Metabolism of Valine by the Filamentous Fungus Arthrobotrys conoides'

Vol. 103, No. 1 Printed in U.S.A.

JOURNAL OF BACTERIOLOGY, July 1970, p. 131-139 Copyright 0 1970 American Society for Microbiology

Metabolism of Valine by the Filamentous Fungus Arthrobotrys conoides' RISHAB K. GUPTA2 AND DAVID PRAMER Department ofBiochemistry and Microbiology, Rutgers-The State University, New Brunswick, New Jersey, 08903

Received for publication 25 March 1970

Uptake of valine by Arthrobotrys conoides was an active process and was independent of its incorporation into cellular protein. Chemical fractionation of cells supplied with '4C-L-valine for different time intervals revealed that the amino acid initially entered a pool of metabolic intermediates and was extractable with cold trichloroacetic acid. After a 4-min interval, some intracellular valine was incorporated into cell proteins, but most underwent metabolic transformation to a variety of products that included carboxylic acids and other amino acids. Carbon derived from valine was not localized in the lipid or nucleic acid fraction of cells, but some was completely oxidized and recovered as metabolic 14CO2. Autoradiograms of paper and thin-layer chromatograms of acid hydrolysates of cellular protein identified the following amino acids as having originated from valine: glutamate, aspartate, alanine, and leucine. Similar analysis of cold trichloroacetic acid extracts established that '4C supplied as L-valine had been transformed also to a-ketoisovalerate, isobutyrate, propionate, succinate, malate, oxalacetate, pyruvate, and c-ketoglutarate. Pathways for transformation of the carbon skeleton of valine to various metabolic products are proposed.

Valine has been shown to induce trap formation by the predaceous fungus Arthrobotrys conoides which captures nematodes on adhesive hyphal loops (L. M. Wootton and D. Pramer, Bacteriol. Proc., p. 75, 1966). Specific and absolute amino acid requirements are uncommon among fungi, but it is possible to induce auxotrophic mutations (1). Furthermore, some apparent amino acid requirements are not necessarily for the amino acid as such; for example, a putative need for methionine and cysteine may be satisfied by sulfur, and cysteine promotes growth of some fungi by maintaining low oxidation-reduction potentials (12). Like most filamentous fungi, A. conoides has no absolute or specific amino acid requirement for growth. It develops in a synthetic medium containing glucose, nitrate-nitrogen, minerals, biotin, and thiamine (5). When an attempt was made to explain the morphogenic activity of valine by reference to the literature, it became evident that little was known of valine metabolism by filamentous fungi. Our previous studies indicated that the

amino acid was actively absorbed and accumulated against a concentration gradient by A. conoides (11), and the present investigations were performed to localize the absorbed valine and to determine its metabolic fate. This work was intended to extend our very limited knowledge of amino acid metabolism by filamentous fungi in general. The use of a nematode-trapping fungus as test organism was expected to broaden the significance of results and provide a biochemical basis for future understanding of morphogenesis as manifested by a change in fungus form. MATERIALS AND METHODS Organism. Germinated spores of A. conoides were used throughout this investigation. The cells (germinated spores) were prepared as described previously (11). Culture conditions. Spores were used at a concentration of 1.0 mg (dry weight) per ml of 0.02 M phosphate buffer at pH 6, and treated with uniformly labeled '4C-L-valine (New England Nuclear Corp., Boston, Mass.). They were incubated at 28 C in a rotary water-bath shaker (New Brunswick Scientific Co., New Brunswick, N.J.), and at appropriate time intervals, samples of the suspensicn were taken and filtered rapidly through membranes with a pore size of 0.45 ,um (Millipore Corp., Bedford, Mass.). Pads

1 Paper of the Journal Series, New Jersey Agricultural Experiment Station, New Brunswick, N.J. 2Present address: Department of Medical Microbiology and Immunology, School of Medicine, University of California, Los

Angeles, Calif. 90024.

131

132

GUPTA AND PRAMER

of fungal cells accumulated on membrane surfaces were collected for extraction and analysis. Collection and measurement of CO2 derived from valine. Production by A. conoides of "CO2 from uniformly labeled 14C-L-valine was measured in a gastrain. A total of 80 ml of cell suspension was placed in a 125-mi Erlenmeyer flask and treated with 4 X 105 M L-valine. The flask was closed with a rubber stopper fitted with two glass tubes. It received air that was scrubbed free of CO2 by 20% KOH and moistened by bubbling through distilled water. The C02-free air entered the flask through a fritted glass cylinder located below the surface of the cell suspension. The atmosphere above the suspension was driven under slight pressure to a receiving tube and bubbled through an alkaline solution (NCS reagent, Nuclear Chicago, Des Plaines, Ill.) to trap metabolic CO. The trap was renewed at frequent intervals and, ultimately, the NCS reagent was tested for radioactivity. Biochemical fractionation of cells. Fractionation was employed to localized valine that entered cells of A. conoides treated with the amino acid at a concentration of 4 X 10-5 M. Treated cells were harvested on membranes by filtration at appropriate time intervals. They were transferred to a centrifuge tube, washed with cold distilled water and fractionated by the following modification of the method of Roberts et al. (20). Washed cells were suspended in 5 ml of 5% trichloroacetic acid and maintained in an ice bath with occasional stirring for 30 min. The preparation was then centrifuged at 2,200 X g for 10 min, and the supernatant fluid tested for radioactivity (cold acidsoluble fraction). The walls of the centrifuge tube were dried with a cotton swab to remove residual supernatant, and the pellet was suspended and extracted with 5 ml of 75% ethanol for 30 min at 50 C. The extract was decanted after centrifugation and examined for radioactivity (alcohol-soluble fraction). The pellet was then suspended in 5 ml of a solution containing 2.5 ml of diethyl ether and 2.5 ml of 75% ethanol and extracted at 50 C for 15 min. The extract (alcohol-ether-soluble fraction) was collected by centrifugation and analyzed. The pellet was extracted with 5 ml of 5% trichloroacetic acid for 30 min in a bath of boiling water, and this hot acid-soluble fraction was separated by centrifugation and analyzed. The pellet was washed free of residual trichloroacetic acid with 5 ml of acidified alcohol (pH 2) which was tested for radioactivity (acidified alcohol-wash fraction). The washed pellet was then suspended in 5 ml of diethyl ether. The cells were sedimented by centrifugation, dried, and extracted with two 2.5-ml portions of 1.0 N NaOH in a bath of boiling water for 30 min. The NaOH-soluble protein fractions were pooled and analyzed for radioactivity. The residue was suspended in 5 ml of distilled water and designated the NaOHinsoluble fraction. A 0.2-ml portion of each of the various fractions was used in tests for radioactivity. Analysis of fractions. Samples were usually concentrated by freeze-drying and analyzed chromatographically. Special treatments were required for analysis of the cold acid-soluble, NaOH-soluble, and NaOHinsoluble fractions; these are described below.

J. BACTERIOL.

Cold acid-soluble fractions were condensed to a volume of approximately 0.5 ml, diluted with an equal volume of 1.0 N HCI, and then extracted twice with 5-ml portions of diethyl ether to remove residual trichloroacetic acid and other nonpolar compounds. The aqueous phase was used for amino acid analyses by paper and thin-layer chromatography. The ether phase was evaporated by a stream of nitrogen to a greenish-yellow oil that was tested for radioactivity. NaOH-soluble fractions were lyophilized, dissolved in an excess of acid so that the final solution was approximately 6 N HCl. This was transferred to vials that were sealed and heated at 110 C for 36 hr to assure complete hydrolysis of protein. The vials were cooled and opened, and the solutions were evaporated under vacuum at room temperature. Each vial then received 1.0 ml of distilled water and this was evaporated to remove all traces of HCI. Finally, the residue was dissolved in 0.5 ml of distilled water, and the solutions were used for chromatographic identification of the various amino acids they contained. The NaOH-insoluble fractions were hydrolyzed with 6 N HCI and were analyzed as described for the NaOH-soluble fraction. Preparation of 2,4-dinitrophenyl hydrazones of aketo acids. a-Keto acids were precipitated as their 2,4-dinitrophenyl hydrazones from culture filtrates. The efficiency of the process was increased by seeding the filtrate with authentic compounds. a-Keto-isovalerate, pyruvate, oxalacetate, and a-ketoglutarate were supplied singly or collectively as follows. A 0.1ml amount of a solution containing 0.5 mg of the authentic keto acid per ml was added to 2 ml of concentrated filtrate in a centrifuge tube. To this solution, 5 ml of 0.2% 2,4-dinitrophenyl hydrazine in 2 N HCl was added with thorough mixing, and the tube was wrapped in aluminum foil and allowed to stand for at least 60 min at 5 C. Precipitates were collected by use of a centrifuge at 2,200 X g for 5 min, washed twice with 10-ml portions of cold distilled water, and recrystallized from ethanol. Mixtures of known and unknown hydrazones were separated and identified by thin-layer and cellulose column chromatography. Radioactivity associated with spots on thinlayer plates was detected by autoradiography or by isolation and analysis of the material as described below. Fractions collected from cellulose columns were also tested for "C activity. Keto acids in the cold acid-soluble fraction of cells of A. conoides were isolated and identified as their 2,4-dinitrophenyl hydrazones by using procedures identical to those just described for analysis of culture filtrates. Paper chromatography. Amino acids were identified by chromatography on Whatman 3MM paper with butanol-acetic acid-water (4:1:5) and phenol-water (8:2) in a two-directional system. When movement in only one direction was employed, the solvent generally selected was butanol-acetic acid-water (2). A frame capable of holding 17 sheets of paper (25 by 25 cm) in a plastic tank was used in this work (8). Whatman 3MM paper and butanol-acetic acid-water (4:1:5) were also used for the separation and identifi-

VOL. 103, 1970

METABOLISM OF VALINE BY A. CONOIDES

cation of ca-ketoisovalerate and L-valine. The keto acid was developed by spraying with semicarbazidehydrochloride (15), and located by viewing under ultraviolet light. The same papers were sprayed with ninhydrin to detect amino acids. Carboxylic acids were identified by chromatography on Whatman 3MM paper with ethanol-NH40H-water (80:5:15) as solvent in an ascending system (4). Thin-layer chromatography. Silica-gel plates (type K301R2 without indicator, Eastman Kodak Co., Rochester, N.Y.) were spotted with approximately 20 1.liters of each sample and developed with butanolacetic acid-water (8:2:2) in an ascending system in one direction only. The plates were air-dried and then heated at 110 C for 10 min. Ninhydrin-positive areas were stabilized by spraying with Cu(NO3)2 in nitric acid-ethanol and treating with ammonia vapor in a desiccator. The stabilizer was prepared by mixing 1.0 ml of saturated aqueous Cu(NO3)2 with 0.2 ml of 10% nitric acid in 100 ml of 96% ethanol (19). 2,4-Dinitrophenyl hydrazones of a-keto acids were separated on silica-gel thin-layer plates by the system described by Dancis et al. (6) and Ronkkainen (21). No-screen medical X-ray films (Eastman Kodak Co., Rochester, N.Y.) were used for autoradiography. When radioactivity was too low to reduce sufficient silver halide to darken an autoradiograph, the area considered suspect was scraped from a thin-layer plate or cut from paper. The silica gel or paper was placed in a vial and extracted with scintillation liquid for 60 min or more with intermittent shaking. The liquid was then tested for radioactivity. The scintillation liquid was prepared by dissolving 7 g of 2, 5-diphenyloxazol, 0.05 g of 1,4-bis-2 (5-phenyloxazolyl) benzene (Nuclear Associates, Inc., Westburg, N.Y.), and 70 g of napthalene in 1,000 ml of p-dioxane (Matheson, Coleman and Bell, East Rutherford, N.J.). Column chromatography. 2,4-Dinitrophenyl hydrazones and carboxylic acids were separated by cellulose column chromatography (7) and silicagel column chromatography (13), respectively.

133

Figure 1 is the result of an autoradiographic analysis of the cold acid-soluble fraction of cells after 4, 8, 10, 15, 20, 30, 40, and 50 min of L-valine uptake. The 4-min sample contained only one major component and that corresponded to valine. This was true also for the 8-min sample, but subsequent samples showed increasing tailing, and evidence for several compounds in addition to valine was visible after 20, 30, 40, and 50 min. From this and similar radioautograms it was clear that, after 10 min, intracellular valine was transformed to products also soluble in cold trichloroacetic acid and mobile in n-butanol-acetic acid-water (4:1:5) on Whatman paper. The fraction of cells soluble in cold trichloroacetic acid contained metabolic intermediates having a high rate of turnover. It displayed almost 50% of the total radioactivity absorbed after 50 min, but only a small portion of this was still in the form of valine. The alcohol-soluble fraction of cells contained very little radioactivity. This represented lipids and alcohol-soluble protein and amounted to only 2 % of that of total uptake after 3 hr. When this fraction was dried under vacuum, dissolved in water, and partitioned with a double volume of diethyl ether, only a negligible amount of radioactivity entered the ether phase. Apparently most of the radioactivity associated with this fraction was in the form of alcohol-soluble protein. Roberts et al. (20) reported that only one-sixth of the total protein of cells was alcohol-soluble, and since it incorporated very little valine, no further tests were made on this fraction. Radioactivity associated with the alcohol-ether-soluble fraction

RESULTS

Distribution of the radioactivity of '4C-L-valine in cells of A. conoides. It was previously demonstrated (11) that cells of A. conoides can accumulate and retain 14C-L-valine against a concentration gradient. Initially, all the absorbed amino acid was recovered by extraction with cold trichloroacetic acid but, after 5 min, it appeared that some of the amino acid was incorporated into the macromolecular fraction of cells. A more detailed view of intracellular distribution was obtained by exposing A. conoides to "4C-L-valine for a total period of 3 hr. At appropriate intervals of 2 3 4 5 6 7 9 .1 time, separate portions of the preparation were FIG. 1. Autoradiogram of paper chromatogram of removed, fractionated, and analyzed. cold trichloroacetic acid-soluble fraction of cells exFor a period of 10 min, more than 80% radioto "4C-L-valine for different intervals of time. activity entering cells was present in the cold acid- posed Numbers 1, 2, 3, 4, 5, 6, 7, and 8 represent 4, 8, 10, 15, soluble fraction. This amount decreased with 20, 30, 40, and 50 min, respectively. B:A: W = butanol time to 36 and 6% in 1.0 and 3 hr, respectively. (4)-acetic acid (1)-water (5), and val = valine.

134

GUPTA AND PRAMER

of cells of A. conoides was negligible. It did not exceed 0.2% of the total, indicating that the carbon of L-valine was not used for lipid synthesis. The hot acid-soluble fraction of cells of A. conoides showed that only 1% radioactivity of total uptake was incorpQrated during a 3-hr incubation period. No further analyses of this fraction were made. The NaOH-soluble fraction of cells is mostly protein. For an initial 4-min period, little radioactivity was associated with this fraction, but after 15 min it contained 14% of the total radioactivity, and eventually it included more than 50% of the activity absorbed as valine. The insoluble residue of the extraction procedures was also radioactive. It first displayed activity after 4 min, and was maximal at about 14% of the total after 3 hr, when the experiment was terminated. A large portion of the 14C supplied as L-valine was completely oxidized by A. conoides and recovered as CO2. The evolution of '4CO2 was preceded by a lag of 10 min, but, once initiated, it continued for the 3-hr experimental period and accounted for 23.4% of total radioactivity absorbed as 14C-Lvaline by cells of the fungus. Some radioactivity was returned to the suspending medium with time and, although this amounted to less than 8% of that absorbed, it was further examined to determine if treated cells were leaking valine or releasing a metabolic product of valine. A likely candidate for the latter possibility was a-ketoisovaleric acid, the deamination product of valine. This possibility was confirmed by treatment of test solutions with 2,4dinitrophenyl hydrazine in the presence of authentic nonradioactive a-ketoisovalerate. The reaction mixture yielded a radioactive precipitate. Autoradiograms of silica gel thin-layer chromatograms of the phenyl hydrazone derivative produced only one spot, and it moved identically to the hydrazone of a-ketoisovalerate. As a further test, 50 ,uliters of the condensed filtrate were chromatographed in the presence of authentic a-ketoisovalerate and valine on Whatman 3MM paper with n-butanol-acetic acid-water (4:1:5). The chromatograms were treated with semicarbazole-hydrochloride to detect a-ketoisovalerate and ninhydrin for valine. The spots that appeared were cut from the papers and extracted with p-dioxane scintillation liquid which was tested for radioactivity. Both a-ketoisovalerate and valine were detected by chromatographic procedures, but about 1,300 counts/min were associated with the a-ketoisovalerate spot (RF 0.8) and only 60 counts/min were associated with the valine spot (RF 0.36). Only the a-ketoisovalerate 14C-activity was adequate to darken the X-ray film so as to be detected by autoradiography;

J. BACTrERIOL.

therefore, there was little doubt that the compound was produced by cells of A. conoides supplied with "C-L-valine. The uptake and distribution with time of radioactivity into the major fractions of cells of A. conoides is illustrated by Fig. 2. Approximately 90% of the valine supplied was absorbed by cells within 6 min (curve A). The accumulation of radioactivity by the protein portion of cells (curve B) was initiated after a lag of about 4 min; it proceeded as a linear function of time for 6 min, and then decreased for an interval of almost 5 min before it was reinitiated at a changed rate. It reached a maximum at 100 min. It was of interest that the initial rate of incorporation of activity into protein was more rapid than the subsequent rate and corresponded to the movement of radioactivity from the cold acid-soluble (curve D) to the insoluble fraction of cells (curve B): the movement of free amino acid into protein. The plateau coincided with the production of 14CO2 (curve C), and the second phase of incorporation may have involved incompletely oxidized products of the metabolic conversion of L-valine. During this latter period, L-valine in the cold acid-soluble pool was certainly undergoing transformation. Autoradiographic analysis of silica gel thinlayer chromatograms of acid hydrolysates of the NaOH-soluble fractions established that L-valine was incorporated into protein without change for 20 min. At the 30-min interval, radioactive leucine was first detected, and at 60 min thereafter, until the experiment was terminated, four distinct ninhydrin-positive radioactive areas were produced. They were identified by finger printing (co-chromatography with authentic amino acids as carrier and matching radioactive areas on autoradiograms with ninhydrin-positive areas on thin-layer plates) as valine, leucine, alanine, and aspartate or glutamate, or both. These are most clearly illustrated in Fig. 3, no. 12. The appearance after 20 min of amino acids other than valine on these autoradiograms supported the possibility that the change in rate of incorporation of radioactivity into the protein fractions of cells was preceded by a conversion of the carbon skeleton of L-valine to other amino acids. Twodimensional paper chromatograms of the 100-min sample, using n-butanol-acetic acid-water (4:1:5) and then phenol-water (8:2), resolved the four into five distinct areas that corresponded to valine, leucine, alanine, aspartate, and glutamate. Autoradiograms enabled the radioactivity of these compounds to be estimated as follows: valine > leucine > alanine = aspartate > glutamate. To quantitate the distribution of radioactivity in these amino acids, protein hydrolysates were separated by two-dimensional paper chromatog-

VOL. 103, 1970

135


raphy in the presence of authentic compounds. The ninhydrin-positive areas were cut from the paper, extracted with p-dioxane scintillation liquid, and tested for radioactivity. This procedure was repeated for samples taken at each of eight time intervals. The results of these analyses (Table 1) indicated that the radioactivity of the NaOH-soluble fraction of cells was mainly due to direct incorporation of L-valine, and this was true throughout the test period. Conversion of valine to leucine required 10 to 15 min. Alanine was the next compound to be synthesized (20 to 30 min) and aspartate and glutamate were last (30 to 40 min). Leucine had about 42% of the radioactivity of L-valine. The radioactivity of alanine was greater than that of aspartate and glutamate, and about one-sixth that of valine. Autoradiograms of thin-layer chromatograms of acid hydrolysates of the NaOH-insoluble fraction of cells of A. conoides were similar to those of the NaOH-soluble fraction except that they lacked aspartate and glutamate. The labeled amino acids in the residue were probably cell wall components or possibly protein which was not soluble in alkali. Analysis of cold acid-soluble fraction. Since the carbon of valine was transformed to other amino acids which were incorporated into protein, it was of interest to analyze the cold acid-soluble fraction of cells to identify intermediates and characterize the reactions involved. For this purpose, 5 mg of cells of A. conoides was supplied with 200 nmoles of uniformly labeled '4C-L-valine (about 5 ,uc) and incubated at 28 C for 2 hr. At the end of the incubation period, the treated cells

MINUTES

FIG. 2. Uptake and distribution with time of radioactivity derived from 14C-L-valine. Initially, the system received 2.4 X 105 counts per min per ml. A, Uptake; B, NaOH-soluble and cold trichloroacetic acid-insoluble fractions of cells; C, C02; D, cold acid-soluble fraction ofcells.

5

s t r s 31

7

6

9

0

2 1

3

FIG. 3. Autoradiogram ofa thini-layer chromatogram of acid hydrolysates of the NaOH-soluble fraction of cells. Number I is authentic valine; nulmbers 2, 3, 4, S, 6, 7, 8, 9, 10, II, 12, and 13 represent samples takent after 10, 20, 30, 40, 50, 60, 70, 80, 100, 120, 140, and 160 mitt, respectively. B: A: W = butanol (8)-acetic acid (2)-water (2); Val = valine; Leu = leuicine; Ala = alanii)e; Glu = glutamate; and Asp = aspartate.

were extracted with cold trichloroacetic acid; the extract was concentrated by evaporation under vacuum and chromatographed on Whatman 3MM paper using n-butanol-acetic acid-water

(4:1:5) and phenol-water (8:2). When the papers were treated with ninhydrin, only four faint purple spots appeared and they were tentatively identified as glutamate, aspartate, alanine, and valine. When autoradiograms of these chromatograms were prepared (exposed to X-ray film for 3 weeks), 14 different regions developed. Some of these corresponded to known amino acids and were identified by finger-printing as glutamate, aspartate, glutamine, alanine, valine, and leucine. The radioactivity of glutamate and aspartate was most prominent. These identifications were confirmed, and much "cleaner" chromatograms were obtained by extracting the cold trichloroacetic acid concentrate with diethyl ether and then using the aqueous phase for analysis. However, the majority of compounds separated by chromatography of the cold acid-soluble fraction of cells supplied with '4C-L-valine were not amino acids and required further investigation. Changes in the cold acid-soluble fraction of cells with time was investigated by preparing a series of two-dimensional chromatograms 10, 20, 40, and 70 min after they had been treated with '4C-L-valine. The extracts were washed with diethyl ether before they were applied to paper to improve the quality of the chromatograms and of the autoradiograms developed from them. Interest here was concentrated on the persistence of L-valine and on its transformation to other amino acids in the cold acid-soluble pool, and results suggested the following sequence of events. Valine remained unchanged for 10 min after uptake was

GUPTA AND PRAMER

136

J. BACTERIOL.

TABLE 1. Appearance of "IC-activity of L-valine with time in amino acids of NaOH-soluble protein 14C activity associated with spots on paper chromatogram (counts/min) Amino acid S min

Valine .................. 66 8 Leucine ................ 0 Alanine ................ Aspartate ............0..O 0 Glutamate .......|.O.|

10 min

15 min

20 min

30 min

40 min

60 min

100 min

167 21

317 100

0 0 0

0 0 2

410 151 19 2

707 200 100 48

937 327 152 72

1, 207 465 210 172

2,000 864 344 256

8

32

48

53

189

initiated. Some conversion to leucine was detected after 20 min, and by 40 min, "IC derived from L-valine was present in alanine, aspartate, and glutamate, in addition to leucine. These same amino acids were detected in 70-min samples, and four additional regions were radioactive but not identified. The intensity of radioactivity decreased with time, but the cold acid-soluble pool was not depleted of valine or any of the newly formed amino acids during a 70-min experimental period. The soluble pool of amino acid appeared to be the source drawn upon for protein synthesis, since 'IC-labeled amino acids in the pool corresponded to those found in the protein hydolysates. Since the immediate precursors of glutamate and aspartate are intermediates in the tricarboxylic acid cycle, it was of interest to search for these compounds as labeled components of the cold acid-soluble fraction of the cells of A. conoides that were exposed to "C-L-valine. For this purpose, 20 ,uliters of cold trichloroacetic acid-soluble extract of cells treated with 14C-Lvaline for 15 min was spotted on a silica gel thin-layer plate, with 10 gliters of 0.05% fumarate, succinate, and oxalacetate, singly and as a mixture. The plate was chromatographed in benzene-methanol-acetic acid (45:8:4) and treated with 0.04% alkaline bromocresol purple in 50% ethanol. Three spots, corresponding to succinate (RF 0.53), fumarate (RF 0.41), and oxalacetate (RF 0.23), appeared on the plate, but a fourth and unknown compound also developed (RF 0.7), and it was the unknown that had the greatest 14C activity (1,600 counts/min). The radioactivity of succinate, fumarate, and oxalacetate was measured at 230, 170, and 30 counts/min, respectively, and it was apparent that the carbon skeleton of valine had entered into the tricarboxylic acid cycle. When the cold trichloroacetic acid extract was partitioned with diethyl ether at acid pH, about 30% of the radioactivity entered the ether phase. The carboxylic acid content of this material was determined by using a silica gel column and benzene-ether as described by Kinnory et al. (13). The sample analyzed was chosen to coincide with

the change in rate of protein synthesis that appeared as a plateau in the course of 14C incorporation into the NaOH-soluble and cold acid-insoluble fractions of cells. From the elution pattern illustrated in Fig. 4, preliminary identification was made of the following compounds: isobutyrate, a-ketoisovalerate, propionate, fumarate, pyruvate, succinate, fi-hydroxyisobutyrate, a-ketoglutarate, and malate. These identities were confirmed when fractions collected from the column were made alkaline, concentrated, chromatographed on paper with ethanol-NH20H-water (80:5:5), and finger printed. Table 2 lists the RF values and radioactivity associated with the various spots that developed. Isobutyrate, propionate, succinate and f-hydroxyisobutyrate were more radioactive than the other substances. a-Keto acid phenyl hydrazones were obtained by co-precipitation from a cold trichloroacetic acid extract of cells treated with "C-L-valine for 20 min and separated on a cellulose column with n-amyl alcohol saturated with NH40H used as eluant. The absorption at 380 nm and radioactivity of fractions were determined (Fig. 5), and a-ketoisovalerate, oxalacetate, and a-ketoglutarate were tentatively identified as products of valine metabolism. Portions of these fractions were concentrated and further characterized by silica gel thin-layer chromatography with two n-butanol-ethanol-0.5 % solvent systems: NH40H (7:1:2) and isoamyl alcohol-0.25% NH40H (20: 1). Radioactive regions were scraped from the plates and their radioactivities were measured. Table 3 lists the RF values and radioactivity of the compounds identified. Oxalacetate and a-ketoglutarate had received little of the 4IC from L-valine. Most was present in a-ketoisovalerate, the product of valine deamination. DISCUSSION

Uptake of L-valine by germinated conidia of A. conoides suspended in 0.02 M phosphate buffer of pH 6 at 28 C was initiated immediately and proceeded rapidly as a linear function of time until the exogenous supply was exhausted. For an

137


VOL. 103, 1970

initial period of approximately 4 min, valine entering cells of A. conoides was localized in the pool of the free amino acids and was recovered completely by extraction with cold trichloroacetic acid. Subsequent to this initial 4-min period, recovery by cold trichloroacetic acid extraction was incomplete and valine was demonstrated to i have been incorporated into cold acid-insoluble 9 fraction of the cells (11). This indicated that inoi corporation of valine into macromolecules began after a lag of about 4 min. An interval between amino acid uptake and incorporation is a general phenomenon that has been noted in work with bacteria, yeast, and filamentous fungi (3, 22, 23). FIG. 5. Elution patter/i of 2,4-dinitrophenyl hiydraChemical fractionation of cells of A. conoides supplied with 14C-L-valine for varying periods of zone of a-keto acids separated on cellulose columnt. time indicated that carbon derived from valine Symbols: 0, counts per minute; +, optical density was not located in lipid (alcohol-ether-soluble) at 380 nm. or nucleic acid (hot acid-soluble) fractions of the cells. Radioactivity present in hydrolysates of the TABLE 3. Characteristics ofa-keto acid hydrazones alcohol-soluble, NaOH-soluble, and NaOH-in- separated on a cellulose column and idenitified by silica gel thini-layer chiromatography with two soluble fractions of cells was associated with solvent systems amino acids. In fact, more than 65 % of the carbon derived from 14C-L-valine was recovered in the Solvent system cellular proteins. A similar observation was made E

Bb

Aa

Phenylhydrazone of Radio2300

RF

no

Radio-

cts ~ (counts RF /min)

activity (counts /min)

2055~~~~~~~~~

a-Ketoisovalerate....0 .72 Oxalacetate .......... 0.23

2000-

c-Ketoglutarate...0.0.22 a

b

80 20 15

0.54 0 0

100 34 21

n-Butanol-ethanol-0.25% NH40H (7:1:2). Isoamylalcohol-0.5% NH40H (20: 1).

by Roberts et al. (20), who supplied Escherichia coli with "4C-valine and reisolated most of the FRACTION radioactivity from protein hydrolysate as leucine FIG. 4. Elution pattern of carboxylic acids separated and valine. In the present investigation, 14C on silica gel column. activity concentrated with time in the protein fraction of cells, but it was present here as leucine, TABLE 2. Characteristics of carboxylic aceids alanine, aspartate, and glutamate, in addition to separated on a silica gel column and idenltified by valine. paper chromnatography and autoradiographly Initially, valine entered a pool of metabolic intermediates that was extracted from cells by Acid RF Counts/min cold trichloroacetic acid. After a 4-min interval, some valine was incorporated unchanged into cell Isobutyric ................ 0.63 2,230 proteins, but most of the accumulated amino acid a-Ketoisovaleric ....... 0... .70 220 underwent metabolic transformation into a Propionic ................ 2,200 .56 0 .20 Fumaric .................. variety of products that included carboxylic acids 300 Pyruvic ................... 0.45 410 and other amino acids. Some valine-carbon was 0 .23 Succinic .................. 2,500 oxidized and recovered as 4CO2 . ,B-Hydroxyisobutyric .....0. .50 2,000 Relatively little is known of the degradation of a-Ketoglutaric ........ 0.... .20 100 branched-chain, aliphatic amino acids by microMalic .................... 0.15 470 organisms, but pathways for the oxidation of 200

20

0

0

100

120

2

138

GUPTA AND PRAMER

glutarate (normal working of the tricarboxylic acid cycle) and to glutamate by transamination. No radioactive citrate or isocitrate was detected, possibly because of their high rate of turnover by the A. conoides system. The appearance of labeled succinate, oxalacetate, and a-ketoglutarate indicated that the carbon of valine was channeled through the complete tricarboxylic acid cycle, and the labeling of aspartate and glutamate indicated that the tricarboxylic acid cycle was generating a-keto acids for biosynthetic purposes. Roberts et al. (20) reported that, in E. coli, '4C-valine is transformed to leucine, but "Cleucine is incorporated unchanged into protein. They suggested that valine is deaminated and that the a-ketoisovalerate formed condenses with acetate to give rise to a-ketoisocaproate, which is the immediate precursor of leucine. Meister (17) summarized a more complete sequence for conversion of valine to leucine that includes the following intermediates: ca-ketoisovalerate, f-carboxy-fl-hydroxyisocaproate, a-isopropylmaleate,

valine, leucine, and isoleucine have been established by using animal tissues (10). Isolation and identification here of radioactive a-ketoisovalerate suggested that valine was deaminated by a transamination reaction, involving a-ketoglutarate as described by Meister (16), and detected in Pseudomonas aeruginosa by Norton and Sokatch (18). Isobutyrate and propionate formation by A. conoides is understandable in view of the findings of Fones et al. (9). They proposed that propionate could originate from valine by the following pathway: deamination of the amino acid to yield a-ketoisovalerate; decarboxylation of a-ketoisovalerate to produce isobutyrate; and conversion of the isobutyrate to propionate. Using specifically labeled valine (4,4'-'4C), Kinnory et al. (14) showed that, in rat kidney, isobutyrate was derived from valine and converted to propionate. Isolation here of labeled ,B-hydroxisobutyrate and propionate by silica gel column chromatography suggests that the transformation of valine to propionate by A. conoides may follow a pathway similar to that proposed for the rat kidney

a-hydroxy-fl-carboxyisocaproate,

It is possible that some propionate was converted to succinate via methylmalonyl-coenzyme A after fixation of CO2. This transformation would introduce 'IC, that originated from valine, into the tricarboxylic acid cycle and account in part for the radioactivity associated with succinate, fumarate, malate, oxalacetate, and a-ketoglutarate. The incorporation of valine-carbon into the tricarboxylic acid cycle provides a basis for understanding the formation of labeled amino acids isolated from extracts of A. conoides supplied with '4C-L-valine. Decarboxylation of oxalacetate yields pyruvate, and the amination of pyruvate and oxalacetate produces alanine and aspartate. Oxalacetate is also converted to a-ketoCit3

,CN3 -CO C

,CH-CHH-C-NCOOH

CH3

CH3

NH2

C t2

C-COO C-COOH

Hu2z0 2

CH3

PROPIONALDEHYDE

.

> CHN-COOH

CH3

$-HYDROXYISOBUTYRATE

METHYLACRYLATE

C H3- CH2- CHO

,CH20N

1023

-2

-2H

CHN-COOH

CH3

°

ISOBUTYRATE

a-KETOISOVALERATE

VALINE

a-keto-j3-car-

boxyisocaproate, and a-ketoisocaproate. It is possible that such a pathway for the transformation of valine to leucine is operative in A. conoides. The conversion of valine to other amino acids may have occurred through tricarboxylic acid cycle intermediates as already discussed. However, Norton and Sokatch (18) reported that P. aeruginosa oxidized DL-valine to isobutyrate and propionate, and that propionate was directly oxidized to pyruvate for alanine biosynthesis. Formation of aspartate from labeled valine occurred by condensation of three-carbon (propionate) and one-carbon (CO2) units. In the present investigation, it is more likely that propionate was formed from valine via isobutyrate and converted to succinate, and that alanine, aspartate, and glutamate biosynthesis proceeded via tricar-

(Fig. 6).

CHt

J. BACTERIOL.

-2N

CHO

CHN3

CN-COOH

METHYLMALONATE SEMIALDEHYDE

CH3-CH2-COOH PROPIONATE

FIG. 6. Pathway of Kinnory et al. (14) for the conversion of valine to propionate.

-CO2

2,


VOL. 103, 1970

VA LINE

aw-K ETOISOVALERATE

SOBUTYRATE

a-K ETOI SOCAPROATE

PROPI ONATE

LEUCINE

SUCCI NATE

FUMARATE

MA

LATE ASPARTATE

OXA

IACETATE

OXA-ACETATUEPYRUVATE

_AALANINE

cl-K ETOGLUTARATE GLUTAMATE

GLUTAMINE

FIG. 7. Proposed pathway for vali/te metabolism in A. conioides.

boxylic acid cycle intermediates. However, the time sequence chosen for this aspect of the work may not be fine enough to make such a decision unequivocal. The biosynthetic pathways summarized by Fig. 7 are suggested as operative in A. conoides when valine is metabolized to various amino and carboxylic acids. ACKNOWLEDGMENTS The senior author gratefully acknowledges the assistance of Dexter H. Howard in the preparation of this manuscript.

LITERATURE CITED

1. Beadle, G. W., and E. L. Tatum. 1941. Genetic control of biochemical reactions in Neurospora. Proc. Nat. Acad. Sci. U.S.A. 27:499-506. 2. Block, R. J. 1950. Estimation of amino acids on paper, chromatograms. Anal. Chem. 22:1327-1332.

139

3. Britten, R. J., and F. T. McClure. 1962. The amino acid pool in Escherichia coli. Bacteriol. Rev. 26:292-335. 4. Cheftel, R. I., R. Munier, and M. Macheboeuf. 1953. Microchromatographie sur papier des acides aliphatique hydrosolubleset non volatils. Bull. Soc. Chim. Biol. 35:1085-1093. 5. Coscarelli, W., and D. Pramer. 1962. Nutrition and growth of Arthrobotrys conoides. J. Bacteriol. 84:60-64. 6. Dancis, J., J. Hutzler, and M. Levitz. 1963. Thin-layer chromatography and spectrophotometry of a-keto-acid hydrazones. Biochim. Biophys. Acta 78:85-100. 7. Drew, F. D., L. M. Marshall, and F. Friedberg. 1952. Separation of keto-acids by cellulose columns. J. Amer. Chem. Soc. 74:1852-1853. 8. Dutta, S. P., C. E. Dent, and H. Harris. 1950. An apparatus for simultaneous production of many two-dimensional paper chromatograms. Science 112:621-623. 9. Fones, W. S., T. P. Waalkes, and J. White. 1951. The conversion of L-valine to glucose and glycogen in the rat. Arch. Biochem. Biophys. 32 :89-95. 10. Greenberg, D. M. 1961. Carbon catabolism of amino acids, p. 79-172. In D. M. Greenberg (ed.), Metabolic pathways, vol. 3. Academic Press Inc., New York. 11. Gupta, R. K., and D. Pramer. 1970. Amino acid transport by the filamentous fungus Arthrobotrys conoides. J. Bacteriol. 103:120-130. 12. Jennison, M. W., M. D. Newcomb, and R. Henderson. 1955. Physiology of woodrotting Basidiomycetes. I. Growth and nutrition in submerged culture in synthetic inedia. Mycologia 47:275-304. 13. Kinnory, D. S., Y. Takeda, and D. M. Greenberg. 1955. Chromatography of carboxylic acids on a silica-gel column with a benzene-ether solvent system. J. Biol. Chem. 212: 379-383. 14. Kinnory, D. S., Y. Takeda, and D. M. Greenberg. 1955. Isotopic studies on the metabolism of valine. J. Biol. Chem. 212:385-396. 15. Magasanic, B., and H. E. Umbarger. 1951. The separation and identification of keto-acids by filter paper chromatography. J. Amer. Chem. Soc. 72:3308-2309. 16. Meister, A. 1955. Transamination. Advan. Enzymol. 16:185246. 17. Meister, A. 1965. Intermediary metabolism of amino acids, p. 748. In Biochemistry of the amino acids, 2nd ed., vol. 2. Academic Press Inc., New York. 18. Norton, J. E., and J. R. Sokatch. 1966. Oxidation of D- and L-valine by enzymes of Pseudomonias aeruginiosa. J. Bacteriol. 92:116-120. 19. Randerath, K. 1966. Thin-layer chromatography, 2nd ed., p. 110. Academic Press Inc., New York. 20. Roberts, R. B., D. B. Cowie, P. H. Abelson, E. T. Bolton, and R. J. Britten. 1955. Studies of biosynthesis in Escherichia coli. Carnegie Inst. Wash. Publ. No. 607. 21. Ronkkainen, P. 1963. Die Dunnschichtchromatographie der Ketosauren. J. Chromatogr. 11:228-237. 22. Shieh, K. Z., and L. R. Hedrick. 1966. Energy requirement for L-glutamate uptake and utilization by Hansenula subpelliculosa cells. J. Bacteriol. 92:1638-1644. 23. Wiley, W. R., and W. H. Matchett. 1966. Tryptophan transport in Neurospora crassa. I. Specificity and kinetics. J. Bacteriol. 92:1698-1705.