Vol. 259, No. 5, Issue of March 10, pp. 2879-2885, 1984
Printed in U.S.A.
THEJOURNAL
OF BIOLOGICAL
CHEMISTRY
0 1984 by The American Society of Biological Chemists. Inc
Two Low K , Hydrolytic Activities on Dinucleoside
5’,5”’-P1,P4-tetraphosphates
in Rat Liver
CHARACTERIZATION AS THE SPECIFIC DINUCLEOSIDE TETRAPHOSPHATASE AND A
PHOSPHODIESTERASEI-LIKE ENZYME*
(Received for publication, March 21, 1983)
Jose C. Cameselle, Maria J. Costas, Maria A. Gunther Sillero, andAntonio Sillero
From the Znstituto de Enzimologik y Patologia Molecular del Consejo Superior de Znvestigaciones Cientificas, Departamento de
Bioquimica, Facultad de Medicina, Universidad de Ertremadura, Badajoz, Spain
Ninety per cent of total rat liver hydrolytic activity the corresponding nucleoside mono- and triphosphates, has a
(1.4 unitslg of fresh tissue) on diadenosine or diguano- molecular weight between 20,000 and 22,000 (rat), and is
sine 5’,5”’-P’,P4-tetraphosphate(Ap4A and Gp4G) strongly inhibited (nanomolarKi)by nucleoside 5”tetraphospresent in isotonic homogenates sedimented at 37,000 phates. The enzyme is equally active on Ap4A’ and GpdG, two
x g. Supernatant activity corresponded to the earlier nucleotides present in biological sources (Zamecnik, 1969;
described, cytosolic and specific, bis(5’-guanosyl) te- Finamore and Warner, 1963; Sillero and Ochoa, 1971). This
traphosphatase or dinucleoside tetraphosphatase (EC activity was originally named diguanosine tetraphosphatase
3.6.1.17; Lobaton, C. D., Vallejo, C. G., Sillero, A., and by the Enzyme Commission on the basis of its earlier reported
Sillero, M. A. G. (1975) Eur. J . Biochem. 50, 495- activity towards Gp4G (Warner and Finamore, 1965). Later
501). Particulate activity, as extracted with Triton
X- studies on its substrate specificity made in our view more
100, is composed of two enzymes separable bygel
appropriate the name of dinucleoside tetraphosphatase (Valfiltration. One of them was a low K , (1p~ Gp4G,5 p~
Ap4A) 22,000-dalton enzyme, strongly inhibited by lejo et al., 1976). This denomination will be adopted through
guanosine 5‘-tetraphosphate ( I C i = 9 nM), and likely this work.
The increasing interest in the potential metabolic roles of
identical to the cytosolic specific enzyme. The other
both GplG and Ap4A (Renart et al., 1976; Sillero et al., 1977;
Triton-extracted form was unspecific, with an estimated molecular weight of 150,000 (sucrose gradient) Grummt et al., 1979; Rapaport et al., 1981; Yamakawa et al.,
or 450,000 (gel filtration), both in the presence of 1982) made desirable the further investigation into the medetergent. Substrate specificity was broad, requiring tabolism of these nucleotides. Previous results from our laba nucleoside 5‘-phosphoryl residue with a free 3’-hy- oratory had shown that thehydrolytic activity on GplG presdroxyl group, and actingon 5’-5’and 5’-3‘ compounds. ent in the total homogenates from several rat tissues was
K , values were 12 MM (Gp4G)and 8 p~ (Ap4A).Guan- higher than that recovered in the 27,000 X g supernatants
osine 5’-tetraphosphate was a competitive inhibitor (Cameselle et al., 1982). With rat liver, two thirds of the
(Ki = 2 pM). It required bivalent cations since a residual activity appearing in homogenates obtained in 50 mM Tris/
activity after dialysis was abolished by EDTA and HC1 buffer, pH 7.5,0.5 mM EDTA, sedimented at 27,000 X g,
enhanced by Mg2+, Mn2+,or Ca2+.In the absence of and one third remained in the supernatant. Thelast activity
other added cations, the enzyme, inhibited by 1 mM was characterized as the specific dinucleoside tetraphosphaEDTA, is fully reactivated by an equimolar amount of tase (see above and Lobath et al., 1975b; Cameselle et al.,
Zn”. The possible identity of this activity with phos- 1982).The purpose of this article is to describe the two distinct
W. E. (1963) activities on Gp4G detected in the particulate fraction. One
phodiesterase I (EC 3.1.4.1;Razzell,
Methods Enzymol. 6, 236-258) is discussed, and its of them could not be distinguished from the specific tetrapotential role in the metabolism of dinucleoside tetraphosphatase present inthe cytosol, whereas the otherone was
phosphates is indicated.
of a higher molecular size and quantitatively predominant.
The broad substrate specificity of the latterform makes likely
that it corresponds to the previously described phosphodiesterase I (EC 3.1.4.1; Khorana, 1961). In order to facilitate the
The presence of bis(5’-guanosyl) tetraphosphatase or di- presentation of the results, this activity will be here named
guanosine tetraphosphatase (EC3.6.1.17) has been described high molecular weight or unspecific dinucleoside tetraphosin the cytosol of Artemia salina (Warner andFinamore, 1965; phatase in contrast to thespecific one.
Vallejo et al., 1974), ascites tumor cells (Moreno et al., 1982),
and several rat tissues (Lobat6n et al., 1975b; Cameselle et al.,
MATERIALS AND METHODS
1982). This enzyme cleaves dinucleoside tetraphosphates to
* This research was supported by Grants 4075 and 993 from the
Comisibn Asesora de Investigacibn Cientifica y Tkcnica and a grant
from the Fondo de Investigaciones Sanitarias de la Seguridad Social.
Preliminary experiments of this work were carried out a t Colegio
Universitario de Alava, Departamento de Bioquimica. The costs of
publication of this article were defrayed in part by the payment of
page charges. This article must therefore be hereby marked “advertisement’’ in accordance with 18 U.S.C. Section 1734 solelyto indicate
this fact.
Proteins, Substrates, and Nucleotides
Auxiliary enzymes, molecular weight standards, NADP, NAD,
NADH, AMP, glucose &phosphate, and fructose 6-phosphate were
The abbreviations used are: Ap,A, diadenosine tetraphosphate or
diadenosine 5’,5”’-P’,P4“tetraphosphate;
ApA, adenosine 5”phospho3”adenosine; ApzA, diadenosine 5’,5”’-P‘,P’-pyrophosphate;Ap3A,
Gp4G,diguanosine tetraphosdiadenosine 5’,5”’-P’,P3-triphosphate;
phate or diguanosine 5’,5”’-P’,P4-tetraphosphate.
2879
This is an Open Access article under the CC BY license.
2880
Dinucleoside Tetraphosphatase Activities in Rat Liver
from Boehringer Mannheim. The rest of the substrates ornucleotides
were obtained from Sigma, except for GpaG which was purified from
A. salina cysts as described by Vallejo et al. (1974).
(lactate dehydrogenase and glucose phosphate isomerase) and
the hydrolytic activity on GpIG were measured in the total
homogenate, in thesuccessive washes, and in the final
precipitate (Table I). As expected, most of the activities of the
Detergents and Buffers
wash, only 3% being
The following detergents were used in this work Triton X-100 cytosolic enzymes were found in the first
(Sigma), sodium dodecyl sulfate (Merck), Zwittergent 3-14 (Calbi- retained in the last precipitate. The distribution of the total
ochem-Behring), and CHAPS (Serva). The
two last products are Gp4G hydrolyticactivity was different sinceonly 10% apzwitterionic surfactants which correspond to N-tetradecyl-N,N-dipeared in washes 1-4 (Table I) and thebulk of it remained in
methyl-3-ammonio-1-propanesulfonate
(Gonenne and Ernst, 1978) the precipitate.As shown later, the totalactivity is due to the
and 3-[(3-cholamidopropyI)dimethylammonio]-l-prop~~~su~f~~~~~
presence of at least two different enzymes in thehomogenate:
(Hielmeland, 19801, respectively. The adjustmentof the pHvalues of
Tris buffers was performed at the same temperature at which they the specific dinucleoside tetraphosphatase and another one,
which unspecifically cleaves these compounds. The distribuwere to be used.
tion of the specific dinucleoside tetraphosphatase among the
Enzyme Assays
fractions of Table I was calculated after its separation by
Sephadex G-100 chromatography (results not shown). Two
Unless otherwise stated, all the measurements were done in a
volume of 1 ml and at 37 "C, with 50 mM Tris/HCl buffer, pH 8.0, 5 peaks of activity were apparent in this experiment, one of
mMMgC1,.
them being partly included and corresponding to thespecific
Direct Discontinuous Assay-To measure the phosphohydrolytic
enzyme. The other peak, appearing in the void volume, was
activities on AMP and glucose &phosphate, the liberation of inorless prominent and amounted 20tomilliunits/g of fresh tissue
ganic phosphate was determined. After incubation, the reaction (0.10.2-ml mixtures) was stopped with 3 ml of a solution prepared by (16%of the activity present in wash 1).Since the magnitude
mixing 0.3 M sodium dodecyl sulfate with 2 volumes of 10 mM of the excluded peak varied with the conditions chosen for
ammonium heptamolybdate in 2.5 N H2S04.After strong shaking, 0.1 centrifugation (filling level of the tubes, rotor speed, or buffer
ml of the Fiske and SubbaRow's reducing reagent (preparedaccording density; results not shown), it can be tentatively assigned to
to Leloir and Cardini, 1957) was added. The tubes were left for 1 h
procedure,
at room temperature before reading the absorbance at 660 nm. the microsomal vesicles which, in our experimental
may
still
contaminate
the
supernatants.
Hence,
we
have not
Sodium dodecyl sulfate removes Triton X-I00 interference and makes
this minor peak which, on the other
deproteinizationunnecessary (Dulley, 1975; Tashima, 1975). The been concerned with
reliability of the procedure was tested and found satisfactory up to at hand, canbe regarded as a modest loss of particulate activity.
least 0.6 mg of protein and 1 2 mg of Triton/reaction mixture.
From the results shown above, it seems that the specific
Alkaline Phosphatase-coupled Method-The procedure was the
and
theunspecific tetraphosphatases behave as cytosolic and
same as for the direct assay (see above) except that alkaline phosphatase was included in the reaction mixtures. Afurther modification particulate enzymes, respectively. This point will be further
was needed for the assay in crude extracts andpellets. In these cases, discussed below.
the reaction was finished with 1 ml of 0.6 M trichloroacetic acid.
Dinucleoside TetraphosphateHydrolyticActivities
from
Inorganic phosphate was determined in the supernatants as above Liver Particulate Fraction: Solubilization with TritonX-100described.
Theprecipitateobtainedafter
successive washes withan
Hyperchromicity Assay-It was carried out aspreviously described
isotonic medium was resuspended in 50 mM Tris/HCl buffer,
in the presence of Gp&, Ap4A, or Ap,A (Vallejo et al., 1974; Lohat6n
et al., 1975b; Lobat6n et al., 1975a). Amolar absorption coefficient of p H 7.5,lO mM MgC12,0.5 mM EDTA (buffer B) supplemented
with 4% Triton X-100. After an overnight incubation, the
4600 M" cm" was found for the hydrolysis of ApzA at pH8.0.
Other Enzyme Assays-The hydrolysis of NAD was followed by suspension was centrifuged at 37,000 x g for 2 h and the
recording the decrease in absorbance at 265 nm in the presence of supernatant was taken (Table I, 4% Triton wash). Most of
alkaline phosphatase and adenosine deaminase. When testing the
the activity present in the liver particulate fraction was obeffect of guanosine 5"tetraphosphate on the hydrolysis of NAD, this
assay was performed in two steps. The first one was an incubation tained in thatwash (Table I and Fig. 1).
Gel Filtration of the Triton-solubilizedTetraphosphatase
without coupling enzymes, which was finished by heating the assay
mixture for 2 min in a boiling bath. After centrifugation, alkaline
Activity from Liver Particulate Fraction-A 4% Triton wash
phosphatase and adenosine deaminasewere added to the supernatant, of the precipitate obtainedas described above was applied to
and adenosine was evaluated by theend point method. Lactate
a Sepharose 4B column and eluted with buffer B (Fig. 2a).
dehydrogenase and glucose phosphate isomerase were quantified by
recording the changes in absorbance a t 340 nm. Catalase was assayed Fractions were collected and the activity on Gp4G was deterassay.
with 50 mM sodium phosphate buffer, pH 7.0, and 10 mM H20,. The mined in them with the alkaline phosphatase-coupled
decrease in absorbance at 240 nm was followed at room temperature.
A rather broad profile of activity appeared between elution
H20, was prepared shortly before use as a 0.3 M solution in 0.1 M volumes 200 and 430 ml and aclear peak was apparent
sodium phosphate buffer, p H 7.0, and 0.05 mM EDTA.
between 430 and 480 mI. The inclusion of 0.5% (w/v) Triton
Throughout this work, 1 unit of enzyme activity is defined as the
X-100 in the elutionbuffer changed the elutionprofile of the
activity transforming 1kmol of substrate/min under the experimental
first peak (Fig. Za), making it sharper, whereas the second
conditions.
peak was not affected by detergent.Similarresults
were
RESULTS
obtained when other gel types (Sephadex G-100, Sephadex
Washing Outof the Activity on
Diguanosine Tetraphosphate G-200, and SephacrylS-300)were utilized. With 0.5% Triton
from a Rat LiverParticulate Fraction-As shown in the Intro- in the elution buffer, two clear peaks were obtained; in the
depended on
duction, about two thirds of the hydrolytic activity on dinu- absence of Triton, thewidth of the first one also
cleoside tetraphosphates present in a total rat liver homoge- the discriminating characteristicsof the gel, i.e. it was sharp
G-100 and
nate (Cameselle et al., 1982), obtained in 50 mM Tris/HCl andbroadafterchromatographyinSephadex
buffer, pH 7.5, 0.5 mM EDTA, precipitated a t 27,000 X g. In Sepharose 4B columns, respectively. In every case,both peaks
the following experiments and in order topreserve the integ- were fairly separated. An example of the elution profile on
rity of subcellular structures, an isotonic extraction medium the Triton-solubilized precipitate in a Sephadex G-200 colwas used. A total liver homogenate was obtained and centri- umn and in thepresence of 0.5%Triton in the elutionbuffer
fuged a t 37,000 X g, and the resulting precipitatewas washed is drawn in Fig. 2b. Altogether, these results show that the
three times. The activitiesof two typically cytosolic enzymes solubilized liver particulate fraction contains two different
Dinucleoside Tetraphosphatase Activitiesin Rut Liver
2881
TABLEI
Washing out of the activity on GplG from a rat liverparticulate fraction
The livers from two female rats (14 g of fresh weight) were homogenized in a motor-driven Potter apparatus
with a glass pestle in the presence of 29 ml of 35 mM Tris/HCl buffer, pH 7.7, 70 RIM KCI, 9 mM MgC12, 0.25 M
sucrose (buffer A). A 3.2-mI sample of the homogenate was taken, and theremaining 40 ml weredistributed equally
among eight 30-ml tubes and centrifuged at 18,000 rpm (37,000 X g average) and 2 "C for 1 h. The supernatants
were decanted, pooled, and kept at 4 "C. Each pellet was resuspended in 3.5 ml of buffer A with the help of a glass
rod and centrifuged as described above. This step was repeated two more times, and four supernatants were
obtained (washes 1-4). After the lastcentrifugation, the pellets were resuspended either in buffer A (precipitate in
the table) or in buffer B containing 4% (w/v) Triton X-100. Aliquots of every step were dialyzed overnight against
200 volumes of buffer A, except for the Triton-resuspended fraction whichwasdialyzed
against buffer B
supplemented with 4% Triton X-100. The latterpreparation, after dialysis, was centrifuged as described above for
2 h, and the supernatant was taken (4% Triton wash in the table). Total hydrolytic activity on Gp,G (0.6 mM)
with the alkaline phosphatase-coupled assay and the activities of lactate dehydrogenase and glucose phosphate
isomerase were measured in those samples. The specific dinucleoside tetraphosphatase was evaluated (Gp,G; same
assay as described above) after chromatography of the extracts in a Sephadex G-100 column, except for the value
assigned to the homogenate, which is the sum of the activities found in washes 1-3.
"
"
Lactate
dehydrogenase
units/g
3
%
143
100 100
Homogenate
426
83
352
Wash 1
5.2
22
Wash 2
1.7
Wash
7.4
1.1
4.6
Wash 4
2.8
Precipitate
12
1.9
4% Triton
8.2 wash
ND, not determined.
Glucose phosphate
isomerase
Total Gp,G hydrolysis
Specific dinucleoside
tetraphosphatase
units/g
%
milliunits/ g
rnilliunits/g
1414
193
160
11
2.9
1.9
5.2
4.1
100
83
5.7
1.5
1.0
2.7
2.1
125
16
4.4 6
5
1275
1600
%
9
1.1
0.4
0.4
90
20113
122
17
ND"
ND
30
%
100
85
12
3
ND
ND
respectively. Theproducts of thereaction, withGp4G as
substrate and characterized by spectrophotometric coupled
methods or by thin layer chromatography, were GTP and
GMP. The enzyme required Mg2+, was maximally active at
pH 7.5, and was inhibited by guanosine 5'-tetraphosphate (Ki
= 9 nM) and by Ca2+; the apparent molecular weight was
22,000 as determined by gel filtration in Sephadex G-100. In
our view, theseresultsare
sufficient tocharacterizethis
[Triton X-1001, rng/rnl
enzyme
as
the
specific
dinucleoside
tetraphosphatase previFIG. 1. Solubilization of particulate hydrolytic activity on
Gp4G by Triton X-100. A liver precipitate was obtained as de- ously described in the rat liver supernatant (Lobatbn et al.,
scribed in the legend to Table I. That pellet was resuspended in buffer 1975b; Cameselle et al., 1982).
B. Six 3-ml aliquots of that suspension received 0.75 ml of buffer B
The presence of this enzyme in the liver supernatant and
supplemented with various amounts of Triton X-100 sufficient to
in
the successive washes of the precipitate was investigated
bring the final detergent concentrations to 0, 0.1, 0.5, 1, 2, and 4%
a Sephadex G-100
(w/v), respectively. The six samples were dialyzed overnight against by chromatography of thesamplesin
100 volumesof buffer B alone or supplemented with Triton X-100 up column to allow for the separationof the low molecular weight
to the same concentration to which each sample had been brought. tetraphosphatase. As shown in Table 1, the percentageof this
After centrifugation in 30-ml SS-34 Sorvall rotor tubes a t 18,000rpm
successive washes was very similar
and 2 "C for 2 h, the hydrolytic activity on GpIG ( 0 )was assayed in activity extracted with the
the supernatantswith the alkaline phosphatase-coupled method and tothose of lactate dehydrogenase and glucose phosphate
0.6 mM Gp,G as substrate. The results are expressed as percentage isomerase, pointing to a cytosolic localization of the specific
of the activity present in the untreated precipitate. Protein content tetraphosphatase. However, an appreciable amount (20%) of
(A) was determined by a procedure whichremoves Triton X-100
is
interference (Wang and Smith,1975), with bovine serum albumin as the specific enzyme remains in the precipitate and liberated
after detergent treatment. This result could favor the possistandard.
bility that a certain amount of this enzyme is also located in
dinucleoside tetraphosphatase activities with dissimilar ap- the rat liver particulate fraction.
parent molecular weights.
Characterization of the High Molecular Weight Form of the
Characterization of the Low Molecular Weight Form of the Triton-solubilized Dinucleoside Tetraphosphatase Activity
Triton-solubilized Dinucleoside Tetraphosphatase Activity fromLiver Particulate Fraction-The apparent molecular
from Liver Particulate Fraction-This enzyme form was ob- weight of thisactivity was studied by gel filtration in a
tained as described above; the washed precipitate (Table I) Sephadex G-200 column. The enzyme samplewas a 4% Triton
was treated with 4% Triton, and the supernatant
was applied
to a Sephadex G-100 column and eluted without detergent. wash of the last liver precipitate (Table I). The elution was
Two fully resolved peaks of hydrolytic activityon Gp,G accomplished with buffer B containing 0.5% (w/v) Triton Xappeared (results not shown). The three fractions with the 100 (Fig. 3). The calibration of the column was performed
maximal activity from the peak corresponding to the
low with markers of known molecular weight (ferritin, 450,000;
molecular weightform were pooled, and the enzymatic activitycatalase, 240,000; lactate dehydrogenase, 140,000; cytochrome
was characterized asdescribed by L o b a t h e tal. (1975b). The c, 12,500) which were chromatographed under the same exmore relevant resultswere as follows. The enzyme was equally perimentalconditionsasthe
sample. The high molecular
active on Gp4G and Ap4A, with K,,, values of 1 and 5 /IM, weight form of dinucleoside tetraphosphatase activity eluted
Dinucleoside Tetraphosphatase Activitiesin Rat Liver
2882
20
-
10 -
E
\
3
E 0r
-
1
z
I
30 -
20
500
100
300
03
-
10 -
L
0
50
100
150
Elution volume, mi
FIG. 2. Gel filtration of the Triton-solubilized tetraphosphatase activity from liver particulate fraction. In these experiments, a 4% Triton wash of a liver particulate fraction was used
(Table I and Fig. 1). a, 15 ml of enzyme preparation were chromatographed in a Sepharose 4B column (2.6 X 100 cm) equilibrated with
either buffer B alone (0)or buffer B plus 0.5% (w/v) Triton X-100
( 0 )The
.
elution was accomplished in both cases with the equilibrating buffer at a flow rate of20 ml/b; fractions were collected and
analyzed for Gp4G hydrolysis by the alkaline phosphatase-coupled
method with 0.6 mM Gp4G.The recovery of activity in the two above
experiments was 62 and 71%. respectively. b, 4 ml of a similar
preparation to the above one were applied to a Sephadex G-200
column (1.6 X 90 cm) equilibrated in buffer B plus 0.5% (w/v) Triton
X-100 and eluted with the same buffer a t a flow rate of 3.2 ml/h.
Fractions were collected and analyzed as above. The recovery of
activity was 77%. Protein ( X ) was determined as described in the
as
legend to Fig. 1. The arrows indicate the column void volume (VO)
determined with dextran blue.
Elutton volume, ml
FIG. 3. Apparent molecular weight of the unspecific dimcleoside tetraphosphatase. The same column and elution buffer of
Fig. 2b were used. The flow rate was 3.2 ml/h and 1.7-ml fractions
were collected. The following samples were run successively: 1) 4 ml
of the 4% Triton wash of a liver precipitate (Table I) and 2) 4 ml of
a solution containing 7.5 mg of ferritin, 4.5 mg of cytochrome c, 0.7
mg of catalase, and 25 units of lactate dehydrogenase, in elution
buffer. The elution profiles are represented in arbitrary units. One
arbitrary unit equals: 6.7 milliunits/ml (Gp4G hydrolysis; alkaline
phosphatase-coupled assay with 0.6 mM Gp4G;O ) , 1 absorbance unit
(ferritin; measured at 400 nm; 0),160 units/ml (catalase; X ) , 0.22
unit/ml (lactate dehydrogenase; A), or 0.4 absorbance units (cytochrome c; measured at 400 nm; A). The arrow marks the elution
volume of the tetraphosphatase activity.
same in the five gradients (Fig. 4 ) . Referring to the control
and to theKCI-, Zwittergent-, CHAPS-, and Triton-supplemented gradients, the calculated molecular masses for the
small form were 30,27,23,32, and 21 kilodaltons, respectively.
The corresponding values for the high molecular weight form
were 224, 213, 152, 225, and 150 kilodaltons. In agreement
with the results presented above using gel filtration, the first
activity corresponds to the specific dinucleoside tetraphosphatase. The molecular mass calculated for the second form
was a third of that obtained by gel filtration in Sephadex G200 (see above), both cases in the presence of Triton X-100.
For purposes of kinetic characterization, the large form of
dinucleoside tetraphosphatase activity, which had been isolated by gel filtration, was subjected to an additional purification step. Fractions corresponding to elution volume 70-83
very near to ferritin, and an apparentmolecular weight of at mlof a Sephadex G-200 column (Fig. 26) were pooled and
applied to a DEAE-cellulosecolumn (1.6 X 13cm) equilibrated
least 450,000 can be assigned to it.
The molecular size of this activity was also studied by with 0.5% (w/v) Triton X-100 in buffer B. The column was
means of sucrose gradient centrifugation. The source of en- then washed with the same buffer until the protein detected
zyme was the same as in the preceding experiment. A 0.4-ml in theeluate was negligible.Further elution was accomplished
portion of that preparation was layered on the top of a with 200 ml of a linear gradient (0-0.35 M) ofKC1 in the
continuous sucrose gradient (10-30%) in buffer B. In separate starting buffer. The hydrolytic activities on several substrates
tubes, the enzyme was applied onto gradients supplemented were determined in the collected fractions (Fig. 5). The activwith one of the following reagents: 1 M KCl, or 0.5% (w/v) ity on Gp4G eluted as a single peak coinciding with another
Triton X-100, or 0.3% (w/v) Zwittergent 3-14, or 0.3%(w/v) one on Ap3A and Ap2A. Two peaks of activity were apparent
CHAPS (see under “Materials and Methods” for detergent when AMP was used as the substrate. This result points to
data). In every case, an inner control of known molecular the existence in those fractions of at least two enzymatic
weight (catalase, 240,000) was added to theenzyme samples. activities, one acting on AMP and theother one with capacity
After centrifugation at 38,000 rpm and 2 “C in a Beckman to hydrolyze molecular structures with innerphosphates.
SW 41 rotor for 14.5 h, the gradients were fractionated and What followsis the characterization of the latteractivity with
catalase and the hydrolytic activity on GplG were measured. a pool of fractions 47-60 in Fig. 5 .
The compounds tested as substrates were: ( a ) dinucleoside
The latter activity sedimented in two peaks, whose apparent
molecular weights were calculated, with catalase as a marker, polyphosphates (Gp4G,Ap,A, Ap3A, Ap2A,NAD, 3-acetylpyraccording to Martin and Ames (1961). It can be remarked, idine adenine dinucleotide, and FAD), ( b ) other molecules
however, that the relative position of catalase was not the with inner phosphates (ApA, ADP-ribose, ADP-Glc, UDP-
Liver
in Rat
Dinucleoside
Tetraphosphatase
Activities
0
0
5
Gradlent volume, ml
10
FIG.4. Sucrose gradient centrifugation of the Triton-solubilized dinucleoside tetraphosphatase activities. Five 10-ml
continuous gradients (sucrose, 10-30%) were accomplished in buffer
B supplemented with one of the following reagents: nothing (a), 1 M
KC1 ( b ) ,0.3% (w/v) CHAPS ( e ) , 0.3% (w/v) Zwittergent 3-14 (d),
0.5% (w/v) Triton X-100 ( e ) (see under “Materials and Methods” for
detergent data). A 0.4-ml sample of a 4% Triton wash of the liver
precipitate (Table I), having 0.25 mg of catalase added, was layered
onto each gradient. After 14.5 h at 38,000 rpm in a Beckman SW 41
rotor, the gradients were fractionated and catalase (0)and theGp,G
hydrolytic activity ( 0 ) were assayed. For thelatter activity, the
alkaline phosphatase-coupled method with 0.6 mM Gp4G was used,
Enzyme activities are expressed inarbitraryunits
equalling 100
milliunits/ml (catalase) or 10 milliunits/ml (GprG hydrolysis).
Glc, 3‘-dephospho-CoA, bis(p-nitrophenyl) phosphate, dT5’-(4-nitrophenylphosphate),dT-3‘-(4-nitrophenyl
phosphate), and CAMP), and ( e ) two molecules with terminal
phosphates (AMP and glucose 6-phosphate). The enzymatic
activities were followed in conditions of linearity with both
time and amount of extract. For each substrate, the concentration at which it was tested, the relative velocity of hydrolysis, and, when pertinent, theK , value are included in Table
11. All the substrates were efficiently hydrolyzed with the
exception of bis(p-nitrophenyl) phosphate, dT-3’-(4-nitrophenyl phosphate), glucose 6-phosphate, and CAMP. It is
relevant that, contrary to the 5’-derivative, d-T3’-(4-nitrophenyl phosphate) was not a substrate of the reaction. Similarly, the molecules which did not have a free 3’-OH end were
poor substrates of the enzyme. This activity most probably
corresponds to phosphodiesterase I (EC 3.1.4.1; Khorana,
1961). The activity towards AMP (Table 11) is likely due to
an enzyme different from that acting on Gp,G (and the rest
of the substrates;Table 11) for two reasons: the DEAEcellulose elution profile (Fig. 5) and differentinactivation
curves. The DEAE-cellulose preparation lost 50% of activity
on GplG (and also on Ap4A,Ap3A, NAD, anddT-5‘-(4-
25
2883
50
75
Fractlon number
100
125
FIG.5. DEAE-cellulose chromatography of thehigh molecular weight, unspecific dinucleoside tetraphosphataseactivity. A 13-ml sample of pooled fractions containing the high molecular
weight tetraphosphatase (Sephadex G-200; Fig. 2b) was applied to a
DEAE-cellulosecolumn (1.6 X 13 cm) equilibrated with buffer B plus
0.5% (w/v) Triton X-100 and washed with the same buffer at a flow
rate of 15 ml/h. In these conditions, 28 fractions of volume 2.9 ml
were collected. Then, a 200-ml linear gradient of KC1 (0-0.35 M) in
the same buffer was used to elute the tetraphosphatase activity, and
97 fractions of 2-ml volume were collected. The activities on Gp,G,
Ap,A, and Ap2A were assayed by the alkaline phosphatase-coupled
method with 0.6 mM substrate. The activity on AMP was measured
by the direct discontinuous assay with 5 mM AMP. Protein was
determined as described in the legend to Fig. 1.
TABLE
11
Substrate specificity of the unspecific dinuckoside tetraphosphatase
activity from the DEAE-cellulose step
The relative rate for each substrate, at theconcentration indicated
in the table, was determined in 0.1-ml samples of the pooled fractions
47-60 (Fig. 51, with the direct discontinuous assay (AMP and glucose
6-phosphate), orfollowing the increase in absorbance a t 405nm
(bis(p-nitrophenyl) phosphate), or with the alkaline phosphatasecoupled method (the rest of the substrates). K,,, values were estimated
at pH 7.5 and 25 “C by the hyperchromicity assay (Gp4G,A p d , A p d ,
and ApzA), or measuring the decrease in absorbance at 265 nmin the
presence of alkaline phosphatase and adenosine deaminase (NAD),
or at. 405 nm (bis@-nitrophenyl) phosphate). The relative rate (5.4)
obtained for AMP (2 mM) was not included in the table because it
did not seem to be substrate of the unspecific tetraphosphatase but
of a contaminatingactivity (see the text). AcPyNAD, 3-acetylpyridine
adenine dinucleotide.
Concentration Relative
rate
1.7
2.0
APA
ADP-ribose
ADP-Glc
UDP-Glc
2.5
3’-Dephospho-CoA
Bis-(p-nitrophenyl) phosphate
dT-5’44-nitrophenyI phosphate)
dT-3’-(4-nitrophenyl phosphate)
CAMP
Glucose 6-phosphate
0.6
0.6
0.6
0.6
2.0
0.6
0.6
1.2
0.6
0.6
0.6
0.6
2.0
2.0
2.0
2.0
2.0
Km
liM
mA¶
1.0
2.5
3.3
3.0
12
8
11
22
10
2.0
”
2.7
2.9
2.2
2.8
0.06 8,000
3.8
c0.3
a.2
c0.2
Dinucleoside
Tetraphosphatase
Activities
2884
Liver
in Rat
nitrophenyl phosphate)) after 5 days a t 4 "C, whereas the
DISCUSSION
activity on AMP remained unchanged through that period.
Two
different
hydrolytic
activities on dinucleoside tetraMichaelian kinetics was obtained when Gp4G,Ap4A, Ap3A,
phosphates
have
been
solubilized
by Triton X-100 from the
ApzA, NAD, or bis(p-nitrophenyl) phosphate was used as
isotonic rat liver precipitate. Both activities differ markedly
substrate. The K , values calculated were 12 p ~8 p, ~ 11, p ~ ,
in abundance, molecular weight, kinetic properties, and, pos22 pM, 10 pM, and 8 mM, respectively. It seemed to us of
sibly,
in subcellular localization. We have obtained strong
interest to test theinhibition of the enzyme by guanosine 5'evidence that the smaller one is the specific dinucleoside
tetraphosphate. This nucleotide is a very strong competitive
tetraphosphatase (EC 3.6.1.17) which was previously reported
inhibitor (Ki 10 nM) of the specific dinucleoside tetraphosin the cytosol of several rat tissues ( L o b a t h et al., 1975b;
phatase from rat liver (Lobath et al., 197513; and thiswork).
Cameselle et al., 1982). The soluble and the detergent-ex:
The effect of this nucleotide was tested onthe high molecular
tracted specific enzymes have the same size as evaluated by
weight form obtained by chromatography on a Sepharose 4B
gel filtration, the same cationrequirements, the same pattern
column in the presence of 0.5% Triton X-100 (Fig. 2a).
and extent of inhibition by guanosine 5'-tetraphosphate and
Guanosine 5"tetraphosphate was also a competitive inhibitor
Ca2+,and very similar K , values for Ap4A and Gp,G. Furof Ap4A and Ap3A hydrolysis by the unspecific tetraphosphathermore, the same anhydride bond is split by both enzymes,
tase with Ki values of 2 p~ in both cases. A fixed concentration
yielding 1 molof nucleoside 5'-triphosphate and 1 molof
(22 p ~ of) guanosine 5'-tetraphosphate was also tested as
nucleoside 5'-monophosphate/mol of dinucleoside tetraphosinhibitor of the hydrolysis of several substrates of the enzyme,
phate hydrolyzed. Hence, it appears that the specific tetraAp4A, Ap3A, NAD, and dT-5'-(4-nitrophenyl phosphate), all
phosphatase could be particulate to some extent. Under our
of them at a 35 p~ concentration. In thoseexperimental
experimental conditions, the detergent-extracted specific enconditions, the inhibitions obtained were 66,59,57, and66%,
zyme represents about20% of the activity found in the cytosol
respectively, inrelation to controlswithout guanosine 5'and can be compared with the corresponding percentage (3%)
tetraphosphate. The cation requirements were studied with
obtained for lactate dehydrogenase and glucose phosphate
the same preparation. After extensive dialysis, the enzyme
isomerase. Relevant to thediscussion is the fact that thelast
exhibited a residual activity in the absence of added bivalent
two enzymes have been reported to be present in the nucleocations (Fig. 6). This activity was completely abolished by 1
plasm (Price and Stevens, 1982). If lactate dehydrogenase and
mM EDTA. Full reactivation was achieved with an equimolar
glucose phosphate isomerase activities which can be extracted
amount of Zn", but not of M$+, showing that probably the
with Triton X-100 are from nuclear origin, it follows that the
enzyme hasa strict requirement for Zn2+. Othercations
specific tetraphosphatase that appears in the same preparabehaved as activators, such as Ca", M$+, or Mn2+.Maximal
tion should be at least considered as a component of the rat
velocity obtained with Ca2+was twice as much as that with liver precipitate. Obviously, both the cytosolic and theparticMg+. Theconcentration giving rise to half-maximal activa- ulate specific dinucleoside tetraphosphatases could corretion was about 0.5 mM for both Ca2+and M e . Quite different spond to thesame protein and, infact, we think likely that it
was the behavior of the enzyme in the presence of Mn2+.A is so. The assumption that this enzyme is present in the
sharp peak of activity was reached at around 50 p ~decreasing
,
nucleoplasm is tempting, since Ap4Ahas been related to DNA
also sharply at a concentration of 0.1-0.5 mM. After a slight synthesis (Grummt,1978). However, this picture must remain
increase in activity, the enzyme was inhibited by higher, up speculative for the moment.
to 4 mM, Mn2+concentrations (Fig. 6). Maximal activity was
The high molecular weight form of the splitting activity on
found at pH 8.5, with both Ap4A or Ap3A as substrates. At Gp4Gand other nucleotides (Table 11) is clearly different from
pH 7.0, the activity was less than 10% of that obtained at pH the specific tetraphosphatase. Its apparent molecular weight,
8.5. In the presence of cysteine, reduced glutathione, or p- as estimated by Sephadex G-200 gelfiltration in the presence
mercaptoethanol, each one at a concentration of 8 mM, the of 0.5% Triton X-100, is very near to 450,000 since ferritin
activity was, respectively, 10,45, and85% of that obtained in and theunspecific tetraphosphatase activity eluted practically
the absence of thiol groups.
at the same volume. Nevertheless, when detergent was not
included in the elution buffer, the latter enzyme seemed to
aggregate, itself spreading towards the void volumeof a Sepharose 4B column (exclusion limit 2 X lo7 dalton). On the
other hand, its apparent molecular weight was smaller when
estimated in sucrose gradient centrifugation. In thiscase and
depending on the detergent used, the values varied from
150,000 to 240,000. Altogether, these results could indicate
that thehigh molecular weight form of the Gp4G-lyticactivity
is an integral membrane protein that can be extracted with
Triton and aggregates when the detergent is eliminated. According to this and to the special features of the proteinI
detergent complexes, the molecular weights quoted above for
0
2
4
the unspecific tetraphosphatase activity are to be considered
[Salt], mM
as an index of its behavior during gel filtration or gradient
FIG. 6. Cation effect on the high molecular weight, unspecific dinucleoside tetraphosphatase activity. Enzyme from the centrifugation, but not as a measure of actual protein size. In
Sepharose 4B step was used (elution with Triton X-100;Fig. 2a). this regard, the amount of lipid and/or detergent bound to
X-100 complexes
That preparation was thoroughly dialyzed for 24 h against 125 vol- protein is not known, and the protein-Triton
umes of 50 mM Tris/HCl buffer, pH 7.5, 0.5 mM EDTA, which was
are featured by a low sedimentation coefficient and a high
renewed %fold during that period. Enzyme samples of 40 pl were
Stokes radius (Heleniusand Simons, 1975), mainly due to the
incubated at 25 "C with 50 mM Tris/HCl buffer, pH 7.5, 26 pM ApsA high partial specific volume of Triton X-100. Hence, factors
as substrate and various concentrations of MgCl, (O), MnC12 (O), or
CaClz ( X ) . The increase in absorbance a t 259 nmwas recorded such as nature andconcentration of detergent can affect size
estimations (e.g. see Ey and Ferber, 1977). It seems reasonable
(hyperchromicity assay).
-
Dinucleoside Tetraphosphatase Activitiesin Rat Liver
2885
REFERENCES
to assume that the true molecular weight of the unspecific
Cameselle,
J.
C.,
Costas,
M.
J., Sillero, M. A. G., and Sillero, A. (1982)
dinucleoside tetraphosphatase complex is between the values
Bwchem. J. 201,405-410
estimated by gel filtration and sedimentation analysis. These
Dulley, J. R. (1975) Anal. Biochem. 67,91-96
results, together with the kinetic data presented above, indi- Ey, P. L., and Ferber, E. (1977) Biochim. Biophys. Acta 480, 163cate that thisactivity may correspond to thephosphodiester177
ase I (EC 3.1.4.1), earlier characterized by others in the rat Finamore, F. J., and Warner, A. H. (1963) J. Biol. Chem. 2 3 8 , 344348
liver particulate fraction (Touster et al., 1970; Prospero et al.,
Futai, M., and Mizuno, D. (1967) J. Biol. Chem. 242,5301-5307
1973). It is known that 50% of the total phosphodiesterase I Goldberg, N. D.,and Haddox, M. K. (1977) Annu. Reu. Biochem. 4 6 ,
is located in the microsomal fraction (Touster et al., 1970).
823-896
Under our experimental conditions, almost all
the enzyme Gonenne, A., and Ernst, R. (1978) Anal. Biochem. 8 7 , 28-38
sedimented in the37,000 X g precipitate (see the first heading Grummt, F. (1978) Proc. Natl. Acad. Sci. U. S. A. 75,371-378
Grummt, F., Waltl, G., Jantzen, H. M., Hamprecht, K., Huebscher,
under “Results”). Phosphodiesterase I can be isolated from
U., and Kuenzle, C.C. (1979) Proc. Natl. Acad. Sci. U. S. A. 7 6 ,
several sources, requires a nucleoside 5“phosphoryl residue
6081-6085
with a 3”hydroxyl group, and is equally active on 5’-5’ and Harshman, S., Conlin, J. G., Stoller, D., and Harshman, D. L. (1979)
J. Membr. Bwl. 5 0 , 177-185
5‘-3’ phosphodiester linkages (Khorana, 1961; Razzell, 1963).
A., and Simons, K. (1975) Biochim. Biophys. Acta415,29The enzyme presents maximal activity at pH8.5 (Schliselfeld Helenius,
79
et al., 1965), is inhibited by EDTA, and requires bivalent Hjelmeland, L. M. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, 6368cations (Prospero et al., 1973), with a strict requirement for
6370
Zn2+(Lau and Carlson, 1981). Thiol groups are also inhibitors Khorana, H. G . (1961) in The Enzymes (Boyer, P. D., Lardy, H., and
Myrback, K., eds) 2nd Ed., Vol. 5, pp. 79-94, Academic Press, New
of the enzyme (Razzell, 1963). The properties summarized
York
above are in good accord with those obtained here for the Lau, J. T. Y., and Carlson, D. M. (1981) J. Biol. Chem. 256, 7142unspecific high molecular weight dinucleoside tetraphospha7145
tase. From Table I1 it could be also inferred that the activity Leloir, L. F., and Cardini, C. E. (1957) Methods Enzymol. 3,840-850
C. D., Sillero, M.A. G., and Sillero, A. (1975a) Biochem.
is inversely related to thelength of the inner phosphate chain, Lobaton,
Biophys. Res. Commun. 6 7 , 279-286
considering the decreasing activitieson Ap2A,Ap,A,
and Lobath, C. D., Vallejo, C. G., Sillero, A., and Sillero, M. A. G . (1975b)
A p a . T h ehydrolysis of ApzA and Ap,A by the kidney phosEur. J. Bwchem. 50,495-501
phodiesterase Ihad been incidentally reported(Razzell, 1963). Martin, R. G., and Ames, B. N. (1961) J. Biol. Chem. 2 3 6 , 13721379
As shown under “Results,” the activity on AMP is due to a Moreno,
A., Lobath, C. D., Sillero, M. A. G., and Sillero, A. (1982)
different enzyme, probably a nucleotidase (EC 3.1.3.5), in our
Int. J. Biochem. 14,629-634
preparation. In the presence of this activity, it is difficult to Price, N. C., and Stevens, L. (1982) Fundamentals of Enzymology, pp.
312-314, Oxford University Press, New York
assess the products of the reaction catalyzed by the phosphoT.D., Burge, M. L. E., Norris, K. A., Hinton, R. H., and
diesterase I on
the different substrates which have been tested. Prospero,
Reid, E. (1973) Biochem. J. 132,449-458
It is also pertinentto recall that confusion existsin the Rapaport, E., Zamecnik, P. C., and Baril, E. F. (1981) J. Biol. Chem.
256,12148-12151
bibliography on this enzyme. Three entries of the Enzyme
Commission may correspond to the same enzyme: phospho- Razzell, W. E. (1963) Methods Enzymol. 6 , 236-258
M. F., Renart, J., Sillero, M. A. G., and Sillero, A. (1976)
diesterase I (EC 3.1.4.1; Khorana, 1961), oligonucleotidase Renart,
Biochemistry 15,4962-4966
(EC 3.1.13.3; Futai and Mizuno, 1967), and nucleotide pyro- Schliselfeld, L. H., Van Eys, J., and Touster, 0. (1965) J. Bwl. Chem.
240,811-818
phosphatase (EC 3.6.1.9; Touster et al., 1970; Harshman et
al., 1979).At the same time, uncertainty exists concerning the Sillero, A., and Ochoa, S. (1971) Arch Biochem. Biophys. 1 4 3 , 548552
truesubstrate(s)andthe
function of this enzyme(s). We Sillero, M. A. G., Villalba, R., Moreno, A., Quintanilla, M., Lobath,
incidentally arrived at it by following the hydrolytic activity
C. D., and Sillero, A. (1977) Eur. J. Biochem. 76,331-337
on Ap,A and/or Gp4G. The family of dinucleoside polyphos- Tashima, Y. (1975) Anal. Biochem. 69, 410-414
phates could be of relevance in metabolic regulation (Renart Terasaki, W. L., Russell, T. R., and Appleman, M. M. (1974) Methods
Enzymol. 3 8 , 257-259
et al., 1976; Sillero et al., 1977; Grummt et al., 1979; Rapaport Touster, O., Aronson, N. N., Jr., Dulaney, J. T., and Hendrickson,
et al., 1981; Yamakawa et al., 1982) and certainly are possible
H. (1970) J. Cell Biol. 47,604-618
physiological substrates for this enzyme. The K,,, values re- Vallejo, C. G., Sillero, M. A. G., and Sillero, A. (1974) Biochim.
Bwphys. Acta 358, 117-125
poked here for Ap4A and Gp,G are around 10 p ~ unequivo,
Vallejo, C. G., Lobath, C. D., Quintanilla, M., Sillero, A., and Sillero,
cally within the range of K , values exhibited by other enzymes
M. A. G . (1976) Biochim. Biophys. Acta 4 3 8 , 304-309
also present in the particulate fraction such as some cyclic Wang, C. S., and Smith, R. L. (1975) Anal. Biochem. 6 3 , 414-417
nucleotide phosphodiesterases (Terasaki et al., 1974; Goldberg Warner, A. H., and Finamore, F. J. (1965) Biochemistry 4,1568-1575
and Haddox, 1977). Also deserving special mention is the Yamakawa, M., Furuichi, Y., and Shatkin, A. J. (1982) Proc. Natl.
Acad. Sci. U. S. A. 79,6142-6146
inhibition of the activity by guanosine 5’-tetraphosphate, Zamecnik, P. C. (1969) Cold Spring Harbor Symp. Quunt. Biol. 34,
which is competitive with a K, value of 2 p ~ .
1-16