Arachidonic acid release: an in vitro alternative for dermal irritancy testing.

Validated in vitro alternatives are being utilized extensively for mutagenicity and ocular irritancy testing. However, validation of alternative assays for dermal irritancy is progressing more slowly. As the irritant response in human skin is mediated, at least in part, by eicosanoids derived from arachidonic acid, the effect of relatively pure anionic surfactants (AS, n=8) and surfactant-containing finished products (FP, n=25) on the release of [3H]arachidonic acid from a prelabelled murine fibroblast cell line (C3H-10T1/2 cells) in vitro was examined. Test substances were administered at various non-lethal concentrations, in triplicate, to 12- and 24-well plates containing preconfluent monolayers (80-90% confluence) of C3H-10T1/2 cells. Because it is impossible to test all concentrations of each test substance in a single assay, statistical techniques were developed to 'standardize' in vitro assay results. In each assay, radiolabel release due to a positive control was also measured, using 0.04, 0.05 and 0.06 mM concentrations of sodium dodecyl sulfate (SDS). Test substance releases were then transformed into 'SDS equivalent' responses, significantly reducing both inter- and intra-assay variability. A straight line was fitted to the test substance responses and compared with that for SDS to calculate the relative potency in vitro for individual AS and FP. Relative potencies correlated with in vivo responses, that is primary dermal irritation indices obtained in rabbits, with Spearman p=0.408 (P<0.03) for 32 tested agents, and p=0.976 (P<0.001) for the eight AS. Exclusion of extremely alkaline or acidic FP (pH>11 or <2, n=4) and those which were insoluble in the aqueous cell culture media at the 1% stock dilution (n=5), improved the overall in vivo-in vitro correlation significantly (p=0.683, P<0.001, n=23) and produced a significant correlation for FP alone (p=0.539, P<0.05, n=15). These results suggest that release of [3H]arachidonic acid from cultured skin cells represents a novel, mechanistically based in vitro screen for dermal irritancy testing.

A physiologically relevant pharmacokinetic model of xenobiotic percutaneous absorption utilizing the isolated perfused porcine skin flap.

A physiologic pharmacokinetic model describing percutaneous absorption of topically applied compounds in the isolated perfused porcine skin flap (IPPSF) is presented. As an extension of a previously reported hybrid physiologically relevant compartmental model of uptake of intra-arterially administered drug in the IPPSF, this percutaneous model should allow experimental results obtained from an in vitro preparation to serve as quantitative input to an in vivo pharmacokinetic system. Model parameters estimated from 8-10-h IPPSF experiments were able to predict 6-day in vivo radiolabel absorptions in pigs for topically applied benzoic acid, caffeine, malathion, parathion, DFP, testosterone, and progesterone. These results compare favorably with those obtained previously using a classical compartmental modeling approach.

Percutaneous absorption and excretion of xenobiotics after topical and intravenous administration to pigs.

Interspecies comparisons suggest that the weaning pig is a suitable surrogate for man in percutaneous absorption studies. Despite known anatomical and physiological similarities between porcine and human skin, very few investigations of percutaneous absorption phenomena have been conducted in pigs. This study examined radiolabel excretion patterns after intravenous (iv) and topical administration of six 14C-radiolabeled compounds in weanling Yorkshire sows. Radiolabel recovery from excrement collected over 6 days following iv doses in physiological saline (200 micrograms, 10 muCi) showed that malathion (M), parathion (P), caffeine (C), and benzoic acid (B) were primarily excreted into urine (greater than 80%), while greater fractions of testosterone (T, 72%) and progesterone (R, 35%) were fecally eliminated. Percutaneous absorption was determined from total urine and fecal excretion of radiolabel after topical application, corrected for incomplete excretion following iv administration. Topical doses in ethanol (200 micrograms, 10 muCi) were applied at a surface concentration of 40 micrograms cm-2 and penetrated in the following rank order (percentage dose): B (25.7%) greater than R (16.2%) greater than C (11.8%) greater than T (8.8%) greater than P (6.7%) greater than M (5.2%). Fecal clearances of radiolabel, expressed as a percentage of total excretion, were greater after topical administration for four of the six compounds (B, C, P, and T, p less than 0.05). Calculations based on urinary excretion alone underestimated percutaneous absorption determined from total excretion by 5-30%, although the difference between the two estimates was statistically significant only for C (p less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)

The isolated perfused porcine skin flap. III. Percutaneous absorption pharmacokinetics of organophosphates, steroids, benzoic acid, and caffeine.

The isolated perfused porcine skin flap (IPPSF) has been developed as an alternative in vitro tool for examining the pharmacokinetics and mechanisms of percutaneous absorption. In this study, dosing solutions of seven 14C-radiolabeled compounds representing three chemical classes--organic acid/base [benzoic acid (B), caffeine (C)], organophosphate (OP) pesticides (diisopropylfluorophosphidate, malathion, parathion), and steroid hormones (progesterone, testosterone)--were prepared in ethanol and applied topically at a surface concentration of 40 micrograms cm-2 to the IPPSF. A three-compartment pharmacokinetic model used to stimulate mass transfer from the surface (C1), diffusion through epidermis and dermis (C2), and transfer into the capillary perfusate (C3), was developed based on flux through the IPPSF from 0 to 8 hr. This basic model accurately stimulated measured IPPSF fluxes for five of seven compounds, including the OPs and steroids. The model was modified to simulate the shunting of drug to fast and slow release pathways, which occurred for B 3-4 hr postapplication, and to account for flow-dependent flux increases seen for C at 6 hr postapplication. The latter may be due to a direct pharmacologic effect, since C is a known vasodilator. Extrapolated (to 6 days) areas under the curve from the model simulations were compared with in vivo percutaneous absorption estimates, obtained from 6-day excretion studies in pigs. The in vivo-in vitro correlation, based on simple linear regression across compounds, was excellent (R2 = 0.88, R = 0.94, p less than 0.002). These results suggest that xenobiotic penetration in the 8-hr IPPSF experiments is highly predictive of in vivo absorption totals (6-day studies). In addition, since pig and human skin are similar physiologically and pharmacologically, the IPPSF may eventually have applications in formulating human dermal risk assessment models.

The isolated perfused porcine skin flap (IPPSF). I. A novel in vitro model for percutaneous absorption and cutaneous toxicology studies.

This article describes the development of a novel in vitro alternative animal model for dermatology and cutaneous toxicology. A single-pedicle, axial-pattern, island-tubed skin flap was created in crossbred Yorkshire weanling pigs in one surgical procedure, then transferred 2 or 6 days later to a computer-controlled temperature-regulated perfusion chamber for 10-to 12-hr studies. Perfusate consisted of Krebs-Ringer bicarbonate buffer (pH 7.4) containing albumin and glucose. Viability was assessed by glucose utilization, lactate production, an absence of significant concentrations of the intracellular enzyme lactate dehydrogenase in the perfusate, and light and electron microscopy. A mean lactate to glucose ratio of 1.6 for flaps harvested 2 days after surgery and 1.8 for flaps taken 6 days after surgery suggested primarily anaerobic glycolysis. This preparation would be a humane alternative animal model for studies in cutaneous toxicology, physiology, oncology, and percutaneous drug absorption and metabolism.

Identification of a subgroup of Sprague-Dawley rats highly sensitive to drug-induced renal toxicity.

Laboratory rats available from breeding facilities are usually assumed to be homogeneous populations within each strain; however, previous studies in our laboratory suggested that there may be a subgroup of Sprague-Dawley rats which are highly sensitive to aminoglycoside nephrotoxicity. The present study clearly identifies a subpopulation of Sprague-Dawley rats which was highly sensitive to nephrotoxicity from supratherapeutic doses (75 mg kg-1 day-1) of the aminoglycoside antibiotic gentamicin. Gentamicin was administered subcutaneously in a divided regimen, 25 mg/kg every 8 hr, for 7 days. Statistical analysis of post-treatment serum creatinine (SCR) and urea nitrogen (SUN) concentrations demonstrated two distinct populations: normally responding rats (SCR = 1.92 +/- 0.54 mg/dl, SUN = 71.5 +/- 18.4 mg/dl, N = 87) and highly sensitive rats (SCR = 4.10 +/- 0.83 mg/dl, SUN = 146.4 +/- 24.9 mg/dl, N = 12) (mean +/- SD). Comparison of predosing blood and serum chemistries between these two populations revealed statistical differences only in initial serum osmolality, oxygen tension, and total protein. Since there is a subpopulation of humans which are at risk for developing aminoglycoside nephrotoxicity due to unknown host factors, these highly sensitive Sprague-Dawley rats may provide an animal model for investigating this human clinical problem.

Exaggerated response to gentamicin-induced nephrotoxicity in Sprague-Dawley rats: identification of a highly sensitive outlier population.

A number of factors have been shown to predispose patients treated with aminoglycosides to nephrotoxicity. In a previous study in our laboratory investigating the interaction of prior renal dysfunction with gentamicin toxicokinetics, 9.4% of rats in all treatment groups were relatively more sensitive to gentamicin-induced nephrotoxicity. To determine if these outliers had an underlying disease or physiological abnormality, serum was collected from 99 Sprague-Dawley rats prior to daily treatment with 75 mg/kg gentamicin for seven days. Urea nitrogen, creatinine, Na, K, Ca, Mg, P, total protein, albumin, aspartate transaminase, serum osmolality, total white and red blood cell count, hematocrit, hemoglobin, blood gases, and thyroxine were measured. Blood was collected one and four hours after the first dose of gentamicin for pharmacokinetic analysis. Elevations in post-treatment creatinine and nitrogen were significantly greater in the outliers (4.10 +/- 0.24 mg/dl (n = 12) vs 1.92 +/- 0.06 mg/dl (n = 87) and 146.4 +/- 7.2 mg/dl (n = 12) vs 71.5 +/- 2.0 mg/dl (n = 87); both p = 0.0001) and served as criteria for identifying this subgroup. Post-treatment creatinine and urea nitrogen were not normally distributed in the entire study population. However, when the population was divided into normal and sensitive subgroups, both subgroup values were normally distributed. The gentamicin pharmacokinetic profiles were similar in both groups. Postmortem histopathology showed significant increases in tubular casts and tubular necrosis (p = 0.01) in the sensitive rats, compared to the normally responding subgroup.(ABSTRACT TRUNCATED AT 250 WORDS)

Dose-response studies of gentamicin nephrotoxicity in rats with experimental renal dysfunction. I. Subtotal surgical nephrectomy.

Gentamicin pharmacokinetics and nephrotoxicity have not been widely studied in animals with preexisting renal dysfunction, despite the fact that nephrotoxicity is a continuing manifestation of clinical therapy. The present study contrasted the dose-response nephrotoxicity of gentamicin in control rats with that of rats with renal insufficiency secondary to subtotal (2/3) surgical nephrectomy. Total daily doses ranging from 0 to 120 mg/kg were given in a divided regimen, every 8 hr and doses were reduced by doubling the interval in subtotally nephrectomized (Nx) rats, in proportion to impaired renal elimination on the first day of gentamicin administration. Estimates of renal function, including creatinine clearance, fractional sodium and potassium excretion, and serum creatinine and urea nitrogen, were collected after 6 and 12 days of dosing. In addition, urinary N-acetyl-beta-D-glucosaminidase excretion (6 days), in vitro renal cortical slice accumulation of tetraethylammonium (TEA) (6 days), quantified morphological lesions (12 days), and renal gentamicin concentrations (6 days) were examined. Pharmacokinetic data collected immediately after the first dose revealed a reduced gentamicin clearance and slightly reduced volume of distribution, with a corresponding prolonged half-life in the Nx rats. Based on statistical analysis of the dose-response relationships, Nx rats were functionally resistant to gentamicin nephrotoxicity after 6 days of dosing. This resistance was partially reversed by 12 days dosing, despite light-microscopic evidence of greater structural damage in the control rats. Renal parenchymal gentamicin concentrations were lower at some doses in the Nx rats, in contrast to the higher fractional reabsorption found in these rats at all doses. TEA transport was depressed at all doses in control rats but not at the lower doses in Nx rats, indicating that resistance was partially mediated at the level of the proximal tubular epithelium. This study demonstrates altered gentamicin pharmacokinetics and nephrotoxicity in a surgical model of renal dysfunction in rats.

Dose-response studies of gentamicin nephrotoxicity in rats with experimental renal dysfunction. II. Polyvinyl alcohol glomerulopathy.

Animal studies involving concurrent pathophysiologic states, including experimental renal dysfunction, are useful for a proper understanding of the mechanisms of gentamicin nephrotoxicity and acute renal failure. This study examined gentamicin nephrotoxicity in a model of glomerular dysfunction in rats. Administration of medium molecular weight polyvinyl alcohol (PVA) produced a glomerulopathy, with characteristic accumulation of macromolecular PVA in the glomerular mesangium without altering glomerular filtration or causing proteinuria. Subsequent daily doses of gentamicin ranging from 0 to 120 mg/kg elicited a dose-response nephrotoxicity in both control and PVA-pretreated rats after 6 or 12 days of drug. Based on statistical analysis of renal clearances of creatinine, urea, sodium, and potassium; serum creatinine and urea nitrogen; urinary N-acetyl-beta-D-glucosaminidase excretion (6 days only); in vitro renal cortical transport of tetraethylammonium (TEA) (6 days); and quantified light-microscopic data (12 days), PVA induced an early (6 days) sensitivity to gentamicin nephrotoxicity. By 12 days, there were no differences in the responses of control and PVA rats to gentamicin. Single-dose gentamicin clearance, volume of distribution, and half-life were not altered by PVA and renal concentrations at 6 days were generally lower in these rats. Results of TEA transport studies tend to rule out PVA-induced metabolic lesions in the proximal tubular epithelium as the mechanism for the early sensitivity. This investigation demonstrates altered gentamicin nephrotoxicity in rats with an otherwise benign glomerulopathy and, combined with similar conclusions from a related study in subtotally nephrectomized rats, presents further evidence that the underlying pathophysiologic state of the kidney is an important factor in the renal response to nephrotoxins.

Pharmacokinetics and comparative nephrotoxicity of fixed-dose versus fixed-interval reduction of gentamicin dosage in subtotal nephrectomized dogs.

There is presently no consensus as to the relative safety of fixed-interval/reduced dose (FI) vs fixed-dose/increased interval (FD) dosage adjustment regimens for use in renal insufficiency. This study compared their nephrotoxic potential using gentamicin in beagle dogs with renal insufficiency secondary to subtotal surgical nephrectomy. Pharmacokinetic analysis in six dogs showed that this surgical procedure resulted in a decreased total body clearance of drug and a marginally contracted volume of the central compartment. An allometric analysis of gentamicin disposition in different species was used to derive a human-equivalent maximum canine nontoxic dose of 9 mg kg-1 day-1. Nephrotoxicity was detected by histopathologic analysis and changes in the pre- and post-drug treatment, creatinine clearance, and daily drug elimination rate constants. This allometric dose did not produce clinical toxicity in a control group of six dogs with intact kidneys given drug for 14 days. Dosage adjustments within the FI and FD groups were based on serum creatinine concentrations 10 days after surgery. Statistical analysis of morphological and functional parameters indicated that the FD method was significantly less toxic than the FI regimen.

Effects of familial hypothyroidism and subtotal surgical nephrectomy on gentamicin pharmacokinetics in beagle dogs.

Gentamicin serum pharmacokinetics were studied in normal, subtotally nephrectomized, and familial hypothyroid beagle dogs. Length of the gamma phase was not affected by subtotal nephrectomy but was shorter or almost nonexistent in hypothyroid dogs.