D-Mannitol Metabolism by Aspergillus candidus

Transcription

1 JOURNAL of BAcrEioLOY, Mar. 1969, p Copyright 1969 American Society for Microbiology Vol. 97, No. 3 Printed In U.S.A. D-Mannitol Metabolism by Aspergillus candidus GERALD W. STRANDBERG Northern Regional Research Laboratory, U.S. Department of Agriculture, Peoria, Illinois Received for publication 3 December 1968 Pathways of mannitol biosynthesis and utilization in Aspergillus candidus NRRL 305 were studied in cell-free extracts of washed mycelia prepared by sonic and French pressure cell treatments. A nicotinamide adenine dinucleotide-linked mannitol-l-phosphate (MIP) dehydrogenase was found in French pressure cell extracts of D-glucose-grown cells, whereas a specific mannitol-l-phosphatase was present in extracts prepared by both methods. The existence of these two enzymes indicated that mannitol may be synthesized in this organism by the reduction of fructose-6- phosphate. A specific nicotinamide adenine dinucleotide phosphate-linked mannitol dehydrogenase was also identified in both extracts. This enzyme may have been involved in mannitol utilization. However, the level of the mannitol dehydrogenase appeared to be substantially reduced in extracts from mannitol-grown cells, whereas the level of MIP dehydrogenase was increased. A hexokinase has been identified in this organism. Fructose-6-phosphatase, glucose isomerase, and mannitol kinase could not be demonstrated. Pathways for the biosynthesis and utilization of mannitol have been described for many organisms. Basically, there are two mechanisms. In some organisms fructose is reduced directly by an NADPH (reduced nicotinamide adenine dinucleotide phosphate)-linked mannitol dehydrogenase. In a second mechanism, fructose-6- phosphate (F6P) is reduced to mannitol-l-phosphate (M1P) via an NADH (reduced nicotinamide adenine dinucleotide)-linked MlP dehydrogenase. MlP is acted upon by a specific phosphatase (mannitol-1-phosphatase) that releases inorganic phosphate and mannitol. Mannitol utilization occurs by reversing these reactions; however, mannitol must be phosphorylated when MlP dehydrogenase is involved (3). Touster and Shaw (11) have reviewed the occurrence and function of polyol metabolizing systems in plants, animals, and microorganisms. Lee (4) reported a biosynth-etic pathway for mannitol in an Aspergillus sp. in which F6P was reduced to MlP by an NADH-linked MlP dehydrogenase. A mannitol-1-phosphatase was also found in this organism. Since a mannitol fermentation was developed at the Northern Laboratory (10) with the white Aspergillus, A. candidus NRRL 305, we wanted to determine the pathway by which mannitol was synthesized in this organism. The results of this work, as well as studies on the pathway of mannitol utilization, are reported here MATERIALS AND METHODS Culture. A. candidus NRRL 305 was grown in a medium consisting of Czapek's salts plus 5% dextrose (Difco) or D-minnitol (Mann Research Laboratories, Inc., New York, N.Y.) and 1% Difco yeast extract. (The sugars and polyols used were of the D configuration.) The spores from a yeast extract-malt extractagar slant were washed into 100 ml of medium in a 500-ml baffled Erlenmeyer flask. The flask was shaken in a New Brunswick rotary water bath shaker at 28 C (200 oscillations per min). After growth for 48 to 72 hr, the culture was added to 500 ml of the same medium in a baffled Fernbach flask which was shaken at 28 to 30 C in a New Brunswick Psychotherm shaker (200 oscillations per min). A 5-ml amount of a 1: 6 dilution of Antifoam 60 (General Electric, Silicone Products Division, Waterford, N.Y.) was added to the larger cultures. Reagents. MlP and sorbitol-6-phosphate were prepared by sodium borohydride reduction of mannose- 6-phosphate and glucose-6-phosphate (G6P), respectively (Sigma Chemical Co., St. Louis, Mo.), according to the method of Wolff and Kaplan (12). The sodium salt of F6P was prepared by treating the barium salt (Sigma Chemical Co. or Schwarz Laboratories, Inc., New York, N.Y.) with Bio-Rad cationic exchange resin (50W-X2-H+ form; Bio-Rad Laboratories, Richmond, Calif.) and neutralizing with NaOH. Detrminations. The product from the mannitol dehydrogenase reaction was determined by descending chromatography on Whatman no. 1 paper after passing the reaction mixture through a 1- by 5-cm column of Bio-Rad cationic exchange resin (50W-X2- H+ form, 50 to 100-mesh) to stop the reaction and remove cations. Mannitol and free sugars were de-

2 1306 STRANDBERG J. BACrRIOL. veloped in the methyl ethyl ketone-boric acid-acetic acid solvent (9:1:1) of Rees and Reynolds (8) or in a phenol-water solution (4:1, w/v). The sugars were detected by spraying the developed papers with 2% p-anisidine phthalate in 95% ethyl alcohol and heating at 110 C for 10 min. The spots were observed and marked under ultraviolet light (366 nm). The paper was then sprayed with a 1% periodate solution. Polyols appeared as white spots on a purple background. This chromatographic procedure was also used to confirm the identity of the sugar and polyol phosphates after treatment with alkaline phosphatase (EC ). Fructose and F6P were quantitatively determined by the method of Roe et al. (9). MIP was assayed by the periodiate oxidation method (12). Glucose was measured by alkaline ferricyanide reduction with an autoanalyzer (10). Protein was assayed by the procedure of Lowry et al. (6). Enzyme preparation and assays. The mycelia from 24-hr cultures were harvested by centrifugation or filtration and washed thoroughly with cold 0.1 M phosphate buffer (ph 7.0) or with cold distilled water (for phosphatase assay). The washed mycelia were broken by treatment with a prechilled French pressure cell (FPC; American Instrument Co., Inc., Silver Spring, Md.). The extracts were centrifuged at 0 to 4 C at 20,000 X g for 40 min, or as indicated. The supernatant fluid (crude extract) was used for enzyme assays. Mannitol and MlP dehydrogenase activities were assayed by following nicotinamide adenine dinucleotide phosphate (NADP) and nicotinamide adenine dinucleotide (NAD) reductions, respectively, at 340 nm. The reaction mixture consisted of 1.0 ml of substrate, 0.2 ml of buffer (0.2 M glycine-naoh, ph 9.8), 0.1 ml of NADP or NAD (0.01 M), and 0.05 ml of extract. The extract was added to start the reaction. NADPH and NADH oxidation were studied with 0.1 M sodium acetate (ph 6.0). The assays were carried out at 25 C. Mannitol-1-phosphatase activity was determined by measuring the inorganic phosphate released by the Fiske and Subbarow method. MlP (19 ;,moles) was incubated at 25 C with 0.7 ml of buffer, 0.2 ml of extract (water-washed cells), and distilled water to a volume of 2.0 ml. At 0 and 15 min, 0.4 ml was removed and pipetted into 1.0 ml of 10% trichloroacetic acid. The protein was removed by centrifugation. NADP reduction was used to detect hexokinase. A reaction mixture containing 5,umoles of MgCla2 7H20, 2 pmoles of adenosine triphosphate (ATP), 33,umoles of glycine (0.1 M glycine-naoh, ph 9.0), and 40 umoles of fructose or glucose; the extract was incubated at 25 C for 15 min. NADP (1.0 pmole) was added and the increase in optical density at 340 nm was observed. RESULTS AND DISCUSSION Yamada et al. (13) reported finding both NADlinked MIP dehydrogenase and mannitol-lphosphatase in crude extracts of A. niger and A. oryzae. Lee (4) found the same enzymes in the Aspergillus strain used in his work. An NAD-linked MlP dehydrogenase was detected in FPC extracts of glucose-grown A. candidus NRRL 305 as well (Fig. 1). It can be seen that the activity of this enzyme was much lower when the KHCOs-EDTA (ethylenediaminetetraacetate) buffer (ph 7.5) of Lee (4) was used. There was no NAD reduction with mannitol as the substrate and no NADP reduction with M1P. In addition, little or no enzyme activity was demonstrable in extracts treated for 10 to 15 min in an ice salt-water bath with a Sonifier (Branson Instruments Inc., Stamford, Conn.). Possibly the sonic treatment destroyed the enzyme or FPC breakage solubilized the enzyme. However, no activity was found in the pellet obtained by centrifuging the crude sonic extract. The level of protein of FPC extracts was between 5 and 6 mg of protein/ml, whereas that in sonic extracts was around 2 to 3 mg of protein/ml. A high level of endogenous NADH oxidation prevented an assay of the reverse reaction. In both sonic and FPC extracts, there was a specific mannitol-l-phosphatase. This enzyme E0@5 co 0.4 C U - A -....Aw.^...wAin A A-m' I I I I I FIG. 1. MIP dehydrogenase: effect of buffer. (A) 9.5 pmoles of MIP and glycine-naoh (0.2 ml; 0.2 m, ph 9.8); (B) water and glycine-naoh as in A; (C) 9.5,&moles of MIP, KHCO3-EDTA (0.2 ml; 0.1 m KHCO % EDTA, ph 7.5); (D) water and KHCO3-EDTA as in C. Each cuvette contained I pmole of NAD and FPC cell-free extract (0.37 mg of protein; extract centrifuged at 10,300 X g, 30 min). Total volume, 1.35 ml. IA B 0

3 .. VOL. 97, 1969 MANNITOL METABOLISM 1307 TABLE 1. Mannitol-l-phosphatase activity- 1.2 Buffer Buffer Time Inorganic Inorganic ~phosphate Phosphate released misn jumoles,umo Ics 0.1 M Sodium acetate a (ph 6.0) M Glycine-NaOH (ph 9.8) C.0.6 Water w=o.4 a Extract (5.4 mg of protein/ml) used was prepared by FPC breakage of water-washed mycelia. _~~*.ieh I*- 1-W= *0F-C* BC was active in water or ph 6.0 acetate buffer, but inactive at ph 9.8 (Table 1). No cofactors, such as Mg+, appeared to be required. An enzyme I I I fraction, precipitating between 0.3 and saturation of ammonium sulfate and dialyzed 16 hr against water, still retained mannitol- 1-phosphatase activity (as well as mannitol FIG. 2. Mannitol dehydrogenase. (A) Mannitol dehydrogenase activity). The crude enzyme was (0.1.mole), NADP (1.0,umole), and glycine-naoh; inactive with glucose-6-phosphate, F6P, sorbitol- (B) water, NaDP, and glycine-naoh; (C) mannitol, NAD (1.0 numole), and 6-phosphate, glycine-naoh; and (D) mannitol, mannose-6-phosphate. NADP, and Since both MlP dehydrogenase and mannitol- KHCOg-EDTA; (E) water, NaDP, and KHCOg-EDTA. The buffers were used in the same 1-phosphatase were present in the extracts from concentration as in the MIP dehydrogenase assay glucose-grown cells, it seemed reasonable to (see Fig. 1). assume that mannitol was being synthesized via F6P reduction as in other aspergilli (4, 13). TABLE 2. Mannitol dehydrogenase assay: However, a high level of NADP-linked mannitol requirements for ketose production dehydrogenase occurred in the extracts. The activity was present in both sonic and FPC extracts. Figure 2 shows that the enzyme is most active with glycine-naoh buffer (ph 9.8). Its activity is diminished considerably in KHCO3- EDTA buffer (ph 7.5). Enzyme activity was also reduced with 0.2 M carbonate-bicarbonate buffer (ph 9.8), 0.05 M tris(hydroxymethyl)aminomethane (ph 8.0 and 9.8), 0.2 M glycine-naoh (ph 8.5), and 0.1 M sodium acetate (ph 6.0). Horikoshi et al. (1) used 0.2 N glycine-naoh (ph 9.8) to detect mannitol dehydrogenase in conidia of A. oryzae. The enzyme was specific for mannitol; NADP reduction did not occur with arabitol, sorbitol, dulcitol, or ribitol. No NAD-linked activity was noted. Horikoshi et al. (1) reported the enzyme from A. oryzae conidia had both NADP and NAD activity. To secure further evidence for the action of this enzyme, ketose production was examined. Table 2 contains the requirements for ketose formation, and a time-course experiment is Sample Ketose5 Minus extractb... 3 Minus mannitol Minus NADP... 9 Minus buffer Complete Complete v Expressed as micrograms of ketose per sample. A 1-ml amount of Resorcinol reagent (9) was added to stop the reaction after 30 min at 25 C. b FPC extract contained 5.4 mg of protein/ml. plotted in Fig. 3. Fructose was identified chromatographically as the end product of the reaction. NADPH oxidation was also demonstrated with fructose as the substrate. That the organism possessed a mannitol dehydrogenase as well as an MlP dehydrogenase raised the question as to which enzyme, if not both, is operable in mannitol biosynthesis. The logical synthetic pathway appeared to be the

4 1STRANDBERG 1308 J. BACrERIoL. 7n- *Uv 60[ 50O 40 m s FiG. 3. Mannitol dehydrogenase: ketose assay. A reaction mixture containing 8.0 ml of mannitol (0.1 M), 1.6 ml of glycine-naoh (0.2 m, ph 9.8), 0.8 ml of NADP (0.01 m), and 0.4 ml of FPC extract (5.4 mg of protein/ml) was incubated at 25 C. Samples (1.0-mi) were pipetted at intervals into Resorcinol reagent (9). reduction of F6P, because of the mannitol-1- phosphatase. The most plausible explanation for the presence of this enzyme would be its participation in mannitol production from MlP since it is specific for this substrate. If mannitol dehydrogenase were active in mannitol biosynthesis, there would have to be a source of fructose. No evidence for fructose-6-phosphatase or glucose isomerase (7) was found. If mannitol dehydrogenase were active in mannitol utilization, then the organism would need a mechanism of incorporating fructose into its metabolism. Evidence is given in Fig. 4 for the existence of such a mechanism in extracts of this organism (i.e., an enzyme capable of phosphorylating fructose). This enzyme, probably a hexokinase, was demonstrated by means of an NADP-reduction assay based on the production of G6P from newly formed F6P by phosphoglucoisomerase (EC ). NADP reduction occurs via G6P dehydrogenase (EC ). The latter reaction was readily demonstrated in the extracts with G6P and NADP. Phosphoglucoisomerase has been reported in Aspergillus (2), along with the other enzymes involved. These 1.2 E 1.0, 0.8 i o.~ <O.2 An Vs -.&-Fructose-Reg. - # 61Glucose-Reg. - // H20-Reg. -s*/fructose-pi I,.,,,.'-GIucose-PI If.; * lo4ructose-pi (minus ATPJ 1 2 Time, 3 min. 4 U.Ul------F Fio. 4. Hexokinase activity. Reg = extract from washed mycelia (0.1 M phosphate, ph 7.0) without incubation before breakage (8.7 mg of protein/ml of extract; 0.09% glucose). PI = extract from washed mycelia which were shaken at 28 C for 2 hr before breakage (3.2 mg of protein/ml; 0.024% glucose). NADP (1.0 mole) was added after 15 min incubation at 25 C. Total volume, 3.0 ml. data present indirect evidence for the existence of phosphoglucoisomerase in A. candidus extracts as well. The reaction was ATP-dependent (Fig. 4). A high level of endogenous activity was removed by preincubating washed mycelia so that excess glucose could be metabolized before breaking the cells (Fig. 4). This procedure reduced the amount of glucose in the extracts by more than 70%. Mannitol kinase could not be demonstrated by a similar procedure. NAD reduction by MlP dehydrogenase was used as an assay for MIP. The alkaline ph (9.0) of the reaction mixture should have prevented the hydrolysis of MIP by mannitol-1-phosphatase. Recently, Lee (5) reported a mannitol acetyl phosphate phosphotransferase in Aspergillus (Upjohn 4177) and in A. candidus NRRL 305. Because there was no mannitol dehydrogenase, mannitol kinase, pyrophosphate phosphotransferase, or mannitol phosphoenol pyruvate phosphotransferase in the former organism, Lee (5) suggested that this new enzyme serves to phosphorylate mannitol so that it can be incorporated via MlP dehydrogenase. A Michaelis constant of 2.5 M for mannitol was found. This unusually high value suggests that mannitol may not be the true substrate for the enzyme. We have been unsuccessful in our attempts to identify this enzymatic activity in extracts of mannitol-grown A. candidus NRRL 305. The enzymes reported so far were demonstrated in glucose-grown cells. The question remained as to which pathway was operable in mannitol biosynthesis. An examination of extracts from man-

5 VOL. 97, 1969 MANNITOL METABOLISM 1309 nitol-grown cells revealed both dehydrogenases and the phosphatase as well. Somewhat more perplexing was the finding that MlP dehydrogenase was present to a higher degree than in glucose-grown cells (Table 3). The relationship of the two dehydrogenase activities is unusual. If MIP dehydrogenase is active in mannitol biosynthesis, then why is it present to a much greater extent in mannitolgrown cells? If it is active also in mannitol utilization, a satisfactory mechanism for mannitol phosphorylation remains to be established. Similarly, mannitol dehydrogenase is reduced in activity where it would be expected to be high (i.e., in mannitol-grown cells). Since a mechanism for fructose formation has not been found, there is no evidence that mannitol dehydrogenase is active in mannitol biosynthesis. The mechanism controlling the production of TABLE 3. Mannitol and M-1-P dehydrogenase activities in FPC extracts from glucose- and mannitol-grown cells Carbon source Enzyme assayed Specific activity" Glucose Mannitol de (0.45)a hydrogenase MlP dehydrogenase 0.18 (0.21) Mannitol Mannitol de- 0. lob hydrogenase MlP dehydrogenase O.58b a Expressed as A optical density per minute per milligram of protein. Data for glucose-grown cells was taken from Fig. 1 and 2. Values in parentheses are averages of three additional experiments. baverage of four experiments. these enzymes and the pathway of either biosynthesis or utilization of mannitol remain undetermined. The answers to these problems lie in the detection of those enzymes that introduce the proper substrates for the respective dehydrogenases. The metabolic regulation of these enzymes should be an interesting phenomenon as well. LITERATURE CITED 1. Horikoshi, K., S. lida, and Y. Ikeda Mannitol and mannitol dehydrogenases in conidia of Aspergillus oryzae. J. Bacteriol. 89: Jagannathan, V., and K. Singh Carbohydrate metabolism in citric acid fermentation. II. The glycolytic enzymes of Aspergillus niger. Enzymologia 16: Klungsoyr, L Mannitol kinase in cell-free extracts of Escherichia coli. Biochim. Biophys. Acta 122: Lee, W. H Carbon balance of a mannitol fermentation and the biosynthetic pathway. Appl. Microbiol. 15: Lee, W. H Mannitol acetyl phosphate phosphotransferase of Aspergillus. Biochem. Biophys. Res. Commun. 29: Lowry, 0. H., N. S. Rosebrough, A. L. Farr, and R. J. Randall Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: Marshall, R. O., and E. R. Kooi Enzymatic conversion of D-glucose to D-fructose. Science 125: Rees, W. R., and T. Reynolds A solvent for the paper chromatographic separation of glucose and sorbitol. Nature 181: Roe, J. H., J. H. Epstein, and N. P. Goldstein A photometric method for the determination of inulin in plasma and urine. J. Biol. Chem. 178: Smiley, K. L., M. C. Cadmus, and P. Liepins Biosynthesis of mannitol from D-glucose by Aspergillus candidus. Biotechnol. Bioeng. 9: Touster, O., and D. R. D. Shaw Biochemistry of the acyclic polyols. Physiol. Rev. 42: Wolff, J. B., and N. 0. Kaplan D-Mannitol-l-phosphate dehydrogenase from Escherichia coil. J. Biol. Chem. 218: Yamada, H., K. Okamoto, K. Kodama, F. Noguchi, and S. Tanaka Enzymic studies on mannitol formation by Piricularia oryzae. J. Biochem. 49: