Inhibition of the early phase of HIV replication by an isothiazolone, PD 161374.

A new class of substituted 2'-benzisothiazolone represented by PD 161374 was discovered with antiviral activity against retroviruses similar to previously described nucleocapsid inhibitor PD 159206 (DIBA-4). In T cell culture, the 50% inhibitory concentrations (EC(50)) of PD 161374 and PD 159206 were on average 2.5 microM (ranges of 1.2-13.5 microM) without any cytotoxic effect up to 100 microM. PD 161374 inhibited acute HIV infection and it was effective when added during the early phase of HIV infection. However, very modest effects were observed in chronically infected H9 cells and the HIV latency model line OM-10.1. Direct PCR analysis of infected cells demonstrated that PD 161374 delayed the appearance of completed HIV-cDNA products including 2LTR circles. Together all these results suggest that PD 161374 exerts its antiviral effect at pre-integration steps in the early phase of the virus life cycle. When combined with a protease inhibitor, PD 161374 did not show any antagonism and combination with a reverse transcriptase inhibitor (AZT) resulted in a synergistic effect.

Functional reconstitution of lost activity of chemically cross-linked or mutant moloney murine leukemia virus nucleocapsid proteins by trans-complementation.

Intermolecular crosslinking of the retroviral gag structural proteins, capsid (CA) and nucleocapsid (NC), by oxidation of cysteine thiols interferes with virus assembly and infectivity. PD156202 is a dithiobisbenzamide with broad antiviral activity against retroviruses. Treatment of cell free Moloney murine leukemia virus (MuLV) particles with PD156202 induced crosslinking of gag structural proteins and inhibited reverse transcription in mellitin permeabilized virions. Site directed mutagenesis of NC cysteines in the zinc finger of MuLV proviral DNA resulted in virus particles that were noninfectious. The NC mutant virus did not display any intermolecular crosslinking following treatment with PD156202 under nonreducing conditions, which supported that NC cysteine residues participated in PD156202 mediated crosslinking. Replication of MuLV NC mutant virus could be restored by independent expression of wild type NC, making it susceptible to PD156202. These results demonstrate that oxidation of NC cysteines are detrimental for virus assembly and that complementation with wild type NC restored the nucleocapsid protein activity.

Inhibitors of protein-RNA complexation that target the RNA: specific recognition of human immunodeficiency virus type 1 TAR RNA by small organic molecules.

TAR RNA represents an attractive target for the intervention of human immunodeficiency virus type 1 (HIV-1) replication by small molecules. We now describe three small molecule inhibitors of the HIV-1 Tat-TAR interaction that target the RNA, not the protein. The chemical structures and RNA binding characteristics of these inhibitors are unique for each molecule. Results from various biochemical and spectroscopic methods reveal that each of the three Tat-TAR inhibitors recognizes a different structural feature at the bulge, lower stem, or loop region of TAR. Furthermore, one of these Tat-TAR inhibitors has been demonstrated, in cellular environments, to inhibit (a) a TAR-dependent, Tat-activated transcription and (b) the replication of HIV-1 in a latently infectious model.

Discovery of selective, small-molecule inhibitors of RNA complexes--I. The Tat protein/TAR RNA complexes required for HIV-1 transcription.

We have developed a therapeutic program focusing on the inhibition of a human immunodeficiency virus-1 specific protein-RNA interaction. This program begins with a search for small organic molecules that would interfere with the binding of Tat protein to TAR RNA. The methodologies chosen to study the HIV-1 Tat-TAR interaction and inhibition include gel mobility shift assays, scintillation proximity assays, filtration assays, and mass spectrometry. These methods helped establish in vitro high-throughput screening assays which rapidly identified Tat-TAR inhibitors from our corporate compound library. Tat-activated reporter gene assays were then used to investigate the cellular activities of the Tat-TAR inhibitors. The cellular activity, selectivity, and toxicity data for select Tat-TAR inhibitors were determined. Evaluation of both the cellular data and the Tat-TAR inhibition results led to further testing in anti-HIV-1 infection assays.

Nonpeptidic HIV protease inhibitors: 4-hydroxy-pyran-2-one inhibitors with functional tethers to P1 phenyl ring to reach S3 pocket of the enzyme.

A systematic study of tethering various groups on 6-phenyl ring of 4-hydroxy-6-phenyl-3-[(2-isopropylphenyl)thio]pyran-2-one was performed to increase the binding affinity with HIV protease. This tethering approach was aimed to fill S3 pocket of the enzyme. Thus, tethering hydrophilic groups resulted in more potent inhibitors. Similarly, various aromatic hydrophobic rings as well as heterocyclic rings were explored as tethering substituents to alter the physical properties as well as to enhance the binding affinity with HIV protease. Inhibitor 24, 4-hydroxy-3-[(2-isopropylphenyl)thio]-6-[4-(3-pyridinylmethoxy+ ++ ) phenyl]-2H-pyran-2-one, was evaluated as a prototypic lead structure to study various physical as well as pharmacological properties of this class of HIV protease inhibitors.

An HIV-1 protease screening assay using a non-infectious proviral clone.

An in-vitro assay was developed to screen for HIV-1 protease inhibitors using a non-infectious proviral clone (X19) with a deletion in the envelope gene (Ratner et al., 1991). The proviral clone, X19, was expressed transiently in the COS 7 cell line. The virus was able to replicate as assessed by the presence of p24 in the supernatants, yet the virions produced did not infect CD4 positive cells. To determine the effect of a known protease inhibitor on p24 antigen production, PD 148310 (a Ro 31-8959 analog) was added immediately after transfection of the COS 7 cells. Virus particles were produced maximally after 24 h and cell supernatants were assayed for p24 antigen production using a p24 ELISA assay. PD 148310 inhibited p24 release in a dose-dependent manner with an IC50 of 23.6 nM. Western blot analysis of the supernatants using a mouse monoclonal antibody against p24 confirmed the presence of a well-defined p24 band in the control lane. At 1000 nM of PD 148310 the p24 band was not detectable, leaving only the unprocessed p55 Gag precursor. This technique is a useful tool to screen for potential HIV-1 protease inhibitors.

Erythromycin contracts rabbit colon myocytes via occupation of motilin receptors.

Erythromycin stimulates gastroduodenal motility via action on motilin receptors. We evaluated erythromycin as a colonic muscle motilin agonist using in vitro rabbit colon studies. Isolated myocytes contracted to erythromycin with a half-maximal effective concentration of 2 pM and peak shortening of 22.4 +/- 2.5% at 1 nM, which was superimposable with the response to motilin. 125I-labeled motilin binding to colon muscle homogenates was saturable and specific with a dissociation constant (Kd) of 0.39 nM and maximal binding (Bmax) of 41 +/- 3 fmol/mg protein. Motilin displaced specifically bound 125I-motilin, with a Kd of 0.31 nM. Erythromycin displaced 125I-motilin but was less potent, with an inhibitory constant of 84.0 nM. Bmax values from displacement studies were similar to the Scatchard data. Motilin receptor protection from alkylation by N-ethylmaleimide preserved contraction to motilin and erythromycin but not acetylcholine or cholecystokinin, whereas protection with erythromycin preserved contraction to motilin but not other agonists. In conclusion, erythromycin binds to colon muscle motilin receptors present in densities similar to reported values for the upper gut. Furthermore, erythromycin contracts colonic myocytes via specific action on motilin receptors. Thus erythromycin may have colonic motor-stimulating properties by action on motilin receptors.

Cisapride acts on muscarinic (glandular M2) receptors to induce contraction of isolated gastric myocytes: mediation via a calcium-phosphoinositide pathway.

The effects of cisapride on guinea pig gastric smooth muscle were characterized. Cisapride induced concentration-dependent contraction of isolated myocytes with an EC50 of 10(-11) M, which was antagonized in concentration-dependent fashion by atropine. 4-Diphenylacetoxy-N-methylpiperidine-methiodide (4-DAMP) (10(-6) M), a selective M2 glandular receptor antagonist, inhibited contraction to cisapride (10(-10) M), whereas pirenzepine, an M1 receptor antagonist, had no effect. For comparison, carbachol induced contraction with an EC50 of 5 x 10(-11) M, which was also atropine-sensitive. Cisapride-induced contraction was not blocked by methysergide (10(-6) M), a nonselective serotonin antagonist. Cisapride displaced specific binding of [3H]N-methylscopolamine at a single receptor population with a Ki of 6.51 x 10(-5) M and a Bmax of 351.0 fmol/mg protein. Cisapride (10(-6) M) evoked a 73.8 +/- 7.6% increase in inositol 1,4,5-trisphosphate at 30 sec and a maximal increase of greater than 130% at 2 min, which was significantly reduced by atropine (10(-6) M). In contrast, cisapride did not enhance or inhibit basal levels of cyclic AMP. Thus, cisapride induces contraction via specific interaction at smooth muscle M2 glandular muscarinic receptors, which is associated with inositol trisphosphate generation, but not adenylyl cyclase activation.

Inhibition of guinea-pig oxyntic cell function by adenosine and prostaglandins.

Adenosine and prostaglandins (PGs) are known inhibitors of oxyntic cell function. Using quantitative cytochemistry of hydroxyl ion production (HIP) in guinea-pig oxyntic cells, we examined the effects of adenosine and PGs on secretagogue-stimulated HIP. Adenosine (10(-6) M) inhibited the actions of histamine (10(-14) M) and gastrin (2.5 X 10(-12) M) by 69 and 67%, respectively, but not that of dibutyryl cyclic AMP (10(-16) M) or carbachol (10(-9) M). These observations suggest that adenosine does not influence the Ca++-dependent pathway of carbachol action and that the adenosine activity precedes the generation of cyclic AMP. Adenosine and related analogs, N6-L-phenylisopropyladenosine and 5-N-ethylcarboxam-idoadenosine (10(-12) to 10(-14) M), inhibited histamine-stimulated HIP (10(-14) M) in the following order: N6-L-phenylisopropyladenosine greater than 5-N-ethylcarboxamidoadenosine greater than adenosin. The adenosine antagonist, 1,3-diethylphenylxanthine (10(-6) M), reversed the inhibitory effects of adenosine. Exogenous PGE2 (10(-6) M) also inhibited histamine- and gastrin-stimulated HIP by 65 and 55%, respectively. Indomethacin (10(-6) M) and flurbiprofen (10(-6) M), PG synthesis inhibitors, potentiated the action of histamine by 175 and 159%, respectively. Adenosine was incapable of reversing this potentiated effect. These data indicate that adenosine and related analogs are inhibitors of oxyntic cell HIP and suggest that these biological properties are mediated by binding to a cell surface receptor and thereby regulating oxyntic cell adenylate cyclase activity. The more potent properties of N6-L-phenylisopropyladenosine as compared to 5-N-ethylcarboxamidoadenosine are consistent with activity at the high-affinity surface adenosine receptor.(ABSTRACT TRUNCATED AT 250 WORDS)

Preliminary characterization of enterooxyntic activity on the guinea pig oxyntic cell.

Stimuli originating in the small intestine enhance gastric acid secretion, an effect termed the intestinal phase of gastric secretion. A humoral agent, enterooxyntin (EO) may mediate this effect. Mechanical distension of an ex vivo-perfused segment of canine jejunum caused the release of EO measured by cytochemical quantification of hydroxyl ion production (HIP) in guinea pig gastric oxyntic cells, an index of acid secretion. The H2 receptor antagonist, cimetidine (10(-5) M), and diamine oxidase, which hydrolyzes endogenous histamine, reduced HIP by 60% and 48%, respectively. Atropine (10(-5) M), the muscarinic cholinergic antagonist, reduced HIP by 75%. EO-induced HIP was also inhibited partially by the Ca2+ antagonist EGTA (10(-6) M) and by blockade of calcium channels with LaCl3 (10(-6) M). EO does not appear to operate through any single pathway. EO may be a single substance, different from gastrin, or a mixture of substances that have stimulatory effects on the oxyntic cell. Its action on the oxyntic cell is apparently mediated by both histaminergic and cholinergic pathways. Since neither cimetidine nor atropine completely inhibited EO-induced HIP, a direct effect of EO on the oxyntic cell seems likely and appears to depend on Ca2+. Isolation and purification of EO will be necessary to better assess the potency, efficacy, and the detailed cellular mechanisms of EO action.

Isolation and partial characterization of an entero-oxyntin from porcine ileum.

Porcine ileal mucosa was homogenized and freeze-thawed in 0.05 M NH4HCO3 + 0.01 M EDTA + 1 mM benzamidine hydrochloride at pH 8.6. Subsequent stepwise precipitation with (NH4)2SO4 followed by fractionation on Sephadex G-50 medium and G-50 fine eluted with alkaline buffer and final fractionation on G-50 superfine in 1.0 M acetic acid yielded a pure protein of 13,000 daltons as determined by sodium dodecyl sulfate electrophoresis. The amino acid composition of the protein has been determined and it contains 126 residues with no tryptophan detectable. Tryptic peptide maps demonstrate that the protein does not contain glucagon and RIA of the peptide did not detect any immunoreactive glucagon or gastrin. The isoelectric point is 6.4. The intact protein is resistant to Edman degradation and the partial N-terminal sequences of two CNBr fragments are: Lys-Arg-Leu-Ala-Leu ...., Glu-Gly-Gly-Thr-Val-Val-Val-Asn-Ser.... The C-terminal residue, alanine was determined using carboxypeptidase Y. The isolated peptide, in the range of 10(-15)-10(-9) M stimulated oxyntic cell hydroxyl ion production in sections of guinea pig gastric fundus. The dose response was linear with biphasic peaks at 10(-14) and 10(-9) M and the maximal response to the peptide was equal to that observed with gastrin. The addition of either atropine (10(-5) M) or cimetidine (10(-5) M) with the peptide (10(-14) M) caused greater than 50% inhibition of oxyntic cell stimulation (P less than 0.005). This peptide is a potent stimulator of the oxyntic cell and its effect is inhibited by muscarinic cholinergic and H2 receptor blockers. Hence, it represents a significant component of the physiological enterooxyntin effect observed in response to intestinal meals.

Action of gastrin in guinea pig oxyntic cells. Studies using quantitative cytochemistry.

The mechanism of action of gastrin was investigated using cytochemical quantitation of hydroxyl ion production (HIP) in guinea pig gastric oxyntic mucosa. The reaction depends upon the trapping of OH ions produced during gastric stimulation and is blocked by the benzimidazole, Hassle 149/94, which inhibits the K+ + H+-ATPase and by acetazolamide, an inhibitor of carbonic anhydrase activity. It is thus a measure of hydroxyl ions produced during stimulation of the oxyntic cell and reflects upon hydrogen ion production. Gastrin (2.5 X 10(-16) -2.5 X 10(-12) M) caused a linear dose-dependent stimulation of HIP in the oxyntic cells. The response was biphasic, with an early peak at 90 s and a secondary rise at 240 s, which persisted for 10 min. Natural human gastrin (sulfated and nonsulfated) and the active COOH-terminal octapeptide fragment of gastrin stimulated HIP, whereas the biologically inert NH2-terminal (1-13) fragment of gastrin had no effect. The activation of oxyntic cell HIP by gastrin was neutralized by an antiserum directed towards the COOH-terminus of gastrin and not by nonimmune serum. Cimetidine (10(-5) M) blocked 25% and atropine (10(-5) M) had no effect on gastrin-stimulated HIP. EGTA (10(-3) M) and LaCl3 (10(-3) M) inhibited the action of gastrin by 67 and 52%, respectively. The calmodulin antagonists, trifluoperazine (10(-5) M), pimozide (10(-5) M), and the naphthalene sulfonamides, W-7 and W-13 (10(-5) M), inhibited gastrin-stimulated HIP by 45.6 38.5, 42.3, and 37.2%, respectively. Higher doses of W-7 and W-13 (10(-4) M) inhibited gastrin-stimulated HIP by 83 and 67%. The Ca2+ ionophore, A23187 (10(-4) M), stimulated HIP. Thus, it appears that gastrin stimulation of HIP is complex. 25% of its action is via a histamine-dependent pathway. 45% of its action is dependent upon extracellular Ca2+. Its action is also in part dependent upon a Ca2+/calmodulin mechanism.

Cytochemical quantification of physiologic regulation of oxyntic cell carbonic anhydrase.

We have studied the physiologic factors regulating oxyntic cell activity using cytochemical quantification of carbonic anhydrase (CA) activity. Gastrin (10 (-16) to 10(-12)M), histamine (10(-17) to 10(-13) M), and carbamylcholine (10(-13) to 10(-8) M) caused a dose-dependent increase in CA in the oxyntic cells in guinea pig gastric fundus, maximal at 90 sec. The stimulation of CA by all three secretagogues was inhibited by the CA inhibitor, acetazolamide. The agonist activities were selectively blocked by respective antagonists. The benzimidazole derivative compound Hassle 149/94 (10(-3)M) abolished the actions of all agonists. Thus, histamine, gastrin and carbamylcholine have independent actions on oxyntic cell CA. The inhibition of the activity of all three secretagogues by H149/94 suggests a close link between CA activity and the functioning of the proton pump H+ + K+-ATPase.

Evidence for histamine H1 and H2 receptors in guinea-pig oxyntic cells.

We have previously shown that histamine in the dose range of 1.0 X 10(-16) to 1.0 X 10(-14) M caused a linear increase in oxyntic cell carbonic anhydrase (CA) activity. A maximal dose of the H2 antagonist, cimetidine, reduced the action of histamine by only 74 +/- 7.8% suggesting a possible H1 component of histamine action on oxyntic cells. Using quantitative cytochemistry of CA activity in guinea-pig oxyntic cells, we compared the dose-response relationships (1.0 X 10(-17)-1.0 X 10(-9) M) of specific H2 (dimaprit, 4-methylhistamine, impromidine) and H1 (2-pyridylethylamine) agonists with histamine (H1 and H2). In addition, we studied the effects of the specific H2 antagonist, cimetidine, and H1 antagonist, mepyramine, on histamine, dimaprit and 2-pyridylethylamine-stimulated CA activity. We found that guinea-pig oxyntic cells respond to both H1 and H2 agonists. H1 agonists were more potent but less efficacious than H2 agonists. H2 agonists were more efficacious, but less potent than H1 agonists. Cimetidine (10(-10)-10(-8) M) was a competitive antagonist of dimaprit-stimulated CA activity. However, at higher doses (10(-5) M), cimetidine antagonism of histamine-stimulated CA activity was of a mixed noncompetitive and competitive type. Cimetidine and mepyramine exhibited both H2 and H1 antagonism, but their affinities for their respective agonists were greater. The data suggests that the oxyntic cell has receptors for both H1 and H2 agonists. H1 agonism is important at low concentrations of agonists. The H1 and H2 receptor components appear to be functionally linked.

Role of calcium and calmodulin in activation of the oxyntic cell by histamine and carbamylcholine in the guinea pig.

The role of calcium in stimulation of the oxyntic cell by histamine and carbamylcholine was studied using a sensitive quantitative cytochemical staining technique that measures oxyntic cell hydroxyl ion production (HIP) as an index of acid secretion. Histamine (10(-17)-10(-14) M), carbamylcholine (10(-12)-10(-9) M), and extracellular calcium (10(-7)-10(-3) M) caused a linear, dose-dependent stimulation of the oxyntic cell. EGTA (10(-6) M) inhibited carbamylcholine by 50% but not histamine-stimulated activity. Lanthanum chloride (10(-6) M) caused 100% inhibition of carbamylcholine-induced activity but did not affect histamine-stimulated activity. A maximally effective dose of calcium (10(-4) M) caused additive effects on HIP at low doses of carbamylcholine without alteration of the maximal effect of carbamylcholine. Calcium (10(-4) M) did not enhance the effects of histamine. The calmodulin antagonists, trifluoperazine (10(-5) M), pimozide (10(-5) M), and a naphthalenesulfonamide (W-7), inhibited the integrated response to histamine by 54, 56, and 53%, and that of carbamylcholine by 65, 64, and 99%, respectively. Thus, extracellular calcium per se, stimulates the oxyntic cell. The action of carbamylcholine is completely dependent upon calcium/calmodulin mediation, supporting the concept that cholinergic actions are mediated via calcium-calmodulin events. Although histamine does not require extracellular or membrane calcium events to stimulate the oxyntic cell, calmodulin appears to participate in histamine action.

Enterooxyntin release from isolated perfused canine jejunum.

UNLABELLED: A humoral factor may mediate the intestinal phase of gastric acid secretion. An ex vivo perfused segment of canine jejunum maintained by an oxygenated asanguinous physiologic perfusate was used to test for release of an enterooxyntin (EO) in response to balloon distention at 30 mm Hg for 15 min. Gastric acid secretion in guinea pig fundic mucosa was determined indirectly by a quantitative cytochemical bioassay (CBA) of oxyntic cell hydroxyl ion production (HIP). An increase in the optical density (OD) caused by the cytochemical stain in the oxyntic cells reflects HIP, an index of acid secretion. Basal OD for segments with distention was 16.6 +/- 0.53 and for those without 15.5 +/- 0.68 (NS). Results are expressed as mean change of OD from basal (mean delta OD +/- SEM). (Table-see text) EO caused greater stimulation of HIP than gastrin or histamine. EO was heat stable. Trichloroacetic acid treatment decreased EO activity as did pronase digestion suggesting that EO is composed of one or more peptides. CONCLUSION: EO, an acid secretagogue, is a humoral agent probably composed of one or more peptides and is released by small bowel distention. Mechanical distention of the small bowel may be an important mechanism for the perpetuation of gastric acid secretion. The ex vivo perfused jejunal segment in conjunction with the CBA are ideal tools with which to study mechanisms of release of EO and the mechanism of action of EO on the oxyntic cell.

Somatostatin is an agonist and noncompetitive antagonist of gastrin in oxyntic cell function.

Somatostatin inhibits gastric acid secretion in vivo by effects on oxyntic cells which may be indirect. Using quantitative cytochemistry, we have investigated the actions of somatostatin on guinea pig oxyntic cell carbonic anhydrase activity, an index of acid secretion. Cyclic tetradecapeptide somatostatin, in the range of 2.9 X 10(-15) to 2.9 X 10(-10) M, stimulated oxyntic cell carbonic anhydrase activity in sections of guinea pig gastric fundus with a linear dose-response between 2.9 X 10(-14) and 2.9 x 10(-11) M. The maximal response to somatostatin was 50% of that to gastrin. Gastrin stimulated oxyntic cell carbonic anhydrase in the dose range of 2.3 X 10(-15) to 2.3 X 10(-11) M. The addition of 2.9 x 10(-13) M somatostatin caused an inhibition of 38-75% of the carbonic anhydrase activity stimulated by various concentrations of gastrin. The addition of 2.3 X 10(-13) M gastrin (which caused 79% of maximal carbonic anhydrase activation), to each dose of somatostatin from 2.9 X 10(-15) to 2.9 X 10(-10) M did not alter the action of somatostatin, and the action of gastrin was reduced to that observed with somatostatin alone. The data indicate noncompetitive inhibition between somatostatin and gastrin. Thus, somatostatin is an intrinsic, albeit weak, stimulator of gastric oxyntic cell carbonic anhydrase activity, and despite this, is a noncompetitive antagonist of the action of gastrin. Gastrin is an agonist and noncompetitive antagonist of the action of somatostatin.