Increased Sensitivity of the Microbiological Assay

Transcription

1 APPLID MICROBIOLOGY, Sept. 197, p Copyright 197 American Society for Microbiology Vol., No. 3 Printed in U.S.A. Increased Sensitivity of the Microbiological Assay for Biotin by Lactobacillus plantarum JAMES R. WALLER Department of Microbiology, University of North Dakota, Grand Forks, North Dakota 81 Received for publication May 197 Addition of Tween 8 to biotin assay medium containing acid-hydrolyzed casein as the amino acid source caused marked growth of Lactobacillus plantarum ATCC 81 in the absence of added biotin. This growth-promoting activity could be eliminated by treating the "vitamin-free" Casamino Acids (Difco) with activated charcoal (Darco G-6) at ph 3. for 3 to 6 min. Incorporation of Tween 8 and charcoal-purified Casamino Acids (PCA) into the assay medium (.8 g and 7 g, respectively, per liter of singl? strength medium) in place of unpurified Casamino Acids resulted in a medium in which L. plantarwn responded to 3 to times less biotin over an extended linear response range (1.3 logs versus 1. log) than was required for similar growth in the standard medium. Endogenous growth in the modified medium was absent if the inoculum used was of low density, if it was prepared from biotin-deficient cells, and if the reagents used were free from contaminating traces of biotin. Assays of biological materials for biotin content using the standard medium and the Tween 8-PCA-modified medium resulted in nearly identical values for all samples tested. The linear growth response of Lactobacillus plantarum ATCC 81 to biotin in a modified Wright-Skeggs medium (7, 1, 1) extends from about.3 to.3 ng of the vitamin per ml of assay medium. Currently, the sensitivity of the system is increased by reducing the assay volume from milliliter to microliter (7) or submicroliter (6) quantities. These methods require specialized equipment and time-consuming techniques. The method reported here achieves marked increases in sensitivity and range of linear response without resorting to complicated techniques or equipment. MATERIALS AND METHODS Organism. Lactobacillus plantarum ATCC 81 was maintained as stabs in APT agar (Difco) and transferred into fresh medium every 1 to 1 days. Preparation of inoculum. The single-strength, basic assay medium ( to 1 ml amounts) described below supplemented with.1 to. ng of biotin per 1 ml was inoculated directly from the APT stock culture, incubated overnight at 3 C, washed three times with ml of sterile distilled water, and resuspended in sterile distilled water to the desired turbidity. The turbidity of the suspension (in 16 mm Pyrex tubes) was read in a Spectronic- at 66 nm and adjusted to 6% i % unless otherwise stated. One drop of this suspension (ca..6 ml) was added to each assay tube from a sterile, plugged.-ml pipette. The same pipette was used to inoculate all tubes. Assay medium. The basic assay medium was that of Wright and Skeggs (1) modified (7, 1) by adding folic acid (. ng/liter), by using vitamin-free acidhydrolyzed casein (Difco Casamino Acids-7. g/liter) as the source of amino acids, and by omitting NaCl since Casamino Acids contain 38% NaCl. A -mg amount of DL-tryptophan and a 1-mg amount of L-cysteine HCI H (Sigma Chemical Co., St. Louis, Mo.) were added per liter since these amino acids are absent from acid-hydrolyzed casein. Purification of acid-hydrolyzed casein. Fifteen per cent solutions of vitamin-free Casamino Acids were adjusted to ph 3. with 6 N HSO, combined with activated charcoal (Darco G-6), mixed continuously on a magnetic stirrer at room temperature for 3 to 6 min, filtered through Whatman no. filter paper on a Buchner funnel, and readjusted to ph 6.8 to 7.. Microbiological assay procedure. Assays were carried out in Pyrex culture tubes (16 by 1 mm) covered with stainless-steel caps (Bellco Glass, Inc., Vineland, N.J.). To each ml of a suitable biotin concentration or unknown sample, ml of double strength assay medium was added. The tubes were sterilized by autoclaving at 11 C for 7 min and cooled rapidly in cold water to avoid excessive "browning"; each tube was then inoculated with one drop of the L. plantarwn inoculum suspension described above. All tubes were incubated in a water bath at 3 C for to hr. Just prior to reading, all tubes were placed in an ice-water bath to stop growth in all tubes at the same time. Turbidities were measured at 66 nm (no. 66 filter, peak transmission 6 to 67 nm) in a Klett-Summerson photoelectric colorimeter adjusted to zero on an uninoculated medium blank. 8

2 86 WALLER APPL. MICROBIOL. Chemicals. Most chemicals were obtained from either Sigma Chemical Co., St. Louis, Mo., or Mallinckrodt Chemical Works, St. Louis, Mo. Avidin was purchased from Nutritional Biochemicals Corporation, Cleveland, Ohio. RESULTS Growth stimulation bytween 8. When Tween 8 was added to the Wright-Skeggs medium containing vitamin-free Casamino Acids as the amino acid source, endogenous growth (see Table 1 for definition of endogenous) of L. plantarum increased from 3 to 38 Klett units. In addition, maximum response to biotin occurred at a much lower vitamin level than was observed in the absence of the Tween (Table 1). It appeared that the presence of Tween 8 in the medium enabled L. plantarum to respond more efficiently to much lower concentrations of biotin. This was particularly noticeable at the biotin levels of.1 and.1 ng. Purification of Casamino Acids with activated charcoal. Since Tween 8 seemed to stimulate growth in the presence of small amounts of biotin rather than replace the biotin requirement as has been suggested by others (3, 13, 1), it seemed logical that the increased endogenous growth in the presence of Tween 8 might be due to small amounts of biotin present in the amino acid preparation. Treatment of Casamino Acids (vitamin free) with activated charcoal significantly reduced the endogenous growth in the presence of Tween 8 (Table ). These data also show that purification was complete in 3 min and that 1 g of activated charcoal (Darco G-6) per 1 g of Casamino Acids effected maximum TABLE 1. Effect of Tween 8 on the response of Lactobacillus plantarum to biolin in the presence of vitamin-free Casamino Acids (Difco)a Biotin concn (ng/1 ml) Without Tween 8 Klett readings With Tween 8a 3h 38b a Basic assay medium (modified Wright-Skeggs) supplemented with Tween 8 (1 g/liter) and biotin as indicated was used in this experiment. b Endogenous growth: the amount of growth supported by a medium, in the absence of added growth factor, which is indicative of the amount of growth factor present as a contaminant in the components of the medium. TABLE. Purification of vitamin-free acidhydrolyzed casein with activated charcoala Time with charcoal at ph 3. Grams of activated charcoalb per 1 g of casein hydrolysatec min d aa 1-ml amount of 1% solution of casein hydrolysate was stirred continuously at room temperature for the designated lengths of time with the indicated amounts of activated charcoal as outlined in Materials and Methods. The purified mixtures were added to single strength basic assay medium devoid of Casamino Acids, but containing 1 mg of Tween 8 per ml, such that the final concentrations were.1 g/ml. b Darco G-6. c Difco Vitamin Free Casamino Acids. d Klett readings. Endogenous was 1 Klett units. Endogenous without Tween 8 was Klett units. purification. However, it was found that some lots of Casamino Acids required more activated charcoal to effect maximum purification. In no case, however, was more than g of Darco G-6 per 1 g of Casamino Acids required. Therefore, treatment of 1 g of vitamin-free Casamino Acids (as a 1 or 1% solution) with g of Darco G-6 at ph 3. for 3 to 6 min at room temperature was established as an adequate purification procedure. Attempts to purify vitamin-free, salt-free casein hydrolysate (Nutritional Biochemicals Corp., Cleveland, Ohio) by the same general procedure were made, but the results obtained were not as satisfactory as those obtained with Difco Casamino Acids. For this reason, the latter product was used routinely as the amino acid source. Similarly, another activated charcoal, Norite A, was used as the purifying agent. Although Norite A removed the biotin as efficiently as Darco G-6, the Casamino Acids purified with Norite A did not always support as much maximum growth as did those purified with Darco G-6. The differences were not great, but it was felt that Darco G-6 was the better choice. The endogenous growth present even after treatment with activated charcoal (Table ) will be discussed later. Titration of purified Casamino Acids (PCA) for growth. Since activated charcoal is a nonselective adsorbant, the possibility existed that substances other than biotin were being removed from the amino acid preparation and that this removal might necessitate addition of more

3 VOL., 197 INCREASED SENSITIVITY OF BIOTIN ASSAY 87 TABLE 3. Titration of purified Casamino Acids (PCA) for growth stimulationa Amt of PCA/ml Klett readings No biotin added Biotin added (.3 ng/1-ml tube) a A 1-ml amount of a 1% solution of Difco Casamino Acids was treated with g of Darco G-6 for 3 to 6 min at room temperature. Basic assay medium was supplemented with.8 g of Tween 8 per liter, and Casamino Acids were omitted. purified Casamino Acids (PCA) to the growth medium than was necessary with the unpurified material. Table 3 shows the results of increasing PCA concentrations on the response of L. plantarum in Tween 8-containing basic assay medium from which Casamino Acids had been omitted. The endogenous response did not increase beyond.18 g of PCA per ml of medium; however, in the presence of added biotin, growth increased until the medium contained.7 g of PCA per ml. This corresponds to 7 g of PCA per liter of single strength assay medium, a 3.6- fold increase over the amount of unpurified Casamino Acids required in the basic assay medium. Subsequently, the modified basic assay medium contained 7 g of PCA per liter of single strength medium. Effects of various Tweens and oleic acid on the growth response of L. plantarum to biotin. Tweens,, and 8 were tested for growth-stimulating activity in basic assay medium containing 7 g of PCA in place of untreated Casamino Acids (Table ). In the absence of biotin, no Tween caused any significant stimulation of growth. When biotin was added to the assay medium, only Tween 8 stimulated growth of L. plantarwn. In the presence of 1 ng of biotin per tube, maximum stimulation of growth was observed with mg of Tween 8 per tube. No inhibition of growth was observed with Tween 8 up to and including mg per tube. Eight milligrams of Tween 8 per tube (.8 g/liter) was chosen as the concentration to be used in the modified basic assay medium. The modified basic assay medium is composed of modified Wright-Skeggs medium (see above) supplemented with.8 g of Tween 8 and in TABLE. Effect of Tweens,, and 8 on stimulation of growtha Amt of biotin per tube ng No biotin.1 1. Tween tested Amt (mg) of Tween per 1-ml tube. 1 1 Ob l a Basic assay medium, in which the Casamino Acids were replaced by 7 g of PCA, was supplemented with biotin and the various Tweens as indicated. b Klett readings. which the Casamino Acids have been replaced with 7 g of Darco G-6-purified Casamino Acids per liter of single strength medium. Since oleic acid is known to stimulate the biotin response of lactic acid bacteria (3, 13, 1), the effects of oleic acid alone and in combination with Tween as a detoxifying agent with and without Tween 8 in the modified basic assay medium were studied (Table ). Maximum response of L. plantarum to added biotin was induced by to 6,tg of oleic acid per tube (1 ml). Also, Tween in combination with oleic acid effected an additional growth-stimulating effect, probably through detoxification of the fatty acid (1). The known toxicity of oleic acid was evident at the higher levels, both in the presence and absence of Tween. Most interesting were the data pertaining to the stimulatory activity of Tween 8 with and without oleic acid and Tween. Quantities of oleic acid causing maximum growth stimulation (,ug/tube) did not affect in any way the stimulatory effect of Tween 8 on the response of L. plantarum to biotin. Although Tween appeared to increase slightly the growth response to biotin in the presence of Tween 8, it was near the margin of experimental error and was not considered significant. Effect of inoculum. Since the Tween 8-PCAmodified medium allowed detection of very low concentrations of biotin, the amount of the vitamin being carried over in the inoculum cells was

4 88 WALLER APPL. MICROBIOL. TABLE. Effects ofoleic acid, Tween, and Tweeni 8 on the growth respontse of L. plantarum to biotina TABLE 6. Effect of inoculum on the growth of L. plantarum in Tweent 8-PCA-modified basic assay mediuma Amt (mg) of oleic acid per 1-ml tube Without Tween Klett readingsb With Tween (1 mg/tube) Biotin levels in inoculum medium (ng/1 ml)b Optical density of inoculum cell suspension , 1,, plus 8 mg of Tween 8 plus 8 mg of Tween 8 plus 1 ng of biotin c C a In the basic assay medium Casamino Acids were replaced by 7 g of PCA, and the medium was supplemented with.1 ng of biotin/1 ml and with oleic acid, Tween, and Tween 8 as indicated. b Since increasing amounts of oleic acid caused the autoclaved medium to become increasingly dark, Klett readings were taken only after adjusting the instrument to zero on uninoculated medium blanks. c Maximum growth attainable at nonlimiting biotin concentrations. of considerable importance. The biotin content of L. plantarum is known to vary with the amount of biotin in the growth medium (8); therefore, cells were grown in various concentrations of biotin and prepared for use as inocula. Also, since the total amount of biotin added with the inoculum cells would depend upon the number of cells added to each tube, the density of the inoculum suspensions was varied. Table 6 summarizes the results of these experiments. It can be seen that growth increased directly with increasing inoculum density. Also, except for the highest level, as the amount of biotin with which the cells were grown increased, so did the growth response in the assay medium. These results occurred both in the absence of any biotin in the assay medium (endogenous growth, a real indicator of the amount of biotin carry-over in the inoculum cells) or when limiting amounts of biotin had been added to the assay medium. It might be argued that at the lower inoculum levels [optical density (OD):.1 and.] the low endogenous readings merely reflected a lagging growth No biotin added to assay medium.1 OC Biotin added (.3 ng per 1 ml of assay medium) a Modified basic assay medium contained, per liter:.8 g of Tween 8 and 7 g of PCA in place of Casamino Acids in addition to all other components of the modified Wright-Skeggs medium. b Inoculum cells were grown with the indicated amounts of biotin, washed, and resuspended in sterile distilled water to the optical density indicated. c Klett readings. response due to the difference in inoculum size. Since turbidity varied only slightly with inoculum size when the medium containing added biotin was inoculated with cells grown in 1 ng of biotin per 1 ml, that posssibility was eliminated. The reason for the lower turbidity values obtained when cells grown in the highest biotin concentration were used as inoculum is not understood, but it likely is related to repression of the biotin transport system (1; Waller, unpublished data) or formation of active bound coenzyme (). Effect of avidin on endogenous growth. Usually endogenous turbidity in Tween 8-PCA-modified assay medium produced Klett readings of to with extremes varying from less than 1 to 1. Since endogenous values near zero were obtained even with moderate inoculum sizes (Table 6), the reason for variable endogenous growth was of interest. Biotin-deficient inoculum cells were so low in intracellular biotin content (Table 7) that the inoculum size used (OD:.1) could not have carried sufficient biotin to support the growth often observed. Thus, it seemed reasonable to believe that some medium component was contaminated with small amounts of biotin. To test this hypothesis, growth in Tween 8- PCA-modified medium was measured in the presence or absence of the specific biotin-binding

5 VOL., 197 INCREASED SENSITIVITY OF BIOTIN ASSAY 89 TABLE 7. Bounid biotin content of Lactobacillus plantarum cells grown with various amounts of biotin Biotin in growth medium Amt (ng) of bound biotin per (ng/1 ml) mg of cells (dry wt) protein avidin. Endogenous growth without avidin was 7 Klett units; when. unit of avidin (one unit of avidin will bind 1 jig of biotin) was added per 1 ml of medium, no growth occurred; endogenous growth was eliminated completely by avidin. Since avidin could not have bound intracellular-bound biotin, it must have combined with free vitamin present in the medium as a contaminant. Comparison of standard and Tween 8-PCA media for the assay of biotin. Various natural, biological materials were assayed for biotin with L. plantarum by using the basic assay medium (modified Wright-Skeggs, see Materials and Methods) or the Tween 8-PCA medium (Table 8). The biotin content of each sample assayed nearly identical with both systems, and drift between sample volumes was, if anything, less with the Tween 8-PCA assay system. Sample volumes required for significant response of the assay organism were 3- to 1-fold less with the Tween 8-PCA medium. Plots of biotin assay standard curves obtained by using the basic assay medium or the Tween 8-PCA-modified medium (Fig. 1) reveal ranges of linearity extending from.3 to 3. ng per 1-ml tube and from.9 to.3 ng per 1-mi tube, respectively. Both range of linearity and sensitivity of response were increased significantly in the Tween 8-PCA-modified biotin assay medium. Effect of Tween 8 on the L. plantarum assay for pantothenic acid. For comparison purposes, the effect of Tween 8 on the L. plantarum assay for pantothenic acid was tested. The basic assay medium (modified Wright-Skeggs) with pantothenic acid omitted and biotin added was used as the assay medium. Plots of the data obtained for pantothenic acid standard curves run in the basic assay medium with or without Tween 8 supplementation (.8 g/liter) are shown in Fig.. As with biotin, Tween 8 intensified the response of L. plantarum to the vitamin resulting in the standard curve linearity being shifted to significantly lower vitamin concentrations. The recommended assay medium () for the microbiological TABLE 8. Comparison of biotin values of biological materials obtained with standard and Tween 8 assays Material assayedc Corn meal Macaroni Rice Bacterial cells Rat cardiac muscle (free biotin) Bacterial cell extracts (free biotin) Sample 1 3 Standard assay' Vol of Amt (ng) Vol of sample of biotin sample sayd mg pgy t Tween 8 assayb Amt (ng) of biotin asper asprm sae Wr t a Basic assay medium (modified Wright-Skeggs) was used for these assays. bthe Tween 8-PCA-modified basic assay medium was used for this assay. c Corn meal, macaroni, and rice were ground to fine powders, hydrolyzed with 6 N HSO at 11 C for I hr to release bound biotin into free, assayable form. A pellet of bacterial cells was suspended in 6 N HSO and hydrolyzed as above. All hydrolysates were neutralized and diluted prior to assay. Free biotin was extracted from cardiac muscle and bacterial cells by immersing aqueous suspensions in boiling water for 1 min. Cell debris was removed by centrifugation, and the clarified extracts were assayed for biotin as indicated. 3 3IQ. - PCA TWEEN 8 O ~~~~~~Enogeu-, 1 V-- e 1 SIC ASSAY NfDIUN S g g hin~~~~~~~~~~~~~m T"9K QF h Pr ng BIOTIN / TUBE ( 1 ml) FIG. 1. Biotin standard curves in basic assay medium and Tween 8-PCA-modified medium. Assays were conducted as described in Materials and Methods. Endogenous readings ( and 1) refer to the amount of growth occurring in inoculated tubes of the respective assay media in the absence of added biotin. Such readings are indicative of the amount of biotin present in the other medium components as a contaminant.

6 9 WALLER APPL. MICROBIOL. y 31 - z + TWEEN 8, -TWEEN 8 - Al *' IC DO,/ IX IO1 1l 1 ng PANTOTHENIC ACID/TUBE (1 ml) FIG.. Pantothenic acid standard curves in basic assay medium with and without Tween 8. The basic assay medium was modified to include biotin and exclude pantothenic acid. Assays were conducted as described in Materials anid Methods. determination of pantothenic acid using L. plantarum contains Tween 8. DISCUSSION Fatty acids, especially oleic acid, long have been known to interfere with microbiological assays for various vitamins (1, 9, 11, 13). The stimulatory effects of fatty acids can be overcome by removing the fatty acid prior to assay or by incorporating it into the assay medium. In the case of pantothenic acid, incorporation of oleic acid or its polyoxyethylene sorbitan derivative (Tween 8) in maximum stimulatory concentrations has effectively overcome most of the nonvitamin-induced growth stimulation (11). The microbiological assay procedures for biotin require that stimulatory fatty acids be removed before assay to avoid nonspecific stimulation. If biotin is present in the assay medium or unknown sample, even in quantities too low to detect under normal conditions, the presence of certain fatty acids, especially oleic acid, induces considerable growth response of the assay organism (3, 13, 1). The stimulatory effect of oleic acid can be overcome if Tween 8 is incorporated into the assay medium (Table ). Also, the data in this report show that L. plantarum has an absolute requirement for biotin, even in the presence of Tween 8 and aspartic acid if all traces of biotin are removed from the medium (Table 3,, 6). Since some reagent grade chemicals, e.g., tryptophan (), contain significant amounts of biotin, chemical stocks must be suspect if high endogenous values are encountered. The condition and density of the inoculum cells are important factors in establishing low endogenous readings in this biotin assay system (Table 6). Although the cell density of the inoculum can be so reduced that each tube is inoculated with only a few hundred cells, thereby avoiding detectable biotin carry-over with the inoculum cells, such a reduction is accompanied by a lag period of nearly hr or more and requires incubation periods of to 8 hr to attain maximum growth. Use of larger inocula reduces the incubation time required to reach near maximum growth to 18 to hr. The amount of bound biotin contained within the bacterial cells depends upon the biotin content of the medium in which they were grown (8; Table 7). For this reason, growth of inoculum cells in a biotin-deficient medium was required to obtain the lowest possible endogenous readings and still utilize a large enough inoculum to bring each inoculated tube into the stationary growth phase within 18 to hr. On the basis of the data presented in this report, it is recommended that the microbiological assay for biotin using L. plantarum ATCC 81 be conducted with the modified Wright-Skeggs medium containing Tween 8 and Darco G-6 purified Casamino Acids as described to increase assay sensitivity and range and to overcome fatty acid-induced stimulation of the biotin growth response. Use of this medium in the micro and ultramicro techniques of Glick et al. (7) and Glick and Ferguson (6) should provide an added dimension to the sensitivity of these techniques. ACKNOWLEDGMENTS The author is indebted to Judy Zimmer-man and Sharon E. Zimmerman for technical assistance and to Ralph Germinario and Gordon Fillipi for independent confirmation of the workability of the system. These studies were supported by a grant from the National Science Foundation (GB-) and a Faculty Research grant from the University of North Dakota (NSF Institutional Funds). LITERATURE CITED 1. Bauernfeind, J. C., A. L. Sotier, and C. S. Boruff. 19. Growth stimulants in the microbiological assay for riboflavin and pantothenic acid. Ind. Eng. Chem., Anal. Ed. 1: Birnbaum, J Coenzyme repression of acetyl-coa carboxylase by (+)-biotin. Arch. Biochem. Biophys. 13: Broquist, H. P., and E. E. Snell Biotin and bacterial growth. I. Relation to aspartate, oleate and carbon dioxide. J. Biol. Chem. 188:31-.. Calcium pantothenate assay, p In Pharmacopeia of the United States, Revision XVI, 196. Mack Publishing Co., Easton, Pa.. Croom, J. A., J. J. McNeill, and S. B. Tove Biotin deficiency and the fatty acids of certain biotin-requiring bacteria. J. Bacteriol. 88: Glick, D., and R. B. Ferguson Histochemistry. LXVII. Microscopic microbiological assay. Deternmination of

7 VOL., 197 INCREASED SENSITIVITY OF BIOTIN ASSAY 91 biotin to 11 gram. Proc. Soc. Exp. Biol. Med. 19: Glick, D., H. C. Lichstein, R. B. Ferguson, and R. M. Twedt Studies in histochemistry. LIII. Microbiological assay in quantitative histo- and cytochemistry. Proc. Soc. Exp. Biol. Med. 99: Lichstein, H. C., and J. R. Waller Factors affecting the accumulation of biotin by Lactobacillus arabinosus. J. Bacteriol. 81: Neal, A. L., and F. M. Strong Microbiological determination of pantothenic acid. Further studies. Ind. Eng. Chem., Anal. Ed. 1: Rogers, T. O., and H. C. Lichstein Regulation of biotin transport in Saccharomyces cerevisiae. J. Bacteriol. 1: Skeggs, H. R., and L. D. Wright. 19. The use of Lactobacillus arabinosus in the microbiological determination of pantothenic acid. J. Biol. Chem. 16: Waller, J. R., and H. C. Lichstein Biotin transport and accumulation by cells of Lactobacillus plantarum. I. General properties of the system. J. Bacteriol. 9: Williams, V. R., and E. A. Fieger Oleic acid as a growth stimulant for Lactobacillus casei. J. Biol. Chem. 166: Williams, W. L., H. P. Broquist, and E. E. Sneil Oleic acid and related compounds as growth factors for lactic acid bacteria. J. Biol. Chem. 17: Wright, L. D., and H. R. Skeggs. 19. Determination of biotin with Lactobacillus arabinosus. Proc. Soc. Exp. Biol. Med. 6:9-98.