Ribonucleotide reductase from Euglena gracilis, a deoxyadenosylcobalamin-dependent enzyme.

Abstract A ribonucleotide reductase requiring 5'-deoxyadenosylcobalamin has been detected in Euglena gracilis var. bacillaris by measuring 3H exchange from labeled coenzyme to water. Extracts were partially purified by centrifugation, dialysis, and treatment with MnCl2. These extracts catalyzed the reduction of ribonucleoside triphosphates to the corresponding deoxyribonucleotides as measured by colorimetric assay with the diphenylamine procedure. The enzyme required 5'-deoxyadenosylcobalamin, a dithiol, and a ribonucleoside triphosphate for exchange activity and reduction. Monothiols could not replace the dithiol requirement. Ribonucleoside diphosphates were slightly active as substrates in both assay systems. Mg++, at low concentrations, did not affect the reaction; at concentrations greater than 2 mm, this ion was inhibitory. The reduction of ribonucleotides was inhibited by adding deoxyribonucleotides other than the immediate products; e.g. ATP reduction was inhibited by adding dGTP, dCTP, or dUTP, but not by dATP. dTTP inhibited the reduction of all ribonucleotides to the greatest extent. The molecular weight of this enzyme, determined by sucrose density gradient centrifugation, was approximately 145,000. Extracts of Ochromonas malhamensis and Ochromonas danica showed no activity in the 3H exchange assay under similar conditions.

F. I< Extracts were partially purified by centrifugation, dialysis, and treatment with MnC12. These extracts catalyzed the reduction of ribonucleoside triphosphates to the corresponding deoxyribonucleotides as measured by colorimetric assay with the diphenylamine procedure.
The enzyme required 5'-deoxyadenosylcobalamin, a dithiol, and a ribonucleoside triphosphate for exchange activity and reduction.
Monothiols could not replace the dithiol requirement. Ribonucleoside diphosphates were slightly active as substrates in both assay systems. Mg++, at low concentrations, did not affect the reaction; at concentrations greater than 2. mM, this ion was inhibitory.
The reduction of ribonucleotides was inhibited by ad.ding deoxyribonucleotides other than the immediate products; e.g. ATP reduction was inhibited by adding dGTP, dCTP, or dUTP, but not by dATP. dTTP inhibited the red.uction of all ribonucleotides to the greatest extent.
The molecular weight of this enzyme, determined by sucrose density gradient centrifugation, was approximately 145,000. Extracts of Ochromonas malhamensis and Ochromonas danica showed no activity in the 3H exchange assay under similar conditions.

Blakley
(I) and co-workers have isolated a ribonucleotide reductase from Lactobacillus leichmannii, which requires 5'-deoxyadenosylcobalamin for activity and which reduces only the ribonucleoside triphosphates at an appreciable rate. Cell-free extracts of Rhizobium meliloti also contain a deoxyadenosylcobalamin-dependent ribonucleotide reductase (2) ; ribonucleoside diphosphates are the preferred substrates for this enzyme. In a preliminary study, Blakley et al. (3) have also shown that a number of strains of Lactobacillus acidophilus have a high level of deoxyadenosylcobalamin-dependent reductase activity, while Abeles and Beck (4) were able to show that cell-free extracts of *This work \\-a~ supported by IJnited States Public Health Service Grants AM-08627 and C;R4 00550 from the National Institutes of ITcalth.
Clostridium sticklandii and Clostridium tetanonzorphum catalyze tritium exchange between 5'-deoxyadenosylcobalamin-5-3Hz and water. This hydrogen exchange reaction is considered to be an integral part of the reduction and thus serves as a qualitative assay for a dcoxyadenosylcobalamin-requiring ribonucleotide reductase.
The evidence accumulated thus far indicates that the cobalamin-dependent reductase occurs in several different types of microorgansims.
On the other hand, ribonucleotide reduction in mammalian systems and in several microbial preparations is not stimulated by 5'-deoxyadenosylcobalamin (5). The requirements for the mammalian systems appear to be very similar to those established for the Escherichia coli reductase by Reichard (5) and co-workers.
Both Euglena gracilis and Ochromonas malhamensis require vitamin Blz for growth.
These algae are routinely used in microbiological assays for vitamin B12 and other cobalamins in blood and various materials.
Whereas the requirement of a cobalamin in L. leichmannii is spared by deoxyribonucleosides, E. gracilis grown in the absence of a cobalamin does not respond to either deoxyribosidcs or methionine (6). However, vitamin B12 depletion in both L. leichmannii and 13. gracilis leads to the formation of abnormally la,rge cells and prolongation of the generation time (7, 81, suggesting that in both organisms vitamin B12 participates in DNA synthesis.
In contrast, the cobalamin requirement of 0. malhamensis is spared by methionine and not affected by deoxyribonuclcosides.
By using the extremely sensitive hydrogen exchange reaction between 5'-deoxyadenosylcobalamin-5-3H2 and water, the presence of a cobalamin-requiring ribonucleotide reductase has been detected in extracts of E. gracilis.
This communication describes several properties of this partially purified ribonucleotide reductase from R. gracilis.

EXPERIhlENTAL PROCEDURES
Materials-Nucleotides were purchased from P-L Biochemicals. DL-Lipoic acid, dithiothreitol, dithiocrythritol, rabbit muscle lactic dehydrogenase, and pig heart diaphorase were obtained from Sigma.
Pig heart malate dchydrogenase was purchased from Boehringer. Dihydrolipoate was prepared by reduction of lipoic acid (9) without distillation of the product. 5'-Deoxyadenosylcobalamin-5'-3H:! (20 &i per ~molc) was synthesized by the method of Gleason and Hogenkamp (10). Ribonucleotide reductase from L. leichmannii was kindly supplied by Dr. R. L. Blakley; this preparation has a specific activity of 70 pmoles per hour per mg of protein.
Cultures of II. gracilis var. Methods-E. gracilis was maintained on a semisolid medium (11) at room temperature and subcultured every month. To obtain large quantities of cells, the algae were grown in 2800-ml Fernbach flasks containing 1 liter of the following medium: 1 g of bovine extract (Baltimore Biological Laboratory), 2 g of Difco t,ryptone, 2 g of Difco yeast extract, and 0.01 g of calcium chloride.
The medium was adjusted to approximately pH 6 with 5 N HCl.
The flasks were autoclaved twice and inoculated either from 15-ml culture tubes or with approximately 100 ml of actively growing cells. The cultures were grown in continuous illumination with banks of fluorescent lamps at 18-20". Magnetic stirrers were used to agitate the medium. 0. malhamensia and 0. danica wcrc grown in a similar manner with the following medium: 1 .O g of glucose, 1.0 g of Bacto-tryptone, 1.0 g of Difco yeast extract, 40 ml of Difco liver infusion, and 960 ml of distilled water.
E. gracilis was also grown holophytically in a defined medium (12). The cultures were continuously flushed with a mixture of 5% CO2 and air under sterile conditions. In all cases, ccl1 growth was followed by recording the absorbance at 650 rnp. The cells were harvested in the logarithmic phase (absorbance at 650 rnp approximately 1.5). Preparation of Cell-free Extract-All operations were carried out at 4". The cells were harvested by centrifugation at 8,000 X g and suspended in a solution containing 0.01 M Tris-HCl buffer (pH 7.4), 0.005 M mercaptoethanol, and 0.3 M sucrose. The cells were again sedimented at 8,000 x g and resuspended in 70 ml of the above solution but without the sucrose.
Six liters of culture yielded about 10 g of packed cells. A small amount of neutral Norit-A charcoal (approximately 50 mg per ml of cell suspension) was added. The cells were then disrupted by sonic oscillation for 10 min in a Raytheon 10 KC sonic oscillator. The cell debris and charcoal were removed by centrifugation at 37,000 x g for 15 min.
These crude extracts usually contain 20 to 25 mg of protein per ml and can be used in the tritium exchange assay without further treatment.
This crude extract slowly loses activity when stored at -10". For further purification, the cell-free extract was centrifuged at 29,000 x g for 1.5 hours.
The supernatant was removed and dialyzed for 48 hours against several changes of 0.01 M Tris-HCl buffer, pH 7.4, containing 0.005 M mercaptoethanol. The dialyzed extract was then treated with MnC12 to remove chlorophyll and nucleoproteins.
The nucleic acid concentration was estimated from the absorbance at 260 rnp (E&b% = 20) and 0.005 ml of a 1 M MIX& solution was added per mg of nucleic acid. The solution was stirred for 5 min and then centrifuged at 12,000 x g for IO min.
The resulting supernatant was dialyzed for 24 hours against 0.05 M sodium dimethylglutarate buffer, pH 7.2, containing 0.005 M mercaptoethanol. Tllis partially purified extract can be stored at 4" and remains active for approximately 1 month. Assay Procedure-Two procedures were used to assay ribonucleotide reductase activity.
The coenzyme was added to the assay mixture in dim light, and the reaction was initiated by the addition of enzyme. The assay tubes were incubated in the dark at 37", and the reaction was terminated by freezing in liquid nitrogen.
Water was removed from the tubes by sublimation, and a O.l-ml aliquot of the thawed sublimate was added to 10 ml of scintillation fluid (14). Radioactivity was determined in a Packard model 3003 Tri-Carb liquid scintillation spectrometer.
Specific activity is defined as the radioactivity released to the solvent (counts per min) per mg of protein after 30.min incubation.
In Method 2, ribonucleotide reductase activity was determined by measuring the amount of deoxyribonucleotide formed by the diphenylamine procedure of Blakley (15). The composition of the reaction mixture was the same as that described in Method 1, except that unlabeled 5'-deoxyadenosylcobalamin was used. When assaying for the reduction of purine nucleotides, the reaction was stopped by placing the tubes in ice and adding 0.4 ml of 0.5 M chloroacetamide in 0.25 M potassium phosphate buffer, pH 7.3. The tubes were heated for 10 min in a boiling water bath.
After cooling the tubes, 2 ml of diphenylamine reagent were added.
The mixtures were then incubated at 37" for 4 hours, and the absorbance at 595 rnp was determined.
The amount of product formed was estimated from standard curves for dATP or dGTP.
In assaying for the reduction of pyrimidine nucleotides, the reaction was terminated by adding 0.2 ml of 1 N I-ICY. The precipitated protein was removed by centrifugation at 50 x g for approximately 5 min. The supernatant solution was extracted twice with 2 ml of petroleum ether to remove lipoic acid, and the ether phase was discarded.
The solution was then mixed with 0.2 ml of saturated bromine water and allowed to react for 10 min; 0.050 ml of 5 N NaOH was added and the solutions were allowed to stand for 10 min for dCMP estimation or 40 min for dUMP estimation.
Two milliliters of diphenylamine reagent were then added, and the tubes were incubated for 4 hours at 50" for color production with dCMP or at 37" with dUM1'.
Absorbance at 595 rnp was determined and the amount of product was estimated from standard curves for the respective nucleotides.

RESULTS
Growth Conditions-In general, cells harvested in the mid to late logarithmic growth phase showed the highest enzyme activity, whereas extracts of cells harvested in the early logarithmic growth phase were only slightly active.
Extracts prepared from cells harvested in the stationary phase were inactive. The effect of specific growth conditions on the level of ribonucleotide reductase activity as measured by Method 1 is shown in Table I (17) were unsuccessful. The more active preparations used in the following procedures were obtained by inoculating the Fernbach flasks with 100 ml of a culture containing cells in the mid logarithmic growth phase. to the corresponding deoxyribonucleoside triphosphates as well as the hydrogen exchange reaction between 5'-deoxyadenosylcobalamin-5'-aHQ and water.
The results presented in Table III indicate that both reactions have an absolute requirement for 5'-deoxyadenosylcobalamin, a reducing agent and a nucleotide.
The effect of various thiols on the rate of both reactions is shown in Table IV.
It is clear that the dithiols such as dihydrolipoate, dithiothreitol, and dithioerythritol are much more active than the monothiols.
It is interesting to note that dithiothreitol and dithioerythritol are superior to dihydrolipoate in promoting the hydrogen exchange, while in the reduction reaction dihydrolipoate and dithiothreitol are equally active. The effect of Mg++ ion on the reduction of ATP by the partially purified reductase of E. gradis is shown in Fig. 1. These results show that this reductase system does not require Mg"+ ion, and, furthermore, that at higher concentrations this metal ion inhibits the reaction.
The effects of various nucleotide substrates on the rate of  reduction and the rate of hydrogen exchange are shown in Table  V. These results indicate that the enzyme from E. gracilis is similar to that from L. Zeichmannii in that both enzyme systems are ribonucleoside triphosphate reductases. This is in contrast to the ribonucleoside diphosphate reductase from E. coli (5) and the enzyme from R. meliloti which reduces nucleoside di-as well as triphosphates (2). The results also indicate that the hydrogen exchange reaction is not only promoted by ribonucleoside triphosphates but also by deoxyribonucleoside triphosphates. Effect of DeoxyribonucleotidesThe ribonucleotide reductase of L. leichmannii catalyzes the reduction of the five ribonucleoside triphosphates at markedly different rates (18). However, in the presence of specific deoxyribonucleotides, the relatively low rates of ATP and CTP reduction are increased to the rate of GTP reduction.
A similar study for the ribonucleotide reductase of E. gracilis shows several fundamental differences between these two reductases (see Table VI).
Unlike the bacterial enzyme, the reductase of E. gracilis reduces all four substrates at approximately the same rate. In addition, the rate of reduction is not stimulated by a specific deoxyribonucleotide, but rather the reduction of a particular purine ribonucleotide is inhibited by all deoxyribonucleotides except its own product.
Thus, as shown in Table VI, the reduction of GTP is inhibited by dATP, dCTP, dUTP, and dTTP but is not affected by dGTP.
On the other hand, the reduction of ATP is not affected by dATP but is inhibited by dGTP, dCTP, dUTP, and dTTP. dTTP was found to be the most effective inhibitor of both reduction and exchange. It was the only deoxyribonucleotide that inhibited the reduction of all ribonucleotides.
Recently, Brown et al. (19) have shown that the ribonucleotide reductase from E. coli is either inhibited or stimulated by dATP, depending on the concentration of the deoxyribonucleotide used. To test for such a dual effect in the E. gracilis ribonucleotide reductase, the inhibition by two deoxyribonucleotides, dTTP and dGTP, was studied in greater detail. These two deoxyribonucleotides were found only to inhibit the reduc-  tion of ATP; the extent of inhibition was directly proportional to the amount of deoxyribonucleotide added, over a concentration range of 10T6 to low3 M. Technical limit'ations of the colorimetric assay precluded the determination of meaningful kinetic constants for this inhibitory effect. Estimation of Molecular Weight-The molecular weight of the partially purified ribonucleotide reductase from E. gracilis was estimated by sucrose density gradient centrifugation. Linear sucrose gradients, 5 to 20%,, with a total volume of 4.6 ml were prepared by the technique described by Martin and Ames (20) except that the mixing chamber was agitated by a magnetic bar rather than a stirring rod. Sucrose was buffered with 0.05 M dimethylglutarate, pH 7.2, containing 0.005 M mercaptoethanol. Approximately 5 to 7 mg of a partially purified extract from E. gracilis as well as varying amounts of standards (total protein less than 10 mg) were layered on the gradients.
The gradients were centrifuged in a SW 50.1 rotor for 12 hours in a Spinco model L ultracentrifuge at 140,000 x g. The rotor was decelerated without brake.
The bottom of the gradient tubes was punctured and drops were collected to give approximately 25 fractions per gradient.
The fractions were assayed for enzyme activity.
Commercial preparations of rabbit muscle lactic dehydrogenase, pig heart malate dehydrogenase, and pig heart diaphorase were used as standards. These enzymes were assayed by published techniques (References 21, 22, and 23, respectively).
L. Zeichmannii ribonucleotide reductase, also a standard, and the E. gracilis enzyme were assayed by the 3H exchange procedure.
A typical sedimentation pattern can be seen in Fig. 2 In this experiment the fractions were collected from the top rather than the bottom of the gradient.
The positions of the two reductases were est,ablished in separate gradients with rabbit mllscle lactate dehydrogenase as a marker. The act,ivity of t,he ribonucleotide reductases was measured by the tritium exchange reaction and is expressed as the count,s per min X 10-z released to the solvent under standard conditions. The activity of pig heart diaphorase was measured spectrophotometrically at 42Q rnp by following the reduction of K,Fe(CN)e and is expressed as the change in absorbance X 10 per min.
The activity of rabbit mllscle lact,ate dehydrogenase was measured by following oxidation of I)PNII at 340 rnp and is expressed as the change in absorbance X 10 per min.
With these values, the molecular weight of E. gracilis ribonucleotide reductase was estimated to be in the range of 140,000 to 150,000.
The results presented here clearly indicate that E. gracilis contains a ribonucleoside triphosphate reductase dependent on 5'-deoxyadenosylcobalamin.
With the extremely sensitive hydrogen exchange reaction between 5'-deoxyadenosylcobalamin-5'-3H2 and water, low levels of enzyme activity could be detected in cell-free extracts prepared from E. gracilis grown under a variety of conditions.
In contrast, extracts of 0. malhamensis and 0. &mica grown under similar conditions are unable to promote this hydrogen exchange.
The requirements for a dithiol as a reducing agent, a ribonucleoside triphosphate as a substrate, and 5'-deoxyadenosylcobalamin as a cofactor are identical with the requirements of the reductase system isolated from L. Zeichmannii.
The enzyme from L. leichmunnii, the only deoxyadenosylcobalamin-requiring ribonucleotide reductase investigated in detail, shows specific stimulation by addition of deoxyribonucleotides (15,28), and it has been shown kinetically that the effecters bind to an allosteric site on the enzyme (29).