Metabolism of biotin and analogues of biotin by microorganisms. II. Further studies on the conversion of D-biotin to biotin vitamers by Lactobacillus plantarum

Birnbaum, Jerome (University of Cincinnati, Cinncinati, Ohio), and Herman C. Lichstein. Metabolism of biotin and analogues of biotin by microorganisms. II. Further studies on the conversion of d-biotin to biotin vitamers by Lactobacillus plantarum. J. Bacteriol. 92:913-919. 1966.-Lactobacillus plantarum growing in excess biotin converts a portion to two vitamers (combinable and uncombinable with avidin) not utilizable for growth. These were detected by differential yeast-lactobacillus assay. In the present study, suspensions of 12- and 72-hr cells showed no converting activity. Vitamer formation by nonproliferating 24-hr cells required glucose and exhibited a lag; 17-hr cells showed neither a lag nor a glucose requirement. Iodoacetate and chloramphenicol inhibited vitamer formation by 24-hr cells, but had no effect on 17-hr cells. Addition of hydrolyzed casein or preincubation in biotin decreased the lag and enhanced vitamer formation in 24-hr cells, but had no effect in 17-hr cells. Apparently, 17-hr cells contain the converting enzymes which degenerate as growth proceeds; the lag exhibited by 24-hr cells represents the time necessary to reform the enzymes. Equal amounts of the two vitamers were formed in 17-hr cells; only the avidin-combinable form was produced initially by 24-hr cells, unless hydrolyzed casein was present. Electrophoresis revealed that the avidin-combinable vitamer has the same charge as biotin,whereas the uncombinable form possesses both positive and negative groups. Column chromatography was used to separate the avidin uncombinable material from biotin and the avidin-combinable form. L. plantarum was unable to accumulate the avidin-uncombinable vitamer under conditions permitting good biotin accumulation. It was concluded that L. plantarum sequentially converts biotin to avidin-combinable and -uncombinable vitamers, the latter being impermeable to the cells.     

Biosynthesis of biotin vitamers by Yarrowia lipolytica

The hydrocarbon utilizing yeast Yarrowia lipolyyica NCYC 1421 produces biotin and its vitamers when grown on glucose in biotin-free media. Levels of production can be influenced by the medium composition. Growth in the presence of longchained fatty acids greatly increases biotin vitamer production. The biotin vitamers produced are normally dethiobiotin and 7-keto, 8-aminopelargonic acid. The addition of succinic acid at 0.5 g per litre causes the vitamer 7, 8-diaminopelargonic acid to be produced at high levels. The biotin antagonist α-dehydrobiotin inhibits the growth of Yarrowia lipolytica . Mutants can be readily isolated which show resistance to α-dehydrobiotin, but these do not produce greater amounts of biotin or its vitamers.     

Studies on Production of Biotin by Microorganisms

During the course of the study on biotin vitamers production by a hydrocarbon-utilizing bacterium, strain 5–2 (Pseudomonas sp.), it was found that crude RNA-alkali-hydrolyzate from yeast increased the accumulaion of biotin vitamers, most of which was determined as desthiobion, and that adenine in the crude RNA-alkali-hydrolyzate was a potent stimulator. Effect of adenine on biotin vitamers accumulation was observed in the medium with either hydrocarbon or glucose as a sole carbon source. The accumulation of total biotin vitamers by some other bacteria was also increased by adenine but that of true biotin was scarcely increased or inhibited by adenine.

The role of adenine on the accumulation of biotin vitamers was investigated with non-proliferating cells of strain 5–2, and it was supposed that adenine would not only inhibit the accumulation of true biotin but, as a result, cause the large accumulation of biotin vitamers which might be intermediates of biotin synthesis. When the medium was supplemented with excess biotin, complete repression occurred even in the presence of adenine.     

Synthesis of 7-oxo-8-aminopelargonic acid, a biotin vitamer, in cell-free extracts of Escherichia coli biotin auxotrophs

The enzymatic synthesis of 7-oxo-8-aminopelargonic acid (7-KAP) from pimelyl-coenzyme A and l-alanine was demonstrated in cell-free extracts of a biotin mutant of Escherichia coli K-12 which excretes only 7-KAP into the growth medium. This biotin vitamer was identified by its chromatographic and electrophoretic properties. The enzyme (7-KAP synthetase) was repressed when the organism was grown in biotin concentrations greater than 0.2 ng/ml. The parent strain and members of other mutant groups that excrete 7-KAP, in addition to other vitamers, also exhibited synthetase activity. A mutant group that failed to excrete 7-KAP was further sub-divided into three groups, one of which lacked synthetase activity. These results are discussed in relation to a previously proposed scheme for biotin biosynthesis in which the formation of 7-KAP is considered the point of entry for pimelic acid into the biotin pathway.     

Studies on Biosynthesis on Biotin by Microorganisms

The accumulation of biotin-vitamers in the culture media of a large number of microorganisms (about 700 strains) was studied. The contents of the biotin-vitamers were quantitatively determined by microbiological assays with Lactobacillus arabinosus and Saccharomyces cerevisiae.

It was found that large amounts of biotin-vitamers were accumulated by various microorganisms such as Streptomyces, molds and bacteria, and that the yield of biotin-vitamers was enhanced by the addition of pimelic acid or azelaic acid to the media. It was also found that the main portion of the vitamers accumulated by many microorganisms did not support the growth of Lactobacillus arabinosus, while it did support that of Saccharomyces cerevisiae. The small amounts of true biotin were observed in the culture media of various Streptomyces and molds, but hardly in the culture media of bacteria.

The identification of biotin-vitamers accumulated by various microorganisms is described, and the distribution of the vitamers in microorganisms is also described.

The results presented in this paper show that the main component of the vitamers accumulated by many microorganisms is identified as desthiobiotin by anion exchange column chromatography, paper chromatography and chemical analysis. Small amounts of fraction B (unidentified vitamers) and Fraction D (biotin) were also detected in the culture media of various molds and Streptomyces. However, these fractions were not observed in the culture media of any bacteria tested.

It was also found that large amounts of an unknown biotin-vitamer was accumulated by various bacteria. The vitamer was avidin-uncombinable, and, from the paper electrophoretic studies, it was assumed that the vitamer might be an analogue of pelargonic acid.     

An early intermediate in the biosynthesis of biotin: Incorporation studies with [1,7-C(2)]pimelic acid

1. An unknown biotin vitamer was obtained in high yields in culture filtrates of Penicillium chrysogenum. 2. Production of this vitamer and desthiobiotin is controlled by the biotin concentration in the medium. 3. The unknown vitamer becomes labelled when the organism is grown in the presence of radioactive pimelic acid. 4. Chromatographic procedures were developed for the purification of the radioactive vitamer. 5. The vitamer is extremely stable in concentrated acid but gives rise to new vitamers under certain conditions. 6. The intermediate role of this vitamer in the synthesis of biotin is discussed.     

Repression of Acetyl-Coenzyme A Carboxylase by Unsaturated Fatty Acids: Relationship to Coenzyme Repression

It has been reported that the level of d-biotin in the growth medium of Lactobacillus plantarum regulates the synthesis of apoacetyl-coenzyme A (CoA) carboxylase; high levels cause repression, and deficient levels effect derepression. In this study, evidence has been obtained which suggests that coenzyme repression by biotin is an indirect effect; i.e., biotin regulates the synthesis of unsaturated fatty acids which are the true repressors of the acetyl-CoA carboxylase. This was observed in an experiment in which long-chain unsaturated fatty acids were added to media containing deficient, sufficient, or excess levels of d-biotin. In every case, independently of the biotin concentration for growth, the unsaturated fatty acids caused a severe repression of the carboxylase. Saturated fatty acids were without effect. The level of oleic acid required to give maximal repression was 50 mug/ml. The free fatty acids had no adverse effect on the activity of the cell-free extracts nor on the permeation of d-biotin into the cell. Saturated and unsaturated fatty acids decreased the rate of holocarboxylase formation from d-biotin and the apoacetyl-CoA carboxylase in the extracts. It is concluded that there are at least three mechanisms that control the acetyl-CoA carboxylase in this organism: (i) indirect coenzyme repression by d-biotin, (ii) repression by unsaturated fatty acids, and (iii) regulation of the activity of the holocarboxylase synthetase by both saturated and unsaturated fatty acids.     

Production of biotin by yeasts

A quantitative screening procedure for biotin and biotin vitamer production was conducted on 129 yeast strains able to grow in a biotin-free medium. Production of biotin and biotin vitamers varied considerably from strain to strain even within a species. The best producers of biotin were strains of Sporobolomyces roseus and Rhodotorula rubra whilst strains of Rhodotorula rubra and Yarrowia lipolytica produced the largest amounts of vitamers.     

Biotin formation by recombinant strains of Escherichia coli: influence of the host physiology.

Strains of Escherichia coli were transformed with different plasmids bearing the gene clusters bioXWF and bioDAYB isolated from the Gram positive bacterium Bacillus sphaericus. These genes encode for the enzymes involved in the metabolic pathway which synthesizes biotin from the precursor pimelic acid. Transformed E. coli strains were grown in bioreactors to reach a biomass of 18 g l-1 cell dry weight in 1 litre batch culture with substrate feeding and approximately 50 g l-1 in 10 l fed batch culture. Improved yields of total vitamers and biotin formed in these processes were achieved after a comparative analysis of different culture conditions, medium compositions, host strains and expression systems. Production of 27 mg l-1 of biotin and 200 mg l-1 of vitamers was achieved in 1 litre batch culture. Using a 10-1 fed batch process, biotin and vitamer concentrations reached maximum values of 45 mg l-1 and 350 mg l-1, respectively.     

Studies on Degradation of d-Biotin by Microorganisms

During the course of our investigations on the metabolism of d-biotin by microorganism, it has been found that some strains of fungi belonging to the genera Rhodotorula, Penicillium and Endomycopsis, are able to degrade d-biotin oxidatively into various biotin vitamers. The present work was undertaken to characterize these vitamers. The vitamers formed were separated by the ion exchange column chromatography, into Fraction A (d-biotin sulfoxide), Fraction B (unknown vitamer II), Fraction C (d-biotin) and Fraction D (unknown vitamer I). Rf values of vitamer I and vitamer II were found to be different from those of the known biotin vitamers. The vitamers I and II did not support the growth of Lactobacillus arabinosus and Saccharomyces cerevisiae, but did support that of Bacillus subtilis. This degradation reaction occurred rather favorably in high aerobic condition.     

Repression of biotin biosynthesis in Escherichia coli during growth on biotin vitamers.

A strain of Escherichia coli in which the lacZ gene was fused to the bioA promoter was constructed. Colonies of this strain formed Lac(+) colonies on low-biotin agar (1.6 to 4.1 nM) and Lac(-) colonies on high-biotin agar (41 nM). This lac-bio fusion strain was used to study the question of whether cells growing on the biotin vitamers d-biotin-d-sulfoxide (BDS) and dethiobiotin (DTB) generate enough biotin to give maximal repression of beta-galactosidase synthesis. Repression by high concentrations (400 nM) of BDS was almost maximal (about 96%), whereas DTB repression reached a saturation level of about 80% with increasing DTB concentrations. The levels of repression obtained with both vitamers were sufficient to cause the colonies to appear Lac(-). When the lac-bio fusion was transduced into lines carrying mutations (bis) that prevent reduction of BDS to biotin, the transductants were not repressed by added BDS. Repression by BDS is unlikely to result from accumulation of extracellular biotin-related substances because (i) washed bis(+) cells were not detectably derepressed when transferred into medium containing BDS and (ii) washed bis cells were not detectably repressed when transferred into medium in which bis(+) cells had grown. Lactose agar plates containing high concentrations of DTB or BDS comprise an efficient selective medium for bioB or bis mutants and were used to isolate spontaneous mutations of these genes. This method should be adaptable to the selection of mutations in any biosynthetic pathway subject to end-product repression.     

The biosynthesis of biotin in growing yeast cells: The formation of biotin from an early intermediate

1. Yeast cells grown in the presence of an unknown radioactive biotin vitamer produced by Penicillium chrysogenum incorporated the vitamer into the newly synthesized biotin. 2. The biotin was isolated as the avidin–biotin complex and after hydrolysis the biological activity and radioactivity were shown to be coincidental. 3. The specific activity of the biotin was identical with that of the pimelic acid used in a previous investigation to label the unknown vitamer. 4. The role of the unknown biotin vitamer as an intermediate in biotin biosynthesis is discussed.     

Characterization of the biotin transport system in Saccharomyces cerevisiae

The characteristics of the biotin transport mechanism of Saccharomyces cerevisiae were investigated in nonproliferating cells. Microbiological and radioisotope assays were employed to measure biotin uptake. The vitamin existed intracellularly in both free and bound forms. Free biotin was extracted by boiling water. Chromatography of the free extract showed it to consist entirely of d-biotin. Cellular bound biotin was released by treating cells with 6 n H(2)SO(4). The rate of biotin uptake was linear with time for 10 min, reaching a maximum at about 20 min followed by a gradual loss of accumulated free vitamin from the cells. Biotin was not degraded or converted to vitamers during uptake. Transport was temperature- and pH-dependent, optimum conditions for uptake being 30 C and pH 4.0. Glucose markedly stimulated biotin transport. In its presence, large intracellular free-biotin concentration gradients were established. Iodoacetate inhibited the glucose stimulation of biotin uptake. The rate of vitamin transport increased in a linear fashion with increasing cell mass. The transport system was saturated with increasing concentrations of the vitamin. The apparent K(m) for uptake was 3.23 x 10(-7)m. Uptake of radioactive biotin was inhibited by unlabeled biotin and a number of analogues including homobiotin, desthiobiotin, oxybiotin, norbiotin, and biotin sulfone. Proline, hydroxyproline, and 7,8-diaminopelargonic acid did not inhibit uptake. Unlabeled biotin and desthiobiotin exchanged with accumulated intracellular (14)C-biotin, whereas hydroxyproline did not.     

Metabolism of biotin and analogues of biotin by microorganisms. 3. Degradation of oxybiotin and desthiobiotin by Lactobacillus plantarum

Birnbaum, Jerome (University of Cincinnati, Cincinnati, Ohio), and Herman C. Lichstein. Metabolism of biotin and analogues of biotin by microorganisms. III. Degradation of oxybiotin and desthiobiotin by Lactobacillus plantarum. J. Bacteriol 92:920-924. 1966.-Lactobacillus plantarum growing in excess oxybiotin degraded a portion to products not utilizable by Saccharomyces cerevisiae. The loss of activity for the yeast suggested that no vitamers of oxybiotin accumulated during the degradation. The initiation of degrading activity was controlled by the pH of the growth medium and appeared during early stationary phase. Only cells grown in excess oxybiotin could degrade this biotin analogue. Nonproliferating cells grown previously in excess oxybiotin were able to convert biotin to vitamers (active for the yeast) as well as to degrade oxybiotin. Those grown in excess biotin also developed the ability to degrade oxybiotin as well as to convert biotin; however, in this case, the enzymes degenerated more rapidly. Cells grown with excessive amounts of either material were able to degrade desthiobiotin to products not available for the yeast. Both biotin conversion and oxybiotin degradation were found to have the same requirements for Mg and Mn ions. It was concluded that conversion of biotin to vitamers, and the degradation of oxybiotin or desthiobiotin are functions of the same on closely related enzyme systems.     

Biosynthesis of Biotin in Microorganisms VI. Further Evidence for Desthiobiotin as a Precursor in Escherichia coli

Biotin auxotrophs were isolated from Escherichia coli K-12. One of the mutants was unable to grow on desthiobiotin and accumulated a large amount of a vitamer in medium when growing on an optimal concentration of biotin. The production of the vitamer was inhibited in the presence of an excess amount of biotin. The vitamer was identified as desthiobiotin on the basis of biological activities, avidin combinability, and chromatographic characteristics. The mutant lacked the ability to convert desthiobiotin to biotin. These results further support the hypothesis that desthiobiotin is a normal intermediate in the biosynthesis of biotin in E. coli.     

Construction of a Biotin-Overproducing Strain of Serratia marcescens

We have isolated mutants resistant to acidomycin, a biotin analog, from Serratia marcescens Sr41. Strain SB304, resistant to 0.5 mg of acidomycin (frequently called actithiazic acid) per ml, produced 5 mg of d-biotin per liter of a medium containing sucrose and urea. Strain SB412, which was isolated from SB304 on a minimal agar plate containing 2 mg of acidomycin per ml and 0.1 mg of 5-(2-thienyl)-valeric acid per ml, produced 20 mg of d-biotin per ml. The two enzymes related to biotin synthesis were found to be released from biotin-mediated feedback repression in these mutants. Transductional analysis revealed that SB412 had acquired at least two mutations, one in the biotin operon locus and the other in an unknown locus distant from the biotin operon locus.     

Formation of Biotinyl-CoA Synthetase,the First Enzyme Involved in Microbial Biotin Degradation

Various bacteria which can grow on biotin as a sole carbon source were isolated from soil samples. These bacteria were classified into three groups according to the biotin-degrading pattern. Cell-free extracts from all these bacteria grown on biotin possessed high biotinyl-CoA synthetase activities. Cultural conditions for strain No. 166, the highest biotinyl-CoA synthetase producer, were examined. Biotinyl-CoA synthetase was induced in the presence of biotin, but not in the presence of any other carboxylic acids or biotin intermediates. The addition of lactose and yeast extract to the medium enhanced both the growth and the enzyme activity. The enzyme reaction for biotinyl-CoA synthesis required ATP, CoA and Mg²⁺ as well as biotin. From taxonomical studies, bacterium No. 166 was identified as a strain of Mycoplana.     

Studies on Production of Biotin by Microorganisms

The utilization of hydrocarbons by microorganisms was studied in many fields, but the production of biotin vitamers by hydrocarbon-utilizing bacteria has never been reported.

We have screened many hydrocarbon-utilizing bacteria which produce biotin vitamers in the culture broth. The effects of cultural conditions on biotin vitamers production by strain 5–2, tentatively assigned to the genus Pseudomonas, were studied.

More than 98% of biotin vitamers produced from hydrocarbons by strain 5–2 was chromatographically determined as desthiobiotin. As nitrogen source, natural nutrients were more effective than inorganic nitrogen sources. The production of biotin vitamers was increased under the condition of good aeration. Exogenous pimelic or azelaic acid enhanced biotin vitamers production by strain 5–2.

The production of biotin vitamers from n-alkanes, n-alkenes or glucose by an isolated bacterium, strain 5-2, tentatively assigned to the genus Pseudomonas, was investigated. Among these carbon sources, n-undecane was the most excellent for biotin vitamers production.

The biosynthetic pathway of biotin vitamers, especially desthiobiotin, from n-undecane was also studied. It was found by thin-layer and gas-liquid chromatographical methods that pimelic and azelaic acids were the main acid components in n-undecane culture.

This result, together with previously reported enhancement of biotin vitamers production by these acids, suggests that pimelic and azelaic acids may be the intermediates of biotin vitamers biosynthesis from n-undecane.     

Stimulation of guanylate cyclase and RNA polymerase II activities in HeLa cells and fibroblasts by biotin

HeLa cells cultured in a biotin-deficient medium showed reduced rates of protein synthesis, DNA synthesis and growth. Continuous synthesis is required for the increase in DNA synthesis observed upon addition of biotin to cells cultured in biotin-deficient medium. The addition of biotin to the biotin-deficient culture medium increased the activity of guanylate cyclase in both HeLa cells and fibroblasts. Both cell types cultured in biotin deficient medium showed reduced activity of RNA Polymerase II. The exogenous addition of biotin to the biotin-deficient cell cultures also resulted in increased activity of RNA Polymerase II in HeLa cells and fibroblasts. The maximal response was observed in 4 hours. Significant increase in enzyme activity was observed at 10^–8 M biotin in the culture medium. The growth promoting effect of biotin seems to involve stimulations of cellular guanylate cyclase and RNA Polymerase II activity.     

Regulation of enzyme synthesis in the arginine biosynthetic pathway of Pseudomonas aeruginosa.

In Pseudomonas aeruginosa the synthesis of only two out of eight arginine biosynthetic enzymes tested was regulated. Comparisons were made between the specific activities of these enzymes in bacteria grown on arginine or on its precursor, glutamate. N2-Acetylornithine 5-aminotransferase (ACOAT), an enzyme involved in both the biosynthesis and catabolism of arginine, was induced about 14-fold during growth of the organism on arginine as the only carbon and nitrogen source, and the anabolic ornithine carbamoyltransferase (aOTC), a strictly biosynthetic enzyme, was repressed 18-fold. Addition of various carbon sources to the arginine medium led to repression of ACOAT and to derepression of aOTC. Fructose, which supported only slow growth of P. aeruginosa, had a weak regulatory effect on the synthesis of the two arginine enzymes while citrate, a good carbon source for this organism, had a strong effect. The repression of ACOAT by citrate was not relieved by adding cyclic AMP to the medium. Under a variety of growth conditions leading to different enzyme activities, a linear relationship between the reciprocal of the specific activity of ACOAT and the specific activity of aOTC was observed. This inverse regulation of the formation of the two enzymes suggested that a single regulatory system governs their synthesis. Such a view was supported by the isolation of citrate-resistant regulatory mutants which constitutively formed ACOAT at the induced level and aOTC at the repressed level.