Systemic and intestinal pharmacokinetics of methotrexate in patients with inflammatory bowel disease.

BACKGROUND: The pharmacokinetics of low-dose subcutaneous methotrexate have not been determined throughout the standard weekly dosing interval. It is not known whether methotrexate concentrations in the gastrointestinal tract are sufficient for pharmacologic activity in inflammatory bowel disease. METHODS: Ten patients with inflammatory bowel disease participated in the study. After the patients started taking 15 or 25 mg subcutaneous methotrexate once a week, erythrocyte methotrexate concentration was measured every 2 weeks. The absorption, rectal distribution, metabolism, and elimination of methotrexate were measured. The effect of methotrexate on proliferation of an intestinal epithelial cell line was determined. RESULTS: After weekly subcutaneous administration of methotrexate was begun, trough erythrocyte concentration rose to reach a plateau after 6 to 8 weeks, ranging from 150 to 300 nmol/L. More than 90% of subcutaneously administered methotrexate was rapidly excreted in the urine. The methotrexate plasma time course after subcutaneous administration fit a 2-compartment first-order model with biphasic elimination and trough concentration of about 1 nmol/L. Trough and peak methotrexate concentrations (mean value +/- SD) were 64 +/- 33 and 206 +/- 64 fmol/mg in the rectal mucosa and 4 +/- 3 and 51 +/- 26 nmol/L in the rectal lumen. These methotrexate concentrations were in the range found to be pharmacologically active against Caco-2 cell growth, that is, a 50% inhibitory concentration from 10 to 46 nmol/L. CONCLUSION: Subcutaneous methotrexate was well absorbed and distributed to the site of the lesions in patients with inflammatory bowel disease. Methotrexate was concentrated intracellularly in blood and in the rectum. The methotrexate concentration in the rectal mucosa remained within a pharmacologically active range throughout the dosing interval. The findings represent a pharmacologic explanation for the sustained efficacy of weekly methotrexate therapy.

The novel Ins(1,4,5)P3 analogue 3-amino-3-deoxy-Ins(1,4,5)P3: a pH-dependent Ins(1,4,5)P3 receptor partial agonist in SH-SY5Y neuroblastoma cells.

We have synthesized the first amino-substituted inositol 1,4,5-trisphosphate [Ins(1,4,5)P3] analogue, D-3-amino-3-deoxy-myo-Ins(1,4,5)P3 (9). Although 9 is a full agonist at the Ca2+ mobilizing Ins(1,4,5)P3 receptor at pH 7.2 and 7.6, it is apparently a high intrinsic activity partial agonist at pH 6.8, releasing only 80% of the Ins(1,4,5)P3-sensitive Ca2+ stores of SH-SY5Y cells. Additionally, 9 was able to fully displace [3H]Ins(1,4,5)P3 from binding sites in rat cerebellum membranes at both pH 6.8 and 7.6, indicating a full interaction with the Ins(1,4,5)P3 receptor. The activity displayed by this amino analogue is unexpected and may be indicative of a pH-dependent conformational change in the amino acid residues comprising the Ins(1,4,5)P3 binding site.

Inhibition of recombinant human mitochondrial aldehyde dehydrogenase by two intermediate metabolites of disulfiram.

Disulfiram is used in aversion therapy for alcoholism. S-Methyl-N,N-diethylthiocarbamate (MeDTC) sulfoxide, a potent inhibitor of the target enzyme mitochondrial aldehyde dehydrogenase (ALDH2), is thought to be the principal active metabolite of disulfiram in vivo. We examined the effects on recombinant human ALDH2 of two intermediate metabolites of disulfiram, S-methyl-N,N-diethyldithiocarbamate (MeDDC) sulfoxide and MeDDC sulfine. MeDDC sulfoxide was a potent inhibitor of ALDH2 with an IC50 of 2.2 +/- 0.5 microM (mean +/- SD, N = 4) after preincubation with enzyme for 30 min. MeDDC sulfine was a relatively weak inhibitor of ALDH2 under the same conditions with an IC50 value of 62 +/- 14 microM. The inhibition of ALDH2 by both compounds was irreversible and did not require the cofactor NAD. The latter finding demonstrates that inactivation of ALDH2 is independent of the dehydrogenase activity of the enzyme. GSH blocked almost completely the inhibition by 20 microM of MeDDC sulfoxide and greatly diminished the inhibition by 200 microM of MeDDC sulfine. Inactivation by MeDDC sulfoxide was time dependent. MeDTC sulfoxide was a more potent inhibitor of recombinant human ALDH2 (IC50 = 1.4 +/- 0.3 microM after preincubation for 15 min) than either of the intermediate metabolites, and its inhibition was unaffected by GSH. Our results suggest that these newer intermediate metabolites of disulfiram, especially the more potent MeDTC sulfoxide, have the potential to inhibit the target enzyme ALDH2 in patients receiving disulfiram. However, until the significance of the interactions of the inhibitors with GSH is more fully understood, the contribution of MeDDC sulfine and MeDDC sulfoxide to the pharmacological effects of disulfiram in vivo is uncertain.

S-methyl N,N-diethylthiocarbamate sulfone, a potential metabolite of disulfiram and potent inhibitor of low Km mitochondrial aldehyde dehydrogenase.

Disulfiram inhibits hepatic aldehyde dehydrogenase (ALDH) causing an accumulation of acetaldehyde after ethanol ingestion. It is thought that disulfiram is too short-lived in vivo to directly inhibit ALDH, but instead is biotransformed to reactive metabolites that inhibit the enzyme. S-Methyl N,N-diethylthiocarbamate (MeDTC) sulfoxide has been identified in the blood of animals given disulfiram and is a potent inhibitor of ALDH (Hart and Faiman, Biochem Pharmacol 46: 2285-2290, 1993). MeDTC sulfone is a logical metabolite of MeDTC sulfoxide. Therefore, we investigated the effects of MeDTC sulfone on the activity of rat hepatic low Km mitochondrial ALDH, the major enzyme in the metabolism of acetaldehyde. MeDTC sulfone inhibited the low Km mitochondrial ALDH in vitro with an IC50 of 0.42 +/- 0.04 microM (mean +/- SD, N = 5) compared with disulfiram, which had an IC50 of 7.5 +/- 1.2 microM under the same conditions. The inhibition of ALDH by MeDTC sulfone was time dependent. The decline in ALDH activity followed pseudo first-order kinetics with an apparent half-life of 2.1 min at 0.6 microM MeDTC sulfone. Inhibition of ALDH by MeDTC sulfone was apparently irreversible; dilution of the inhibited enzyme did not restore lost activity. The substrate (acetaldehyde, 80 microM) and cofactor (NAD, 0.5 mM) together completely protected ALDH from inhibition by MeDTC sulfone; substrate alone partially protected the enzyme. Addition of either thiol-containing compound glutathione (GSH) or dithiothreitol (DTT) to MeDTC sulfone before incubation with the enzyme increased the IC50 of MeDTC sulfone by 7- to 14-fold. Neither GSH nor DTT could restore lost ALDH activity after exposure of the enzyme to MeDTC sulfone. Results of these studies indicate that MeDTC sulfone, a potential metabolite of disulfiram, is a potent, irreversible inhibitor of low Km mitochondrial ALDH.

Cellular pharmacology of D-3-azido-3-deoxy-myo-inositol, an inhibitor of phosphatidylinositol signaling having antiproliferative activity.

D-3-Azido-3-deoxy-myo-inositol (3AMI) is an inhibitor of the growth of v-sis-transformed NIH 3T3 cells but not of wild-type NIH 3T3 cells, whose effects may be mediated through the phosphatidylinositol-3'-kinase pathway. We studied some properties of the cellular pharmacology of 3AMI using high-specific-activity [3H]-3AMI. The uptake of [3H]-3AMI by wild-type NIH 3T3 and v-sis NIH 3T3 cells was similar. [3H]-3AMI was a substrate for phosphatidylinositol synthetase, with the maximal velocity (Vmax) being 1.0 nmol min-1 mg-1 and the Michaelis constant (Km) being 23 mM. Corresponding values obtained for [3H]-myo-inositol as a substrate were 5.5 nmol min-1 mg-1 and 3.2 mM. [3H]-3AMI was incorporated into the cellular inositol lipids of v-sis NIH 3T3 cells to a similar extent as that observed for [3H]-myo-inositol but was not incorporated into the inositol lipids of wild-type NIH 3T3 cells. The [3H]-3AMI incorporated by the v-sis NIH 3T3 cells was present in the phosphatidylinositol and phosphatidylinositol phosphate fractions but not in bisphosphorylated phosphatidylinositol. myo-Inositol antagonized the growth-inhibitory effects of 3AMI. The v-sis NIH 3T3 cells were found to be more sensitive than the wild-type NIH 3T3 cells to growth inhibition (without 3AMI) caused by the removal of myo-inositol from the medium. The results of the study suggest that 3AMI is an antimetabolite of myo-inositol. The relative sensitivity of v-sis NIH 3T3 and some other cells to 3AMI may be a reflection of increased myo-inositol requirements for the growth of these cells as compared with wild-type NIH 3T3 cells.

D-3-deoxy-3-substituted myo-inositol analogues as inhibitors of cell growth.

A number of unnatural D-3-deoxy-3-substituted myo-inositols were synthesized and their effects on the growth of wild-type NIH 3T3 cells and oncogene-transformed NIH 3T3 cells were studied. The compounds were found to exhibit a diversity of growth-inhibitory activities and showed selectivity in inhibiting the growth of some transformed cells as compared with wild-type cells. Remarkably, D-3-deoxy-3-azido-myo-inositol exhibited potent growth-inhibitory effects toward v-sis-transformed NIH 3T3 cells but had little effect on the growth of wild-type cells. The growth-inhibitory effects of the myo-inositol analogues were antagonized by myo-inositol. Since [3H]-3-deoxy-3-fluoro-myo-inositol was shown to be taken up by cells and incorporated into cellular phospholipids, we suggest that these unnatural myo-inositol analogues may act as antimetabolites in the phosphatidylinositol intracellular signalling pathways. Because cells transformed by oncogenes often exhibit elevated phosphatidylinositol turnover, the inhibition of signalling pathways that mediate oncogene action could offer new opportunities for controlling the growth of cancer cells.

[125I]thienylphencyclidine, a novel ligand for the NMDA receptor.

We have monitored the binding of [125I]thienylphencyclidine ([125I]TCP), a novel high affinity radioiodinated ligand that specifically recognizes the NMDA (N-methyl-D-aspartate) receptor in rat brain membranes. [125I]TCP binds with an affinity of about 30 nM, and recognizes a similar number of binding sites to previously employed ligands for this receptor. [125I]TCP binding is characterized by slow association and dissociation rates, and the latter can be modified by the addition of Mg2+ or Zn2+, as previously described for [3H]dizocilpine ([3H]MK801). Other phencyclidine-like ligands displaced [125I]TCP binding with the order of potency dizocilpine greater than thienylphencyclidine greater than ITCP greater than phencyclidine greater than ketamine. The binding of [125I]TCP was also increased by NMDA and glycine-site agonists and inhibited by antagonists of these sites. Surprisingly, however, the polyamines spermidine and spermine did not increase [125I]TCP, even though the polyamine antagonist arcaine was an effective inhibitor of binding. These results show that [125I]TCP is a useful ligand for the NMDA receptor complex that binds to the receptor in a manner that is qualitatively distinct from previously described ligands.

S-methyl-N,N-diethylthiocarbamate sulfoxide and S-methyl-N,N-diethylthiocarbamate sulfone, two candidates for the active metabolite of disulfiram.

The mechanism of action of disulfiram involves inhibition of hepatic aldehyde dehydrogenase (ALDH). Although disulfiram inhibits ALDH in vitro, it is believed that the drug is too short-lived in vivo to inhibit the enzyme directly. The ultimate inhibitor is thought to be a metabolite of disulfiram. In this study, we examined the effects of S-methyl-N,N-diethylthiocarbamate (MeDTC) sulfoxide and S-methyl-N,N-diethylthiocarbamate sulfone (confirmed and proposed metabolites of disulfiram, respectively) on rat liver mitochondrial low K(m) ALDH. MeDTC sulfoxide and MeDTC sulfone, in 10-min incubations with detergent-solubilized mitochondria, inhibited ALDH activity with an IC50 (mean +/- SD) of 0.93 +/- 0.04 and 0.53 +/- 0.11 microM, respectively, compared with 7.4 +/- 1.0 microM for the parent drug disulfiram. Inhibition by MeDTC sulfone and MeDTC sulfoxide, both at 0.6 microM, was time-dependent, following apparent pseudo-first-order kinetics with a t1/2 of inactivation of 3.5 and 8.8 min, respectively. Dilution of ALDH inhibited by either sulfoxide or sulfone did not restore activity, an indication of irreversible inhibition. Addition of glutathione (50 to 1000 microM) to ALDH before the inhibitors did not alter the inhibition by MeDTC sulfoxide. In contrast, the inhibition by MeDTC sulfone was decreased > 10-fold (IC50 = 6.3 microM) by 50 microM of glutathione and almost completely abolished by 500 microM of glutathione. The cofactor NAD, in a concentration-dependent manner, protected ALDH from inhibition by MeDTC sulfoxide and MeDTC sulfone. In incubations with intact mitochondria, the potency of the two compounds was reversed (IC50 of 9.2 +/- 3.6 and 0.95 +/- 0.30 microM for the MeDTC sulfone and sulfoxide, respectively). Our results suggest that MeDTC sulfone is highly reactive with normal cellular constituents (e.g., glutathione), which may protect ALDH from inhibition, unless this inhibitor is formed very near the target enzyme. In contrast, MeDTC sulfoxide is a better candidate for the ultimate active metabolite of disulfiram, because it is more likely to be sufficiently stable to diffuse from a distant site of formation, such as the endoplasmic reticulum, penetrate the mitochondria, and react with ALDH located in the mitochondrial matrix.

Cloning of a cDNA for short/branched chain acyl-Coenzyme A dehydrogenase from rat and characterization of its tissue expression and substrate specificity.

The acyl-CoA dehydrogenases are a family of related enzymes which catalyze the alpha,beta-dehydrogenation of acyl-CoA esters, transferring electrons to electron-transferring flavoprotein. A cDNA for human short/branched chain acyl-CoA dehydrogenase has recently been cloned, and it has been suggested that this enzyme represents the human homolog for the previously reported 2-methyl branched chain acyl-CoA dehydrogenase purified from rat liver. We now report the cloning and expression of rat short/branched chain acyl-CoA dehydrogenase and characterization of its substrate specificity. The rat enzyme is more active toward longer carbon side chains than its human counterpart, while the human enzyme can utilize substrates with longer primary carbon chains. In addition, short/branched chain acyl-CoA dehydrogenase can utilize valproyl-CoA as a substrate. Northern blotting of mRNA shows ubiquitous tissue expression of both the rat and human enzyme. Further study of these enzymes will be helpful in understanding structure/function relationships in this gene family.

Molecular recognition at the myo-inositol 1,4,5-trisphosphate receptor. 3-position substituted myo-inositol 1,4,5-trisphosphate analogues reveal the binding and Ca2+ release requirements for high affinity interaction with the myo-inositol 1,4,5-trisphosphate receptor.

Several novel D-myo-inositol 1,4,5-trisphosphate (Ins(1,4,5)P3] analogues equatorially substituted at the 3-position have been synthesized to probe the structure-activity relationship of the Ins(1,4,5)P3-receptor subsite adjacent to the native 3-hydroxy (3-OH) of Ins(1,4,5)P3. This study was prompted, in part, by our observation that myo-inositol 1,3,4,5-tetrakisphosphate (Ins(1,3,4,5)P4), the 3-position phosphorylated product of Ins(1,4,5)P3 was a full agonist at the Ca(2+)-mobilizing Ins(1,4,5)P3 receptor of SH-SY5Y cells (Wilcox, R.A., Challiss, R. A. J., Liu, C., Potter, B. V. I., and Nahorski, S. R. (1993) Mol. Pharmacol. 44, 810-817). The 3-position Ins(1,4,5)P3 analogues were equatorially substituted with groups spanning the steric range between the 3-OH of Ins(1,4,5)P3 and the 3-phosphate of Ins(1,3,4,5)P4; in order of increasing 3-position steric bulk these were: 3-fluoro-, 3-chloro-, 3-amino-, 3-bromo-, 3-methoxy-, and 3-phosphorothioate-Ins(1,4,5)P3. The analogues were assessed at the specific Ins(1,4,5)P3 binding-site of bovine adrenal cortex and for Ca2+ mobilizing activity in saponin-permeabilized SH-SY5Y human neuroblastoma cells. A correlation was observed between increasing molecular volume of the 3-position substituent and respective decreases in both affinity and Ca2+ mobilizing efficacy. Further analysis of the data also revealed that Ins(1,4,5)P3 analogues with equatorial 3-OH, 3-phosphate, and 3-phosphorothioate substituents interacted more favorably with Ins(1,4,5)P3 recognition sites than would be predicted by purely steric considerations. In contrast, 3-C-trifluoromethyl-Ins(1,4,5)P3 (which is axially substituted, but retains the native 3-OH of Ins(1,4,5)P3) interacted with Ins(1,4,5)P3 recognition sites with virtually the same potency as Ins(1,4,5)P3, indicating that the binding pocket of the Ins(1,4,5)P3-receptor was not sterically restrictive with respect to axially oriented 3-position substituents. We conclude that the Ins(1,4,5)P3 receptor has favorable non-covalent binding interactions with the equatorial 3-position substituents of Ins(1,4,5)P3 and Ins(1,3,4,5)P4 and that these interactions significantly ameliorate the steric constraints of the Ins(1,4,5)P3 receptor binding pocket.