Absorption and metabolism of (all-trans-) retinol acetate, retinol, and retinoic acid in the single-pass, albumin-perfused rabbit ear.

BACKGROUND: Retinoic acid, a metabolite of retinol, is formed in the skin of various species. PURPOSE: Formation rates have not been determined in a dynamic skin perfusion model which may show the dehydrogenation of retinol to be rate limited. METHODS: (all-trans-) Retinol acetate, retinol, or retinoic acid was applied dermally with isopropyl myristate. The ears were single-pass perfused with 6% hetastarch or 5% bovine serum albumin in the buffer solution. The effluent was analyzed by HPLC for these substances. RESULT: No ester hydrolysis of retinol acetate was observed, nor did this substrate appear in the effluent. Retinol or retinoic acid were detected in the effluent. The absorption rate was linearly related to the dermally applied concentration density. The absorption rate of retinoic acid was 70 times larger than that for retinol. The formation rate of retinoic acid from retinol was rate limited (apparent Vmax: 0.08 pmol/min/cm2 skin). CONCLUSION: Since retinoic acid is a recognized pharmacologically active ingredient in the skin, the dehydrogenation of retinol or retinal was also observed in the effluent of this skin perfusion model. The percutaneous retinol adds to the endogenous pool of retinoic acid by cytosolic dehydrogenation or microsomal oxidation elsewhere. The steady-state absorption rate of 0.1% retinoic acid in isopropyl myristate, referred to 500 cm2 of the human skin, would contribute about 20% of the daily requirement of vitamin A, without increase in endogenous retinoic acid plasma concentration.

Dehydrogenation of 3-phenoxybenzyl alcohol in isolated perfused rabbit skin, skin homogenate and purified dehydrogenases.

The formation of 3-phenoxybenzoic acid from 3-phenoxybenzyl alcohol was determined in (a) rabbit ears, single-pass perfused with a protein-free buffer, pH 7.4; (b) the microsomal fraction and its supernatant from homogenized rabbit skin; and (c) purified alcohol dehydrogenase from horse liver and baker's yeast. The inhibition of product formation in (a) was about 60% by various 4-methylpyrazole concentrations, but metyrapone had no effect. Following ultracentrifugation, only the supernatant of homogenized skin showed product formation (apparent Vmay: 32 pmol/min per cm2 skin; apparent Km: 64 microM). 3-Phenoxybenzyl alcohol and ethanol dehydrogenation was similar by alcohol dehydrogenase from horse liver (apparent Km: 0.7 vs. 0.4 mM; apparent Vmax: 0.3 vs. 0.2 U/ microg protein). In baker's yeast, the apparent Km of 3-phenoxybenzoic acid formation was several times larger than that for ethanol dehydrogenation. The KI of 4-methylpyrazole for alcohol dehydrogenase from horse liver was 0.6 (3-phenoxybenzyl alcohol) vs. 0.04 microM (ethanol). The KI for ethanol in baker's yeast was 470 microM. In conclusion dehydrogenation is an important metabolic pathway in the skin for xenobiotics with an aliphatic alcohol at a side chain.

Permethrin absorption not detected in single-pass perfused rabbit ear, and absorption with oxidation of 3-phenoxybenzyl alcohol.

Isolated rabbit ears were single-pass perfused with a protein-free medium. Permethrin (0.05-23.5%, w/w) was applied in four distinct ointments. Permethrin, 3-phenoxybenzyl alcohol, 3-phenoxybenzaldehyde, and 3-phenoxybenzoic acid were analysed by HPLC. Permethrin was not detected in the effluent. The permeation coefficient, calculated from the detection limit was < 7.3 x 10(-12) (cm/sec). The appearance rate of the 3-phenoxybenzyl moieties in the effluent agreed with the absorption of the corresponding impurities in the various ointments. In supernatant of homogenised skin, the hydrolysis rate of permethrin was linear; about 4 pmol/min per cm2 at 10 microM substrate concentration. The proportion of 3-phenoxybenzoic acid, a further metabolite of 3-phenoxybenzyl alcohol increased when an oxidizing co-factor system was added. The appearance rate in the effusate of 3-phenoxybenzyl alcohol following the lipophobic ointment was five times faster than from isopropyl myristate. The formation rate of 3-phenoxybenzoic acid followed saturation kinetics. Occupational systemic poisoning by dermal absorption of permethrin seems very unlikely since humans bear more epithelial cell layers than rabbits. These experiments do not contradict, however, possible paraesthesia during systemic poisoning after inhalation or ingestion of the pyrethroid-containing aerosols used in agriculture.

No effect of albumin on the dermal absorption rate of hydrocortisone 21-butyrate, permethrin or diflunisal in the isolated, single-pass perfused rabbit ear.

Rabbit ears were single-pass perfused with a buffer solution containing either 6% hetastarch or 5% bovine serum albumin. Hydrocortisone 21-butyrate (5 mM), diflunisal (17 mM) or permethrin (33 mM) was added to isopropyl myristate with 5% polyethylene, and applied to about 40% of the epithelial surface area of the ear. Hydrocortisone 21-butyrate or permethrin were not found in the effluent with hetastarch or albumin. Following cutaneous ester hydrolysis, the appearance rate of hydrocortisone was about 4 pmol/min per cm2 in the hetastarch- or the albumin-containing buffer solution. No hydrolysis of permethrin was detected; the appearance rate of 3-phenoxybenzyl alcohol with 3-phenoxybenzoic acid corresponded to the absorption rate of the substrate impurities. During ex vivo perfusion of intact skin, serum albumin in the perfusion fluid may not enhance the appearance rate of xenobiotics in the effluent following dermal application when the distribution coefficient n-octanol/water is > 2,000 or when the xenobiotic is ionized at physiological pH. In general, for all substances investigated with our perfusion model thus far, the appearance rates decreased with rising distribution coefficient (n-octanol/buffer pH 7.4). High lipophilicity hinders the release from isopropyl myristate and the penetration through the skin.

Absorption and metabolism of hydrocortisone 21-butyrate, 21-hemisuccinate and hydrocortisone by skin of the rabbit ear during single-pass perfusion.

1. Rabbit ear was single-pass perfused with protein-free buffer solution at a rate of 0.02 ml/min per cm2 surface area for up to 6 h. 2. Hydrocortisone or hydrocortisone 21-butyrate or hydrocortisone 21-hemisuccinate was applied dermally (0.01-1.0% w/w) in suspension with isopropyl myristate or dissolved in 1,2-propanediol. The ointments were stiffened with 5% polyethylene 1500 or 1.2% methylcellulose respectively. 3. Hydrocortisone was slowly absorbed and did not reach a steady-state absorption rate during the experimental period. 4. No metabolites of hydrocortisone were found during ear perfusion. In the supernatant of skin homogenate, however, cortisone, hydrocortisone-20 alpha-ol and another, as yet unidentified metabolite, were observed. 5. Hydrocortisone 21-butyrate was completely hydrolysed during dermal absorption, showing steady-state concentrations of hydrocortisone in the effluent. Hydrocortisone 21-hemisuccinate did not reach steady-state concentrations in the effluent. About two-thirds was hydrolysed during absorption. 6. During arterial perfusion approximately 30% of hydrocortisone 21-butyrate was hydrolysed whereas 97% of hydrocortisone 21-hemisuccinate remained intact. 7. First-pass ester hydrolysis in skin may be complete for the poorly absorbed glucocorticoids leading to metabolites of less lipophilicity.

Metabolism of ethyl 4-aminobenzoate in the isolated single-pass perfused rabbit ear.

Isolated rabbit ears were single-pass perfused with a protein-free Tris buffer solution. Ethyl 4-amino[U-ring-14C]benzoate or 4-amino[U-ring-14C]benzoic acid was applied to the epithelial surface in isopropyl myristate, or perfused arterially. Metabolites in the effusate were extracted and separated by high-performance liquid chromatography. Five metabolites were separated from ethyl aminobenzoate and identified as aminobenzoic acid, or acid-labile conjugates from either substrate. Their chemical structure has not yet been established. One metabolite, perhaps a glycolic acid N-conjugate of ethyl 4-aminobenzoate, was found at about 100 times higher concentration after arterial than after dermal drug application, indicating that the quantitative metabolic patterns in skin differ depending on whether the xenobiotic reaches the enzymes of the skin by cutaneous absorption or by the blood circulation. The metabolite pattern of the skin was compared with that from rabbit or rat liver cell preparations.

Ester hydrolysis and conjugation reactions in intact skin and skin homogenate, and by liver esterase of rabbits.

1. Procaine, 2-chloroprocaine, ethyl aminobenzoate and methyl salicylate were added at various concentrations to liver esterase, supernatant of skin homogenate, or single-pass perfused ears of rabbits. 2. Vmax of product formation by purified liver esterase correlated with the rank order of the distribution coefficients (n-octanol/buffer) of the substrates and ranged between 11 and 1100 pmol/min per micrograms protein. Km values were between 20 and 50 microM. 3. No correlation was observed when the apparent enzyme kinetics, calculated by non-linear adaptation, were compared with each other after substrate administration to skin, arterial influx, or incubation with skin homogenate. 4. An acid labile conjugate of ethyl 4-aminobenzoate was found, mainly during arterial perfusion and in supernatant of skin homogenate, after administration of ethyl 4-aminobenzoate. 5. Acetamidobenzoic acid was observed in quantities of about 10% of the free 4-aminobenzoic acid during dermal or arterial application of procaine. This metabolite was not found with ethyl 4-aminobenzoate. 6. The results from isolated rabbit ear perfusion differ quantitatively and qualitatively with those obtained from supernatant of skin homogenate or purified liver esterase.

Dermal metabolism of 4-nitrophenol and 4-nitroanisole in single-pass perfused rabbit ears.

1. 4-Nitrophenol and 4-nitroanisole were applied either dermally or arterially to isolated rabbit ears perfused under single-pass conditions with protein-free buffer solution. 2. 4-Nitroanisole yielded only phase II metabolites of 4-nitrophenol. 3. The apparent Vmax values for 4-nitrophenol glucuronidation and sulphation were about 20 pmol and 10 pmol/min per cm2, respectively. 4. The difference in apparent Km between dermal and arterial drug application is a measure of first-pass metabolism by the epidermal layer. 5. The amount of 4-nitrophenyl conjugate detected after 4-nitroanisole administration was assumed to represent O-dealkylation of 4-nitroanisole; the capacity of this reaction was one order of magnitude lower than the direct conjugation of 4-nitrophenol.

Absorption and ester cleavage of methyl salicylate by skin of single-pass perfused rabbit ears.

1. Rabbit ears were single-pass perfused with a cell-free medium. Arterial pressure, oxygen consumption, and lactate production increased with flow rate. 2. Methyl salicylate was hydrolysed with an apparent Vmax of 1.5 nmol/min per cm2 surface area, a rate about 25 times greater than after arterial administration. 3. Estimation of the Km was not possible, due to oedema developing at arterial concentrations greater than 100 microM methyl salicylate. 4. The model appears suitable to compare absorption, as well as metabolic rates, of xenobiotics in rabbit skin.

Cutaneous absorption of indomethacin from two topical preparations in volunteers.

The percutaneous absorption of indomethacin in 0.5 % or 1 % solution or 1 % gel at a dose of 50 mg or 100 mg indomethacin was compared in a randomized complete block design in seven healthy volunteers. The formulations were applied over an area of 12 dm(2) under an 8 h occlusion dressing. In addition, in the same volunteers the plasma concentration curves were determined after a single oral dose of 50 mg indomethacin. Indomethacin and some of its metabolites were determined with modified, existing assays using HPLC-fluorescence or gas chromatography-mass spectrometry. On the basis of a newly developed method, it was possible to separate and quantify O-desmethylindomethacin and N-deschlorobenzoyl-O-desmethylin-domethacin. After cutaneous administration of the two drug formulations, peak indomethacin plasma concentration of 95 ng/ml and 130 ng/ml were found between 4 and 8 h; the cutaneous bioavailability was approximately 20 % of the oral dose, as judged by comparing the areas under the plasma concentration time curves (AUC) and the amount of metabolites excreted into the urine. Percutaneous absorption did not change the metabolic pattern in the urine that is obtained after oral administration.

Effects and metabolic pathway of 4-dimethylaminophenol during kidney perfusion.

1. Isolated rat kidneys were perfused (single pass) with 4 to 40 microM soln. of 4-dimethylamino[14C]phenol (DMAP). 2. Nephrotoxicity was not detected at concn. of 14C-DMAP up to 18 microM; higher conc. resulted in irreversible loss of physiological functions with simultaneous five-fold increase in covalently bound 14C. 3. At all concn., 85% of the post-renal 14C was due to unchanged DMAP, while 15% corresponded to DMAP conjugates. These conjugates were identified as DMAP-glucuronide (90% total), DMAP-sulphate and DMAP-thioethers. 4. All DMAP conjugates were conc. in the urine with urine/perfusate concn. ratios in the range 5-7 for the glucuronide, 50-100 for the sulphate, and 10-20 for the thioethers. 5. Rates of formation fo metabolites vs DMAP concn. followed Michaelis-Menten kinetics for DMAP-sulphate (Km 25 microM, Vmax 2.2 nmol/min per g) and DMAP-thioethers (Km range 10-100 microM, Vmax 4 nmol/min per g). DMAP-glucuronide formation rate increased linearly with DMAP concn. and showed a four-fold increase at toxic DMAP doses. 6. From the data for DMAP conjugation capacity, urine/perfusate concn. ratios of DMAP conjugates and microautoradiography, it is suggested that DMAP conjugation is located mainly in the proximal convoluted tubule, where deterioration of renal functions originates.

Serum concentrations of ampicillin and probenecid and ampicillin excretion after repeated oral administration of a pivampicillin-probenecid salt (MK-356).

Twenty male volunteers received oral doses (2100, 1050, and 525 mg) of a pivampicillin-probenecid salt in a 1 to 1 molar ratio (MK-356) at 12 hour intervals. After each dose peak serum concentrations of probenecid were observed 2 hours later than peak concentrations of ampicillin. Following the first dose of MK-356 the apparent elimination rate of ampicillin was dose-dependent and did not follow first order kinetics, as it showed a longer apparent half life after a higher dose. An equal dose of MK-356 administered 12 hours later caused an increase in the peak serum ampicillin level greater than expected from the concentration of ampicillin after the preceding dose. In twelve male volunteers who received at random 525 mg of MK-356 or 350 mg of pivampicillin, each three times daily for 4 days, the areas under the ampicillin concentration curve were the same after the first or last dose of either drug. When 2100 or 1050 mg of MK-356 was taken as an initial dose, 30 to 40 per cent of the ampicillin was recovered from urine in the ensuing 12 hours. The results indicate that when at least 400 mg probenecid was coadministered twice daily with 700 mg pivampicillin (MK-356), the peak serum concentrations of ampicillin were increased and its elimination rate slowed following successive doses.