Disturbances in the formation of FAD and covalently bound flavins in Novikoff hepatoma from riboflavin-deficient rats

The incorporation of radiolabeled riboflavin into flavin mononucleotide, flavin adenine dinucleotide, and flavin covalently bound to protein was determined in Novikoff hepatoma grown in both riboflavin-deficient and normal chow-fed rats. In Novikoff hepatoma, the incorporation of [14C]riboflavin into covalently bound flavins relative to that into FAD was substantially greater than that in host liver, and the turnover rate of riboflavin was also accelerated in tumor compared with the liver. The magnitude of incorporation of [14C]riboflavin into each of the various flavin fractions was substantially greater in tumors from riboflavin-deficient animals than in tumors from control animals. These data support the hypothesis that in conditions of riboflavin deprivation, Novikoff hepatoma maintains the levels of the physiologically important flavin coenzymes at the expense of the free riboflavin fraction. The incorporation of riboflavin into covalently bound flavins relative to that into FAD is substantially greater in Novikoff hepatoma than in liver. Accordingly, covalently bound flavins are either present in greater amounts or regulated differently in tumor than in normal tissue. Because the flavin moiety cannot be reutilized, the covalently bound flavin fraction in Novikoff hepatoma theoretically should be able to sequester riboflavin and thereby deplete the body reserves of this vitamin when dietary intake is marginal.

Electron-transferring flavoprotein from pig kidney: flavin analogue studies

Apo-electron-transferring flavoprotein from pig kidney (apo-ETF) has been prepared by an acid ammonium sulfate procedure and reconstituted with FAD analogues to probe the flavin binding site. The 8-position of the bound flavin is accessible to solvent as judged by the reaction of 8-Cl-FAD-ETF with sodium sulfide and thiophenol. A series of 8-alkylmercapto-FAD analogues containing increasingly bulky substituents bind tightly to apo-ETF and can be reduced to the dihydroflavin level by octanoyl-CoA in the presence of catalytic levels of the medium-chain acyl-CoA dehydrogenase. Bulky substituents severely slow the rate of these interflavin electron-transfer reactions. In the case of the 8-cyclohexylmercapto derivative, this decrease reflects a sizable increase in the Km for ETF (approximately 14-fold) with only a 20% decrease in Vmax. Reduction of all of these 8-substituted derivatives involves the accumulation of ETF anion radical intermediates. Dihydro-5-deaza-FAD dehydrogenase, unlike the corresponding 1-deazaflavin substitution, is unable to reduce native ETF despite a strongly favorable redox potential difference. These results, together with data from the native proteins, are consistent with obligatory 1-electron transfer between dehydrogenase and ETF possibly involving the exposed dimethylbenzene edge of ETF. Irradiation of apo-ETF reconstituted with the photoaffinity analogue 8-azidoflavin leads to approximately 10% covalent incorporation of the flavin. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of apo-ETF labeled with tritiated 8-azido-FAD shows preferential labeling of the smaller subunit (88%, Mr 30,000 subunit; 12%, Mr 33,000 subunit).(ABSTRACT TRUNCATED AT 250 WORDS)

New approaches to the possible prevention of side effects of chemotherapy by nutrition

In an effort to develop new methods for preventing side effects of chemotherapy, the authors initiated studies to determine whether Adriamycin (doxorubicin) inhibits the metabolism of riboflavin (vitamin B2). Adriamycin has been shown to form a 1:1 stoichiometric complex with riboflavin, as well as to compete for binding to tissue proteins. Adult rats treated with Adriamycin in clinically relevant doses were compared to control animals in ability to convert riboflavin into flavin adenine dinucleotide (FAD), the active flavin coenzyme derivative, in heart, skeletal muscle, liver, and kidney. Rats treated with Adriamycin exhibited diminished formation of carbon 14 (14C)FAD in skeletal muscle to nearly 50% that of controls, and in heart to about 70% to 80% of controls. Under these conditions, (14C)FAD formation in liver and kidney was largely unaffected by Adriamycin. In preliminary studies, riboflavin-deficient animals treated with Adriamycin had accelerated mortality rates compared to those of food restricted controls treated with similar doses of Adriamycin. The data as a whole suggest a potential mechanism for Adriamycin-induced cardiac and skeletal myopathy, i.e., inhibition of synthesis of FAD, a flavin coenzyme which is involved in electron transport, lipid metabolism, and energy generation. These findings in an animal model raise the possibility that defects of riboflavin nutriture, either dietary or drug-induced, may be a determinant of Adriamycin toxicity. Further studies are required to explore the potential for preventing side effects due to Adriamycin by administration of this vitamin.

Riboflavin-responsive defects of beta-oxidation

The key reaction in the beta-oxidation of fatty acids is the acyl-CoA dehydrogenation, catalyzed by short chain, medium chain, and long chain acyl-CoA dehydrogenases. Acyl-CoA dehydrogenation reactions are also involved in the metabolism of the branched chain amino acids, where isovaleryl-CoA and 2-methylbutyryl-CoA dehydrogenases are involved and in the metabolism of lysine, 5-hydroxylysine and tryptophan, where glutaryl-CoA dehydrogenase functions. In all of these dehydrogenation systems reducing equivalents are transported to the main respiratory chain by electron transfer flavoprotein (ETF) and electron transfer flavoprotein dehydrogenase (ETFDH), which are common to all the dehydrogenation systems. The acyl-CoA dehydrogenation enzymes are dependent on flavin adenine dinucleotide (FAD) as coenzyme, for which riboflavin is the precursor. Patients with multiple acyl-CoA dehydrogenation deficiencies have been found in whom the defect has been located to ETF and/or ETFDH. A few patients with multiple acyl-CoA dehydrogenation deficiencies have been described, in whom no defects in acyl-CoA dehydrogenases, ETF or ETFDH have been found but who respond clinically and biochemically to pharmacological doses of riboflavin. This indicates a defect related to the metabolism of FAD. An uptake defect of riboflavin or a synthesis defect of FAD from riboflavin have been excluded by in vivo and in vitro studies. A mitochondrial transport defect of FAD or a defect in the binding FAD to ETF and/or ETFDH remains possible.

Glutaric aciduria type II: evidence for a defect related to the electron transfer flavoprotein or its dehydrogenase

Incubation of intact fibroblasts from a patients with glutaric aciduria type II with [2-14C]riboflavin showed normal synthesis of flavin mononucleotide and flavin adenine dinucleotide. This is taken as evidence for normal transport of riboflavin into the cells and normal activity of riboflavin kinase (EC 2.7.1.26) and flavin mononucleotide adenylyltransferase (EC 2.7.7.2). The ability of intact fibroblasts to oxidize 1-14C-fatty acids and [6-14C]lysine is impaired in the patient which together with the urinary excretion pattern of organic acids indicates a defective dehydrogenation of fatty acid acyl-CoAs and glutaryl-CoA. However, dehydrogenation of (C6-C10) fatty acid acyl-CoA derivatives and glutaryl-CoA was normal when the dehydrogenases were measured in fibroblast homogenate with artificial electron acceptors. In vivo, these dehydrogenases transfer their electrons to CoQ10 in the main electron transport chain via electron transfer flavoprotein and electron transfer flavoprotein dehydrogenase. Glutaric aciduria type II fibroblasts showed very diminished activity when the glutaryl-CoA dehydrogenase activity was measured without artificial electron acceptor but with intact endogenous electron transport system. As the NADH and succinate oxidation seems normal in glutaric aciduria type II patients, this is strong evidence for a defect in either the electron transfer flavoprotein or the electron transfer flavoprotein dehydrogenase.

Enzymatic synthesis of flavin adenine dinucleotide

A new and simple enzymatic method for the synthesis of flavin adenine dinucleotide (FAD) from flavin mononucleotide by the transadenylylation reaction using microbial cells is described. Among various microorganisms tested, Artherobacter globiformis IFO 12138 and two soil bacteria were selected as useful enzyme sources. Under suitable reaction conditions, the amount of FAD synthesized was 2.25 mumol/mL with cells of A. globiformis. The transadenylylation reaction could be coupled with the ATP supplying system through a glycolysis process with yeast.

Cardiac sensitivity to the inhibitory effects of chlorpromazine, imipramine and amitriptyline upon formation of flavins

Chlorpromazine, imipramine and amitriptyline, drugs structurally related to riboflavin, each inhibited the formation in vivo of flavin adenine dinucleotide (FAD) from riboflavin in rat heart at 2-5 mg/kg body weight, doses comparable on a weight basis to those used clinically. All three drugs inhibited FAD formation in heart within 5 hr after a single dose of 25 mg/kg. Chlorpromazine under these conditions also inhibited FAD formation in liver, cerebrum and cerebellum. A series of psychoactive agents structurally unrelated to riboflavin did not inhibit flavin formation in the organs tested. These findings indicate that the inhibitory effects of the drugs studied have organ specificity with respect to FAD formation.

Inhibition of riboflavin metabolism in rat tissues by chlorpromazine, imipramine, and amitriptyline

Prompted by recognition of the similar structures of riboflavin (vitamin B(2)), phenothiazine drugs, and tricyclic antidepressants, our studies sought to determine effects of drugs of these two types upon the conversion of riboflavin into its active coenzyme derivative, flavin adenine dinucleotide (FAD) in rat tissues. Chlorpromazine, a phenothiazine derivative, and imipramine and amitriptyline, both tricyclic antidepressants, each inhibited the incorporation of [(14)C]riboflavin into [(14)C]FAD in liver, cerebrum, cerebellum, and heart. A variety of psychoactive drugs structurally unrelated to riboflavin were ineffective. Chlorpromazine, imipramine, and amitriptyline in vitro inhibited hepatic flavokinase, the first of two enzymes in the conversion of riboflavin to FAD. Evidence was obtained that chlorpromazine administration for a 3- or 7-wk period at doses comparable on a weight basis to those used clinically has significant effects upon riboflavin metabolism in the animal as a whole: (a) the activity coefficient of erythrocyte glutathione reductase, an FAD-containing enzyme used as an index of riboflavin status physiologically, was elevated, a finding compatible with a deficiency state, (b) the urinary excretion of riboflavin was more than twice that of age- and sex-matched pair-fed control rats, and (c) after administration of chlorpromazine for a 7-wk period, tissue levels of flavin mononucleotide and FAD were significantly lower than those of pair-fed littermates, despite consumption of a diet estimated to contain 30 times the recommended dietary allowance. The present study suggests that certain psychotropic drugs interfere with riboflavin metabolism at least in part by inhibiting the conversion of riboflavin to its coenzyme derivatives, and that as a consequence of such inhibition, the overall utilization of the vitamin is impaired.

Aldosterone stimulation of riboflavin incorporation into rat renal flavin coenzymes and the effect of inhibition by riboflavin analogues on sodium reabsorption

This study was designed to investigate a possible relationship between the effect of aldosterone upon urinary electrolytes and the incorporation of [(14)C]riboflavin into renal [(14)C]flavin mononucleotide (FMN) and [(14)C]flavin adenine dinucleotide (FAD). Adrenalectomized Sprague-Dawley rats that weighed between 185 and 210 g were pretreated with 15 mug/100 g body wt dexamethasone intraperitoneally. 16 h later they were administered aldosterone (1.5 mug/100 g body wt) and [(14)C]riboflavin (5.0 muCi/200 g body wt). The urethra of each rat was ligated, and the rats were sacrificed by decapitation 3 h later. The urine was aspirated from the bladders of each rat and analyzed for total Na(+) and K(+) excretion while the kidneys were removed and the formation of [(14)C]FMN and [(14)C]FAD was determined for each kidney. There was a significant increase in the formation of renal [(14)C]FMN and [(14)C]FAD (27.3 and 14.4%, respectively) after aldosterone treatment. Aldosterone significantly decreased the excretion of Na(+) by 50%, and increased that of K(+) by 55%. To determine if the increased incorporation of [(14)C]riboflavin into renal [(14)C]FMN and [(14)C]FAD was an important intermediary step in the aldosterone-induced alterations in urinary Na(+) and K(+), the riboflavin analogues 7,8-dimethyl-10-formylmethyl isoalloxazine or 7,8-dimethyl-10-(2'-hydroxyethyl) isoalloxazine were given to the animals i.p. to diminish the conversion of riboflavin to FMN by competitively inhibiting the enzyme flavokinase (EC 2.7.1.26). These analogues were found to significantly counteract the decreased urinary excretion of Na(+) as a result of aldosterone from 26+/-9% to 124+/-58% (SEM) with a dose-related response when administered from 10 to 25 mug/100 g body wt. The same doses of the analogues that significantly increased the urinary output of Na(+) when administered simultaneously with aldosterone also significantly decreased the formation of renal [(14)C]FMN from 15+/-4 to 38+/-3% when compared with the effects of aldosterone alone. The analogues exerted no significant effect on the increased urinary excretion of K(+) by aldosterone. The analogues alone had no influence on urinary Na(+) and K(+) output or the formation of renal [(14)C]FMN and [(14)C]FAD at the dose levels that we investigated. These data strongly suggest that the enhanced synthesis of renal FMN and FAD may be a causative factor in the increased reabsorption of Na(+) as a result of aldosterone; and, consequently, riboflavin analogues may function as a novel class of antimineralocorticoids.

Metabolism of injected flavins studied by using double-labeled [14C]flavin adenine dinucleotide and [14C, 32P]flavin mononucleotide

The metabolism of flavins in mouse was studied with [F-(2)-14C, A-(2,8)-14C]FAD and [F-(2)-14C, 32P]FMN. Ninety minutes after injection, radioactive isoalloxazine nucleus of double-labeled FAD was markedly incorporated into FAD, FMN and riboflavin in the liver, whereas a small amount of radioactive adenine nucleus of double-labeled FAD was found in FAD in the liver. In the case of FMN, radioactive isoalloxazine nucleus of double-labeled FMN was markedly incorporated into FAD, FMN and riboflavin in the liver, whereas only a minute amount of radioactive phosphorus was incorporated into FMN and FAD in the same organ. These results indicate that FMN and FAD injected are rapidly hydrolyzed and resynthesized in animal body.

[Flavinogenesis in methylotrophic yeasts]

Methylotrophic yeasts Candida boidinii and Hansenula polymorpha requiring thiamine and biotin accumulate more flavins in the cells during growth on media containing methanol than during oxidation of ethanol and glucose, at the account of increased production of FAD whose percentage in the total flavin content of the cell rises sharply. The level of flavin production depends on growth conditions: the intracellular content of flavins and FAD during utilization of methanol by the yeast cells is higher under conditions of continuous cultivation than in periodic cultures. The content of NAD, another coenzyme of the respiration chain, in the cells during their growth on media containing methanol almost does not differ from its concentration in the cells cultivated on media containing glucose and ethanol. Elevated production of flavins and FAD by the yeast cells on media containing methanol may be caused by an increased requirement in FAD and its specific participation in the first stage of methanol oxidation.

Neurospora crassa conidial germination: role of endogenous amino acid pools

The levels of the endogenous amino acid pools in conidia, germinating conidia, and mycelia of wild-type Neurospora crassa were measured. Three different chromatographic procedures employing the amino acid analyzer were used to identify and quantitatively measure 28 different ninhydrin-positive compounds. All of the common amino acids were detected in conidial extracts except proline, methionine, and cystine. The levels of these three amino acid pools were also very low in mycelia. During the first hour of germination in minimal medium, the levels of most of the free amino acid pools decreased. The pool of glutamic acid, the predominant free amino acid in conidia, decreased 70% during the first hour. Very little glutamic acid or any other amino acid was excreted into the medium. During the first 20 min of germination, the decrease in the glutamic acid pool was nearly equivalent to the increase in the aspartic acid pool. The aspartic acid and lambda-aminobutyric acid pools were the only amino acid pools that increased to maximum levels within the first 20 min of germination and then decreased. It is proposed that an important metabolic event that occurs during the early stages of conidial germination is the production of reduced pyridine nucleotides. The degradation of the large glutamic acid pool existing in the conidia (2.5% of the conidial dry weight) could produce these reduced coenzymes.

Deficient cytochrome b5 reductase activity in nontoxic goiter with iodide organification defect

A 37-yr-old woman with nontoxic goiter is presented. The thyroid 131I uptake at 3 and 24 hr were, respectively, 77.1% and 81.4% dose. Thiocyanate discharged 65.5% of the accumulated 131I in 30 min. In vitro organification of iodine in the thyroid homogenate from the patient was impaired and it was restored to normal by the addition of H2O2, glucose, and glucose oxidase system, FAD, or reduced cytochrome b5. Riboflavin, FMN, oxidized cytochrome b5, oxidized or reduced cytochrome c, NAD(H), and NADP(H) were ineffective in the reaction. The microsomal NADH-cytochrome b5 reductase activity was definitely low in the patient's thyroid. It was augmented to a normal level by incubation of the microsomes with FAD for 30 min or more. The activities of thyroid peroxidase, G6-PD, 6-PGD, catalase, protease, and NADPH-cytochrome c reductase were within normal limits. The major thyroid protein was normal thyroglobulin which could be readily iodinated in the presence of H2O2 and horse radish peroxidase. These findings suggest the correlation of an iodide organification defect with a cytochrome b5 reductase deficiency. Administration of high doses of FAD led to the restoration of thyroidal iodide organification mechanism associated with an increased thyroid hormone production and to a marked decrease of the goiter. Riboflavin was given without effect even at a high dosage level. Consequently, it seems likely that the deficient cytochrome b5 reductase activity in this patient is due to a defect in the biosynthesis of FAD, the coenzyme of the reductase, from riboflavin.