In vivo and in vitro metabolic studies of furazolidone

The experiments described in this thesis were undertaken to get a better understanding of the kinetics and biotransformation of the nitrofuran furazolidone upon oral administration to swine. Such information forms an essential prerequisite for making an appropriate assessment of the consumer hazards of edible products originating from animals treated with this compound. The experiments were conducted according to two different approaches: - kinetic studies in the target animal to determine the elimination kinetics of furazolidone and its metabolites from plasma and tissues upon oral administration of furazolidone - in vitro biotransformation studies using swine and rat liver microsomes to elucidate biotransformation routes and to identify reactive intermediates responsible for the interaction with biological macromolecules. Part I of this thesis starts with a description of some developments in animal husbandry during the last decades. This is followed by a review of literature data on physical/chemical and antimicrobial characteristics, toxicity, elimination kinetics and biotransformation of furazolidone (chapter 1). Part II of this thesis deals with the present study. In chapter 2 a fast, sensitive method has been described for the determination of furazolidone in swine plasma, muscle, liver, kidney, fat and urine based on high-performance liquid chromatographic separation after solid-phase extraction on Extrelut R 1. The sensitivity of the method was 1-2 ng/ml (g) for plasma and tissues and 25 ng/ml for urine. From the kinetic studies the conclusion can be drawn that no accumulation of furazolidone occurs in blood after oral administration of furazolidone to both piglets (chapter 3) and adult swine (chapter 4); the half life time was respectively 45 and 60 minutes. Furazolidone was rapidly and almost completely metabolized and urine proved to be the major excretion pathway of formed metabolites: 61% of the radioactive dose administered to piglets had been exreted via the urine and 18% via faeces, while in urine of adult swine only traces of total dosed furazolidone could be recovered. In tissues of piglets and adult swine no residues of furazolidone could be detected at all. However, the experiments with piglets showed that relatively high levels of radioactivity were present in all tissues studied. After a withdrawal period of 14 days the concentrations varied from 0.9-4.3 pg-equivalents per gram tissue. Up to 56% of the radioactivity detected in organs and tissues appeared to be non-extractable and was partly associated with DNA. This may either be indicative for endogenous incorporation in physiologically occurring compounds or for covalent binding of reactive intermediate metabolites of furazolidone to biological macromolecules (chapter 3). Studies with the adult swine further revealed the formation of a cyano-derivative from furazolidone, namely 3-(4-cyano-2- oxobutylidene amino)-2-oxazolidone. In studies in other animal species, it was claimed that this cyano-derivative was a major metabolite in urine. However, this compound proved to be a minor metabolite in plasma (half life time: 4 hours) and tissues of adult swine. This may be explained by species differences in biotransformation of furazolidone, by a very fast elimination of the cyano-derivative via urine or by an effective trapping of reactive intermediate metabolites of furazolidone by biological macromolecules or agents like glutathione (chapter 4). Two major ethyl acetate extractable metabolites of furazolidone could be observed upon incubation in rat liver microsomes: the cyano-derivative referred to before and a reaction product of furazolidone with its open-chain acrylonitrile derivative, namely 2,3-dihydro-3cyanomethyl-2-hydroxy-5-nitro-lα,2-di-(2-oxo-oxazolidin-3-yl)iminomethylfuro[2,3-b]furan. Approximately 2-7% of the totally formed metabolites proved to be covalently bound to microsomal protein. This covalent binding could be inhibited by addition of glutathione, which also resulted in an almost complete shift from non-polar to watersoluble metabolites. No binding to DNA was detected when calf thymus DNA was added to a microsomal incubation mixture with furazolidone in contrast to the in vivo DNA interaction observed in the piglets. This may be explained by differences between the in vitro and in vivo situation in the availability of reactive intermediate metabolites of furazolidone for reaction with DNA (chapter 5). The above mentioned ethyl acetate extractable metabolites of furazolidone proved to be only minor conversion products of furazolidone upon incubation in swine liver microsomes, while the percentage of the protein bound metabolites formed was much higher, namely 12-22%. The site of attack on microsomal protein is probably formed by the thiol-group of cysteine as indicated by results of experiments performed with amino acids. Using mercapto-ethanol as a trapping agent for reactive intermediates, a new metabolite could be isolated upon incubation of furazolidone in swine liver microsomes. This compound was identified as a conjugation product of the open-chain acrylonitrile derivative of furazolidone with mercapto-ethanol, namely 3-(4-cyano-3-β-hydroxyethyl-mercapto2-oxobutylidene amino)-2-oxazolidone. This conjugate accounted for approximately 50% of total amount of metabolites formed. In addition, neither the above mentioned ethyl acetate extractable metabolites nor covalent binding to microsomal protein could be observed when mercaptoethanol was added to the incubation mixture, indicating that the openchain acrylonitrile derivative plays a central role in reductive biotransformation of furazolidone by swine liver microsomes (chapter 6). The open-chain acrylonitrile derivative of furazolidone binds reversibly with thiolgroup containing agents such as glutathione and mercapto-ethanol or with microsomal protein. The reversibility of the exchange reaction is dependent on pH as is demonstrated for the mercapto-ethanol conjugate. Below pH 2 this conjugate is stable; optimal exchange to microsomal protein is found between pH 7 and 10. The mercapto-ethanol conjugate gives a distinct direct positive response in the Salmonella/microsome test using tester strain TA 100. This is probably due to an interaction of the acrylonitrile moiety with DNA (chapter 7). It can be concluded that furazolidone is rapidly and almost completely metabolized upon oral administration to piglets. A major part of the formed metabolites proved to be non-extractable from the tissues. It can not be excluded that these non-extractable metabolites are the result of covalent binding of a reactive intermediate metabolite of furazolidone to biological macromolecules such as protein or DNA. From in vitro studies using swine liver microsomes evidence has been obtained showing that the open-chain acrylonitrile derivative plays a central role in the biotransformation of furazolidone. It has been shown that this reactive intermediate can bind reversibly to microsomal protein. Whether this non-enzymatic reaction also occurs in vivo is not yet known. As the possible presence of such covalently bound residues in edible tissues could have serious consequences for the acceptibility of the drug for massmedication of food producing animals, the identity of the non-extractable radioactive material in tissue of piglets should be further investigated, together with its bioavailability and toxicity. Efforts should also be made to identify the extractable radioactivity in tissues of piglets as only a small part of this fraction as so far has been identified.