Thick filter neutrophil chemotaxis performed in the absence of albumin.

Migration through a filter is a technique widely used for the study of cell chemotaxis. Since the original description of the technique by Boyden in 1962 (J. Exp. Med. 115, 453) using neutrophils and thick filters prepared from cellulose nitrate, albumin has been required in the incubation medium to support chemotaxis. However, the binding to albumin of compounds affecting chemotaxis can reduce their free concentration. We developed two procedures for studying neutrophil chemotaxis under reduced or albumin-free conditions. In one, cellulose nitrate filters were pretreated with albumin by a novel procedure and chemotaxis was carried out in albumin-free medium. As tested with the chemoattractant LTB4 and human neutrophils, the procedure resulted in full chemotaxis, measured by the number of cells crossing the filter, with an EC50 of 0.43 nM for LTB4. The LTB4-receptor antagonist LY 223982 inhibited the chemotactic action of LTB4 with a Ki of 62 nM in the albumin-pretreated filter system, thus showing 58 times greater potency than in medium containing 0.5% albumin. The second procedure makes use, for the first time, of a relatively new filter (Hydrophilic Durapore). This filter has the same dimensions and pore rating of the cellulose nitrate filter but did not require pretreatment with albumin to support chemotaxis in the albumin-free medium. LTB4 stimulated neutrophil chemotaxis across this filter with an EC50 of 0.29 nM. LY 223982 had a Ki of 11 nM, thus exhibiting a potency even greater than in the albumin-pretreated cellulose nitrate filter system. fMLP and C5a also stimulated chemotaxis in the absence of albumin. These results suggest that albumin is not obligatory for neutrophil chemotaxis through thick filters. The role of albumin in the chemotaxis assay using cellulose nitrate filters may be to counteract the adherence of cells or chemotactic agents to the filters.

Influence of standard and novel LTB4 analogs on human neutrophil chemotaxis measured by the multiwell cap assay.

Standard and novel LTB4 analogs were tested for neutrophil chemoattractant activity using the multiwell cap assay (Evans et al. (1986) Biosc. Rep. 6, 1041). The assay uses disposable equipment and measures chemotaxis by the number of cells able to migrate across the full thickness of cellulose nitrate filters. Under standard conditions (90 min incubation at 37 degrees C in buffer containing 2% bovine albumin), LTB4 and 6-cis-LTB1 had EC50 values of 3.5 and 15,000 nM, respectively. 20-hydroxy-LTB4 was equipotent with LTB4 and exhibited a similar biphasic chemotactic response, however, only one third of the number of cells migrated through the filter. 20-carboxy-LTB4 was inactive up to 1,000 nM. 5-desoxy-((6,7)-cis-cyclopropyl)-LTB2, (6,7)-benzo-LTB2 and 5-desoxy-(8,10)-LTB2 had EC50 values of 11,300, 50,000 and 84,000 nM, respectively. Checkerboard analysis indicated a chemokinetic component of 42% for LTB4 at a concentration causing peak chemotaxis. Reduction of albumin in the buffer to 0.5% increased the apparent potencies of LTB4 and 6-cis-LTB1 five-fold. Since LTB4 is a mediator of inflammation, various anti-inflammatory agents were tested at peak concentrations observed in vivo for in vitro inhibition of LTB4-stimulated chemotaxis in the presence of 0.5% albumin. Under the conditions of the assay, chloroquine diphosphate, dexamethasone, indomethacin, penicillamine, piroxicam and diclofenac sodium were inactive; gold sodium thiomalate was inhibitory (IC50 = 20 microM).

Comparison of LTB4- and C5a-stimulated chemotaxis of isolated human neutrophils: difference revealed by cell migration in thick filters using the multiwell cap procedure.

Under comparable conditions (90 min incubation in 2% albumin buffer) using 3 micron pore cellulose nitrate filters and the multiwell cap procedure (Evans et al., Bioscience Reports 6:1041, 1986), C5a was more potent than LTB4 as a chemoattractant (EC'50 s = 0.5 and 4 nM) and caused 5 times as many cells to completely traverse the filter. During the 90 min incubation, no cells traversed the filter in the absence of chemoattractant. The cap assay was modified to study migration of cells within the filter. During the 30 min incubation, EC50 values for C5a and LTB4 were comparable (1.2 and 1.6 nM) in causing cells to enter the filters but C5a was superior in causing the cells to move as measured by the mean distance of migration (EC'50 s = 0.25 and 1.8 nM). Our studies support the view that LTB4 acts primarily to cause cell adhesion (and penetration) of the endothelium and that C5a plays a major role in cell migration (McMillan and Foster, Agents and Actions 24: 114, 1988).

Enhancement of N-formyl-methionyl-leucyl-phenylalanine (fMLP) binding to isolated human neutrophils.

Short chain aliphatic alcohols have previously been found to enhance fMLP binding to human neutrophils, presumably by non-specific mechanisms involving increased membrane hydrophobicity or decreased microviscosity. We report the discovery of highly potent fMLP binding enhancers with 30,000 times the potency of the alcohols. Activity of the compounds is specific since slight changes in structure drastically alter their ability to enhance binding and since closely related compounds inhibit rather than enhance binding. The effects of experimental compounds on enhancement or inhibition of fMLP binding are paralleled by similar actions on fMLP-stimulated shape change of neutrophils.

Structure-activity profile of a series of novel triazoloquinazoline adenosine antagonists.

During a search for benzodiazepine receptor modulators, a highly potent adenosine antagonist (CGS 15943) was discovered. The compound was defined as a resonance-stabilized hybrid of the canonical structures 9-chloro-2-(2-furyl)[1,2,4]triazolo[1,5-c]quinazolin-5-amine (2a) and 9-chloro-2-(2-furyl)-5,6-dihydro[1,2,4]triazolo[1,5-c]-quinazolin- 5-imine (2b). Spectroscopic evidence and chemical reactivity in polar media favor the amine form 2a as the major contributor of the two canonical structures. The synthesis of 2 and some of its analogues and the structure-activity relationships in four biological test systems are described. Replacement of the 9-chloro group by hydrogen, hydroxyl, or methoxyl gave compounds with comparable binding potency at the A1 and A2 receptors but much less activity as antagonists of 2-chloroadenosine in guinea pig tracheal strips. Alkylation of the 5-amino group caused, in general, a loss of binding activity, particularly at the A2 receptor, as well as complete loss of activity in the tracheal model. Modification of the 2-furyl group caused a pronounced loss of activity in all of the test systems.

Biochemical characterization of the triazoloquinazoline, CGS 15943, a novel, non-xanthine adenosine antagonist.

CGS 15943A, a triazoloquinazoline, is a potent and selective adenosine receptor antagonist as assessed by its effects on radioligand binding and adenosine-stimulated adenylate cyclase activity in guinea pig synaptoneurosomes. At the adenosine A-1 receptor labeled with [3H]cyclohexyladenosine, CGS 15943A had an IC50 value of 20 nM. At the striatal A-2 receptor labeled with [3H]5'-N-ethylcarboxamidoadenosine in the presence of a low concentration of cyclopentyladenosine to block A-1 receptors labeled by this nonselective adenosine agonist, CGS 15943A had an IC50 value of 3 nM, indicating that the compound had some degree of selectivity for the A-2 receptor. Analysis of the effect of the compound on the saturation isotherms for each of the receptors indicated that it was a competitive antagonist at the brain A-1 receptor but that it was noncompetitive at the striatal A-2 receptor. CGS 15943A was a potent adenosine antagonist in the adenosine-stimulated adenylate cyclase system in guinea pig synaptoneurosomes, where the compound was found to have an IC50 value of 30 to 70 nM against the increase in cyclic AMP evoked by 5 microM adenosine. CGS 15943A had no effect on the binding of [3H]nitrobenzylthioinosine, a ligand thought to bind to adenosine uptake sites, and, at a concentration of 10 microM, had no effect on beef heart type III phosphodiesterase activity.(ABSTRACT TRUNCATED AT 250 WORDS)

Differences in Ca2+ entry blockers revealed by effects on adenosine- and adrenergic- receptors and cyclic AMP levels of [2-3H]adenine-prelabeled vesicles prepared from guinea pig brain.

Three major classes of Ca2+ entry blockers, classified according to effects on cardiac and vascular smooth muscle, were tested. Vesicles prepared from cerebral cortex and stimulated by adenosine and epinephrine constituted adenosine and alpha-adrenergic receptor systems respectively. Vesicles prepared from cerebellum and stimulated by epinephrine constituted the beta-adrenergic receptor system. Experiments with adenosine were also performed with vesicles formed or incubated in the absence of exogenous Ca2+. The results indicate that Ca2+ entry blockers had a variety of effects, even within classes of drugs. Vascular-selective group A Ca2+ entry blockers such as nifedipine and nisoldipine antagonized adenosine, but the structurally-related drug nitrendipine was inactive. Inhibition was competitive with adenosine and independent of exogenous Ca2+. In contrast to receptor-binding studies requiring high ratios of the drugs to adenosine receptor radioligands, nifedipine and nisoldipine were inhibitory at equimolar concentrations with adenosine. Non-selective group A Ca2+ entry blockers such as diltiazem and verapamil were inactive against adenosine. Group B Ca2+ entry blockers, prenylamine and perhexilene, increased cyclic AMP (cAMP) levels of vesicles stimulated by adenosine but not by epinephrine or under basal conditions. This effect was observed only in vesicles that had been formed in the presence of Ca2+. Ca2+ entry blockers also exhibited effects on adrenergic receptors unrelated to effects on adenosine. Verapamil and prenylamine acted as alpha-adrenergic antagonists and only prenylamine acted as a beta-adrenergic antagonist. However, the vesicle system also revealed indirect blocking actions of nifedipine on adrenergic receptor systems. The actions of the Ca2+ entry blockers are discussed in relation to the special usefulness of nifedipine in the treatment of patients with defective atrioventricular conduction and also in relation to the unique ability of group B Ca2+ entry blockers to selectively inhibit Ca2+ and calmodulin activated phosphodiesterase. However, some caution must be applied in drawing conclusions relating to the cardiovascular actions of these drugs from data generated in a neuronally-derived model.

Highly activatable adenylate cyclase in [2-3H]adenine-prelabeled vesicles prepared from guinea pig cerebral cortex by a simplified procedure.

An improved procedure utilizing a simple fractionation technique is described for the preparation and use of [2-3H]adenine-prelabeled vesicles from guinea pig cerebral cortex containing highly responsive adenylate cyclase activity. Adenosine consistently increased activity 1500--2000%, contrasted with activations of 200--300% previously reported by other investigators. Adenosine at 5 microM was more active in our system than at 20 times this concentration in studies by other investigators, increasing activity by 580--840%. Experimental conditions were explored, and Ca2+ was found to be necessary during tissue homogenization, but not during subsequent vesicle incubation. However, neither the higher Ca2+ concentration used by us (2.5 mM) nor the method of tissue homogenization could adequately explain the high activity of our preparations. The size of the incubation vessel was critical for both low basal activity and high activity in the presence of adenosine. Our preparations were similar to others in that combinations of neurohormones, which included histamine and epinephrine, elicited higher activities than individual neurohormones. Inspection of our vesicle preparations by scanning electron microscopy also showed them to be compatible with previously described preparations.

CGS 8216: receptor binding characteristics of a potent benzodiazepine antagonist.

CGS 8216 is a novel nonbenzodiazepine that inhibited 3H-flunitrazepam (3H-FLU) binding to rat synaptosomal membranes in vitro at subnanomolar concentrations. It prevented the in vivo labeling of brain benzodiazepine receptors by 3H-FLU with the same potency as diazepam when given orally to mice. Pharmacologic tests showed that it was devoid of benzodiazepine-like activity but it antagonized the actions of diazepam in these tests. It did not interact with alpha- or beta- adrenergic, H1-histaminergic or GABA receptors but it inhibited adenosine-activation of cyclic AMP formation. Studies with 3H-CGS 8216 demonstrated that it bound specifically and with high affinity to rat forebrain membranes and was displaced by drugs with an order of potencies similar to that observed when 3H-diazepam and 3H-FLU were used as radioligands. The regional distribution of 3H-CGS 8216 binding sites in the brain was also similar to that of 3H-FLU. Dissociation of 3H-CGS 8216 binding was slow at 0 degrees C but increased with temperature and was almost complete within 1 min at 37 degrees C. Scatchard analyses were linear, yielding KD values of 0.044, 0.11 and 0.18 nM at 0, 25 and 37 degrees C, respectively; the Bmax value did not change appreciably with temperature and was approximately 1000 fmoles/mg protein. Using 3H-FLU, the value for Bmax as well as for the KD increased with temperature. The total number of binding sites determined for 3H-FLU was greater than that for 3H-CGS 8216 at each temperature. CGS 8216 exhibited mixed-type inhibition of 3H-FLU binding. GABA did not stimulate 3H-CGS 8216 binding whereas it enhanced 3H-FLU binding. CGS 8216 may be a useful ligand for probing the antagonist properties of the benzodiazepine receptor and is likely to exhibit interesting therapeutic effects.