Enzymatic Synthesis of Deoxyribonucleotides

Full text

THE JOWNAL OP Bromorca~ CHEYI~TBY
Vol. 238, No. 10, October 1963
Printed in U.S.A.
Enzymatic Synthesis of Deoxyribonucleotides
III. REDUCTION OF PURINE RIBONUCLEOTIDES ?VITH AN ENZYME SYSTEM
FROM ESCHERICHIA COLI B*
AGNE LMSSON
From the Department of Medical Chemistry, Uppsala University, Uppsala, Sweden
(Received for publication, May 20, 1963)
The reduction of pyrimidine ribonucleotides to deoxyribo-
nucleotides was demonstrated with soluble enzymes of msm-
malian (1,2), avian (3)) and bacterial (4-7) origin. This reaction
occurred without cleavage of the bond between the pyrimidine
and the sugar.
Reichard has purified an enzyme system from Escherichiu
coli B, called Fraction B, which catalyzed the formation of pyrim-
idine deoxyribonucleotides from the corresponding ribonucleo-
tides (4, 5). The reduction of both cytidine and uridine phos-
phates showed absolute requirements for adenosine triphosphate,
Mg++, and reduced lipoic acid and took place at the diphosphate
level. With partially purified enzymes from Novikoff hepatoma,
Moore and Hurlbert demonstrated a reduction of cytidine nucleo-
tides to deoxycyticlme nucleotides (l), and some evidence was
obtained that crude extracts of this tumor could catalyze the
formation of deoxyuridine, deoxyguanosine, and deoxyadenosine
phosphates from the corresponding ribonucleotides. A closer
study of the latter reactions was not made.
So far very little information has been obtained about the
transformation of purine ribonucleotides to deoxyribonucleo-
tides. Reichard has shown that extracts of chick embryos could
form deoxyguanosine phosphates from guanosine phosphates,
and that this reaction was a direct reduction of the ribonucleo-
tide (3). The reaction was stimulated by ATP and Mg*, but
the level of phosphorylation at which the reduction occurred
was not determined.
In the present paper, evidence is given that both Fraction B
from E. coli B and partially purified enzymes from Novikoff
hepatoma catalyze the reduction of adenosine and guanosine
phosphates to deoxyribosyl compounds. A study of the require-
ments of these reactions was made with the bacterial enzyme
system.
EXPERIMENTAL PROCEDURE
AMP, deoxyribosyl compounds, and TPNH were obtained
from Sigma Chemical Company; the other nucleosides and
nucleotides, from California Corporation for Biochemical Re-
search.
Some commercial CDP preparations (Schwarz and Sigma)
were found to be contaminated with up to 2.2 To deoxyribonucleo-
tides and could not be used in the present experiments. The
l
This investigation was supported by grants from the Swedish
Medical Research Council, from the Damon Runyon Memorial
Fund for Cancer Research (DRG-47OC), and from the United
States Public Health Service (CA
06144)
to Dr. P. Reichard.
ribonucleotides used contained less than 0.1% of deoxyribosyl
compounds.
Tritium-labeled CDP (6), reduced &lipoic acid (4), and Fac-
tor& from E. coli B (7) were preparations available in this lab-
oratory.
Fraction B was prepared as described earlier (4). Fraction
Pd from Novikoff hepatoma (7) was a generous gift from Dr.
E. C. Moore. Crude dried snake venom from Crotulus ado-
m&us was obtained from Ross Allen’s Reptile Institute, Silver
Springs, Florida.
Standard Condi&ms of Incubation-The standard incubation
mixture contained the following components: 0.1 ml of Fraction
B (0.3 to 0.9 mg of protein), 0.1 pmole of ADP or GDP or CDP
or UDP, 1.0 pmole of ATP, 2.5 pmoles of MgClz, 0.5 Ctmole of
reduced dl-lipoic acid, 0.2 pmole of EDTA, and 7.5 Ctmoles of
Tris bulfer, pH 7.6, in a total volume of 0.2 ml. After incubation
at 37’ for 30 minutes the reaction was stopped by boiling. When
changes important for the interpretation of the results were
made, this will be recorded in the text.
Assay of Deoxyribosyl Compou&3---The amount of deox-
yribosyl compounds formed was determined by a microbiological
assay. The organisms used were Lac.!obacillw leichmunnii 313
(ATCC 7830),’ requiring for growth either deoxyribonucleosides
or vitamin B1a (8), and Laetobacillus ucidophilus R-26 (ATCC
11506), having a specific requirement for deoxyribonucleosides
(9.
Before each assay the pH of the boiled incubation mixture
was adjusted to around 9 with M NaOH, and crude dry venom
of Crotalus adomanteus (1 mg per 0.2 ml of incubation mixture)
was added. The solution was incubated at 37” for 1 hour in
order to dephosphorylate 5’-nucleotides. The reaction was
stopped by boiling the solution, and the pH was adjusted to
around 7 with M acetic acid.
Since the growth of L. kichmunnii is supported by both deox-
yribonucleosides and vitamin B1z, it was necessary to eliminate
the latter compound when this organism was used for the assay
of deoxyribonucleosides. For this purpose, 0.1 ml of M NaOH
was added to the incubation mixture directly after the dephos-
phorylation with the snake venom enzyme to give a final pH of
about 12. The solution was boiled for 15 minutes, and the pH
was then adjusted to around 7 with M acetic acid.
The sample to be assayed was diluted to 1.5 ml, and the de-
natured protein was removed by centrifugation. In general,
1 Generous gift from Civil Engineer A. Bolinder, Division of
Food Chemistry, Royal Institute of Technology, Stockholm.
3414
This is an Open Access article under the CC BY license.

October 1963
A. Larsson
3415
1.0
2 4
thymidine (mpmoks)
FIG. 1
(left). Typical standard curves for thymidine obtained
with L. leichmannii (O- --O) and L. acidaphilus (o-o).
FIG.
2 (right). Comparison between isotope and microbiological
assays for deoxyribonucleotide formation. The standard mixture
containing Ha-CDP was incubated at 37”. The reaction was
stopped by heating the vessels to 100”. One portion of the mixture
three different aliquots of each sample were taken for the micro-
biological assay. The deoxyribonucleoside content of the sample
was calculated from the average of the three values.
Deoxyribosyl compounds were determined by a turbidimetric
tube assay, mainly as described by Seidler, Nayder, and Schwei-
gert (10). This method is a modification of HoffJZrgensen’s
original method (9). The following further changes were made.
(a) Acid-hydrolyzed casein (Difco Bacto-C&amino acids, tech-
nical) was used in place of enzyme-hydrolyzed casein (Difco
Bacto-C&one) in the assay medium, since the latter contained
quite large amounts of “deoxyriboayl” compounds (10 to 20
pmoles per g). (b) The final volume of the assay tubes was
reduced from 10 to 3 ml. Each tube wss inoculated with 2.5
x 10” bacteria and incubated in a 37” water bath for 24 hours.
Maximal growth was
always
reached in less than 20 hours. After
24 hours, the absorbancy at 640 rnp was determined in
a
Beck-
man B spectrophotometer.
RESUIRS AND DISCUSSION
Microbi~lo~ Determiruztion of
Deaxyibo~l Compounda
In agreement with earlier work (ll), it was found that, equi-
molar amounts of all deoxyribonucleosides tested showed identi-
cal growth-promoting effects for both
L.
ldchmannii and
L.
acidophi2u.s.
For practical rea~ns, thymidine wsa chosen for
the standard curves.
Fig. 1 shows typical thymidine standard curves obtained with
either of the two organisms. The amount of thymidine per
tube (3 ml) is plotted against the absorbancy at 640 rnp. Good
proportionality was obtained with both lactobacilli in the in-
terval from 0 to 2.0 mrmoles of deoxyribonucleoside. Another
finding demonstrated by Fig. 1 is that a given concentration of
deoxyribonucleosides supported higher growth of
L.
leichmannii
than of L. acidophilus, although this difference was not constant.
Deoxyribonucleoside polyphosphstes cannot be utilized by
la&bacilli (12). Since the majority of the deoxyribosyl com-
pounds formed during the reduction of ribonucleotides were
expected to be present as di- and triphosphatea, they had to be
6
was analyzed for deoxycytidine phosphates by the isotope assay
(6) (Curve b). The content of deoxyribosyl compounds of the
remaining part was determined with L. ldchmannii (Curve a). In
parallel a control without CDP was incubated and assayed with
L. leichmannii (Curve c). All vessels contained ATP as a com-
ponent of the usual assay system.
TABLE I
Recovery of added dsoxyribonucleotides
Fraction B (1 mg) was incubated at 37” with !8 wmoles of
each deoxyribonucleotide, 2 pmoles of ATP, 5
pmoles
of MgClt,
15 rmoles of Tris buffer, pH 7.6, and 0.4 pmole of EDTA. The
final volume was 0.4 ml. After different times of incubation,
aliquota were removed, dephoaphorylated with snake venom, and
assayed with L. leichmannii.
Deoxyribosyl compounds present
Added deoxyrfbo- after incubation for
nucleotide Recovery
Omio ) Smin 1 10min 1 20min
mflolu %
dAMP.. . 26.6 27.2 26.7 27.6 98-102
dGMP . . . . . . . . . . . . 29.1 29.0 28.5 25.7 95-108
dCMP . 26.6 25.5 26.8 26.9 9PlO9
dUMP . . 27.2 25.1 27.8 27.9 93-103
dephosphorylated. For this reason all nucleotides were trans-
formed to nucleosides by treatment with snake venom. The
mixture ta be analysed by the microbiological method therefore
contained large amounts of ribonucleosides besides the presump-
tive deoxyribonucleosidea.
Schneider (13) and others (14, 15) found that ribonucleotides
inhibited the growth of
L.
aeidqphi1u.s when deoxyribonucleotides
were used as growth factors. The results shown in Table I
exclude a similar &ect in the present assays
with L. kichmunnii.
Mixtures containing standard concentrations of Fraction B, ATP,
and Mg++ were incubated with known amounts of dAMP,
dGMP, dCMP, and dUMP, respectively. The amounts of
deoxyribosyl compounds present after different times of incuba-
tion were assayed with
L.
leichnaannii with 93 to 103yo recovery.
The good recovery shows that ribonucleosides at‘ the concentra-
tions used did not influence the growth of
L.
leichmannii with
deoxyribonucleosides. Similar experiments with L. ac&?ophilus
gave the same result. The experiment, shown in Table I, also
demonstrates that Fraction B was not contaminated with en-
zymes that degraded deoxyribosyl compounds.

Enzymatic Synthesis of Deoxyribonucleotides. III Vol. 238, No. 10
TABLE
II
Comparison of L. leichmannii and L. acidophilua under
different assay
conditions
The standard mixture containing CDP was incubated at 37”
for 15 minutes. Two aliquots were assayed with
L. leichmannii
and
L.
acidophilus, respectively, without further treatment. The
rest of the mixture was incubated for 1 hour with snake venom and
two aliquots were assayed again. Finally the remainder of the
mixture was boiled for
15
minutes at pH 12 before analysis.
Treatment after incubation
“Deoxyribosyl” compounds
L. laichmannii L. acido)hilur
mjmwles
None................................... 6.0 0
Dephosphorylation . 9.4 4.4
Dephosphorylation + boiling at pH 12. 5.7 4.4
Fraction B itself contained some material permitting the
growth of L. tihmonnii but not that of L. acidophilus. This
point is illustrated by another experiment, shown in Table II.
Fraction B was incubated with CDP under standard conditions
at 37” for 15 minutes, and the reaction was stopped by boiling
the solution. Aliquots were taken and assayed, without further
treatment, with both lactobacilli. Potent stimulation of L.
bichmannii was observed, but no growth of L. acidophilus occur-
red. This part of the experiment shows that the growth of L.
leichmunnii could not be due to deoxyribosyl compounds but
possibly could be attributed to vitamin Brr.
The remaining phosphate esters of the incubation mixture
were dephosphorylat-ed with snake venom, and aliquots were
again assayed with both lactobacilli. Now growth was obtained
in both cases, but L. Zeichmannii gave higher results than L.
actibphilus.
Since the growth of the latter was supported only
after dephosphorylation of the products in the incubation mix-
ture, the experiment shows that deoxycytidine pyrophosphates
were formed during the initial incubation. These are not utilized
by the la&bacilli without dephosphorylation (12).
Finally, the remaining part of the original incubation mixture
was boiled at pH 12 for 15 minutes and again assayed with both
lactobacilli. This resulted in a considerable decrease of the L.
leichmannii- supporting activity, but no change with L. acid-
ophilus
was observed. After treatment with alkali both la&o-
bacilli were supported to about the same extent. The results
strongly indicate that the material which specifically promoted
the growth of L. kichmannii was vitamin Bia or a closely related
compound (16). In control experiments it was found that
vitamin Brr, when added to the incubation mixture, was com-
pletely eliminated by boiling the solution at pH 12, whereas
deox-yribonucleosides were not affected.
In order to test the accuracy of the microbiological assay, it
was compared with a previously used isotope method (6) (Fig. 2).
A mixture containing Fraction B, ATP, Mg*, and reduced
lipoic acid was incubated with t&urn-labeled CDP. After 15
and 30 minutes, respectively, the reaction was stopped by heat-
ing the vessels to 106”. One portion was analyzed for the forma-
tion of isotopic deoxycytidine phosphates, and another part was
used for a microbiological assay with L. leiehmunnii. Fig. 2
shows that the formation of deoxyribonucleotides, as determined
by the microbiological method (Curse a), was somewhat larger
than the corresponding values derived from isotope data (Curve
b). Curve c in Fig. 2 was obtained by omitting CDP from the
incubation mixture. In this case, a much smaller but definite
increase with time of the deoxyribonucleoside content of the
system was observed with the microbiological assay. Since
adenosine nucleotides were routinely included in all vessels in
this experiment, the results of experiments in which CDP was
omitted were taken as evidence for the reduction of adenosine
phosphates to deoxyadenosine phosphates. It is now possible to
understand why the two different assays apparently gave dif-
ferent results in the CDP experiment (Curses a and b). The
isotope method measured only the formation of deoxycytidine
phosphates, whereas the microbiological assay gave the sum of
the reductions of cytidine and adenosine phosphates. In sup-
port for this interpretation was the finding that the sum of Curves
b and c equals Curve a.
Formation of Deoxyribonucleotides from
Aoknosiw Phosphutes
Requirement for AoTerwsiw Phosphates--Fig. 3 shows an ex-
periment in which increasing amounts of ATP with and without
a constant amount of ADP were incubated with standard con-
centrations of the other components. Optimal formation of
deoxyribosyl compounds was dependent on ATP in both the
presence and absence of ADP. The optimal concentration of
ATP was found to be about 2.5 mM.
In these experiments the relative amounts of AMP, ADP, and
ATP in the incubation mixture were determined by paper chro-
matography with isobutyric acid-ammonia (17). It was found
that the ATP used was contaminated with up to 10% of ADP.
Furthermore, Fraction B contained enzymes, such as pyrophos-
phatases and myokinase, which changed the initial distribution
of AMP, ADP, and ATP. It was therefore very difficult to
control the concentration of ADP in the system throughout the
incubation.
As demonstrated by Fig. 3, the addition of ADP considerably
increased the formation of deoxyribosyl compounds at all concen-
trations of ATP. The highest stimulation of the reaction by
ADP amounted to 90%. This increase was much larger than
would be expected if the only effect of ADP on the reaction were
to increase the concentration of ATP, for instance via the myo-
kinase reaction. The mechanism by which ADP influenced the
reaction was apparently more direct. The
amount
of stimulation
observed varied, however, to a large extent with different prepa-
rations of Fraction B. With some preparations hardly any
stimulation by ADP was observed.
The reduction of pyrimidine ribonucleotides occurs at the
diphosphate level (4,5). Evidence is given below that GDP is
the primary substrate of the reducing enzymes. The results
obtained for the reduction of adenosine compounds suggest to
some extent that in this case the reduction also occurs with the
ribonucleoside diphosphate. It must be pointed out, however,
that this idea depends to a large extent on the analogy with the
three other nucleotides, and that the experimental results by no
means are conclusive on this point.
Require for Mfl and Reduced Lip& Acid-The reduc-
tion of adenosine phosphates showed an absolute requirement
for Mg++ (Fig. 4). On the addition of increasing amounts of
MgC12, an initial lag period was observed, after which pro-
nounced stimulation occurred. The optimal concentration of
Mg++ was found to be 15 mM; higher concentrations were in-

October 1963 A. Larssm
mM (reduced dl-lipcctc)
3417
4
1.5 - I I
__-- ----- r--- -----
1.5 -
p.-.--7
16
I I I
2 4 10 20 30
mM (ATPI mM (Mg++)
FIG.
3 (left). ATP requirement for the reduction of adenosine phosphates. The curves were obtained with (O-O) and without
(O---O) addition of 50 mpmoles of ADP.
FIG.
4 (right). Requirements for Mg ++ (0-O) and dl-lipoate (O- - -0) for the reduction of adenosine phosphates.
hibitory. The initial lag of the curve can be explained to some
extent by the presence of EDTA in the reaction mixture.
The reaction was found to be strictly dependent on the addi-
tion of reduced lipoic acid (Fig. 4). The optimal concentration
of dl-lipoate was around 3.0 mM.
pH Optimum-The
reduction of adenosine compounds showed
a rather sharp pH optimum between 7.4 and 7.7 (Fig. 5). The
two curves were obtained by incubating identical mixtures with
two different concentrations of a series of Tris buffers. With
both concentrations, clear optima were obtained in the indicated
interval. On the other hand, if the absolute values are com-
pared, large differences are found. The inhibition at the higher
concentration of Tris is believed to be an unspecific inhibition
by neutral salts.
The experiments with adenosine phosphates demonstrate that
Fraction B catalyzed the formation of adenine deoxyribonucleo-
tides in a system containing ATP, Mg*, and reduced lipoic
acid. The requirements for Mg* and reduced lipoic acid were
absolute. Although it was found that ADP played an important
role in the reaction, it was not possible to determine definitely
whether this compound was the most direct substrate for the
recluctase system. There is a close similarity between the re-
duction of adenosine phosphates and that of pyrimidine ribo-
nucleotides. The two types of reactions also have in common
the inhibition by neutral salts as exemplified by the two pH
curves (Fig. 5) and by the inhibition by excess MgClz (Fig. 4).
Similar pH optima were also observed.
Formation
of
Deoxyribonucleotides
from
Guanosine Phosphates
Requiremen&
for
Cofactors-By substitution of GDP for ADP
in the
standard incubation mixture, the formation of deoxyribo-
nucleotides was greatly increased (Table III). This was taken
as evidence for the reduction of guanosine phosphates. Table
III shows that ATP, Mg*,
and reduced lipoic acid were required
for optimal formation of deoxyribonucleotides from GDP. With
this preparation of Fraction B approximately 5 times more
deoxyribosyl compounds were formed from GDP as compared
PH
FIG.
5. pH curve for the reduction of adenosine phosphates.
The pH values given are those of the buffers added. Two different
final concentrations of Tris buffer were used: O-O. 0.18 M:
o- -
-0.0.33
M.
to ADP. On the other hand, the activity with CDP was ap-
proximately 3 times higher than with GDP. A more thorough
comparison of the reductions of different ribonucleotides is given
below.
Ccnnparkm
of
Da&vent Guano&e Phosphatds
a8 Substrates-
Fig. 6 shows the results of two experiments in which different
guanosine phosphates were incubated with Fraction B under
conditions favorable for the formation of deoxyribonucleotides.
In Experiment
A
of Fig. 6, the standard mixture was incubated
with increasing amounts of either GMP, GDP or GTP. Experi-
ment
B
of Fig. 6 is identical with Experiment
A with
one ex-
ception: TPNH and Factor-& (7) were substituted for reduced
lipoic acid. Other work (7) has shown that these compounds,
and not reduced lipoic acid, are the physiological reducing agents
in the CDP reductase system. Experiments
A
and
B
were done
with the same preparation of Fraction B, but on different oc-
casions, and are therefore not absolutely comparable.
In both experiments of Fig. 6, GDP was a considerably better
substrate for deoxyribonucleotide formation than either the

3418 Enzymatic Synthesis of Deoxydwnudeotides. III Vol. 233, No. 10
TABLE
III
Requirements for reduction of guanosine phosphates
The standard incubation mixture containing GDP was assayed
with L. leichmannii.
Alteration in procedure
None ................................
Incubation for 0 minute at 37”. ......
Boiled Fraction B. ..................
ATP omitted. .......................
Mg++ omitted ........................
Reduced lipoic acid omitted. ........
GDP omitted, ADP added ............
GDP omitted, GDP added ............
.......
.......
.......
.
,.....
. . . . . .
. . . . . .
2.6
0.2
0.2
0.5
0.1
0.2
0.6
5.9
GTP
GMP
I I
.2 .4
I I
B
GDP
GTP.
GMP
mM (gwnosinc phorplvate)
FIQ.
6. Comparison of different guanosine phosphates as sub-
strates. In Experiment A, standard conditions were employed
except for guanosine phosphates. In Experiment B, conditions
were like Experiment A, with 0.2 Hmole of TPNH
and about 0.01
mg of Factor-S* replacing reduced lipoate.
mono- or ixiphosphate. The results here are not as clear cut as
those found previously with pyrirnidine ribonucleotides, but
nevertheless provide strong evidence in favor of a reduction of
guanosine compounds at the diphosphate level. The demonstra-
tion that TPNH and Factor-S could substitute for reduced
lipoic acid further stresses the similarity between the require-
ments for the reduction of GDP and that of CDP.
One weakness of the
assay method used is that the product of
the reaction is defined as a deoxyribosyl compound but no in-
formation is obtained with respect to the base bound to deox-
yribose. The finding of a stimulation of deoxyribonucleotide
formation by the addition of GDP to the standard incubation
system, which contained adenosine phosphates, therefore could
be interpreted in two different ways: either (a) GDP itself serves
as a substrate for the reductive system or (b) GDP in some un-
known way stimulates the reduction of adenosine phosphates.
The second alternative is made very unlikely by some results in
this paper. In favor of the first alternative we may note the
rather specific stimulative effect of GDP as compared to GMP
and GTP, and the finding, demonstrated below in Fig. 7, that
certain preparations of Fraction B were almost completely devoid
of the ability to reduce adenosine phosphates but nevertheless
showed good activity toward GDP. It therefore seems safe to
conclude that the results obtained with GDP indeed represent a
reduction of this ribonucleotide to a deoxyribonucleotide.
Com@on of Reductions of Di@rent RibonucleoEidss wii%
Enzymes from E. coli and Novikoff Hepatoma
The microbiological method could be used to measure the
reductions of all four ribonucleotides with different preparations
of Fraction B (Fig. 7). In all csses the most rapid formation of
deoxyribonucleotides was observed with CDP, and ADP was
always found to be least active as a substrate. GDP and UDP
were usually reduced at about equal rates. Earlier results (Table
I) demonstrate a good recovery of deoxyribonucleotides in-
cubated with Fraction B. Therefore, the differences observed
CDP
GOP-
UIJP
l AADP
‘Oi
15 30
FIQ. 7. Comparison of the reductions of ribonucleotides by
three different preparations of Fraction B. In these experiments
the volume of the incubation mixture was 0.4 ml.
TABLE
IV
Comparison of reduction of difletent ribonueleoside diphosphates
with Fraction Pd from Notikoff hepatoma
Fraction Pd (1.05 mg of protein) was incubated with 0.25
pmole of ADP or GDP or CDP or UDP, 1.2 qoles of ATP, 2.5
rmoles of MgClr, 0.05 pmole of FeCh, 0.75 rmole of reduced dl-
lipoic acid, 20 pmoles of Tris buffer, pH 7.0, and 0.025 pmole of
EDTA. The fmal volume was 0.6 ml. After 30 minutes at 37”
the reaction was stopped by boiling the solution. The formation
of deoxyribosyl compounds ww assayed with L. acidophilus.
Deoxyribosyl compounds formed
Added ribonuclwtide After correctfop
mlrnrolds
ADP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7 1.7
GDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.8 3.1
CDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 2.7
UDP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 2.8

October 1963 A. Larsson 3419
in the experiments of Fig. 7 must represent differences in the
rates of deoxyribonucleotide formation and cannot be due to
differential breakdown of deoxyribosyl compounds.
From Novikoff hepatoma, Moore and Hurlbert (1) and Moore
and Reichard (7) purified an enzyme system, called Fraction
Pd, which catalyzed the reduction of CDP. With the proper
assay method now available, it was of great interest to determine
whether Fraction Pd could catalyze the reduction of the other
ribonucleoside diphosphates as well (Table IV). In these ex-
periments, each of the four ribonucleoside diphosphates was
incubated under conditions found to be optimal for cytidine
nucleotides. Table IV demonstrates that about the same
amounts of deoxyribosyl compound were formed from either
CDP, UDP, or GDP. ADP was less active. However, part of
the deoxyribosyl compounds formed from CDP, UDP, and GDP
were in all probability deoxyadenosine phosphates, since ATP
was added to the incubation mixture in these experiments.
An approximate correction for this effect is made in the second
column of Table IV, where the amount of deoxyribosyl com-
pounds formed from the ADP experiment is subtracted from
the tots1 values. In addition, all values were corrected for a
background of 0.5 mpmole of deoxyribosyl compounds originating
from the enzyme preparation.
Fraction B was purified 100 to 200 times with respect to the
reduction of CDP (4, 5), and Fraction Pd was also purified with
respect to this reaction (7). The observation that both enzyme
systems could also reduce uracil, adenine, and guanine ribonu-
cleotides raised the question whether one single enzyme system
could catalyze all four reactions.
The reductions of cytidine and adenosine phosphates catalyzed
by Fraction B showed the same requirements for cofactors, i.e.
ATP, Mg++, and reduced lipoic acid. In the presence of these
compounds, Fraction B could also transform uridine and guano-
sine phosphates to the corresponding deoxyribonucleotides. All
these reactions, possibly with the exception of the reduction of
adenosine compounds, took place at the diphosphate level.
These observations indicate that the same general type of reduc-
tive process was involved in all cases.
If the rates of the four reduction reactions catalyzed by d8er-
ent preparations of Fraction B were compared, very large varia-
tions were obtained between diflerent preparations (Fig. 7).
Thus Experiments
A
and C of Fig. 7 illustrate results from two
preparations which showed only marginal reduction of adenosine
phosphates but an efficient reduction of CDP. The values for
GDP and UDP were always somewhat lower than those for CDP.
On the other hand, the enzyme system from Novikoff hepatoma
(Table IV) showed an equal reduction of CDP, UDP, and GDP.
All these experiments can be explained on the aszumption that
at least one component of the enzyme system reducing ribonu-
cleotides is specific for each of the four ribonucleoside diphos-
phates. The evidence for ADP is strongest in this direction, and
it is tentatively concluded that the ribonucleotide reductase sys-
tem of
E.
coli contains at least one enzyme specific for ADP.
Further enzyme purification is required, however, before the
question of enzyme specificity with respect to the other three ri-
bonucleotides can be decided.
SUMMARY
Enzyme systems from Escherichia coli B (Fraction B) and
Novikoff hepatoma, purified earlier for the reduction of cytidine
diphosphate, were shown to catalyze the formation of deoxyri-
bosyl compounds from adenosine and guanosine phosphates, re-
spectively.
With bacterial Fraction B these reactions required the addition
of adenosine triphosphate, Mg++, and reduced lipoic acid, and
thus showed a close similarity to the earlier studied reductions of
pyrimidine ribonucleoside diphosphates. The formation of de-
oxyguanosine phosphates probably occurred with guanosine
diphosphate as the primary substrate of the reductive enzymes,
and also in the case of adenosine phosphates the available evi-
dence tentatively points to a reduction at the diphosphate level.
REFERENCES
1. MOORE,
E. C.,
AND HURLBERT,
R. B.,
Biochim. et Biophys.
Acta, 66,
651 (1962).
2. ABRAMS,
R.,
LIBENSON,
L.,
AND EDMONDS,
M.,
Biochem. and
Biophys. Research
Communa., 3,
272 (1960).
3.
REICEARD,
P., J.
Biol. Chem., 336.
2511 (1961).
4. REICHARD,
P., J.
Biol. Chem., 337, 3613 (1962).
6. BERTANI,
L. E.,
HHQGMARK,
A.,
AND REICHARD,
P.,
233, 3407
(1963).
6.
REICHARD,
P.,
BALDESTEN,
A.,
AND RUTBERG,
L.,
J. Biol.
Chem., 236,
1150 (1961).
7. MOORE,
E. C.,
AND REICHARD,
P.,
J. Biol. Chem., 336, PC2244
119631.
8. KITAY,
E., McNmrr, W. S.,
AND SNELL,
E. E.,
J. Biol. Chem.,
177, 993 (1949).
9. HOFF-J$R&N&N,
E.,
Biochem. J., 60.400 (1962).
10.
SEIDLER,
A. J.,
NAYDER,
F. A.,
AND S~HWEIGEBT,
B. S.,
J. Bacterial., 73, 670 (1967).
11.
KITAY,
E., MCNUTT, W. S.,
AND SNELL,
E. E.,
J. Bactet-iol.,
69, 727 (1960).
12.
SCHNEIDER,
W. C.,
AND POTTER,
R. L.,
Proc. Sot. Exptl. Biol.
Med., 94, 798 (1967).
13.
SCHNEIDER,
W. C., J.
Biol. Chem., 237, 1406 (1962).
14.
SEIDLER,
A. J.,
AND SCHWEIQERT,
B. S.,
J. Bacterial., 77, 614
(1969).
16.
SEIDLER,
A. J.,
AND SCHWEIQERT,
B. S.,
J. Biol. Chem., 236,
768 (1960).
16.
SIZHEID,
H. E.,
AND SWWEIGERT.
B. S.. J. Biol. Chem.,
136,
1
(i96oj. .
17.
KBEBS,
H. A.,
AND HEMS,
R., Biochim. et Biophya. Acta, 12,
172 (1963).