Prevention of basement membrane thickening in retinal capillaries by a novel inhibitor of aldose reductase, tolrestat.

A diabetic-like thickening of retinal capillary basement membranes induced in rats fed for 207 consecutive days a diet containing 50% galactose was prevented by the addition to the diet of tolrestat, a potent, structurally novel inhibitor of aldose reductase. Analysis of electron micrographs (X 25,000) of capillaries from the outer plexiform layer of the retina by computer planimetry showed that the basement membranes were approximately twofold thicker in rats fed galactose than in those fed either a standard diet or a diet containing galactose and tolrestat in doses of 43 or 57 mg/kg/day. The thickening of basement membranes in galactose-fed rats was accompanied by other ultrastructural alterations mimicking changes typical of diabetic microangiopathy, such as multilamination and the formation of vacuoles and dense inclusions. Therefore, the galactosemic rat represents a useful model for studying basement membrane-related complications of diabetes and their possible prevention by aldose reductase inhibitors.

Propranolol dosage, plasma concentration, and beta blockade.

One hundred sixty-nine normal men received varying propranolol dosage regimens and placebo. Dose level and frequency were compared with plasma propranolol levels and beta blockade, as assessed by reduction of exercise tachycardia. Propranolol levels above 20 ng/ml induced significant beta blockade. An average daily propranolol dose slightly in excess of 160 mg led to a minimum plasma level above 20 ng/ml. Approximately 50% of subjects achieved 20 bpm or greater decrease in exercise tachycardia with 160 mg per day. The degree of beta blockade at the daily minimum propranolol level was related to dose and not dose frequency. The relation of propranolol dose and plasma levels to beta blockade in normal subjects appears to reflect observations in large clinical trials.

Tolrestat kinetics.

The kinetics of tolrestat, a potent inhibitor of aldose reductase, were examined. Serum concentrations of tolrestat and of total 14C were measured after dosing normal subjects and subjects with diabetes with 14C-labeled tolrestat. In normal subjects, tolrestat was rapidly absorbed and disappearance from serum was biphasic. Distribution and elimination t 1/2s were approximately 2 and 10 to 12 hr, respectively, after single and multiple doses. Unchanged tolrestat accounted for the major portion of 14C in serum. Radioactivity was rapidly and completely excreted in urine and feces in an approximate ratio of 2:1. Findings were much the same in subjects with diabetes. In normal subjects, the kinetics of oral tolrestat were independent of dose in the 10 to 800 mg range. Repetitive dosing did not result in unexpected cumulation. Tolrestat was more than 99% bound to serum protein; it did not compete with warfarin for binding sites but was displaced to some extent by high concentrations of tolbutamide or salicylate.

Dynamics of propranolol dosing schedules.

Kinetic and dynamic data from 27 healthy male subjects were evaluated in a double-blind, randomized, double-crossover study to test the hypothesis that 180 mg/day propranolol twice and three times a day would provide much the same plasma levels and beta 1-blockade. The data indicate that propranolol twice rather than three times a day should be favored when beta 1-blockade is needed in therapy. The dynamic efficacy of the two schedules was the same and maximum concentration, AUC0-24, and 0-hr plasma propranolol values were higher after twice-than after three-times-daily dosing. The degree of beta 1-blockade (reduction in exercise tachycardia) was much the same on both dosing regimens at trough concentrations. These data indicate that both twice- and three-times-a-day dosing schedules provide well-sustained 24-hr beta-blockade and are probably interchangeable for therapeutic purposes.

Effect of tolrestat on red blood cell sorbitol levels in patients with diabetes.

The effect of the aldose reductase inhibitor, tolrestat, on red blood cell (RBC) sorbitol levels was studied in 23 patients with diabetes after oral dosing with tolrestat, 25 or 100 mg b.i.d. The mean (+/- SE) RBC sorbitol levels (measured 12 hours after the preceding dose) after 3, 7, and 13 days of dosing decreased after both dose levels. After 25 mg tolrestat the RBC sorbitol levels fell from 25.1 +/- 4.0 to 20.0 +/- 5.7 nmol/gm hemoglobin (21%) and after 100 mg tolrestat the level fell from 26.7 +/- 3.7 to 11.4 +/- 1.7 nmol/gm hemoglobin (57%; P less than 0.001). This latter RBC sorbitol concentration is similar to levels in individuals without diabetes. At both dosage levels the maximum decrease in RBC sorbitol levels occurred after only 3 days of dosing. Tolrestat had no effect on plasma glucose or hemoglobin A1 concentrations. The overall mean plasma unbound drug concentration measured 12 hours after 100 mg tolrestat (11.7 +/- 3.0 ng/ml; 3.3 X 10(-8) mol/L) was similar to the median inhibitory level (3 X 10(-8) mol/L) of tolrestat for sorbitol accumulation in human RBCs incubated in a high-glucose medium. Our results demonstrate the systemic bioavailability of tolrestat and its aldose reductase inhibitory activity in erythrocytes of patients with diabetes.

Combined effects of clofibrate and diosgenin on cholesterol metabolism in rats.

Groups of male rats were fed various doses of clofibrate and diosgenin, both alone and in combination for 1 week. Clofibrate suppressed the diosgenin-induced increase in hepatic cholesterol synthesis but did not alter the effectiveness of diosgenin in reducing cholesterol absorption. Diosgenin did not affect the bioavailability of CPIB. Clofibrate reduced the diosgenin induced increase in biliary levels of cholesterol; none of the regimens altered biliary bile acids. The combination produced greater decreases in LDL cholesterol than did either compound alone; the diosgenin-induced elevation in HDL cholesterol was partially reversed by clofibrate. The data provide a basis for the combined use of clofibrate and diosgenin in the control of hyperlipoproteinemia.

Studies on the disposition of diosgenin in rats, dogs, monkeys and man.

Rats, dogs and squirrel monkeys were given a single oral dose of [4-(14)C]diosgenin. Virtually all of the radioactivity was excreted in the feces. All of the absorbed radioactivity was eliminated via the bile. The percent of dose absorbed decreased with increasing dose. The amount of radioactivity in livers of rats given [4-(14)C]diosgenin was less than that after [4-(14)C]cholesterol, but more than after [4-(14)C]beta-sitosterol. Absorbed radioactivity in rats distributed into tissues, most notably the liver, adrenals, and walls of the gastrointestinal tract. No serum diosgenin was detected after a single large dose to rats and dogs. After multiple doses (100 mg/kg/day for 4 weeks) of diosgenin to dogs, up to 15 micrograms/ml of unchanged diosgenin was found in serum. Serum from human subjects receiving 3 g/day of diosgenin for 4 weeks contained less than 1 microgram/ml of unchanged drug. After a single dose of [14C]diosgenin, several metabolites were detected in the bile of rats and dogs; the pattern of metabolites was dissimilar in the two species. No diosgenin or 7-hydroxydiosgenin was found. One of the major biliary metabolites was diosgenin monohydroxylated in the F ring, but the location of the hydroxyl group was different in the two species. Although rat caecal contents were capable of reducing diosgenin to smilagenin in vitro, no smilagenin was present in the feces of rats given chow supplemented with diosgenin. It was concluded that diosgenin is poorly absorbed in the species tested, and that the amount which is absorbed undergoes extensive biotransformation.

Effect of AY-25,712 on fatty acid metabolism in rats.

The effect of AY-25,712 [2-methyl-2-phenyl-3(2H)-furanone-5-carboxylic acid] on various aspects of free fatty acid (FFA) and triglyceride metabolism was studied in male rats. Serum triglycerides were lowered by a single oral dose of AY-25,712 or nicotinic acid, but not of clofibrate. Unlike with clofibrate, when AY-25,712 or nicotinic acid was given in the diet, serum triglycerides were not affected. In vitro, both AY-25,712 and nicotinic acid suppressed the theophylline-induced FFA release by epididymal fat pads, but had no effect on lipolysis induced by norepinephrine. Both AY-25,712 and nicotinic acid enhanced the activity of adipose tissue lipoprotein lipase. The initial decrease in plasma FFA and triglycerides, and in liver triglycerides after a single oral dose of nicotinic acid was followed by a rebound to levels which, at later time intervals, wee significantly higher than in controls. AY-25,712 was more potent than nicotinic acid in lowering plasma FFA and triglycerides as well as liver triglycerides, but produced no such rebound effect. The data show that, except for the absence of this rebound effect, the mode of action of AY-25,712 in rats resembles that of nicotinic acid and differs from that of clofibrate.

Evaluation of the lipid-lowering activity of AY-25,712 in rats.

The effect of AY-25,712 (2-methyl-2-phenyl-3(2H)-furanone-5-carboxylic acid) on serum lipids, hepatic lipogenesis and biliary cholesterol was investigated in male rats. Based on one-week treatment, the minimal effective dose of AY-25,712 which lowered serum triglycerides was 1 mg/kg/day, and LDL-cholesterol, 5 mg/kg/day. Nicotinic acid produced a similar lipid-lowering profile albeit at 5 times higher doses. AY-25,712 at doses of 2 mg/kg/day and higher significantly increased the ratio of HDL to total cholesterol. Unlike clofibrate, AY-25,712 did not increase liver weight or liver mitochondrial alpha-glycerophosphate dehydrogenase, nor increase biliary cholesterol levels in rats fed a diet containing 2% cholesterol and 0.5% cholic acid. AY-25,712 lowered serum cholesterol, triglycerides and phospholipids in rats rendered hyperlipidemic with Triton WR-1339 and decreased the elevated serum triglycerides in streptozotocin-diabetic rats. [14C]Acetate incorporation into cholesterol by liver homogenate was suppressed in rats given AY-25,712 p.o. for 1 week. The results show that AY-25,712 is a potent LDL-cholesterol- and triglyceride-lowering agent in rats, and that its lipid-lowering profile differs from that of clofibrate but resembles that of nicotinic acid.

The effects of a new aldose reductase inhibitor (tolrestat) in galactosemic and diabetic rats.

Tolrestat(N-[[5-(trifluoromethyl)-6-methoxy-1-naphthalenyl] thioxomethyl]-N-methylglycine; AY-27,773; Alredase) is a potent, structurally novel inhibitor of aldose reductase (AR). In vitro, tolrestat inhibited in dose-dependent fashion the AR from bovine lenses (IC50, 3.5 X 10(-8) mol/L) and the formation of sorbitol in human RBC incubated with glucose (IC50, 3 X 10(-8) mol/L). Upon administration with the diet to rats made galactosemic or diabetic, tolrestat decreased, in a dose-related fashion, the accumulation of galactitol or sorbitol in the sciatic nerve and lens. The effectiveness of tolrestat depended upon the experimental conditions and tended to be higher in less severe galactosemia and after suitable pretreatment, particularly in galactosemic rats, resulting in ID50 of 5 mg/kg/d in the sciatic nerve and 12-15 mg/kg/d in the lens. Tolrestat also decreased, in dose-related manner, the RBC sorbitol levels in normal and in streptozotocin diabetic rats; in the latter, at less than 2 mg/kg/d, the RBC sorbitol was reduced to control levels.

Comparative bioavailability of propranolol: twice-daily versus four times-daily administration.

Nineteen healthy male volunteers were given daily 160 mg propranolol hydrochloride in divided doses, either four 40-mg tablets or two 80-mg tablets, and the plasma propranolol concentration profiles were compared after one and seven consecutive days of drug administration. The results indicate that the relative rate and extent of propranolol absorption were greater after 80 mg given twice daily than after 40 mg given four times per day. Both differences were statistically significant at steady state attained with the seven-day treatment. The variability in the areas under the concentration-time curves of propranolol appeared to be smaller after the 80-mg twice-a-day dosing schedule. The results are in accordance with the observed therapeutic equivalence of the two dosing regimens.

A gas-liquid chromatographic method for the determination of butriptyline in serum.

1. A gas-liquid chromatographic (GC) method for the analysis of butriptyline in serum has been development. Quantitation is based on the peak height ratio between butriptyline and promazine used as internal standard. A triple partition provides a "clean" extract. A detection limit of 10 ng/ml is achieved. 2. The usefulness of the method has been demonstrated in bioavailability studies in dogs.

A gas-liquid chromatographic determination of p-chlorophenoxyisobutyric acid and its comparison to an ultra violet method.

1. A gas-liquid chromatographic procedure for the determination of p-chlorophenoxyisobutyric acid (CPIB) has been elaborated and compared to the UV spectrophotometric procedure of Barrett and Thorp. 2. Both methods are specific when used to determine CPIB levels in normal sera from laboratory animals or humans treated with clofibrate (Atromid-S). Serum from patients treated with other drugs or abnormal sera from patients affected with a variety of diseases will often contain high and fluctuating levels of non-specific UV absorbing substances and this usually precludes the use of the UV procedure. 3. The GLC method is also applicable to the analysis of CPIB in urine samples.

Etodolac: effect on prostaglandin concentrations in gastric mucosa of rats.

Etodolac is a structurally novel compound exhibiting potent analgesic and anti-inflammatory activity in laboratory animals and man, with excellent G. I. tolerance. Like other nonsteroidal anti-inflammatory drugs (NSAIDs) etodolac inhibits prostaglandin (PG) biosynthesis. In view of the cytoprotective role of PGE2, we have investigated in normal rats the effect of etodolac on the gastric mucosal concentration of PGE2 as well as of 6-keto-PGF1 alpha, the stable metabolite of prostacyclin; naproxen and piroxicam served as reference NSAIDs. The orally effective anti-inflammatory doses in the chronic arthritic rat model (3 mg/kg for etodolac and naproxen; 0.5 mg/kg for piroxicam), and their arbitrarily selected multiples of 10 were used. Rats were killed at 1, 2, 6 and 24 hr after single doses and the PG concentrations were measured by RIA. With the low dose, 2 and 6 hr after dosing, etodolac diminished the PGE2 concentration by 20-25% (vs control) while naproxen and piroxicam caused a fall of 53-65%; the difference between etodolac and the untreated control group is not statistically significant but the difference between etodolac and both piroxicam and naproxen is significant (p less than 0.001). At the high doses, the lowering in PGE2 was similar after all three drugs, i.e. about 70% at 1 and 2 hr; 50% at 6 hr, and 20-50% at 24 hr after dosing. Except for the consistently smaller reduction of concentrations after etodolac, the effects on 6-keto-PGF1 alpha concentration followed a similar pattern but the differences are not significant. The lack of the G.I. irritation of etodolac in rats and man at therapeutically effective doses may be attributed to the benefits of the relatively short-lived and slight decrease in gastric mucosal PGE2 concentrations found in this study.

Aldose reductase inhibitors as pathobiochemical probes.

In tissues susceptible to damage from chronic diabetes, excess glucose is metabolized by aldose reductase (AR) to sorbitol. Originally, AR-catalyzed sorbitol formation (and accumulation) was found in the diabetic lens; the cataractogenicity of this process was proven by preventing cataract formation with an AR inhibitor (ARI). These findings were extended to the hypothesis that, in diabetic tissues, excessive intracellular sorbitol formation initiates a cascade of metabolic abnormalities which gradually progress to loss of functional and structural integrity. The pivotal role of AR as a trigger for such abnormalities was established by preventing their occurrence in diabetic animals treated with an ARI. By inference, this led to the concept that inhibition of AR should prevent, arrest, and, possibly, reverse the development of late diabetic sequelae. In addition to motivating drug-oriented research, the ARI concept provided a rationale for the use of ARIs as experimental tools to probe the pathogenesis of diabetic complications. By helping to elucidate the metabolic, functional, and structural ramifications of the AR-catalyzed disposal of excess glucose in diabetic schemes, and in addition, by helping to define the applicability of animal models for the study of early functional pathogenic alterations occurring in diabetic subjects, ARIs may enable the discrimination in diabetic tissue of arrestible and reversible from the irreversible abnormalities.

Aminoguanidine does not inhibit aldose reductase activity in galactose-fed rats.

Aminoguanidine, nucleophilic hydrazine derivative, has been shown to inhibit diamine oxidase, the formation of advanced glycation endproducts, nitric oxide synthase, and catalase. Prompted by the reports that aminoguanidine also inhibits aldose reductase (AR), we have investigated the effect of aminoguanidine, 1,3-diaminoguanidine, and methylguanidine on AR activity in vitro, and in vivo. In vitro, we have measured the inhibition of AR isolated from bovine lenses; in vivo, we have examined the effect on the galactitol levels in the red blood cells, sciatic nerve, retina, and lens of rats administered the test compounds for 11 days in the drinking water and, for the last 4 days, given access to a 20% galactose diet. Two known, structurally distinct AR inhibitors, tolrestat and compound WAY-121,509, were used as reference. In vitro, at concentrations up to 1.0 mmol/L, none of the tested guanidine derivatives had any effect on AR. As a corollary, in vivo, at doses ranging from 201 to 349 mg/kg/day, none of the guanidine derivatives affected tissular galactitol levels. We conclude that, in short-term galactose-fed rats, at the doses tested, aminoguanidine, 1,3-diaminoguanidine, and methylguanidine do not inhibit AR.

Tolrestat pharmacokinetics in rat peripheral nerve.

The clinical efficacy of an aldose reductase (AR) inhibitor in diabetic polyneuropathy depends on its bioavailability at the site(s) of AR in peripheral nerves. Accordingly, the link between the concentration of the AR inhibitor, tolrestat, and the extent of its inhibition of the AR-catalyzed polyol production was investigated in sciatic nerves of galactosemic rats. Tolrestat was administered by gavage (1 x 150 mg/kg, or 5, and 15 mg/kg/day for 15 days to attain steady state as estimated from the 53-h half-life of tolrestat determined in rat nerve); subsequently, at six time intervals, ranging from 4 to 59 days, rats were given access for 4 days to a 20% galactose diet, and killed. At every time point, the composite tolrestat concentration in the nerve correlated with the percentage decrease in nerve galactitol (r = 0.857, p = 0.0015). Because the latter should reflect the extent of nerve AR inhibition by tolrestat, the concentration of "free" tolrestat available at the site(s) of AR in the nerve was estimated from the tolrestat concentration/percent AR inhibition plot obtained in vitro. The estimated amount of tolrestat present at the site(s) of nerve AR represented 0.4% of the composite tolrestat concentration measured in the nerve. The results support the view that the effectiveness of an AR inhibitor in peripheral nerve depends on its pharmacokinetics in the nerve, i.e., on its uptake, nonspecific binding to cellular constituents, and elimination.

Agents affecting lipid metabolism. XXVI. Specificity of some inhibitors of the late stages of cholesterol biosynthesis.

The capacity of 22,25-DAC, AY-9944 and triparanol to inhibit cholesterol biosynthesis from three precursors, mevalonate, 7-dehydrocholesterol and desmosterol, has been studied in rat liver homogenates.Evidence is presented that, in vitro, 22,25-DAC, a potent inhibitor of the sterol Delta(24)-reductase, also inhibits the 7-dehydrocholesterol-Delta(7)-reductase system.

Effect of tibric acid on hepatic cholesterol synthesis in rats.

Male albino rats were administered various oral doses of tibric acid daily for 1 week. Serum cholesterol and triglyceride levels were reduced, but total liver content of cholesterol, phospholipids, and triglycerides was increased. Tibric acid treatment suppressed the incorporation of both [14C] acetate and [3H] mevalonate into cholesterol by liver homogenates.

Absorption and excretion of isosorbide dinitrate and isosorbide-2-mononitrate in dogs.

In a crossover design four male dogs were given orally or i.v. [14C]isosorbide dinitrate (ISDN) or [14C]isosorbide-2-mononitrate (2-ISMN) at a dose of 1 mg kg-1 (70-80 microCi). Virtually all of the oral dose was absorbed and all of the radioactivity was excreted in the urine. The profile of serum radioactivity was similar after all drug administrations. ISDN was rapidly denitrated, giving rise to isosorbide-5-mononitrate (5-ISMN) as a major metabolite, and 2-ISMN as a minor metabolite. The apparent elimination half-life from serum of 2-ISMN and 5-ISMN was 2-3 h. More than 50% of the serum radioactivity after [14C]2-ISMN was due to unchanged drug. The apparent volume of distribution of 2-ISMN averaged 8.3 litres. The results show that, in contrast to ISDN, administration of 2-ISMN resulted in relatively high unchanged drug levels in the serum; the disposition of radioactivity after [14C]ISMN was however similar to that after [14C]ISDN. The findings support the concept that the concentrations of ISDN, 2-ISMN and 5-ISMN in the blood are inversely related to the rates of denitration, and that the vascular activity of the nitrates of isosorbide relates to the rates of their dinitration.