Serum epoxide hydrolase (preneoplastic antigen) in human and experimental liver injury.

Reports of an increase in a serum epoxide hydrolase (sEH), immunochemically related to microsomal EH in humans and rats with hepatocellular carcinoma (HCC), suggested its use as a serum marker for this disease. We have now measured sEH levels (as either immunochemically determined content or enzyme activity) in a number of human and experimental models of liver disease. sEH was elevated above the normal range in at least 50% of individuals with HCC, including: 3 of 6 northern Californians; 4 of 7 Koreans with hepatitis B-associated HCC; hepatitis B-associated HCC in woodchucks; and male rats receiving chronic treatment with aflatoxin B1 or ciprofibrate. sEH was rarely elevated in other forms of chronic liver disease. Only 2 of 9 Koreans with hepatitis B-associated cirrhosis, 1 of 8 carriers, but none with chronic active hepatitis or infection with no apparent liver disease had elevated sEH. In addition, no elevations were found in woodchucks with noncancerous viral hepatitis. In aflatoxin B1- and M1-treated rats sEH was not elevated in those with only hyperplastic foci or hepatocellular adenomas, and in two rat initiation-promotion protocols sEH was elevated only in those rats which received the entire set of treatments. sEH was also increased during acute hepatotoxicity in rats treated with CCl4 or 1,2-dibromo-3-chloropropane. The mechanism of increase in sEH during hepatocarcinogenesis appears to be different from that of other markers of HCC, for in the Korean patients, there was no correlation between sEH concentrations and those of alpha-fetoprotein or ferritin, nor was there a correlation with alpha-fetoprotein concentrations in the aflatoxin-treated rats. Furthermore, the increase in sEH does not correlate with induction of microsomal EH in the liver of experimental animals. Studies to date indicate that sEH is selective for HCC and severe hepatonecrotic injury, and may be of some use in the diagnosis of HCC, particularly as a complement to other serum markers.

Different densities of 5-HT3 receptors are labeled by [3H]quipazine, [3H]GR 65630 and [3H]granisetron.

The binding of three, structurally distinct, 5-hydroxytryptamine3 (5-HT3) receptor radioligands was characterized in rat cerebral cortex, rabbit ileum myenteric plexus and NG-108-15 neuroblastoma cells. The density of sites labeled by the three ligands in rat cortex or in rabbit ileum was markedly different. [3H]Quipazine labeled more sites than [3H]GR 65630 in rat cortex (4.0-fold) and rabbit ileum (1.8-fold), but not in NG-108-15 cells. [3H]Quipazine also labeled a greater density of sites than [3H]granisetron in rat cortex (7-fold) but not in NG-108 cells. [3H]Quipazine binding in rat cortex and rabbit ileum, but not in NG-108-15 cells, was displaced by non-radiolabeled GR 65630 in a manner consistent with an interaction with more than one site. These data indicate that not all 5-HT3 receptor radioligands recognize the same population of 5-HT3 binding sites with equivalent density and further suggest the existence of subtypes of 5-HT3 receptor binding sites in rat cortical and rabbit myenteric plexus preparations.

The beta 1 sodium channel subunit modifies the interactions of neurotoxins and local anesthetics with the rat brain IIA alpha sodium channel in isolated membranes but not in intact cells.

Mammalian brain sodium channels consist of an alpha subunit and two smaller beta subunits. The role of the beta 1 subunit in modulating ligand interactions at these channels was examined using a cell line stably expressing human beta1 and rat brain IIA alpha subunits. Coexpression of the beta 1 subunit had no effect on the potencies of sodium channel blockers in inhibiting whole cell [3H]batrachotoxinin A benzoate ([3H]BTX) binding or veratridine-stimulated [14C]guanidinium influx. Coexpression of the beta 1 subunit also had no effect on the potencies of alpha scorpion toxin, brevetoxin, or RU 39568 in stimulating [14C]guanidinium influx. By contrast, coexpression of the beta 1 subunit had dramatic effects on ligand interactions in isolated membranes. In isolated membranes of cells expressing only the alpha subunit, the neurotoxins had no stimulatory effect on [3H]BTX binding and the potencies of local anesthetic-like channel inhibitors were 10-100-fold lower than those at native sodium channels. Whereas in membranes of cells coexpressing the beta 1 subunit, the neurotoxins increased [3H]BTX binding 30-fold and the potencies of the sodium channel inhibitors closely matched those found at native sodium channels. These findings indicate that the beta 1 subunit is not required for the binding of sodium channel activators or inhibitors but rather, that the beta 1 subunit may stabilize the alpha subunit in a functional conformation thereby allowing detection of these interactions in disrupted membranes.

[3H]RS 57639, a high affinity, selective 5-HT4 receptor partial agonist, specifically labels guinea-pig striatal and rat cloned (5-HT4S and 5-HT4L) receptors.

RS 57639, by being a partial agonist in rat esophagus but a competitive antagonist in guinea-pig ileum, is one of several ligands which operationally discriminate among 5-HT4 receptors in different tissues. The discovery of splice variants of the 5-HT4 receptor, 5-HT4S and 5-HT4L, raises the possibility that this functional heterogeneity among 5-HT4 receptors may be due to differences in the interaction of ligands with different isoforms of the receptor. To test this idea, the functional and binding interactions of RS 57639 with rat 5-HT4S and 5-HT4L receptors were characterized. RS 57639 stimulated adenylate cyclase in cells expressing 5-HT4S or 5-HT4L receptors with similar potency (pEC50 = 7.9 +/- 0.1 and 7.6 +/- 0.1) and efficacy (71 +/- 3 and 59 +/- 4% of 5-HT). [3H]RS 57639 also bound to 5-HT4S and 5-HT4L receptors with similar affinity (Kd = 0.09 +/- 0.01 and 0.11 +/- 0.01 nM) and specificity (SB204070 > GR113808 > SDZ 205557 > cisapride > renzapride > alpha me-5-HT > 5-CT). Therefore, the operational differences among 5-HT4 receptors, detected with RS 57639, are not explained by differences in the interaction of the ligand with 5-HT4S and 5-HT4L receptors. [3H]RS 57639 binding to guinea-pig striatal membranes was also characterized. [3H]RS 57639 bound with high affinity (Kd = 0.25 +/- 0.07 nM) and a specificity similar to that of the 5-HT4 receptor antagonist, [3H]GR113808. Therefore, while the mechanism by which RS 57639 operationally distinguishes among 5-HT4 receptors was not determined, [3H]RS 57639 was shown to specifically label native and cloned 5-HT4 receptors. As the first selective agonist radioligand to be described for this receptor, [3H]RS 57639 may prove useful in further studies of receptor coupling and ligand interactions.

Identification of a single alpha 1-adrenoceptor corresponding to the alpha 1A-subtype in rat submaxillary gland.

1. The alpha 1-adrenoceptors present in membranes of rat liver, cortex and submaxillary gland were labelled with [3H]-prazosin and the affinity of 15 ligands for these receptors was determined. 2. In saturation studies, [3H]-prazosin bound with high affinity (Kd = 30-39 pM) to a single population of sites in all three preparations. 3. In competition studies using rat cortex, evidence for heterogeneity of the alpha 1-adrenoceptor binding sites was obtained. Displacement isotherms for amidephrine, benoxathian, oxymetazoline, phentolamine and WB 4101 were biphasic and were consistent with the presence of both alpha 1A- and alpha 1B-adrenoceptor subtypes as described by Morrow & Creese (1986) and Han et al. (1987). 4. The rat liver and submaxillary gland membrane preparations both possessed homogeneous populations of alpha 1-adrenoceptors. However, there were pharmacological differences between the receptors in these two preparations. Rat submaxillary gland alpha 1-adrenoceptors displayed high affinity for amidephrine, benoxathian, oxymetazoline, phentolamine and WB 4101 and therefore appeared to represent alpha 1A-adrenoceptors. Rat liver alpha 1-adrenoceptors possessed lower affinity for these ligands (6-65 fold) suggesting that these receptors were of the alpha 1B-subtype. 5. Spiperone exhibited 12.9 fold higher affinity for rat liver alpha 1B-adrenoceptors than for rat submaxillary gland alpha 1A-adrenoceptor and may therefore represent the first alpha 1B-adrenoceptor selective ligand.

Differences between the alpha 2-adrenoceptor in rat submaxillary gland and the alpha 2A-and alpha 2B-adrenoceptor subtypes.

1. The alpha 2-adrenoceptors of rat submaxillary gland, labelled with [3H]-rauwolscine, were characterized by use of a range of subtype selective ligands and were compared to rabbit spleen alpha 2A-adrenoceptors and rat kidney alpha 2B-adrenoceptors. 2. In rat submaxillary gland, [3H]-rauwolscine labelled an apparently homogeneous population of binding sites with relatively low affinity (Kd = 11.65 nM) compared to the affinity in rat kidney (Kd = 2.18 nM) and rabbit spleen (Kd = 4.64 nM). 3. In competition studies using 16 ligands the alpha 2-adrenoceptors in rat submaxillary gland appeared to differ from both the alpha 2A-adrenoceptor of rabbit spleen (r = 0.62) and also the alpha 2B-adrenoceptor of rat kidney (r = 0.28). 4. The affinity data obtained with benoxathian, imiloxan and WB 4101 indicated the presence of an alpha 2B-adrenoceptor in rat submaxillary gland. However, data for chlorpromaxine, oxymetazoline, spiroxatrine and xylometazoline indicated that submaxillary gland alpha 2-adrenoceptors were of the alpha 2A subtype. The affinity estimate for prazosin in rat submaxillary gland was intermediate between its affinity at the alpha 2A- and alpha 2B-adrenoceptors while affinity estimates for idazoxan and phentolamine in rat submaxillary gland were greater than those obtained at either the alpha 2A- or alpha 2B-adrenoceptor. 5. These data indicate that rat submaxillary gland alpha 2-adrenoceptors differ from the alpha 2A- and alpha 2B-adrenoceptors found in rabbit spleen and rat kidney, respectively.

Assessment of imiloxan as a selective alpha 2B-adrenoceptor antagonist.

1. The alpha 2-adrenoceptor binding sites of rabbit spleen and rat kidney, labelled with [3H]-rauwolscine, were characterized using a range of subtype selective ligands. 2. In rabbit spleen, the alpha-2-adrenoceptor binding sites displayed high affinity for oxymetazoline and WB 4101 and low affinity for prazosin and chlorpromazine suggesting the presence of an alpha 2A subtype. 3. There was evidence for heterogeneity of the alpha 2-adrenoceptor binding sites present in rabbit spleen. The results obtained with oxymetazoline and WB 4101 indicated that at least 75% of the [3H]-rauwolscine binding sites in this preparation displayed a pharmacology consistent with the presence of an alpha 2A subtype. 4. In rat kidney, the alpha 2-adrenoceptor binding sites displayed high affinity for prazosin and chlorpromazine and low affinity for oxymetazoline and WB 4101 suggesting the presence of an alpha 2B subtype. 5. The inclusion of guanylylimidodiphosphate (Gpp(NH)p, 0.1 mM) did not modify the pharmacology of the alpha 2-adrenoceptor binding sites present in the two preparations. Furthermore, when the two membrane preparations were combined, the resultant pharmacology was still consistent with the presence of two receptors that retained the characteristics of the alpha 2A and alpha 2B subtypes. 6. Imiloxan was identified as a selective alpha 2B ligand while benoxathian displayed a high degree of selectivity for the alpha 2A-adrenoceptor binding site. The selectivity of imiloxan for the alpha 2B-adrenoceptor binding site, coupled with its specificity for alpha 2-adrenoceptors, should make it a valuable tool in the classification of alpha 2-adrenoceptor subtypes.

Characterization of the muscarinic receptor in isolated uterus of sham operated and ovariectomized rats.

1. The pharmacological characteristics of muscarinic receptors in rat isolated uterus were studied in ovariectomized (ov.) and sham operated (sh.) animals. 2. Competition radioligand binding studies, using uterine membranes and [3H]-NMS, were undertaken with several muscarinic receptor antagonists. Most of the antagonists indicated a one-site fit with apparent affinity estimates (pKi) unchanged by ovariectomy. The selective M2 antagonist, tripitramine revealed high (representing 33+/-8 and 38+/-2%) and low (67+/-8 and 62+/-2%) affinity binding sites in both sh. and ov. rat uterus, respectively. These sites likely represented muscarinic M2 and M3 receptors and the proportions were not significantly different in the two conditions. 3. Carbachol induced concentration-dependent contractions which were surmountably antagonized by several muscarinic receptor antagonists (pKB, sh.; ov.): zamifenacin (9.19; 9.18), p-F-HHSiD (8. 50; 9.06), tripitramine (7.23; 7.54), himbacine (7.21; 7.41), methoctramine (6.79; 7.49), pirenzepine (6.48; 7.21), AF DX 116 (6. 26; 6.61), MTx 3 (<7.00; <7.00) and PD 102807 (<7.00; <7.00). 4. The apparent affinity values obtained in functional studies using the uteri from both sh. and ov. animals correlated most closely with values reported at human recombinant muscarinic M3 receptors. This suggests that the muscarinic M3 receptor mediates contraction under both conditions. 5. Radioligand binding experiments indicate the presence of M2 receptors, in addition to M3 receptors, which probably explains the discrepancies between functional and binding affinities. These data further suggest that the pharmacological profile and proportions of the two populations of muscarinic receptors are unaffected by ovariectomy.

Characteristics of 5-HT3 binding sites in NG108-15, NCB-20 neuroblastoma cells and rat cerebral cortex using [3H]-quipazine and [3H]-GR65630 binding.

1. The biochemical and pharmacological properties of 5-HT3 receptors in homogenates of NG108-15 and NCB-20 neuroblastoma cells and rat cerebral cortex have been ascertained by the use of [3H]-quipazine and [3H]-GR65630 binding. 2. In NG108-15 and NCB-20 cell homogenates, [3H]-quipazine bound to a single class of high affinity (NG108-15: Kd = 6.2 +/- 1.1 nM, n = 4; NCB-20: Kd = 3.0 +/- 0.9 nM, n = 4; means +/- s.e.means) saturable (NG108-15: Bmax = 1340 +/- 220 fmol mg-1 protein; NCB-20: Bmax = 2300 +/- 200 fmol mg-1 protein) binding sites. In rat cortical homogenates, [3H]-quipazine bound to two populations of binding sites in the absence of the 5-hydroxytryptamine (5-HT) uptake inhibitor, paroxetine (Kd1 = 1.6 +/- 0.5 nM, Bmax1 = 75 +/- 14 fmol mg-1 protein; Kd2 = 500 +/- 300 nM, Bmax2 = 1840 +/- 1040 fmol mg-1 protein, n = 3), and to a single class of high affinity binding sites (Kd = 2.0 +/- 0.5 nM, n = 3; Bmax = 73 +/- 6 fmol mg-1 protein) in the presence of paroxetine. The high affinity (nanomolar) component probably represented 5-HT3 binding sites and the low affinity component represented 5-HT uptake sites. 3. [3H]-paroxetine bound with high affinity (Kd = 0.02 +/- 0.003 nM, n = 3) to a site in rat cortical homogenates in a saturable (Bmax = 323 +/- 45 fmol mg-1 protein, n = 3) and reversible manner. Binding to this site was potently inhibited by 5-HT uptake blockers such as paroxetine and fluoxetine (pKi s = 8.6-9.9), while 5-HT3 receptor ligands exhibited only low affinity (pK; < 7). No detectable specific [3H]-paroxetine binding was observed in NG108-15 or NCB-20 cell homogenates. 4. [3H]-quipazine binding to homogenates of NG108-15, NCB-20 cells and rat cortex (in the presence of 0.1 microM paroxetine) exhibited similar pharmacological characteristics. 5-HT3 receptor antagonists competed for [3H]-quipazine binding with high nanomolar affinities in the three preparations and the rank order of affinity was: (S)-zacopride > quarternized ICS 205-930 2 granisetron > ondansetron > ICS 205-209 (R)-zacopride > quipazine > renzapride > MDL-72222 > butanopride > metoclopramide. 5. [3H]-GR65630 labelled a site in NCB-20 cell homogenates with an affinity (Kd = 0.7 + 0.1 nms n = 4) and density (B__ = 1800 + 1000 fmol mg- protein) comparable to that observed with [3H]-quipazine. Competition studies also indicated a good correlation between the pharmacology of 5-HT3 binding sites when [3H]-GR65630 and [3H]-quipazine were used in these cells. 6. In conclusion, [3H]-quipazine labelled 5-HT3 receptor sites in homogenates of NG108-15 cells, NCB-20 cells and rat cerebral cortex. In rat cortical homogenates, [3H]-quipazine also bound to 5-HT uptake sites, which could be blocked by 0.1 microM paroxetine. The pharmacological specificity of the 5-HT3 receptor labelled by [3H]-quipazine was similar in the neuroblastoma cells and rat cortex and was substantiated in NCB-20 cells by the binding profile of the selective 5-HT3 receptor antagonist, [3H]-GR65630.

Effect of dietary clofibrate on epoxide hydrolase activity in tissues of mice.

The effects of dietary clofibrate on the epoxide-metabolizing enzymes of mouse liver, kidney, lung and testis were evaluated using trans-stilbene oxide as a selective substrate for the cytosolic epoxide hydrolase, cis-stilbene oxide and benzo[a]pyrene 4,5-oxide as substrates for the microsomal form, and cis-stilbene oxide as a substrate for glutathione S-transferase activity. The hydration of trans-stilbene oxide was greatest in liver followed by kidney greater than lung greater than testis. Its hydrolysis was increased significantly in the cytosolic fraction of liver and kidney of clofibrate-treated mice and in the microsomes from the liver. Isoelectric focusing indicates that the same enzyme is responsible for hydrolysis of trans-stilbene oxide in normal and induced liver and kidney. Clofibrate induced glutathione S-transferase activity on cis-stilbene oxide only in the liver. Hydrolysis of both cis-stilbene oxide and benzo[a]pyrene 4,5-oxide was highest in testis followed by liver greater than lung greater than kidney. Hydration of cis-stilbene oxide was induced significantly in both liver and kidney by clofibrate but that of benzo[a]pyrene 4,5-oxide was induced only in the liver. These and other data based on ratios of hydration of benzo[a]pyrene 4,5-oxide to cis-stilbene oxide in tissues of normal and induced animals indicate that there are one or more novel epoxide hydrolase activities which cannot be accounted for by either the classical cytosolic or microsomal hydrolases. These effects are notable in the microsomes of kidney and especially in the cytosol of testis.

Sodium cholate extraction of rat liver nuclear xenobiotic-metabolizing enzymes.

DNA is the purported target of several carcinogenic and mutagenic agents. Nuclear enzymes which could generate or detoxify reactive metabolites are of major concern. Several such enzymes have been identified within nuclei, but obtaining samples with enriched content or activity is difficult, time-consuming, and uses harsh isolation techniques. Extraction of rat liver nuclear suspensions with cholate-containing buffer results in solubilization of 25-30% of the protein. Linear extraction was obtained for total protein and cytochromes P-450 and b5, NADPH-cytochrome P-450 reductase, NADH-cytochrome b5 reductase, DT-diaphorase, and microsomal-like epoxide hydrolase with specific activities comparable to values reported for isolated nuclear membrane, while the yield was five to ten times greater. Detergent extracts of rat liver nuclei were employed to study the comparative response of microsomal and nuclear enzymes to chemical treatment. While the responses to acute inductive (phenobarbital and 3-methylcholanthrene) and toxic (carbon tetrachloride and dibromochloropropane) treatments were qualitatively similar, an initiation-promotion protocol (diethylnitrosamine with phenobarbital promotion) resulted in divergent responses between the enzymes in the two subcellular fractions. Detergent extracts of nuclei offer an efficient means of recovering xenobiotic-metabolizing enzymes from rat liver nuclei, and have been utilized to demonstrate a differential response of nuclear enzymes during preneoplastic development.

Effects of chronic exposure to aflatoxin B1 and aflatoxin M1 on the in vivo covalent binding of aflatoxin B1 to hepatic macromolecules.

Induction of resistance to aflatoxin B1 (AFB1) binding to cellular macromolecules in the rat by chronic exposure to AFB1 and aflatoxin M1 (AFM1) was investigated. The binding of [14C]AFB1 to liver macromolecules was measured in F-344 rats fed 0.5 ppb or 50 ppb AFM1 or 50 ppb AFB1 for 41 wk. The animals then received an intragastric dose of [14C]AFB1 at 5 micrograms/kg and were sacrificed 6 h later. Hepatic DNA, RNA, and protein were isolated by chloroform-phenol extraction and hydroxylapatite chromatography. In animals preexposed to 50 ppb AFB1, labeled AFB1 binding to DNA, RNA, and protein was decreased by 72%, 74%, and 61%, respectively. Preexposure to AFM1 resulted in a small reduction in binding to nucleic acids. Glutathione transferase activity was increased by 133% in animals fed 50 ppb AFB1, by 48% in those preexposed to 50 ppb AFM1, and remained at control values in rats fed 0.5 ppb AFM1. These results suggest that the induction of detoxification enzymes following chronic exposure to aflatoxin might contribute to the reduction in covalent binding of AFB1 to macromolecules.

Epoxide metabolism in the liver of mice treated with clofibrate (ethyl-alpha-(p-chlorophenoxyisobutyrate)), a peroxisome proliferator.

An increase in cytosolic epoxide hydrolase (cEH) activity occurs in the livers of mice treated with peroxisome proliferating-hypolipidemic-nongenotoxic carcinogens. As increases in activity of epoxide metabolizing enzymes may reflect the carcinogenic mechanism, a detailed comparison of the response of cEH, microsomal epoxide hydrolase (mEH), and cytosolic glutathione S-transferase (cGST) activities using the geometrical isomers trans- and cis-stilbene oxide as substrates has been performed in livers from mice treated with clofibrate (ethyl-alpha-(p-chlorophenoxyisobutyrate]. The maximal increase of cEH activity occurred at lower dietary doses of clofibrate (0.5%) and within a shorter time (5 days) than mEH and cGST (2%, 14 days) activity. After 14 days at 0.5% clofibrate, cEH, mEH, and cGST activities were 250, 175, and 165% and 290, 220, and 75% of control values in male and female mice, respectively. Withdrawal of clofibrate from the diet resulted in a reversion of activities to control values within 7 days. Clofibrate treatment shifted the apparent subcellular compartmentation of all three enzymatic activities with an increase in the ratio of soluble to particulate activity. In particular, the relative specific activity of all three enzymes decreased in the light mitochondrial (peroxisomal) cell fraction, and an increase of a mEH-like activity (benzo[a]pyrene-4,5-oxide and cis-stilbene oxide hydrolysis) in the cytosol occurred. Both the increase of cEH activity and the appearance of mEH-like activity in the cytosol are novel responses of epoxide metabolizing enzymes, which may be related to the novel cellular responses that follow clofibrate treatment, peroxisome proliferation, hypolipidemia, and nongenotoxic carcinogenesis.

Allosteric interactions among agonists and antagonists at 5-hydroxytryptamine3 receptors.

Cooperation in the action of agonists suggests that there are multiple binding sites on 5-hydroxytryptamine3 (5-HT3) receptors. The purpose of this study was to characterize these binding sites and their interactions on both native and cloned 5-HT3 receptors. The affinities of competitive 5-HT3 receptor antagonists were similar regardless of whether the receptors were labeled with [3H]RS-42358, [3H]granisetron, or 1-(m-[3H]chlorophenyl)biguanide ([3H]mCPG). By contrast, the affinities of the agonists 5-HT, mCPG, and phenylbiguanide were approximately 10-fold higher when the receptors were labeled with [3H]mCPG. The dissociation of [3H]mCPG, [3H]RS-42358, and [3H]RS-25259, but not [3H]granisetron, from both cloned and native 5-HT3 receptors was markedly slower in the presence of 5-HT or 2-methyl-5-HT than in the presence of antagonists such as RS-42358. This suggests that the binding of these agonists to unoccupied sites on the receptor can increase the receptor's affinity for prebound ligands and thereby slow their dissociation. These data support previous indications of positive cooperation among multiple binding sites on both native and cloned 5-HT3 receptors, and they extend this idea by demonstrating that agonists can modify the interaction of some, but not all, antagonists with the receptor.

Cyclopropyl oxiranes: reversible inhibitors of cytosolic and microsomal epoxide hydrolases.

A series of aryl- and alkyl-substituted cyclopropyl oxiranes were synthesized as potential suicide inhibitors of mouse liver epoxide hydrolase (EH). The inhibitory potency of each compound and its corresponding alkene precursor was determined with mouse liver EHs using [3H]-cis-stilbene oxide as substrate for microsomal EH (mEH) and for glutathione-S-transferase, and using [3H]-trans-stilbene oxide for cytosolic EH (cEH). The cyclopropyl oxiranes all showed low (26-60% at 5 X 10(-4) M) inhibition of glutathione transferase and moderate inhibition (I50 = 5 X 10(-4) to 6 X 10(-6) M) for cEH and mEH. cis-Phenylcyclopropyl oxirane had an I50 for mEH near that for a commonly used inhibitor, 1,1,1-trichloropropene oxide. Inhibition appeared competitive and reversible, and the cyclopropyl oxiranes appeared to function as alternate substrates. Absence of irreversible inhibition is evidence against a strongly electrophilic epoxide-opening mechanism involving a cyclopropyl carbinyl-homoallyl cation rearrangement. Instead, a concerted mechanism is favored, in which electrophilic opening and hydroxide attack occur in a concerted fashion.

Inhibition of epoxide metabolism by alpha,beta-epoxyketones and isosteric analogs.

Chalcone oxides and several isosteric compounds have been prepared to examine the importance of the alpha,beta-epoxyketone moiety in the inhibition of the hydrolysis of [3H]-trans-stilbene oxide to its meso-diol by mouse liver cytosolic epoxide hydrolase (cEH). Inhibition of microsomal EH and glutathione S-transferase were also examined. For cEH, replacement of the carbonyl by methylidene reduces inhibitor potency by a factor of 44, while replacement of the epoxide ring with a cyclopropyl ring reduces inhibition by a factor of 450. A 2'-hydroxyl also reduces cEH inhibition by 100 times. These observations are consistent with a model of the active site in which the carbonyl is hy-hydrogen-bonded to an acidic site presumed to be involved in initiating epoxide hydrolysis. The chalcone oxides thus bind tightly but are not readily turned over as substrates.

[3H]BIMU-1, a 5-hydroxytryptamine3 receptor ligand in NG-108 cells, selectively labels sigma-2 binding sites in guinea pig hippocampus.

The binding of [3H]endo-N-(8-methyl-8-azabicyclo[3.2.1.]oct-3-yl)- 2,3-dihydro-3-ethyl-2-oxo-1H-benzimidazole-1-carboxamide hydrochloride ([3H]BIMU-1) a benzimidazolone with high affinity for 5-hydroxytryptamine (5-HT)3 and 4 5-HT3 and 5-HT4 receptors, was characterized in NG-108 cells and guinea pig hippocampus. Specific, heat-sensitive, binding of [3H]BIMU-1 was detected in both NG-108 cells and guinea pig hippocampus. In NG-108 cell membranes, a portion of the specific binding was displaced by 5-HT3 receptor ligands with affinities and specificity consistent with the labeling of 5-HT3 receptors. The residual specific binding was insensitive to serotonin (Ki > 1 mM) but was displaced by haloperidol (Ki of 50 nM). In guinea pig hippocampal membranes [3H]BIMU-1 binding was insensitive to serotonin but was displaced by haloperidol, and 1,3-di-o-tolyl-guanidine with affinities appropriate for the labeling of a sigma binding site (Ki of 6.3 and 31 nM, respectively). The affinity profile of ligands displacing [3H] BIMU-1 binding in guinea pig hippocampus was consistent with the selective labeling of a sigma-2 binding site because the sigma-1 selective benzomorphans, (+)-pentazocine and (+)-N-allylnormetazocine, only weakly displaced the binding (Ki greater than 1 microM). The affinity of BIMU-1 for sigma-2 binding sites (Ki = 32 nM) was 200-fold greater than that for sigma-1 binding sites (Ki = 6.3 microM), dopamine (D1 and D2), other serotonin (5-HT1A, 5-HT2A, 5-HT2C) and muscarinic (M1, M2, M3 and M4) receptors (Ki > 10 microM). The distribution of haloperidol-sensitive [3H]BIMU-1 binding was also consistent with the labeling of sigma-2 binding sites. These data suggest that [3H]BIMU-1 selectively labels sigma-2 binding sites in guinea pig hippocampus. [3H]BIMU-1, under appropriate experimental conditions, is thus the first sigma-2 binding site radioligand to be characterized.

A methodology for the analysis of the preneoplastic antigen.

A highly sensitive assay for the epoxide hydrolase activity associated with the preneoplastic antigen (PNA) has been developed based on the synthesis of cis-stilbene oxide labeled with tritium at approximately 15 Ci/mmol. This assay allows the detection of elevated epoxide hydrolase activity in the serum of humans and rodents as well as in the culture medium bathing isolated hepatocytes. The integrity of the enzymatic assay was confirmed in rodents by precipitating the serum PNA activity using an antibody raised against the rat microsomal epoxide hydrolase. Methodology for the detection of PNA in serum will facilitate evaluation of this antigen as a marker for hepatic neoplasia in man and in experimental animals.

Comparison of crude and affinity purified cytosolic epoxide hydrolases from hepatic tissue of control and clofibrate-fed mice.

An affinity purification procedure was developed for the cytosolic epoxide hydrolase based upon the selective binding of the enzyme to immobilized methoxycitronellyl thiol. Several elution systems were examined, but the most successful system employed selective elution with a chalcone oxide. This affinity system allowed the purification of the cytosolic epoxide hydrolase activity from livers of both control and clofibrate-fed mice. A variety of biochemical techniques including pH dependence, substrate preference, kinetics, inhibition, amino acid analysis, peptide mapping, Western blotting, analytical isoelectric focusing, and gel permeation chromatography failed to distinguish between the enzymes purified from control and clofibrate-fed animals. The quantitative removal of the cytosolic epoxide hydrolase acting on trans-stilbene oxide from 100,000g supernatants, allowed analysis of remaining activities acting differentially on cis-stilbene oxide and benzo[a]pyrene 4,5-oxide. Such analysis indicated the existence of a novel epoxide hydrolase activity in the cytosol of mouse liver preparations.