the relationship of cobalt requirement to propionate ... · ments at 540 rnp were converted to...

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 239, No. 8, August 1964

Printed in U.S.A.

The Relationship of Cobalt Requirement to

Propionate Metabolism in Rhizobium*

A. A. DE HERTOGH,~ PATRICIA A. MAYEUX, AND HAROLD J. EVANS

From the Department of Botany and Plant Pathology, Oregon State University, Corvallis, Oregon

(Received for publication, February 3, 1964)

Cobalt is known to be essential for symbiotic nitrogen fixation by legumes (l-5), but a requirement of the element for growth of leguminous plants supplied with adequate fixed nitrogen has not been shown. Ahmed and Evans (3) provided evidence that cobalt was required for the bacteria within the nodules of inoculated legumes but was not required for the leguminous plants per se. Lowe, Evans, and Ahmed (6) and Lowe and Evans (7) subsequently showed that cobalt is necessary for normal growth of five species of Rhixobium. More recently, Kliewer and Evans (8-10) analyzed nodules of a variety of symbionts and cells from several species of Rhizobium and observed the presence of cobamide coenzymes in each of the organisms. In addition, they (9) isolated and identified the dimethylbenaimidazolylcobamide coenzyme in extracts of cells isolated from soybean nodules and in cells of a pure culture of Rhizobium meliloti and presented data (10) showing that cobalt deficiency not only limited the growth of Rhizobium japonicum and R. meliloti but also limited the synthesis of cobamide coenzymes in these organisms.

The Bn coenzyme has been shown to be involved in propionate metabolism in animals (11-13) and in Propionibacterium sher- ma&i (14, 15). In addition, Arnstein and White (16) have shown that Bn deficiency in Ochromonas malhamensis resulted in a reduced capacity of the organism to oxidize propionate. Ayers (17) has described similar effects of Blz deficiency in a Flavobac- terium species. On the basis of these reports, experiments were conducted to determine whether Rhizobium would metabolize propionate and the possible effect of cobalt deficiency on propionate utilization in these bacteria. In a preliminary communica- tion (18), evidence was presented showing that the rate of oxidation of propionate by cell suspensions of R. meliloti was strikingly affected by the concentration of cobalt in the culture medium. In contrast, the oxidation of glutamate was not affected by cobalt deficiency. In this paper, evidence is presented showing that enzymes necessary for the catalysis of the conversion of propionate to succinyl coenzyme A by a pathway involving methylmalonyl coenzyme A are widely distributed in Rhizobium and that cobalt deficiency in R. meliloti results in insufficient Blz coenzyme for the function of methylmalonyl coenzyme A mutase in cell-free extracts.

* This is technical paper 1776 of the Oregon Agricultural Ex- periment Station. This-investigation was supported in part by Research Grant G18556 to Harold J. Evans from the National Science Foundation.

t Present address, Boyce Thompson Institute, Yonkers, New York.

EXPERIMENTAL PROCEDURE

Biological Materials-The biological materials utilized in this investigation were: soybeans (Glycine mm (L.) Merr. var. Chippewa), peas (Pisum sativum L. var. American Wonder), lupine (Lupinus angustifblius L. var. Borre Blue), cowpeas (Vigna sinenis Savi, var. Iron Clay), Rhizobium meliloti F-29, Rhizobiumjaponicum 61876, Rhizobium phaseoli K-l 1, Rhizobium leguminosarum C-56, and Rhizobium trifolii P-28. The seeds of the various legumes were purchased from commercial sources and the Rhizobium species were kindly supplied by Dr. J. C. Burton of the Nitragin Company, Milwaukee, Wisconsin.

Reagents-ATP (dipotassium salt) and GSH were obtained from Nutritional Biochemical Corporation, coenzyme A from Sigma Chemical Company, and propionate-1-14C, propionate-2- 14C, and propionate-3-14C from New England Nuclear Corpora- tion. The adenylcobamide coenzyme was a gift from Dr. Mark Kliewer, Department of Viticulture and Enology, University of California, Davis, California. The benzimidazolylcobamide coenzyme was a gift from Dr. David Perlman, Squibb Institute for Medical Research, New Brunswick, New Jersey, and the dimethylbenzimidazolylcobamide coenzyme was kindly supplied by Dr. Karl Folkers, Merck Institute for Therapeutic Research, Rahway, New Jersey. All other chemicals were reagent grade and were purchased from commercial sources.

Cultural Procedures-Seedlings of the legumes, inoculated with the appropriate Rhizobium species, were grown in a greenhouse according to the cultural method described by Ahmed and Evans (2, 3). Nodules were harvested from roots of plants at ages listed in the legends of figures and tables that present the results of the various experiments.

Medium A, used for the culture of R. japonicum, R. phaseoli, R. leguminosarum, and R. trifolii, contained the following components in a final volume of 1 liter: K2HP04, 1.00 g; KH2P04, 1.00 g; MgS04.7Hz0, 0.36 g; CaS04.2Hz0, 0.13 g; KNO,, 0.70 g; FeC1,.6HzO, 4 mg; yeast extract (Difco), 1.00 g; L-arabinose, 1 g; glycerol, 4 ml. The components of the medium were dis- solved in distilled water and the pH was adjusted to 6.3 with 1 N

HCl or 1 N NaOH. Medium B, which was used in certain experiments for the cul-

ture of R. meliloti, was the same as Medium A except that L-arabinose, glycerol, and KNOI were omitted and 3.00 g of n-mannitol and 0.06 mg of CoC12.6Hz0 were added to each liter. The pH was adjusted to 6.8.

Media C and D were identical with Media A and B, respec-

2446

by guest on October 13, 2017

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

August 1964 A. A. De Hertogh, P. A. Mayeux, and H. J. Evans 2447

tively, except that 100 mg each of succinic acid, methylmalonic acid, and sodium propionate were added per liter.

The method employed for purifying components of the culture medium and the cultural procedure used for the growth of R. meliloti with varying concentrations of added cobalt have been described in previous communications (2,3,7, 18).

The various Rhixobium species were transferred from slants to 50-ml culture flasks containing 20 ml of the appropriate medium. When the organisms reached the log phase of growth, a series of l-liter flasks, each containing 400 ml of the appropriate medium, were inoculated with 4 ml of the appropriate culture. For the study of cobalt deficiency, it was necessary to transfer at least twice to 50-ml flasks containing 20 ml of cobalt-deficient medium before the final inoculation of the cobalt-deficient medium in the l-liter flasks.

Cells were cultured on a rotary shaker at 30” and were harvested during the log phase of growth. R. meliloti, R. leguminosarum, R. phaseoli, and R. trifolii were harvested after 18 to 24 hours, R. japonicum was harvested after 72 to 96 hours, and R. meliloti, grown in the cobalt-deficient medium, was harvested 36 to 48 hours after inoculation.

Preparation of Cell Suspensions and Extracts-The operations necessary for the preparation of cell suspensions and cell-free extracts were performed at a temperature of O-4”. Bacteroids were collected from legume nodules by the procedure described by Cheniae and Evans (19). Cells were harvested from culture media by centrifugation at 11,000 X g for 10 minutes. Four l-liter flasks containing 400 ml of Medium A, B, C, or D yielded about 1 g of fresh cells. Cells that were collected for radiorespirometric experiments were washed twice in 0.067 M potassium phosphate buffer, pH 6.8, and those used for preparation of cell- free extracts were washed twice with 0.01 M Tris-Cl buffer, pH 8.5. Cells from the cobalt-deficient medium were prepared by the same procedure with the exception that centrifuge tubes were acid-washed and the buffers used for the radiorespirometric experiments were extracted with l-nitroso-2-naphthol (2, 3) to decrease cobalt contamination. For the radiorespirometric experiments, the bacteroids and cells were suspended in sufficient 0.067 M potassium phosphate buffer, pH 6.8, to obtain 20 mg of wet cells per ml. The nitrogen content of cell suspensions was determined by the micro-Kjeldahl procedure described by Um- breit, Burris, and Stauffer (20).

Cell-free extracts for enzyme experiments were obtained by grinding the bacteroids or cells with alumina and 0.01 Tris-Cl buffer (0.001 M with respect to GSH), pH 8.5 (19). The supernatant solution remaining after the slurry was centrifuged at 34,000 x g for 20 minutes was used for all enzyme assays. Throughout this procedure, exposure to direct light was avoided to prevent destruction of endogenous Bn coenzyme. The protein content of the extracts was determined by the Folin-Ciocalteau method described by Layne (21). Crystalline serum albumin was used as the protein standard.

Purified propionyl-CoA carboxylase was prepared according to the method described by Lane and Halenz (22). The procedure was followed up to the first ammonium sulfate fractionation step. At this point, the carboxylase was found to be essentially free of methylmalonyl-CoA mutase.

Radimespirometric Procedures-The radiorespirometric experiments were performed according to the method described by Wang et al. (23) and Wang and Krackov (24). The contents of

each flask in a final volume of 11 ml were made up of 10 ml of cell suspension in 0.067 M potassium phosphate buffer at pH 6.8 and 1.0 ml of 14C-propionate (3.5 pmoles containing 1.0 PC). When suspensions of R. meliloti were used that had been grown with varying concentrations of cobalt, the flasks contained one- half the quantity of all components listed. The flasks were shaken continuously in a controlled temperature bath at 30”. The air flow was maintained at 60 ml per minute and the 14C02 was trapped in 10 ml of a mixture composed of 2 parts absolute ethanol and 1 part 2-aminoethanol. Samples were collected at 20-minute intervals. To determine the amount of ‘*CO2 recov- ered, the trapping solution was collected and diluted to 15 ml with absolute ethanol. A 5-ml aliquot was placed in a 20.ml counting vial containing 10 ml of phosphor solution (0.3 y0 2,5- diphenyloxazole and 0.003 $& 1,4-bis-2-(4-methyl-5-phenyloxa- zolyl)benzene in toluene). This was counted in a Packard Tri-Carb model 314 DC liquid scintillation counter. Counting was usually carried out to a standard deviation of no greater than 2%. Counting efficiency was determined by use of 14C-toluene as an internal standard.

Enzyme Assays-The procedure used for the determination of propionate and acetate activation is a modification of that described by Berg (25) for the activation of acetate. The complete reaction mixture in a final volume of 1.0 ml consisted of the following components in micromoles: Tris-Cl buffer, pH 8.0, 10; Tris-acetate or Tris-propionate, pH 8.0,10; MgCl,, 20; ATP, 10; GSH, 10; NHzOH.HCI, 20; KCl, 50; coenzyme A, 0.25; and enzyme extract containing 1 to 3 mg of protein. In those experiments in which acetate was used as a substrate, the absorbance at 540 rnp was converted to micromoles of acethydroxamate with the use of the conversion factor reported by Berg (25). Where Tris-propionate was used as the substrate, absorbance measure- ments at 540 rnp were converted to concentrations of propion- hydroxamate by use of a standard curve prepared with propionic anhydride (26).

Propionyl-CoA carboxylase was assayed by the procedure described by Lane and Halenz (22). The complete reaction mixture in a final volume of 1.0 ml consisted of the following components in micromoles: Tris buffer, pH 8.5, 160; KH14C03, 15; ATP, 4.0; MgC12, 4.0; GSH, 5.0; propionyl-CoA, 1.0; and enzyme extract containing 1 to 3 mg of protein. For some experiments, other acyl-CoA compounds were substituted for propionyl-CoA. All acyl-CoA compounds were prepared by the method of Simon and Shemin (27).

Methylmalonyl-CoA mutase activity was determined by the procedure of Hegre, Miller, and Lane (28), which involves coupling of a purified propionyl-CoA carboxylase from beef liver and the methylmalonyl-CoA mutase of the enzyme extracts. The complete reaction mixture in a final volume of 1.0 ml consisted of the following components in micromoles: Tris buffer, pH 8.5, 160; KH14C03, 15; ATP, 4.0; MgC12, 4.0; GSH, 5.0; propionyl-CoA, 1.0; propionyl-CoA carboxylase (20 units per mg of protein), 1 to 3 mg; and enzyme extract containing 0.1 to 1.0 mg of protein. The enzyme extract was added following a lo- minute preincubation at 37” of all other reaction components. An amount of eoenzyme listed in the legend of the tables was added to the reaction mixture immediately after the enzyme extract. The complete system was then incubated for an addi- tional 20 minutes. The permanganate oxidation and plating procedures were identical with those described by Hegre et al.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

2448 Cobalt in Rhizobium Vol. 239, No. 8

P v 8 > 100 B g 80 Y oz 60

8 Y 40

I= a 20

2 2 40 80 120

1. MINUTES

s s v v B B

MEDIUM C

& & 100 100

%! %! 80 t

80

8 8 : 60 : 60

E E w w 40 40 2 2

2 20 2 5 5

20

2. MINUTES

40 80 40 80 120 120 40 80 40 80 I20 I20 2. MINUTES MINUTES MINUTES

FIG. 1. The radiorespirometric patterns for the oxidation of A and C. Medium A is the basic medium for R. japonicum propionate-l-, 2, and -3-W by bacteroids from soybean nodules. Medium C is the same medium modified by the addition of 100 The nitrogen content of cells in each flask was 3.4 mg (See “Ex- mg per liter each of sodium propionate, succinate, and methyl- perimental Procedure” for further details). W----W, propio- malonate. The nitrogen content of cells in each flask was 5.0 nate-l-r%; O-O, propionate-2-l%; O-0, propionate-3- and 5.2 mg for experiments involving Media A and C, respectively. 14C. Symbols are as in Fig. 1. See “Experimental Procedure” for

FIG. 2. The radiorespirometric patterns for the oxidation of propionate-1-, -2., and -3.r4C by R. japonicum grown on Media

further details.

f

Y 40

F a 20 <

40 80 120 40 80 120 MINUTES MINUTES

MEDIUM D

FIG. 3. The radiorespirometric patterns for the oxidation of propionate-1-, -2., and -3-l% by R. meliloti grown on Media B and D. Medium B is the basic medium for R. meliloti. Medium D is the same, except that it is modified by the addition of 100 mg per liter each of sodium propionate, succinate, and methylmalonate. The nitrogen content of cells in each flask was 4.3 and 4.1 mg for experiments in which Media B and D were used, respectively. See “Experimental Procedure” for further details.

(28). The radioactivity of permanganate-stable compounds was measured with a Nuclear Chicago model D-47 gas flow counter. The activity of methylmalonyl-CoA mutase in a cell- free extract of R. meliloti was proportional to amounts of extract protein in the range of 0.1 to 1.0 mg.

Products of the methylmalonyl-CoA mutase reaction were isolated and chromatographed by the procedure described by Flavin and Ochoa (29). The radioactive areas of the chromato- grams were detected with a Vanguard model 880 strip counter.

RESULTS

Radiorespirometric Experiments-The radiorespirometric patterns for the oxidation of specifically labeled propionate by bacteroids isolated from soybean nodules are presented in Fig. 1. These data show that propionate is oxidized rapidly after an initial slower rate of oxidation. The curves in Figs. 2 and 3 indicate that cell suspensions from pure cultures of R. japonicum and R. meliloti also oxidize propionate. Fig. 3 shows that propionate oxidation by R. meliloti was enhanced by the addition of propionate, succinate, and methylmalonate to the culture medium.

40 00 120 160 200 240 MINUTES

FIG. 4. The radiorespirometric patterns for the oxidation of propionate-1-14C by R. meliloti grown with varying concentrations of cobalt in the culture medium purified for cobalt. Each flask contained 5 ml of cell suspension and 0.5 ml of propionate (1.75 rmoles containing 0.5 PC). The nitrogen contents of flasks were: 1.68, 1.75, and 1.65 mg. O-O, 0.00; O-O, 0.01; and +--¤, 1.0 p.p.b. cobalt levels, respectively. The abbreviation p.p.b. refers to 1 rg per liter. See “Experimental Proce- dure” for further details.

This effect was not observed with cell suspensions of R. japonicum. Regardless of the organism tested or the medium used for culture, the patterns of 14C02 recovery from propionate labeled in different positions were similar. The rate of recovery of r4C02 from the oxidation of propionate-1-r4C was always greater than rates of recovery of 14C02 from propionate-2- and -3J4C. The rates of recovery of 14C02 from the latter two compounds were similar.

Since methylmalonyl-CoA mutase from animal tissues (11-13) or from P. shermanii (14, 15) has been reported to require cobamide coenzymes, the effect of varying concentrations of cobalt on the capacity of R. meliloti to oxidize propionate-l-, -2., and -3-14C was studied. The results presented in Fig. 4 show trends in propionate-l-r4C oxidation that are very similar to those previously obtained with manometric experiments (18). Cobalt- deficient cells exhibited a greatly reduced capacity to oxidize propionate as indicated by the rate of evolution of 14COa from propionate-1-r4C. The increased rate of formation of 14COa from deficient cells that occurred after a lag period may have been caused by contamination of cell suspensions with cobalt, resulting


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


in the synthesis of cobamide coenayme. The effect of the intermediate concentration of cobalt on the rate of oxidation of propionate was not as striking as that previously observed with the manometric experiments (18). The results of radiorespirometric experiments in which propionate-2J*C and -3-14C were utilized as substrates showed trends very similar to those obtained with propionate-lJ4C and are not presented.

Enzymes of Propionate Oxidative Pathway-Propionate utilization by organisms in all known cases is initiated by a reaction with ATP and coeneyme A yielding propionyl-CoA, adenylic acid, and pyrophosphate. The enzyme catalyzing this reaction will function with either propionate or acetate (30). In the experiments in which the enzyme activity of cell-free extracts of soybean bacteroids and pure cultures of R. japonicum and R. meliloti were studied, both acetate and propionate were utilized as substrates. Data are presented in Table I which show that these extracts contained an enzyme system capable of activating both substrates.

The second reaction in the conversion of propionate to succinate is the carboxylation of propionyl-Coil to form methylmalonyl-CoA. With the use of extracts prepared from soybean bacteroids and pure cultures of R. japonicum and R. meliloti, experiments were conducted to determine if a carboxylase was present which would utilize propionyl-CoA as a substrate.

TABLE I

Activation of propionate and acetate by cell-free extracts of soybean bacteroids, R. japonicum, and R. meliloti

Bacteroids were isolated from nodules of 26.day-old soybean plants. R. japonicum and R. meliloti were grown in Media C and D, respectively. Enzyme activity was measured as described in “Experimental Procedure.” The complete reaction mixture minus acetate or propionate served as a negative control. Es- sentially no activity was obtained when either CoA, enzyme, acetate, or propionate was omitted from reactions.

Enzyme source Acetate activation

/mole acethydroxamate formed/mg grotein/zo min

Soybean bacter- aids............ 0.48

R. japonicum 0.23 R. meliloti. 0.08

TABLE II

Propionate activation

jlmole proQionhydronm?mte formed/mg protein/20 ntin

0.34 0.20 0.09

Requirements of propionyl-CoA carboxylase in cell-free extracts of soybean bacteroids, R. japonicum, and R. meliloti

Bacteroids were isolated from nodules of 31-day-old soybean plants. R. japonicum and R. meliloti were grown in Media C and D, respectively. Enzyme activity was measured as described in “Experimental Procedure.” The complete reaction mixture minus enzyme served as a negative control.

Enzyme activity

system ,“,F$z& R. japonicum

I I R. meliZoti

pmoles HWOa- jked/mg firoteiajhr

Complete...................... 0.89 1.10 0.16

ATP-MgClz omitted. 0.00 0.00 0.00 Propionyl-CoA omitted. 0.00 0.06 0.02 GSH omitted. . 0.72 1.29 0.13

TABLE III

Substrate specificity of propionyl-CoA carboxylase in cell-free extracts of soybean nodule bacteroids, R. japonicum,

and R. meliloti

Bacteroids were isolated from nodules of 31-day-old soybean plants. R. japonicum and R. meliloti were grown in Media C and D, respectively. Enzyme activity was measured as described in “Experimental Procedure.” The complete reaction mixture minus acyl-CoA served as a negative control.

Substrate

Acetyl-CoA Butyryl-CoA Propionyl-CoA.

I

Enzyme activity

Soybean bacteroids R. japonicum R. mliloti

,,moles E’“COa- $xed/mg protein/hr

0.02 0.01 0.04 0.05

I 0.23

I 0.09

1.06 1.61 0.21

TABLE IV

Distribution of propionyl-CoA carboxylase and methylmalonyl-CoA mutase in cell-free extracts of baeteroids from nodules of legumes

and of cells from pure cultures of Rhizobium

Propionyl-CoA carboxylase and methylmalonyl-CoA mutase activity was measured as described in “Experimental Procedure.” For the carboxylase assays, the complete reaction mixture minus propionyl-CoA served as negative control. In the methylmalonyl-CoA mutase assays, a complete reaction was grown in Medium B. R. japonicum, R. phaseoli, R. leguminosarum, and R. trijolii were grown in Medium A. Bacteroids were collected from nodules of the various legumes grown for the number of days indicated in the table.

Soybean bacteroids Soybean bacteroids Pea bacteroids . . Pea bacteroids Lupine bacteroids. Lupine bacteroids.. Cowpea bacteroids Cowpea bacteroids R. meliloti R. japonicum R. phaseoli.. R. leguminosarum R. trifolii

--

Age of xganism

diZYS

35 32 35 32 37 54 37 54

1 4 1 1 1

-

c .-

I

Propionyl-CoA mboxylase activity

mole zmC08 fixed/ mg proteis/lw

0.19

0.10

0.08

0.13

0.10 0.40 0.06 0.07 0.06

-

--

Methylmalonyl-CoA mutase activity

/mole H’~COafixcd into pemanganate-

stable com$ounds/mg puotein/hr

0.84

0.15

0.69

0.69 0.70 0.17 0.33 0.36 0.20

From the data presented in Table II, it is apparent that propionyl-CoA was carboxylated and that the enzyme from these organisms required an ATP-Mg* mixture for activity. The addition or deletion of GSH had no significant effect on the over- all carboxylation reaction. Since Halenz and Lane (31) have described a carboxylase from beef liver that is specific for propionyl-CoA, experiments were conducted to determine the substrate specificity of the carboxylase in extracts of soybean bacteroids and of pure cultures of R. japonicum and R. meliloti.

The results (Table III) provide evidence that propionyl-CoA is carboxylated at a greater rate than either butyryl-CoA and


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


TABLE V TABLE VI

Coenzyme specijicity of methylmalonyl-CoA mutase in cell-free

extract of R. meliloti Effect of varying concentrations of cobalt in growth medium on

The cell-free extract was prepared from R. meliloti grown in Medium D. Enzyme activity was measured by the method described in “Experimental Procedure.” A complete reaction mixture (see “Experimental Procedure”) without enzyme extract served as a negative control for Reactions 1 and 5. In Reactions 2, 3, and 4 and Reactions 6, 7, and 8, a complete reaction mixture without enzyme extract but with AC, BC, or DBC coenzyme as indicated served as the negative controls. The quantity of AC and BC coenzyme added to reaction mixtures (final volume, 1.1 ml) as indicated was 2.6 X 10-S pmole, and the quantity of DBC coenzyme was 2.3 X 10e3 pmole. Certain reaction mixtures indicated in the table contained extract that had been illuminated in direct sunlight for 30 minutes at 0”.

activity of propionyl-Coil carboxylase and methylmalonyl-CoA mutase in cell-free extracts of R. meliloti

Propionyl-CoA carboxylase and methylmalonyl-Coil mutase activity was measured as described in ‘Experimental Procedure.” For the propionyl-CoA carboxylase assays, the complete reaction mixture without propionyl-CoA served as negative control. For the methylmalonyl-CoA mutase assays in which the coenzyme was omitted, the complete reaction mixture without enzyme extract served as negative contro1. For the reaction mixture to which DBC coenzyme was added, the complete reaction mixture without enzyme extract plus coenzyme served as the negative control. The quantity of DBC coenzgme added to the reaction mixture (final volume, 1.1 ml) was 1.65 X 1O-3 Fmole. The abbreviation p.p.b. refers to micrograms per liter.

Reaction Enzyme activity

,moles HWOs fixed into permangenate-

stable com@unds/mg protein/hr

Concentration of Methylmalonyl-CoA mutate activity cobalt added to Propionyl-CoA culture medium carboxylase activity

-DBC I

+DBC

1. Complete.................................... 1.17 2. Complete + AC. .,....................__..._ 1.06 3. Complete + BC............................. 1.26 4. Complete + DBC. 1.23 5. Complete, extract illuminated.. 0.09 6. Complete, extract illuminated, + AC.. 0.03 7. Complete, extract illuminated, + BC 0.48 8. Complete, extract illuminated, + DBC. . 0.78

j.p.6.

None 0.01 1.00

pmoles HWOa- $xed/mg ,moles H’VX)a-j5xed into @rmanganate protein/hr stable compound/mg protein/hr

1.83 0.00 1.22 2.75 0.05 1.15 2.41 1.07 1.07

,

acetyl-CoA. Since Waite and Wakil (32) have isolated a carboxylase from chicken liver that has an absolute requirement for Mn++ and preferentially carboxylates acetyl-Co.4, a preliminary experiment was designed to determine if manganese ions would increase the carboxylation of acetyl-CoA. The results indicated that the carboxylation of acetyl-CoA was not increased when MnClz was used in place of MgC12 in the reaction mixture. With propionyl-Coil as the substrate, extracts of bacteroids from the nodules of four legumes and of five Rhizobium species were then assayed to determine whether or not they have the capacity to carboxylate propionate. A propionyl-CoA carboxylase was found in extracts of each of the organisms studied (Table IV).

As previously mentioned, methylmalonyl-Cob mutase from animal tissues (11-13) and from P. shermunii (14, 15) has been shown to require Blz coenzyme for activity. The results of a survey to determine the occurrence of this enzyme in extracts of bacteroids from nodules of four legumes and of five Rhizobium

species are presented in Table IV. Since Hatch and Stumpf (33) have reported that r4Cmethylmalonate was not metabolized at an appreciable rate by plant tissues which they investigated, an experiment was conducted to determine the localization of methylmalonyl-CoA mutase within the root system of nodulated soybean plants. The results obtained showed that the enzyme was present in cell-free extracts of the isolated bacteroids, but was absent in preparations of the nodular debris, nodular supernatant, and extracts of soybean roots. These results are consistent with the conclusion of Hatch and Stumpf (33).

prepared as described in “Preparation of Cell Suspensions and Extracts” with reduced light conditions. A portion of the extract, maintained in an ice bath at 0”, was subsequently sub- jected to direct sunlight for a period of 30 minutes. This is the procedure utilized by Barker et al. (34) to destroy the AC coenzyme in cell-free extracts of Clostridium tetanomorphum. That portion of the extracts which was not illuminated was used in Reactions 1,2,3, and 4 in Table V. As illustrated by the data in Table V, the addition of the DBC and BC coenzymes to the complete reaction mixtures resulted in a slight stimulation of enzyme activity. The addition of the AC coenzyme caused a slight decrease in enzyme activity of the complete reaction mixture. The addition of the DBC,’ BC, and AC coenzyme to reaction mixtures containing the illuminated enzyme extract resulted in 59, 33 and 0% restoration of the activity, respectively. This evidence, together with that of Kliewer and Evans (lo), is consistent with the conclusion that the Bi2 coenzyme (DBC) is the natural cobamide coenzyme associated with the methylmalonyl-CoA mutase in R. meliloti.

Since Kliewer and Evans (10) have isolated and characterized the DBC coenzyme from R. meliloti, experiments were conducted to determine the coenzyme specificity of the methylmalonyl-CoA mutase in extracts of R. mel;Zoti. For these experiments, R. meliloti was cultured in Medium D, and a cell-free extract was

The effects of varying concentrations of cobalt in the growth medium on the activities of propionyl-Coil carboxylase and methylmalonyl-Cob mutase in extracts of R. meliloti were studied, and the results are presented in Table VI. The activity of propionyl-CoA carboxylase was less in extracts of deficient cells than in extracts of normal cells, but the difference was not striking. In contrast, the activity of methylmalonyl-CoA mutase in extracts of the deficient cells is strikingly reduced when compared with the activity of extracts of normal cells. The addition of 1.65 x 10d3 pmole of DBC coenzyme to each reaction mixture restored full activity to the enzyme extracts of cells grown with no cobalt and with 0.01 p.p.b. of the element. The addition of DBC coenzyme to extracts of cells grown with 1.0 p.p.b. of cobalt, however, failed to stimulate enzyme activity.

1 The abbreviations used for the cobamide coenzymes are: DBC, dimethylbenzimidazolylcobamide; BC, benzimidazolylcobamide; AC, adenylcobamide.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

August 1964 A. A. De Hertogk, P. A. Mayeux, and H. J. Evans 2451

FIG. 5. Radiogram of chromatogram of organic acids isolated solution was passed through a Dowex 50-H+ column (1.5 X 6 from R. meliloti grown on Medium D (see “Experimental Pro- cm) containing 50 to 100 mesh resin. The column was washed cedure”). The reaction mixture consisted of 0.4 g of fresh cells with 4 bed volumes of water, and the effluent was reduced to a in 10 ml of 0.067 M potassium phosphate buffer at 6.8 and 1 ml of volume of 0.5 ml. Aliquots consisting of 100 ~1 were chromato- sodium-propionate-l-14C (3.5 pmoles containing 1 PC). The graphed with isoamyl alcohol saturated with 4 M formic acid (29), mixture was incubated for 10 minutes at 30”, and the reaction was terminated by pouring the mixture into 90 ml of 957, ethanol

and radioactive areas were detected with a strip counter. Com- pounds were identified by cochromatography with known com-

at 78”. The mixture was heated at 78” for 5 minutes and then pounds. RF values were 0.42, 0.69, 0.79, and 0.88 for peaks centrifuged to remove the cell debris. Ethanol was removed from labeled malate, succinate, methylmalonate, and fumarate, the supernatant by evaporation at 75” and the remaining aqueous respectively.

These results suggest that formation of the apoenzyme is inde- pendent of the cobalt supply, but that the catalytic activity of the enzyme may be limited by coenzyme concentration and that synthesis of the coenzyme is directly dependent on cobalt. The data obtained from the radiorespirometric (Fig. 4) and manometric experiments (18) support this conclusion.

Identification of the products of the methylmalonyl-CoA mutase reaction (complete reaction effected with cells grown with 1 p.p.b. of cobalt, Table VI) showed that after the permanganate treatment only succinate was radioactive; however, if the reaction mixture was extracted and chromatographed prior to the permanganate treatment methylmalonate also was found to be radioactive.

mediates as the higher plant system would be expected to result in the most rapid recovery of the methyl carbon of propionate followed in turn by the carboxyl carbon and then the methylene carbon. In contrast, in the pathway whereby propionate is converted to succinate via methylmalonate (29), the expected rate of recovery is greatest for carbon 1. The recovery of carbon atoms 2 and 3 is expected to be equal since they become the methylene carbon atoms of succinate and should become ran- domized in the citric acid cycle.

An experiment was conducted with normal R. meliloti cells to determine the fate of 1% from sodium propionate-1-W when this compound was supplied to intact cells. As shown in Fig. 5, the major compounds that were radioactive after a lo-minute incubation period with sodium propionate-1-i4C corresponded with malate, succinate, methylmalonate, and fumarate in cochromatography experiments. These data are consistent with the other evidence indicating that propionate is converted to succinate via the propionyl-CoA carboxylase and methylmalonyl- CoA mutase route.

To obtain information on the capacity of rhizobia to oxidize propionate and on the possible pathway whereby it is utilized, a series of radiorespirometric experiments were performed with propionate labeled in specific positions. It was found (Figs. 1, 2, and 3) that the rate of recovery of 1% as 14C02 from propionate labeled in different positions was greatest when the cells were incubated with propionate-1-r4C. In addition, the radiorespirometric patterns for W02 formation from propionate-2-14C and propionate-3J4C were essentially the same regardless of the conditions or the Rhizobium species used. These patterns of recovery would be expected only if propionate were utilized by a sequence of reactions in which it was converted to succinate and then oxidized via the citric acid cycle.

DISCUSSION

There are three known pathways whereby propionate is oxidized in biological materials. These pathways result in differ- ences not only in the intermediates formed but also in the patterns of W02 evolution from specifically labeled carbon atoms of propionate. In the P-oxidation pathway reported by Giovanelli and Stumpf (35) and Hatch and Stumpf (33) to operate in certain plant tissues, the expected rate of recovery would be greatest for carbon 1 of propionate following in turn by carbon 3 and then carbon 2. The pathway that has been reported to be operative in Clostridium kluyverii (36) and which involves the same inter-

When R. meliloti was grown with varying concentrations of cobalt in the culture medium, the capacity to oxidize propionate was significantly reduced in cobalt-deficient cells (Fig. 4). The rate of oxidation of propionate by the cells was strikingly increased by an increasing cobalt supply in the culture medium. Arnstein and White (16), using 0. malhamensis, and Ayers (17), using Flavobacterium sp., have conducted similar experiments in which they showed an increase in the rate of propionate oxidation of vitamin BITdeficient cells by the addition of vitamin Blz and certain Biz analogues to cell suspensions. Experiments were conducted, therefore, to determine whether Blz compounds would restore the capacity of cobalt-deficient R. meliloti cells to oxidize propionate. In each instance, experiments of this type proved


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


to be negative. In contrast, the addition of Blz coenzyme to extracts of cobalt-deficient R. meliloti restored activity of methylmalonyl-CoA mutase to a normal level (Table VI). It is apparent, therefore, that cobalt-deficient cells contained a normal level of the apoenzyme of methylmalonyl-CoA mutase and that the content of Blz coenzyme was the limiting factor. These findings are consistent with the report of Kliewer and Evans (10) showing that cobalt deficiency decreased the cobamide coenzyme content of Rhizobium.

The results presented (Tables II and IV) show that cells from pure cultures of R. m&lo& R. japonicum, and bacteroids from soybean nodules have the necessary enzymes to convert propionate to succinate via a pathway in which methylmalonate is an intermediate. Detailed studies of the properties of the propionyl-CoA carboxylase from these sources provided evidence (Table III) that the enzyme preferentially carboxylated propionyl-CoA. The enzyme from these organisms, therefore, exhibits properties similar to those of the propionyl-CoA carboxylase purified from beef liver (31) and pig heart (37). Pro- pionyl-CoA carboxylase is widely distributed in bacteroids from nodules of leguminous species and in pure cultures of Rhizobium

(Table IV), and it is suggested that the enzyme may play an important role in the metabolism of these organisms.

Methylmalonyl-CoA mutase activity was observed in a variety of nodules of legumes and in pure cultures of Rhixobium (Table IV). An examination of various fractions from soybean nodules and soybean roots indicated activity only in the nodule bacteroids. It is of interest to compare the properties of the methylmalonyl-CoA mutase in a crude extract from R. meliloti with those reported for the methylmalonyl-CoA from other sources. Lengyel, Mazumder, and Ochoa (13) made a comparative study of the properties of methylmalonyl-Coil mutase isolated from both sheep kidney and P. shermanii. They observed that activity was restored to the coenzyme-free enzyme from sheep kidney by the addition of the DBC and BC coenzymes. The apoenzyme from sheep kidney showed the strongest affinities for the DBC and BC coenzymes but was inactive with the AC coenzyme. In contrast, the activity of the coenzyme-free enzyme from P. shermanii was restored by the addition of each of the three of the coenzymes. From our experiments with the coenzyme-free methylmalonyl-Cob mutase from R. meliloti, it is apparent that the greatest activation was obtained by the DBC coenzyme and less activation by the BC coenzyme (Table V). The R. meliloti enzyme, therefore, behaved like the enzyme from sheep kidney in that the AC coenzyme was inactive. The methylmalonyl-Coil mutase isolated from R. meliloti was similar to the enzyme from P. shermanii in that the endogenous Blz coenzyme of both was easily destroyed by light. In contrast, the enzyme from the sheep kidney could be resolved into a coenzyme- free apoenzyme only by acid precipitation of the enzyme from ammonium sulfate solution (13). Further characterization of the methylmalonyl-Cod mutase from Rhizobium must be de- layed until a highly purified preparation is obtained from this source.

SUMMARY

An investigation was conducted to determine biochemical sites at which cobalt, in the form of Blz coenzyme, functions in the metabolism of legumes and their associated Rhizobium species. The results of these experiments may be summarized as follows.

1. Data are presented which indicate that propionate is oxi-

dized by cell suspensions of bacteroids from soybean nodules and of Rhizobium japonicum and Rhizobium me&Iota’. The oxidation of propionate by cell suspensions of R. meliloti was enhanced by the addition of sodium propionate, succinate, and methylmalonate to the culture medium, whereas, the addition of these organic acids to the culture media of R. japonicum failed to produce this effect.

2. The rate of propionate oxidation by cell suspensions of R.

meliloti grown with adequate cobalt was markedly greater than that of cells grown in a cobalt-deficient medium.

3. The radiorespirometric patterns obtained by incubation of propionate labeled in specific positions with cell suspensions of R. meliloti, R. japonicum, and soybean bacteroids are consistent with expected patterns if propionate was converted to succinate via methylmalonate and was oxidized by the citric acid cycle.

4. Cell-free extracts of R. meliloti, R. japonicum, and soybean bacteroids exhibit the capacity to catalyze (a) the activation of propionate and acetate, (b) the carboxylation of propionyl coenzyme A, and (c) the conversion of methylmalonate to succinate.

5. The methylmalonyl coenzyme A mutase in extracts of R.

meliloti was easily inactivated by exposure of the extracts to direct light. Enzyme activity was restored by the addition of dimethylbenzimidazolylcobamide or benzimidazolylcobamide coenzyme to the extracts but the adenylcobamide coenzyme was ineffective in restoring activity.

6. Propionyl coenzyme A carboxylase activity in extracts of R. meliloti cells was not greatly affected by the concentration of cobalt in the medium in which cells were grown. In contrast, the activity of the methylmalonyl coenzyme A mutase in the extracts was strikingly influenced by the cobalt content of the culture medium. Methylmalonyl coenzyme ,4 mutase activity, comparable to that obtained with extracts from cells grown with adequate cobalt, was obtained by the addition of dimethylbenzimidazolylcobamide coenzyme to enzyme extracts of cobalt- deficient cells or to extracts of cells grown with 0.01 p.p.b. of cobalt.

It is concluded, therefore, that cobalt deficiency in R. meliloti

prevents the synthesis of quantities of Blz coenzyme adequate for the normal function of methylmalonyl coenayme A mutase and that the inactive mutase results in the failure of the organism to oxidize propionate.

1. AHMED, S., AND EVANS, H. J., Biochem. and Biophys. Research Communs., 1, 271 (1959).

2. 3.

AHMED, S., AND EVANS, H. J., Soil Sci., 90,205 (1960). AHMED, S., AND EVANS, H. J., Proc. Natl. Acad. Sci. U. S.,

47, 24 (1961). 4. DELWICHE, C. C., JOHNSON, C. M., AND REISENAUER, H. M.,

Plant Physiol., ‘36, 73 (1961). 5.

6.

7. 8.

9. 10. 11.

12.

13.

HALLSWORTH. E. G.. WILSON. S. B., AND GREENWOOD. E. A. N., Nat&e, 187, $9 (1960).’ ’

LOWE, R. H., EVANS, H. J., AND AHMED, S., Biochem. and Biophys. Research Communs., 3, 675 (1960).

LOWE, R. H., AND EVANS, H. J., J. Bacterial., 83, 210 (1962). KLIEWER, M., AND EVANS, H. J., Nature, 194, 108 (1962);

Arch. Biochem. Biophys., 97, 428 (1962). KLIEWER, M., AND EVANS, H. J., Plant Physiol., 38, 55 (1963). KLIEWER, M., AND EVANS, H. J., Plant Physiol., 38, 99 (1963). SMITH, R. M., AND MONTY, K. J., Biochem. and Biophys.

Research Communs., 1, 105 (1959). GURNANI, S., MISTR~, S. P.,. AND. JOHNSON, B. C., Biochim.

et Biophys. Acta, 38, 187 (1960). LENGYEL, P., MAZUMDER, R., AND OCHOA, S., Proc. Natl.

Acad. Sci. U. S., 46, 1312 (1960).

REFERENCES


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


14. STADTMAN, E. R., OVERATH, P., EGGERER, H., AND LYNEN, F., Biochem. and Biophys. Research Communs., 2,l (1960).

15. STJERNHOM, R., AND WOOD, H. G., Proc. NatZ. Acad. Sci. U. S., 47,303 (1961).

16. ARNSTEIN, H. R. V., AND WHITE, A. M., Biochem. J., 33, 264 (1962).

17. AYERS, W. A., Arch. Biochem. Biophys., 96, 210 (1962); Can. J. Microbial., 8, 861 (1962).

18. DE HERTOGH, A. A., MAYEUX, P. A., AND EVANS, H. J., J. Bacterial., 87, 746 (1964).

19. CHENIAE, G., AND EVANS, H. J., Biochim. et Biophys. Acta, 36, 140 (1959).

20. UMBREIT, W. W., BURRIS, R. H., AND STAUFFER, J. F., Mano- metric techniques, Ed. 3, Burgess Publishing Co., Minneapo- lis, 1957, p. 274.

21. LAYNE, E., in S. P. COLOWICK AND N. 0. KAPLAN (Editors), Methods in enzymology, Vol. III, Academic Press, Inc., New York, 1956, p. 447.

22. LANE, M. D., AND HALENZ, D. H., in S. P. COLOWICH AND N. 0. KAPLAN (Editors), Methods in enzymology, Vol. V, Academic Press, Inc., New York, 1962, p. 576.

23. WANG, C. H., STERN, I., GILMOUR, C. M., KLUNGSOYR, S., REED, D. J., BIALY, J. J., CHRISTENSEN, B. C., AND CHELDE- LIN, V. H., J. Bacterial., 76, 207 (1958).

24. WANG, C. H., AND KRACKOV, J. K., J. Biol. Chem., 237, 3614 (1962).

25. BERG, P., in S. P. COLOWICK AND N. 0. KAPLAN (Editors), Methods in enzymology, Vol. V, Academic Press, Inc., New York, 1962, p. 461.

26. GROSSOWICZ, N., WAINFAN, E., BOREK, E., AND WAELSCH, H., J. BioZ. Chem., 187,111 (1950).

27. SIMON, E. J., AND SHEMIN, D., J. Am. Chem. Sot., 76, 2520 (1953).

28. HEGRE, C. S., MILLER, S. J., AND LANE, M. D., Biochim. et Biophys. Acta, 66, 538 (1962).

29. FLAVIN, M., AND OCHOA, S., J. BioZ. Chem., 229, 965 (1957). 30. BEINERT, H., GREEN, D. E., HELE, P., HIFT, H., VON KORFF,

R. W., AND RAMAKRISHNAN, C. V., J. BioZ. Chem., 203, 35 (1953).

31. HALENZ, D. R., AND LANE, M. D., J. BioZ. Chem., 236, 878 (1960).

32. WAITE, M., AND WAKIL, S. J., J. BioZ. Chem., 237, 2750 (1962). 33. HATCH, M. D., AND STUMPF, P. K., Arch. Biochem. Biophys.,

96, 193 (1962). 34. BARKER, H. A., SMYTH, R. D., WEISSBACH, H., MUNCH-

PETERSEN, A., TOOHEY, J. I., LADD, J. N., VOLCANI, B. E., AND WILSON, R. M., J. BioZ. Chem., 236, 181 (1960).

35. GIOVANELLI, J., AND STUMPF, P. K., J. Am. Chem. Sot., 79, 2652 (1957); J. Biol. Chem., 231, 411 (1958).

36. VAGELOS, P. R., AND EARL, J. M., J. BioZ. Chem., 234, 2272 (1959).

37. KAZIRO, Y., OCHOA, S., WARNER, R. C., AND CHEN, J., J. BioZ. Chem., 236, 1917 (1961).


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

A. A. De Hertogh, Patricia A. Mayeux and Harold J. EvansRhizobium

The Relationship of Cobalt Requirement to Propionate Metabolism in

1964, 239:2446-2453.J. Biol. Chem.

http://www.jbc.org/content/239/8/2446.citation

Access the most updated version of this article at

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/239/8/2446.citation.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/content/239/8/2446.citation

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;239/8/2446&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/239/8/2446.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=239/8/2446&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/239/8/2446.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/239/8/2446.citation.full.html#ref-list-1

http://www.jbc.org/

the relationship of cobalt requirement to propionate ... · ments at 540 rnp were converted to...

Documents