Antibiotics and renal branching morphogenesis: comparison of toxicities.

BACKGROUND: Many premature born neonates receive antibiotic drugs to treat infections, which are applied during active nephrogenesis. We studied the impact of clinical concentrations of gentamicin and alternatives, ceftazidime and meropenem, on ureteric branching. METHODS: Mice metanephroi were dissected at embryonic day 13 and cultured in media with or without various concentrations of gentamicin, ceftazidime, or meropenem. Zero and 24 h kidney size were assessed by surface area measurements, and the ureteric tree was visualized by whole mount staining and confocal microscopy. Branching was evaluated by counting and gene expression levels of Wt1, Sox9, Bmp7, Fgf8, and Gdnf were investigated. RESULTS: A concentration of 2,000 µmol/l ceftazidime impaired ureteric development. In addition, a 4.5-fold and a 2.5-fold downregulation was noted in Fgf8 and Gdnf, respectively. No adverse effects were noted after gentamicin or meropenem treatment. No relationship was noted between surface area expansion and ureteric bud formation, but surface area at explantation related to bud count after 24 h of culture. CONCLUSION: Ceftazidime, but not gentamicin or meropenem reduced ureteric branching in mice and suggest a role for Fgf8 and Gdnf in its mechanism. Metanephros surface area measurements can be used to reduce intra- and inter-litter variation.

Evaluation by monte carlo simulation of the pharmacokinetics of two doses of meropenem administered intermittently or as a continuous infusion in healthy volunteers.

Meropenem is a broad-spectrum carbapenem antibacterial agent. In order to optimize levels in plasma relative to the MICs, the ideal dose level and dosage regimen need to be determined. The pharmacokinetics of meropenem were studied in two groups, each comprising eight healthy volunteers who received the following doses: 500 mg as an intravenous infusion over 30 min three times a day (t.i.d.) versus a 250-mg loading dose followed by a 1,500 mg continuous infusion over 24 h for group A and 1,000 mg as an intravenous infusion over 30 min t.i.d. versus a 500-mg loading dose followed by a 3,000-mg continuous infusion over 24 h for group B. Meropenem concentrations in plasma and urine were determined by liquid chromatography-mass spectrometry/mass spectrometry and high-performance liquid chromatography with UV detection, respectively. Pharmacokinetic calculations were done by use of a two-compartment open model, and the data were extrapolated by Monte Carlo simulations for 10,000 simulated subjects for pharmacodynamic evaluation. There were no significant differences in total clearance and renal clearance between group A and group B or between the intermittent treatment and the continuous infusion. The analyses of the probability of target attainment by MIC for the high- and low-dose continuous infusions were robust up to MICs of 4 mg/liter and 2 mg/liter, respectively. The corresponding values for intermittent infusions were only 0.5 mg/liter and 0.25 mg/liter. When these observations were correlated with MICs obtained from the MYSTIC database, intermittent infusion results in adequate activity against two of the most common nosocomially acquired pathogens, Klebsiella pneumoniae and Enterobacter cloacae. However, against Pseudomonas aeruginosa, the evaluation shows a clear advantage of high-dose therapy administered as a continuous infusion. We believe that in the empirical therapy situation, the continuous-infusion mode of administration is most worth the extra efforts. We conclude that clinical trials for evaluation of the continuous infusions of meropenem in critically ill patients are warranted.

[Drug interactions between nonsteroidal anti-inflammatory drug and pazufloxacin mesilate, a new quinolone antibacterial agent for intravenous use: convulsions in mice after intravenous or intracerebroventricular administration].

Convulsant activity of pazufloxacin mesilate (PZFX mesilate), a new quinolone antibacterial agent for intravenous use, in combination with nonsteroidal anti-inflammatory drug (NSAID) was investigated in mice after intravenous or intracerebroventricular administration. Following results were obtained. 1. In combination with 4-biphenylacetic acid (BPAA) at an oral dose of 100 mg/kg, PZFX mesilate did not induce any convulsions at intravenous doses up to 200 mg/kg. Reference quinolones induced convulsions at the following intravenous doses: Enoxacin (ENX), 3.13 mg/kg or more; norfloxacin (NFLX) and lomefloxacin (LFLX), 6.25 mg/kg or more; ciprofloxacin (CPFX), 50 mg/kg or more; sparfloxacin (SPFX) and temafloxacin (TMFX), 100 mg/kg or more; fleroxacin (FLRX), 200 mg/kg. 2. PZFX mesilate at an intravenous dose of 50 mg/kg did not induce convulsions in mice after oral administration of any of 14 kinds of NSAIDs. It induced convulsions at 200 mg/kg in combination with aspirin at an oral dose of 600 mg/kg, while it did not with the other 13 kinds of NSAIDs. 3. Convulsion-inducing dose of PZFX mesilate after intracerebroventricular administration was 100 mg/body, which was higher than those of reference quinolones (NFLX, CPFX, ENX, LFLX, TMFX, levofloxacin, ofloxacin, FLRX and SPFX) and beta-lactam antibiotics (penicillin G, cefazoline, imipenem/cilastatin and panipenem/betamipron). In addition, concurrent dosing of BPAA (1 microgram/body) did not reduce the convulsion-inducing dose of PZFX mesilate. These results suggest that PZFX mesilate has remarkably weak convulsant activity.

Low convulsive activity of a new carbapenem antibiotic, DK-35C, as compared with existing congeners.

Since carbapenems and cephalosporins have been suggested to induce convulsive side effects through an inhibitory action on the central gamma-aminobutyric acid (GABA)-mediated inhibitory transmission, the present study evaluated the convulsive activity of a new carbapenem antibiotic (1R,5S,6S)-6[(R)-1-hydroxyethyl]-2-[(3S,5S)-5(S-methyl-4thiomorpholin ylcarbonyl)pyrrolidin-3-thio]-l-methylcarbapen-2-em-3- carboxylic acid (DK-35C) in in vitro and in vivo experiments, in comparison with cefazolin, imipenem and meropenem. In in vitro experiments, their abilities to inhibit [3H]muscimol (5 nM) binding to GABA(A) receptors were measured using crude synaptic membranes prepared from the rat cerebral cortex. The concentrations (mM) of the antibiotics which inhibit 50% of the specific binding, were 0.6 for imipenem, 1.8 for cefazolin, 15.4 for DK-35C and 27.6 for meropenem. In in vivo experiments, intracerebroventricular (i.c.v.) injections of cefazolin, imipenem and DK-35C induced convulsions in a dose-dependent manner in rats. The doses (nmol/rat) of the antibiotics which induce convulsions in 50% of rats, were 57 for imipenem, 96 for cefazolin, 377 for DK-35C and >3000 for meropenem. In the mouse pentylenetetrazole (PTZ) convulsive model, intravenous pretreatment with cefazolin (800 mg/kg) or imipenem (200 mg/kg) shifted the dose-response curve of PTZ (i.p.) to the left, indicating enhancement of the convulsive activity of PTZ. However, pretreatment with cefazolin, meropenem or DK-35C at a dose of 400 mg/kg did not produce any marked effects on the convulsive activity of PTZ compared with the saline vehicle-pretreated control. The results clearly demonstrate a good correlation between in vitro GABA(A) receptor binding assay and in vivo i.c.v. convulsive model using rats, and suggest that DK-35C may possess a relatively weak convulsive activity mediated through an interaction with GABA(A) receptors.

Nephrotoxic and peroxidative potential of meropenem and imipenem/cilastatin in rat and human renal cortical slices and microsomes.

BACKGROUND: Carbapenems are a relatively new class of beta-lactam antibiotics characterized by a broad spectrum of antibacterial activity. Meropenem (MER), a new carbapenem has shown a lower nephrotoxic potential compared to imipenem (IMI). IMI is used in a fixed one-to-one combination with the nephroprotective agent cilastatin (CIL). The present studies examined whether MER and IMI/CIL produce peroxidative and nephrotoxic alterations including oxidative changes in rat and human renal cortical slices and microsomes. MATERIALS AND METHODS: Untreated slices and microsomes were incubated in vitro for various periods of time in phosphate-buffered media containing various concentrations of MER, IMI/CIL or for comparison cephaloridine (CPH). Lipid peroxidation was monitored by the determination of malondialdehyde (MDA) in incubation media and slices in the presence or absence of antioxidants. Total glutathione, oxidized glutathione (GSSG), pyruvate-stimulated gluconeogenesis and paraaminohippurate (PAH) accumulation were measured in slices. RESULTS: In rat renal cortical slices, MER, IMI/CIL and CPH induced a time- and concentration-dependent MDA production (content in incubation media plus slices). 5 mM MER, 5 mM IMI/CIL and 3 mM CPH were the lowest concentrations which caused a significant MDA production after 3 hs compared to control (control 61.5+/-15.3 nmol MDA/g tissue, MER 75.4+/-10.9, p<0.001; control 48.0+/-8.7, IMI/CIL 65.1+/-11.7, p<0.001; control 61.5+/-15.3, CPH 113.0+/-28.2, p<0.001). 20 mM MER, 20 mM IMI/CIL and 12 mM CPH revealed marked MDA production after 3 hs in human renal cortical slices (control 29.8+/-4.2 nmol MDA/g tissue, MER 49.4+/-8.7, p<0.01; control 27.6+/-7.0, IMI/CIL 68.3+/-9.9, p<0.001; control 32.5+/-7.7, CPH 93.8+/-31.6, p<0.001) and in human renal microsomes (control 1.0+/-0.9 nmol MDA/mg protein, MER 2.9+/-1.0, p<0.05; IMI/CIL 6.8+/-2.2, p<0.001; CPH 8.4+/-2.2, p<0.001), respectively. The corresponding MDA production was about 2-fold higher in rat renal cortical slices and almost the same in rat renal microsomes. Antioxidants reduced the MER-induced increase in MDA content in rat renal cortical slices by 48% (alpha-tocopherol, 10(-4) M), 72% ((+)-cyanidanol-3, 10(-5) M) and 100% (DPPD, N, N'-diphenyl-p-phenylendiamine, 10(-6) M). In rat renal cortical slices, MER and IMI/CIL induced an increase up to 50% in the content of GSSG and a corresponding %-decrease in reduced glutathione (GSH). In rat renal cortical slices, MER and IMI/CIL induced a time- and concentration-dependent decrease in PAH accumulation and gluconeogenesis. PAH accumulation was already reduced by 5 mM MER after 1 h (control slice to medium ratio 18.3+/-6.8, MER 10.7+/-1.9, p<0.05) and by 10 mM IMI/CIL after 3 h (control 16.9+/-5.6, IMI/CIL 5.5+/-1.3, p<0.001). Pyruvate-stimulated gluconeogenesis after 3 hs was already reduced by 2.5 mM MER (control 5.7+/-2.1 micromol glucose/g tissue/h, MER 3.9+/-1.1, p<0.05) and by 10 mM IMI/CIL (control 5.7+/-2.1, IMI/CIL 2.8+/-1.0, p<0.001). CONCLUSION: Thus, MER and IMI/CIL (at concentrations more than 10-fold higher as peak plasma concentrations achieved in humans) revealed an oxidative change (depletion of GSH, production of GSSG), a peroxidative potential (production of MDA) and a nephrotoxic potential (reduction in pyruvate-stimulated gluconeogenesis and PAH accumulation). Human kidney seems to be less susceptible to beta-lactam antibiotic-induced lipid peroxidation than rat kidney.

[Nephrotoxicity and drug interaction of vancomycin].

We examined drug interactions of vancomycin hydrochloride (VCM) in the rabbit kidney. VCM has an antibacterial action against Gram positive bacteria, but composite infection patients must be jointly treated with antibiotics that are effective on Gram negative bacteria, e.g., imipenem (IPM)-cilastatin sodium (CS) compounding agent. Both VCM and IPM have the adverse reaction of nephrotoxicity, whereas CS restrains the nephrotoxicity of IPM. To clarify the interactions, we examined the nephrotoxicity and pharmacokinetics of VCM in the rabbit and compared them with those in rabbits administered VCM with CS or IPM-CS. Symptoms of nephrotoxicity such as an increase of serum creatinine concentration and BUN and a morphological change of the kidney were observed with iv. injection of VCM at 300 mg/kg. However, no abnormality of clinical data and morphological alteration were observed in the groups injected with VCM plus CS or IPM-CS. Clearance and urinary excretion of VCM obviously increased in the groups injected with VCM plus CS or IPM-CS. In addition, it was estimated that VCM was actively transported by observation of the uptake in rabbit renal slices. Furthermore, the uptake rate of VCM in the renal cortex was significantly decreased by CS. Together with the above findings, it is suggested that the restraint effect of VCM uptake into nephrocytes by CS is one of the decreasing mechanisms of the nephrotoxic effect of VCM.

Structural features resulting in convulsive activity of carbapenem compounds: effect of C-2 side chain.

The neurotoxicity of meropenem was much lower than that of both imipenem and panipenem after intraventricular administration to mice. To clarify the major structural features responsible for the induction of convulsions by carbapenem antibiotics, the structure-activity relationship on convulsant activity was investigated in N-acetyl-2-pyrroline and cyclopentene derivatives which correspond to the 5-membered ring containing the C-2 side chain of carbapenem antibiotics. Among these derivatives, compounds with strong basicity in the side chain showed convulsant activity similar to that of the parent carbapenem compounds. In addition to the strength of the basicity of the amino group, the distance from the carboxyl to the amino group and steric crowding around the amino group also appeared to play an important role in the induction of convulsions. The results of gamma aminobutyric acid (GABAA) receptor binding assays indicated that the induction of convulsions was caused predominantly by the inhibition of GABAA-mediated inhibitory transmission. However, the in vivo convulsant activity of some of these compounds did not correlate with their in vitro inhibitory effect on GABAA receptor binding.

Correlation between in vitro and in vivo models of proconvulsive activity with the carbapenem antibiotics, biapenem, imipenem/cilastatin and meropenem.

The present study evaluated the proconvulsant liability of biapenem, a novel carbapenem antibiotic, in in vitro and in vivo experiments, in comparison with the carbapenems, imipenem/cilastatin and meropenem. Imipenem/cilastatin is a carbapenem antibiotic with known proconvulsive liability in man and in animal experiments. In in vivo studies imipenem/cilastatin, at doses of 400/400 mg/kg i.v., significantly lowered the convulsive threshold of pentylenetetrazol (PTZ) in mice and shifted the dose-response curve of PTZ. The effects of biapenem (400 mg/kg i.v.) and another reference carbapenem, meropenem (400 mg/kg i.v.), in the mouse PTZ model were not significantly different from control. In in vitro experiments the carbapenems were tested for their ability to inhibit [3H]muscimol (1.3 mM) binding to rat brain homogenates at concentrations of 1-10 mM. Similar to in vivo results, when compared to imipenem/cilastatin, biapenem and meropenem did not inhibit [3H]muscimol binding to the GABAA receptor complex in brain homogenates while imipenem/cilastatin exhibited significant inhibition (IC50 = 4.6 mM). These results further confirm the correlation between in vitro GABAA binding and in vivo PTZ convulsive testing with carbapenem antibiotics, and suggest that biapenem possesses a low proconvulsive liability.

Preventive effect of betamipron on nephrotoxicity and uptake of carbapenems in rabbit renal cortex.

The preventive effect of betamipron (N-benzoyl-3-propionic acid: BP) on the renal uptake and nephrotoxicity of carbapenems (panipenem and imipenem) was studied in rabbits. Panipenem, a new carbapenem antibiotic, induced nephrotoxicity at a dose of 200 mg/kg, i.v., but this was less severe than that caused by a single dose of imipenem or cephaloridine. Along with the significant reduction of nephrotoxicity, the uptake of these carbapenems in the renal cortex was remarkably inhibited by simultaneous treatment with BP (200 mg/kg, i.v.). These results suggest that BP reduces the nephrotoxicity of carbapenems through inhibiting the active transport of carbapenems in the renal cortex. Because of the low toxicity of BP (LD50 in the rat, more than 3,000 mg/kg, i.v.), it was concluded that BP might be a good candidate for reducing the nephrotoxicity induced by panipenem or imipenem.

Low neurotoxicity of LJC 10,627, a novel 1 beta-methyl carbapenem antibiotic: inhibition of gamma-aminobutyric acidA, benzodiazepine, and glycine receptor binding in relation to lack of central nervous system toxicity in rats.

The toxicity of LJC 10,627 to the central nervous system of rats was evaluated by examining the effects of the compound on gamma-aminobutyric acidA, benzodiazepine, and glycine receptor binding in rat synaptic membranes and on the induction of behavioral convulsions by intraventricular administration to rats. The concentrations of this compound needed to inhibit specific [3H]muscimol binding, specific [3H]diazepam binding, and specific [3H]strychnine binding were greater than those of imipenem, as demonstrated by the 50% inhibitory concentrations (IC50S of LJC 10,627, greater than 10 mM for each; IC50S of imipenem, 0.6, 1.9, and 0.2 mM, respectively). These results reflect the fact that LJC 10,627 does not evoke severe convulsions or cause death, even when it is administered intraventricularly at a high dose (300 micrograms per rat), and suggest that the low neurotoxic potential of LJC 10,627 may be attributed to the chemical structure of this compound, which has a methyl radical at the 1 beta site and a triazolium radical at the side chain of the second site.

Neurotoxicity of panipenem/betamipron, a new carbapenem, in rabbits: correlation to concentration in central nervous system.

The neurotoxic potential of panipenem/betamipron (PAPM/BP), a new carbapenem antibiotic, was compared with that of imipenem/cilastatin (IPM/CS). The drug concentration in cerebrospinal fluid (CSF) at the onset of epileptogenic electroencephalographic (EEG)-activity and the drug distribution into the central nervous system (CNS) were evaluated. Epileptogenic reactions correlated well with drug levels in CSF, but not with drug levels in circulating plasma. The concentration of PAPM in CSF at the onset of epileptogenic EEG-activity was almost twice that of IPM, suggesting that neurotoxic activity of PAPM is about half that of IPM. In addition, in terms of incidence percent for the epileptogenic EEG-activity, PAPM/BP was found to be less toxic than IPM/CS within the dose of 1.0-1.2 g/kg. Concentrations of PAPM in CSF and brain extracellular fluid after PAPM/BP i.v. infusion were comparable with those of IPM after IPM/CS infusion, indicating the similar characteristics of distribution into the CNS for the two antibiotics. From these results of pharmacologic effects and drug distributions, it is suggested that the neurotoxicity of PAPM/BP is less than half that of IPM/CS.

Thienamycin nephrotoxicity. Mitochondrial injury and oxidative effects of imipenem in the rabbit kidney.

The nephrotoxic cephalosoprins cephaloridine and cephaloglycin both produce mitochondrial respiratory toxicity in renal cortex. Recent work has provided evidence that this respiratory toxicity is caused by acylation and inactivation of mitochondrial anionic substrate transporters. While cephaloridine also causes significant lipid peroxidative injury in cortical mitochondria and microsomes, cephaloglycin causes little or no oxidative damage under identical conditions. The recently released thienamycin antibiotic, imipenem, like the toxic cephalosporins, produces acute proximal tubular necrosis which can be prevented completely by prior administration of probenecid. The ability of imipenem to block mitochondrial substrate uptake and respiration and produce oxidative changes has not been examined. We therefore evaluated the effects of imipenem in rabbit renal cortex on the following: (1) mitochondrial function [respiration with and uptake of succinate, and uptake of ADP]; and (2) evidence of oxidative change [depletion of reduced glutathione (GSH), production of oxidized glutathione (GSSG), and production of lipid peroxidative injury, as reflected in microsomal conjugated dienes (CDs)]. The mitochondrial effects of 300 mg/kg body wt of imipenem, given i.v. 1 and 2 hr before killing the animals, were comparable to those of the nephrotoxic cephalosporins. There was significant reduction of respiration with, and unidirectional uptake of, succinate at both times, while mitochondrial ADP transport was comparatively unaffected. Imipenem also depleted GSH and increased GSSG and CDs at 1 hr. These effects, however, were considerably smaller than those of a comparably nephrotoxic dose of cephaloridine, and this evidence of oxidative stress had resolved by 2 hr. We conclude that imipenem and the nephrotoxic cephalosporins have similar effects on mitochondrial substrate uptake and respiration, but differ significantly in their production of oxidative injury.

Meropenem: evidence of lack of proconvulsive tendency in mice.

The proconvulsive tendency of the novel carbapenem, meropenem was compared to that of imipenem, alone and in combination with cilastatin. The potentiation of metrazole-induced convulsions in mice was measured. Both imipenem and imipenem/cilastatin caused significant potentiation of metrazole-induced convulsions at a dose of 200 mg/kg, i.v. In contrast, meropenem (50-400 mg/kg, iv) failed to exhibit any significant potentiation of metrazole-induced seizures.

Safety evaluation of meropenem in animals: studies on the kidney.

The effect of meropenem on animal kidneys has been assessed in rats (5 of each sex/group), rabbits (3 of each sex/group) and monkeys (3 of each sex/group) in comparative iv studies with ceftazidime, cefotaxime, cephaloridine and imipenem (without cilastatin). Diarrhoea occurred in rabbits and monkeys dosed with imipenem or meropenem. Emesis occurred only after the administration of imipenem to monkeys. After 14 days administration to rats evidence of nephrotoxicity was seen only in males dosed with cephaloridine (850 mg/kg); no changes were seen with ceftazidime, cefotaxime or meropenem (all at 1000 mg/kg). Four days after a single dose to rabbits renal tubular necrosis was seen in all animals receiving imipenem (150 mg/kg) and cephaloridine (250 mg/kg). Minimal histopathological changes to the kidneys were seen with cefotaxime, ceftazidime and meropenem (all at 400 mg/kg). After seven days' administration to cynomolgus monkeys imipenem (180 mg/kg) caused moderate to severe tubular necrosis. No tubular damage was seen with meropenem at 180 mg/kg or with cefotaxime or ceftazidime (both at 500 mg/kg). At 500 mg/kg meropenem caused mild tubular regeneration and/or fat accumulation in 3/6 animals, with mild tubular necrosis in one of these. The data from these three species indicate that meropenem has a low nephrotoxic potential in these animal models.

Animal model for evaluating the convulsive liability of beta-lactam antibiotics.

The beta-lactam antibiotics imipenem-cilastatin, BMY-26225, and cefazolin significantly lowered the convulsive threshold of pentylenetetrazole in mice. In addition, imipenem-cilastatin and cefazolin were found to inhibit 3H-labeled gamma-aminobutyric acid binding to synaptic membranes from rat brains. Our results suggest that the pentylenetetrazole convulsive model may be useful in evaluating the proconvulsive liabilities of new carbapenems and other beta-lactam antibiotics and that the mechanism of imipenem-cilastatin and cefazolin toxicity may involve interaction with gamma-aminobutyric acid receptors.

Toxicity of imipenem in vitreous replacement fluid.

We evaluated the ocular toxicity of imipenem when administered in the infusion fluid during vitrectomy. Imipenem concentrations of up to 16 micrograms/mL were nontoxic to the ocular structures.

Inhibition of the mammalian beta-lactamase renal dipeptidase (dehydropeptidase-I) by (Z)-2-(acylamino)-3-substituted-propenoic acids.

The title enzyme deactivates the potent carbapenem antibiotic imipenem in the kidney, producing low antibiotic levels in the urinary tract. A series of (Z)-2-(acylamino)-3-substituted-propenoic acids (3) are specific, competitive inhibitors of the enzyme capable of increasing the urinary concentration of imipenem in vivo. Many of the compounds were prepared in one step from an alpha-keto acid and a primary amide. The optimum R2 groups are 2,2-dimethyl, -dichloro, and -dibromocyclopropyl. With R2 = 2,2-dimethylcyclopropyl (DMCP), a wide variety of R3 groups including alkyl, oxa- and thiaalkyl, and alkyl groups containing acidic, basic, and neutral substituents give effective inhibitors with Ki values of 0.02-1 microM and a range of pharmacokinetic properties. By resolution of enantiomers and X-ray crystallography, the enzyme-inhibitory activity of the DMCP group was found to reside with the 1S isomer. The cysteinyl compound 176 (cilastatin, MK-0791) has the desired pharmacological properties and has been chosen for combination with imipenem.