, therapeutic index > 1450). Compounds described in this series were generally poor inhibitors of protein kinase C betaII, and no correlation was found between the ability to inhibit HCMV and the enzyme PKC.

Effects of the indolocarbazole 3744W on the tyrosine kinase activity of the cytoplasmic domain of the platelet-derived growth factor beta-receptor

The cytoplasmic domain of the platelet-derived growth factor (PDGF) beta-receptor was expressed in insect cells by using a baculovirus system. The resulting protein was a constitutively active tyrosine kinase that could phosphorylate both protein and peptide substrates. A recently identified potent and selective inhibitor of intact PDGF receptor autophosphorylation, 3744W, inhibited the autophosphorylation of the cytoplasmic domain both in vitro (IC50 1.8+/-0.12 microM) and within intact insect cells (IC50 2.0 microM). However, under identical assay conditions, 3744W did not inhibit the phosphorylation of the synthetic polymeric peptide poly(Glu4Tyr1) even at concentrations as high as 100 microM. These results suggest that, although 3744W inhibits PDGF receptor autophosphorylation directly, it can discriminate between phosphate acceptor substrates.

Indolocarbazoles: potent and selective inhibitors of platelet-derived growth factor receptor autophosphorylation

A quantitative assay for measuring the autophosphorylation of platelet-derived growth factor (PDGF) receptors in intact vascular smooth muscle cells has been developed and used to screen for novel tyrosine kinase (TK) inhibitors. Several novel inhibitors of PDGF receptor autophosphorylation have been identified from the indolocarbazole series, including the 3,9 dimethoxy derivative, 3744W (IC50 = 14.5+/-2 nM). Tested against a panel of tyrosine and serine/threonine kinases, 3744W is at least 1,000 fold selective for the PDGF receptor tyrosine kinase and was found to inhibit autophosphorylation of both the alpha and beta isoforms of the PDGF receptor in human smooth muscle cells. PDGF-BB-stimulated DNA synthesis in quiescent cultures of human smooth muscle cells was blocked in a concentration-dependent manner by 3744W, IC50 = 10 nM. Binding studies showed that 3744W did not block the binding of PDGF-BB to cell surface receptors on human airway smooth muscle cells. Furthermore, inhibition of bone marrow stem cell proliferation by 3744W was only observed at concentrations 100-1,000 times greater than those needed to block PDGF-driven DNA synthesis in human smooth muscle cells. 3744W represents a novel, potent and selective inhibitor of PDGF receptor autophosphorylation and a powerful biochemical probe for investigating PDGF-dependent responses in vitro.

Phospholipase D: regulation and functional significance

PLD is a major route for hydrolysis of PC in most tissues, consistent with it playing an important role in signal transduction. The enzyme appears to be activated by a variety of different mechanisms in different tissues, suggesting there might be several different isoforms. Little, however, is known at present about its enzymology and molecular biology. There is little direct evidence to indicate the functional significance of PLD activation but an accumulation of indirect evidence links PLD with prolonged changes in cell function. In particular, two areas where there is strong evidence for a role for PLD are mitogenesis and leukocyte hyperresponsiveness. An important area for future work will be the investigation of how products from the PLD pathway exert these effects. Current evidence suggests an important role for Ca(2+)-independent PKC isoforms and probably also for novel cellular targets for the putative second messenger PA.

Tyrosine phosphorylation is involved in receptor coupling to phospholipase D but not phospholipase C in the human neutrophil

The tyrosine kinase inhibitors ST271, ST638 and erbstatin inhibited phospholipase D (PLD) activity in human neutrophils stimulated by fMet-Leu-Phe, platelet-activating factor and leukotriene B4. These compounds did not inhibit phorbol ester-stimulated PLD, indicating that they do not inhibit PLD per se, but probably act at a site between the receptor and the phospholipase. In contrast, the protein kinase C inhibitor Ro-31-8220 inhibited phorbol 12,13-dibutyrate- but not fMet-Leu-Phe-stimulated PLD activity, arguing against the involvement of protein kinase C in the receptor-mediated activation of PLD. ST271 did not inhibit Ins(1,4,5)P3 generation, but did inhibit protein tyrosine phosphorylation stimulated by fMet-Leu-Phe. The phosphotyrosine phosphatase inhibitor pervanadate increased tyrosine phosphorylation and stimulated PLD. These results suggest that tyrosine kinase activity is involved in receptor coupling to PLD but not to PtdIns(4,5)P2-specific phospholipase C in the human neutrophil.

A quantitative investigation into the dependence of Ca2+ mobilisation on changes in inositol 1,4,5-trisphosphate levels in the stimulated neutrophil

1. The coupling of N-formyl-methionyl-leucyl-phenylalanine (fMet-Leu-Phe) receptor stimulation to Ca2+ mobilisation has been investigated in the human neutrophil by measuring the concentration-effect curves for inositol 1,4,5-trisphosphate (IP3) formation and Ca2+ mobilisation. 2. fMet-Leu-Phe-dependent mobilisation of intracellular Ca2+ has been monitored in fluo-3-loaded human neutrophils by measuring increases in the cytoplasmic free Ca2+ concentration ([Ca2+]i) in the presence of extracellular EGTA. Fluo-3 was used in preference to fura-2 because it was found to be more sensitive to the high Ca2+ levels seen in stimulated neutrophils. 3. fMet-Leu-Phe induced a rapid mobilisation of intracellular Ca2+ (EC50 = 2.9 +/- 0.1 nM) and increased [Ca2+]i to a maximum of 1286 +/- 184 nM. 4. The amount of IP3 in fMet-Leu-Phe-stimulated neutrophils was determined by competition with [3H]-IP3 for a specific IP3 binding protein isolated from bovine adrenocortical microsomes. Basal IP3 levels of 13.3 +/- 2.0 pmol per 10(7) cells were increased nearly 4 fold by maximally effective concentrations of fMet-Leu-Phe. 5. The EC50 for the IP3 response (95 +/- 18 nM) was much higher than that for mobilisation of intracellular Ca2+, such that only a doubling in the concentration of IP3 was required to fully mobilise intracellular Ca2+. 6. As a result of this relationship IP3 production was more sensitive than Ca2+ mobilisation to inhibition by demethoxyviridin, an inhibitor of phospholipase activation.

Demethoxyviridin and wortmannin block phospholipase C and D activation in the human neutrophil

1. The fungal metabolite, wortmannin, has recently been shown to inhibit fMet-Leu-Phe-stimulated superoxide production and phospholipase D (PLD) activation in the human neutrophil. 2. We have found that a close structural analogue of wortmannin, demethoxyviridin, has a similar inhibitory profile but in addition blocks phosphatidylinositol 4,5-bisphosphate-specific phospholipase C and hence inositol 1,4,5-trisphosphate (IP3) formation. 3. Inhibition of fMet-Leu-Phe-stimulated PLD by demethoxyviridin was characteristically non-competitive (IC50 = 31 +/- 10 nM). 4. Inhibition of fMet-Leu-Phe-stimulation IP3 formation required concentrations almost 10 times higher (IC50 = 250 +/- 130 nM). 5. Surprisingly, demethoxyviridin only inhibited fMet-Leu-Phe-induced intracellular calcium mobilization at concentrations 100 times greater than those needed to block IP3 formation. 6. Demethoxyviridin also inhibited PLD activation induced by sodium fluoride or phorbol myristate acetate (PMA) but the concentrations required were 100 times those needed to block fMet-Leu-Phe-stimulated PLD. 7. These observations support the contention that PLD plays an important role in signal transduction in the human neutrophil and indicate that wortmannin and demethoxyviridin inhibit PLD activation at a common step in the signalling pathway. 8. Furthermore, these results suggest that demethoxyviridin may block the interaction between the chemotactic peptide receptor and a GTP-binding protein that is intimately involved in PLD activation.

The temporal relationship between phospholipase activation, diradylglycerol formation and superoxide production in the human neutrophil

Fluctuations in the amounts of choline, inositol 1,4,5-trisphosphate (IP3) and diradylglycerol have been used to monitor phospholipase activation in the human neutrophil. Stimulation of human neutrophils by formylmethionyl-leucylphenylalanine (fMet-Leu-Phe) resulted in a rapid activation of both phosphatidylinositol 4,5-bisphosphate breakdown by phospholipase C and phosphatidylcholine breakdown by phospholipase D. Diradylglycerol accumulation occurred more slowly than that of either choline or IP3 and was inhibited by 30 mM-butanol, suggesting that the bulk was derived from the phospholipase D pathway via phosphatidate phosphohydrolase. Consistent with this is the observation that choline and diradylglycerol are produced in similar amounts. 1,2-Diacylglycerol (DAG) and 1-O-alkyl-2-acyl-sn-glycerol species accumulated with different time courses, indicating that one or more steps in the phospholipase D pathway was selective for the diacyl species. Superoxide production by fMet-Leu-Phe-stimulated neutrophils paralleled DAG accumulation over the first 5 min, but thereafter this production stopped, despite the fact that DAG remained elevated. We conclude that DAG derived from the phospholipase D pathway is only one of the second messengers important in controlling this functional response.

A novel and sensitive assay for phospholipase D in intact cells

A novel and sensitive assay for phospholipase D (PLD) that measures the incorporation of high specific activity [3H]butan-1-ol into [3H]phosphatidylbutanol has been developed. The assay has been used to measure PLD activation in human neutrophils and platelets. Both the chemotactic peptide fMet-Leu-Phe and opsonised-zymosan stimulated PLD in the human neutrophil. In the platelet, PLD was stimulated by thrombin and collagen but responses were small and only occurred at high agonist concentrations. This assay has a number of advantages over existing techniques and should be valuable for investigating PLD activation in a variety of isolated cells and possibly intact tissues.

Selectivity of protein kinase inhibitors in human intact platelets

The specificity of commonly used protein kinase inhibitors has been evaluated in the intact human platelet. Protein kinase C (PKC) and cyclic AMP-dependent protein kinase (PKA) were activated selectively by treating platelets with phorbol dibutyrate (PDBu) or prostacyclin (PGl2). PKC activity was quantitated by measuring PDBu-specific phosphorylation of a 47,000 molecular weight protein, and PKA activity monitored by measuring prostacyclin-dependent phosphorylation of a 22,000 molecular weight protein. Staurosporine and 1-(5-isoquinolinylsulphonyl)-2-methyl-piperazine (H-7) were found to be non-specific inhibitors in the intact platelet, consistent with their effects on the isolated enzymes. Tamoxifen inhibited PKC activity (IC50 = 80 microM) but increased PKA-dependent protein phosphorylation. These results support the use of human platelets for measuring the specificity of protein kinase inhibitors and indicate that tamoxifen might have value for experimental purposes as a relatively selective PKC inhibitor.

Phospholipase D activation is functionally linked to superoxide generation in the human neutrophil

Neutrophils stimulated with formylmethionyl-leucylphenylalanine (fMet-Leu-Phe) in the presence of butanol and ethanol formed phosphatidyl alcohols through a phospholipase D mechanism. The alcohols inhibited phosphatidic acid and diradylglycerol (DRG) formation, but did not block inositol 1, 4, 5-trisphosphate release. fMet-Leu-Phe-stimulated superoxide production was inhibited by alcohol concentrations which blocked DRG formation, whereas opsonized-zymosan-stimulated superoxide production was only partially decreased. These results suggest that phospholipase D activation is functionally linked to superoxide production in the human neutrophil.

Ca2+-independent binding of [3H]phorbol dibutyrate to protein kinase C is supported by protamine and other polycations

The activity of the Ca2+- and phospholipid-dependent protein kinase, protein kinase C (PKC), can be modulated by diacylglycerols and phorbol esters. The association of these agents with PKC is, in turn, generally understood to be dependent on Ca2+ and phospholipids. Certain substrates, e.g. protamine sulphate, are known to undergo cofactor-independent phosphorylation by PKC. We report here that, in the presence of such substrates, PKC bound 1,2-dihexanoylglycerol and phorbol dibutyrate in a Ca2+-independent manner. Histone IIIs, which is phosphorylated by PKC only in the presence of Ca2+ and phospholipid, also supported Ca2+-independent binding of 1,2-dihexanoylglycerol and phorbol dibutyrate to PKC, but to a lesser extent than did protamine. Support for Ca2+-independent binding was also exhibited by non-peptide polycations (e.g. DEAE-cellulose DE52), indicating that recognition of the catalytic site is not a prerequisite for this effect. The natural polyamines spermine and putrescine did not have this property, however. The affinity of PKC for phorbol dibutyrate and 1,2-dihexanoylglycerol was found to be unchanged by the presence of substrates or DE52. It is proposed that, in the absence of Ca2+, certain polycations favour expression of the diacylglycerol/phorbol ester binding site by stabilizing the active conformation of PKC.

Evidence that a second stereochemical centre in diacylglycerols defines interaction at the recognition site on protein kinase C

The interaction of novel diacylglycerol analogues at the recognition site on protein kinase C has been evaluated using a modified [3H]phorbol dibutyrate binding assay and an established kinase activation assay. Studies with the 3-methyl analogues of 1,2-dihexanoyl-sn-glycerol have revealed a preferred stereochemical configuration at the C-3 position. Other chemical modifications have extended existing structure/activity relationships by showing that carbamates and sulphonyl esters cannot substitute for carboxylate esters and that cyclic acyl groups are active. Thus, most, if not all of the functionalities in the diacylglycerol molecule are required for interaction at the receptor on protein kinase C. Stereochemical specificity is required at C2 and C3.

Decrease of cellular ATP by dihexanoylglycerol may limit responses to protein kinase C activation

1,2-sn-Dihexanoylglycerol (HHG) reduced the ATP content of HL-60 cells. This concentration-related (10-100 microM) effect reached a maximum of over 90%, was enantiomerically specific and not accompanied by release of lactate dehydrogenase. Oleoylacetylglycerols (3-100 microM) had no effect on ATP levels while phorbol dibutyrate (PDBu, 0.01-1 microM) decreased ATP content of HL-60 cells by up to 40%. Responses stimulated by HHG became limited as the concentration was increased above 10 microM, this being manifest as either a low maximum response compared to PDBu (superoxide release) or a bell-shaped concentration-effect curve (degranulation). HHG (30-100 microM) inhibited PDBu-stimulated superoxide release, this inhibition being enantiomerically specific. It is probable that the effect of HHG on ATP content impairs cellular responses mediated through protein kinase C activation.

Inhibition of phorbol ester stimulated superoxide production by 1-oleoyl-2-acetyl-sn-glycerol (OAG); fact or artefact?

OAG-stimulated superoxide (O2) production by HL-60 granulocytes showed enantiomeric specificity but reached a maximum of only 5% of that produced by either phorbol myristate acetate (PMA) or phorbol dibutyrate (PDBu). At 10-100 microM, OAG displaced specifically-bound [3H]PDBu from intact HL-60 cells by only 25%, suggesting limited cell penetration. OAG (10-100 microM) also inhibited PDBu-stimulated O2 production by 25%; this inhibition was enantiomerically specific. However, at a lower concentration (3 microM), both enantiomers of OAG fully blocked O2 production stimulated by PMA (0.5 microM). This inhibition is probably artefactual, due to the hydrophobic PMA physically associating with OAG in the extracellular fluid.

Stimulation and inhibition of secretion by phorbol myristate acetate in different cell types

In washed human platelets and in HL60 granulocytes phorbol myristate acetate (PMA, 1-2000nM) synergised with threshold concentrations of secretogogues to induce a sustained maximum secretory response. Likewise, superoxide production from HL60 cells maintained a maximal response at PMA concentrations between 30-300nM. At concentrations up to 10nM PMA also augmented calcium ionophore, A23187, stimulated histamine release from rat peritoneal mast cells. However, in the mast cell PMA concentrations above 10nM reduced maximum histamine release in a dose-dependent manner.

Esterification of an endogenously synthesized lipoxygenase product into granulocyte cellular lipids

The human promyelocytic leukemia cell line HL60 can be induced to differentiate into mature granulocytes by exposure to Me2SO. [1-14C]Arachidonic acid incubated overnight with these cells was incorporated mainly into membrane phospholipids. Stimulation of these cells with the calcium ionophore, A23187, resulted in a rapid release of esterified arachidonic acid from phosphatidylethanolamine and phosphatidylcholine. The released arachidonic acid was metabolized via both the cyclooxygenase and lipoxygenase pathways into three major hydroxylated products, 12-L-hydroxy-5,8,10-heptadecatrienoic acid (HHT), 5(S)-hydroxy-6,8,11,14-icosatetraenoic acid (5-HETE), and 5-(S),12(R)-dihydroxy-6,8,10,14-icosatetraenoic acid (leukotriene B). Arachidonic acid was also incorporated into triacylglycerols and phosphatidylinositol. The lipoxygenase product, 5-HETE, was rapidly esterified into cellular lipids. Thirty minutes after ionophore stimulation, 55% of the total 5-HETE synthesized was esterified into phospholipids and 35% incorporated into acylglycerols. In contrast, the other hydroxylated derivatives of arachidonic acid (HHT and leukotriene B) were not incorporated into acylglycerols or phospholipids. Esterification of hydroxylated metabolites of arachidonic acid into membrane phospholipids may serve to regulate a number of granulocyte functions.

The production of 5-HETE and leukotriene B in rat neutrophils from carrageenan pleural exudates

Rat neutrophils isolated from three-hour carrageenan pleural exudates actively metabolize arachidonic acid into three major metabolites, HHT, 11-HETE and 15-HETE. However, in the presence of the calcium ionophore, A23187, or the non-ionic detergent, BRIJ 56, these cells also produce 5-HETE and LTB. The production of these lipoxygenase products is calcium dependent. While non-steroidal anti-inflammatory drugs do not affect 5-HETE or LTB production, BW 755C and ETYA inhibit formation of these metabolites from exogenously added arachidonic acid.

Role of extracellular arachidonate in regulation of prostaglandin biosynthesis in cultured 3T3 fibroblasts

The stimuli that activate phospholipase releasing substrate fatty acids for PG biosynthesis also activates the reuptake of liberated fatty acids by the cell. The substrate released by these stimuli is accessible to albumin in 3T3 cells, and it is therefore presumably released into the extracellular space before it undergoes enzymatic conversion. Added exogenous arachidonate competes freely with endogenous arachidonate released by a stimulus, suggesting no apparent coupling mechanism between phospholipase and cyclooxygenase enzyme systems.