Vol. 24, No. 4, pp. S-55-371, 1992 Printed in Great Britam. All rights reserved
0020-71 IX/92 $5.00 + 0.00 Copyright Q 1992 Pergamon Press plc
Inr. J. Biochem.
CHARACTERISATION OF A ~~~(5‘-NUCLEOSIDYL) TRIPHOSPHATE PYROPHOSPHOHYDROLASE FROM ENCYSTED EMBRYOS OF THE BRINE SHRIMP ARTEMIA MARK PRESCOTT,N. Department
MATTHEWH,THORNE,ANDREWD. MILNEand ALEXANDER G. MCLENNAN*
of Biochemistry,
University of Liverpool, P.O. Box 147, Liverpool L69 3BX, U.K. [Tel. (051) 794-4369; Fax (051) 7944349) (Receiued I7 July 1991)
Abstract- 1. A P’,P’-bis(S-nucleosidyl)triphosphate pyrophosphohydrolase (Np,Nase) has been partially purified from Artemiu embryos. 2. The Np,Nase has a native M, of 115,000 and preferentially hydrolyses substrates of the form Np,N. Relative rates of hydrolysis are Ap,A (V, = I.O), Gp,G (V,, = 0.71), AplA (V,,, = 0.08), ApJA (V,, = O.OY),Gp,G (V,, = 0.3) and Gp,G (V,,, = 0.33). An NMP is always one of the products. 3. The K,,, values for Ap,A and Gp,G are 15 and 10 PM respectively. 4. Mg*+, Mn’+ and Ca2+ Ions all stimulate the activity, while Zn2+, Coz+ and Ni2+ ions are inhibitory. 5. The activity of the Np,Nase remains constant during pre-emergence development of encysted embryos but decreases slightly after hatching.
INTRODUCTION Bis(S’-nucleosidyl)oligophosphates of genera1 structure Np,N have been detected at submicromolar to micromolar concentrations in all organisms studied so far. Of the tetraphosphates, the most widely studied has been ~~~(S’-adenosyl)tetraphosphate, Ap,A. This ubiquitous nucleotide has been ascribed roles in DNA replication (Grummt, 1988; Zamecnik, 1983) and, along with other adenylylated tetraphosphates, e.g. Ap,G, in the response of cells to heat shock, oxidative stress and DNA damage (Baker and Ames, 1988; Baker et al., 1988; Bochner et al., 1984; Garrison et al., 1986; Gilson et al., 1988). Gp,G, on the other hand, is found at millimolar levels in the yolk platelets of eggs and embryos of a few branchiopod crustacea where it serves as a source of purines for development and as a potential source of phosphate bond energy (Sillero et al., 1987; Warner, 1979; Warner, 1992). All cells possess specific hydrolases or phosphorylases whose functions appear to be to regulate the catabolism of these compounds (Guranowski and Sillero, 1992). Less is known about the possible functions of the related bis(5-nucleosidyl) triphosphates (Np,N). Adenine-containing members of this family, e.g. Ap,A, Ap,G, also respond to metabolic stress in prokaryotes and, to a lesser extent, in eukaryotes with an increase in their concentrations (Bochner et al.,
1984; Miller and McLennan, 1986), but no specific roles have yet been suggested for them. Again, Gp,G is found in appreciable quantities in the yolk platelets of embryos of the brine shrimp Artemia and in other branchiopod crustacea, where it is presumed to have functions similar to those of Gp,G (Warner, 1992; Warner and Finamore, 1965). One approach to understanding the biological roles of these nucleotides is to study the enzymes responsible for their synthesis and degradation. Hydrolytic enzymes displaying a high degree of specificity for ~~~(5-nucleos~dyl)triphosphates have been isolated from rat liver (Costas ef ai., 1985, 1986; Sillero et al., 1977) and other rat tissues (Costas et ai., 1984; Garcia-AgGndez, 1991; Montero et al., 1986), lupin seeds (Jakubowski and Guranowski, 1983) and Escherichia coli (Hurtado et al., 1987). The properties of these enzymes have recently been reviewed (Guranowski and Sillero, 1992). Here, we describe the pu~fication and properties of a ~~~(5-nucleosidyl) triphosphate pyrophosphohydrolase (Np, Nase) from encysted embryos of the brine shrimp Artemia. This enzyme was originally reported to be absent from Artemiu (Sillero et al., 1977). Some of these results have been reported in a preliminary form (Prescott et al., 1989a). MATERIALSAND METHODS Materials
‘To whom all correspondence should be addressed. ~~~re~j~r~ff~~: Ap,A, Pj,P3-~~~(S’-adenosyl)triphosphate; P’,~j-~i~(S-guanosyl)triphos~hate~ &,N, GP, G, P’,P”-bis(S’-nudeosidvl)oligophosphate; other 6is(Snucleosid$)oligophosphateH are abbreviated in a similar fashion; Np, Nase, P’,P’-bis(S’-nucleosidyl)triphosphatepyrophosphohydrolase (EC. 3.6.1.29); Np,Nase, P’,P“-bis(S’-nucleosidyl)tetraphosphate pyrophosphohydrolase (EC 3.6.1.17).
Great Salt Lake Artemia cysts were from the Sanders Brine Shrimp Co., Ogden, Utah, U.S.A. and were obtained in 1980. Calf intestine alkaline phosphatase. pyruvate kinase and molecular weight calibration proteins were from Boehringer. All dinucleoside oligophosphates were from Sigma. [P-32P]Ap,A was prepared by a two-step method involving the phosphorylation of AMP to v-32P]ADP using [Y-)~P]ATP (Amersham) and adenylate kinase (Boehringer) and the condensation of the labelled ADP with AMP using a water-soluble carbodiimide (Ng and Orgel, 1987; Prescott 565
MARK PRESCOTT et al.
566
60
80
100
120
140
Fraction number Fig. 1. Ultrogel AcA44 chromatography of a Q-Sepharose eluate containing both Np,Nase (0) and Np,Nase (0) activities. Approximately 30 ml of a concentrated Q-Sepharose eluate (Prescott et al., 1989b) were applied to a 5 x 95 cm column of Ultrogel AcA44 in 50 mM potassium phosphate buffer, pH 7.5, 10% (v/v) glycerol, 5 mM 2-mercaptoethanol, 1 mM EDTA and the column eluted at 80 ml/hr with the same buffer. Fractions (12.4 ml) were collected and 12.5 ~1 portions assayed for both activities using the luminescence assay (method 1). For molecular weight determinations (m), the column was calibrated with (1) rabbit muscle aldolase (ISS,OOO), (2) bovine serum albumin (68,000), (3) ovalbumin (45,000), (4) chymotrypsinogen A (25,000) and (5) cytochrome c (12,500). Absorbance at 280 nm (...).
and McLennan, 1990). The final product was purified on a Mono-Q anion-exchange column. The carbodiimide condensation procedure was also used to prepare [3H]Ap,A. Enzyme assays Several different assay procedures were employed for the detection of Np,Nase. They were employed under different circumstances depending on the substrate or when either speed, lack of auxiliary enzymes or precise quantitation was most important. One unit of Np,Nase activity is defined as the amount that degrades 1 nmol of AplA per min at 30°C. Np,Nase activity was assayed as previously described (Prescott et al., 1989b). Method l-Luminescence assay. This was based on a previous procedure (Ogilvie and Jakob, 1983) and contained 23 mM Hepes-KOH, pH 7.8, 4 mM acetate, 100 PM Ap,A, 2 mM phosphoenolpyruvate, 2 units pyruvate kinase and 25~1 ATP-monitoring reagent (LKB) in a total volume of 125 ~1. The increase in luminescence was monitored over a 2min period at 25-C with an LKB 1250 Luminometer. Method 2-Radiolabel assays (DEAE). These contained 20mM bisTris+HCI, pH 7.0, lOO~g/ml BSA, 1 mM Mg acetate, 5 units alkaline phosphatase and 100 PM [‘H]Ap,A (40 mCi/mmol) in a total volume of 50 ~1. After incubation for 10 min at 3O’C, assays were chilled on ice and 300 ~1 of a 25% (v/v) suspension of DEAE-Sephacel in IOmM Tris-HCl, pH 7.5, were then added followed by centrifugation at 16,000g for 5 min. 200~1 samples of the supernatants were added to 4 ml Optiphase-MP scintillant (LKB) and the radioactivity due to [‘Hladenosine released from the products by the alkaline phosphatase was determined. Method 3-Radiolabel assays (t.i.c.). Np,Nase was assayed for 15 min at 30’C in a total volume of 25 ~1 containing 20 mM bisTris+HCl, pH 7.0, 100 pg/ml BSA, 1 mM Mg acetate and [/?-‘2P]ApjA (20 mCi/mmol). Five ~1 samples were mixed with 0.2~1 each of IOmM ADP and AplA markers and spotted on to a PEI-cellulose plate which was then run its full length in 1.6 M LiCI. The ADP spots were cut out and counted directly in Optiphase MP scintillant. Method I-H.p.l.c. assays. Np,Nase was incubated for 10min at 30°C with 100 or 500pM nucleotide substrates in 20 mM bisTris-HCl, pH 7.0, 100 pg/ml BSA, 1 mM Mg acetate in a total volume of 50~1. Seventy ~1 of 50mM
ammonium phosphate, pH 5.2, were added and 50 ~1 of the mixture injected on to a 4.6 x 250mm Partisil lo-SAX column. The column was developed isocratically for 10 min with 5% buffer B followed by a 44min gradient from 5
to 80% buffer B where buffer A = 50 mM ammonium phosphate, pH 5.2, and buffer B = 1M ammonium phosphate,
pH 5.7 (Garrison
et al., 1982).
Protein determination This was assayed by the bicinchoninic acid (Smith et al., 1985) and silver-binding (Krystal, 1987) methods.
RESULTS Pkjication
of Np,Nase
from
Artemia
cysts
During the early stages of purification (extraction, ammonium sulphate precipitation and chromatography on Q-Sepharose), the Np,Nase co-purified with the Np,Nase activity which we and others have previously described. Therefore the details of purification of the Np,Nase from 250 g dry cysts up to and including the running of the Ultrogel AcA-44 column are exactly as described for the Np,Nase (Prescott et al., 1989b). Fractions eluting from Ultrogel AcA-44 and which contained the Np,Nase activity separated from Np,Nase were pooled and dialysed overnight against 2 x 10 vol. Mono-Q loading buffer (20 mM Tris-acetate, pH 8.0, 10% glycerol, 1 mM 2-mercaptoethanol). This fraction was then applied at 4 ml/min to an 8 ml Mono-Q HRlOjlO column and the column eluted with a 160 ml gradient of (M.75 M sodium acetate in loading buffer. Fractions containing the major peak of Np,Nase activity were dialysed overnight against 50 vol 10mM potassium phosphate, pH 6.8, 10 PM CaCl, and applied to a 100 x 7.8 mm Bio-Gel HPHT column (Bio-Rad) at 0.7 ml/min. After eluting unbound protein, the column was eluted with a 31.5 ml gradient of l&500 mM potassium phosphate, pH 6.8, containing 10 p M CaCl, . The active fractions were pooled and
Artemia dinucleoside triphosphatase
s
J
600
1.L
500
1.0
._ 3E
400
Z 0.8 ‘:
!Z B
300
0.6 fB
-2
200
0.4 9
i
100
0.2 4B
&
0
0.0 0
10
20
30
40
1
B 0.5 g g E 0
50
Fraction number Fig. 2. Chromatography of Np,Nase on Mono-Q. Activity eluting from the Ultrogel AcA44 column was chromatographed on an 8 ml Mono-Q HRlO/lO column as described in the Experimental section. Three ml fractions were collected and 12.5 ~1 portions assayed for Np,Nase activity (0) with the luminescence assay (method 1). Fractions 17-24, which comprised 87% of the eluted activity, were pooled. Na acetate gradient (-); absorbance at 280 nm (. .). diluted with water to reduce the phosphate concentration of approx 100 mM and then concentrated to
0.55ml with an Amicon C-10 Centricon. Glycerol was added to 10% (v/v) and 2-mercaptoethanol to 1 mM and the final preparation stored at 4°C. The final yield was 2.7% with a purification factor of 215-fold over the crude extract. This low figure reflects the substantial loss of activity during purification. The final Bio-Gel fraction was used for all the studies described below, unless otherwise stated. Throughout the purification, fractions were assayed immediately with the luminescence assay (method 1). Pooled fractions were subsequently quantitated with the radiolabel assay (method 2). The relatively mild extraction conditions employed appeared to avoid the extraction of non-specific phosphodiesterases which would otherwise interfere with the quantitation of the enzymes at early states of the purification. Np,Nase activity co-eluted from Q-Sepharose with the Np,Nase but could not be assayed conveniently with the luminescence assay owing to the presence of endogenous ADP-binding proteins, hence its elution profile is not shown. The
two pyrophosphohydrolases were, however, completely separated by chromatography on Ultrogel AcA-44 (Fig. 1). As previously shown for the Artemia Np,Nase activity, the Np,Nase exhibited a degree of heterogeneity when purified on Mono-Q and displayed one major and one minor peak (Fig. 2). The major peak had a higher affinity for Bio-Gel HPHT and coincided with a major protein peak (Fig. 3). Unfortunately, despite repeated attempts, further purification of the enzyme proved impossible because of the instability of the preparation at this stage. A similar instability has been reported for the other eukaryotic Np,Nases (Guranowski and Sillero, 1992). Molecular mass Calibration of the Ultrogel AcA44 column with standard proteins gave a native molecular mass of 115,000 for the Np,Nase (Fig. 1). Since the final preparation was not homogeneous it has not been possible accurately to assign particular polypeptides to this enzyme. However, analysis of the peak
200
0.8 0.6
100
0.4 0.2 c 0 Fraction number Fig. 3. Chromatography
of Np,Nase on Bio-Gel HPHT. Activity eluting from Mono-Q was chromatographed on a Bio-Gel HPHT column as described in Materials and Methods. OSml fractions were collected and 12.5 ~1 portions assayed for Np,Nase activity (0) with the luminescence assay (method 1). Fractions 23-33, which comprised 92% of the eluted activity, were pooled. Potassium phosphate gradient (-); absorbance at 280 nm (. . .).
MARK PRESCWTet al.
568
0’
Of
0.0
1.0
0.5
2.0
1.5
I
0
20
40
Fig. 4. Activation of Np,Nase activity by added divalent cations. Np,Nase activity was determined with Ap,A as substrate in the presence of different concentrations of Mg*+ (a), Mn2+ (0) and Ca*+ (8) ions using method 3 in order to avoid any effect on auxiliary enzymes. 42.4 mU of enzyme were used in each assay.
Bio-Gel HPHT fractions by SDS-polyacrylamide gel electrophoresis showed at least two major polypeptides within the M, range 5060,000 (not shown). The enzyme may therefore be a homo- or heterodimer, but any firmer conclusion would be premature. and pH opt~~la
Although the dependence on temperature was not very critical within the range 2545’C, optimal activity of the Np,Nase was observed at 30°C and it was most active at neutral pH in the absence of any auxiliary enzymes in the assay (assay methods 3 and 4, not shown). All measurements were made at a constant ionic strength of 0.1. In these respects, this enzyme is very similar to other dinucleoside oligophosphate pyrophosphohydrolases that have been described. It was found to be particularly unstable in the presence of Tris buffers, and quickly lost activity at 4°C when stored in Tris-HCI. Eflect
ofdivalent cations
The activity of the Np,Nase was stimulated equally by Mg2+ ions at 0.5 to 2 mM and MnZ+ ions at 0.1 mM, while 1 mM Ca’+. in contrast to its inhibitory effect on the Artemia Np,Nase (Prescott et al., 1989b), was 45% as effective (Fig. 4). Identical properties have been reported for the rat enzyme (Costas et al., 1985; Sillero et al., 1977) while the lupin Np,Nase shows little activity with Mn*+ and none
120
with Ca*+ (Jakubowski and Guranowski, 1983). Interestingly, 14% activity was displayed in the absence of added divalent cations. This activity was abolished by pre-incubation with 1 mM EDTA but only the basal level (12% maximum) could be restored by the addition of excess Mg*+ suggesting the involvement of a different endogenous ion in the activity of this enzyme. Zn *+ ions are powerful inhibitors of the lupin and rat liver (Costas et al., 1985; Jakubowski and Guranowski, 1983). Np,Nases and were also found to inhibit the Artemia enzyme with an I& of 8 PM in the presence of 2 mM Mg*+ (Fig. 5). Co*+ and Ni*+ ions are also effective inhibitors of both the lupin (Jakubowski and Guranowski, 1983) and Artemia enzymes, with IC, values of 35 and 50 pM respectively for the Artemia enzyme (Fig. 5). Substrate speczjicity and products
The enzyme had an absolute specificity for dinucleotide substrates of the form Np,N (n > 2). Using the h.p.1.c. assay, no hydrolysis was detected with the following substrates: AMP, ADP, ATP, GMP, GDP, GTP, UMP, TTP, CAMP, NAD+, FAD, p-nitrophenyl phosphate and bis-fp-nitrophenyl)pyrophosphate. With saturating concentrations of diadenosine substrates (Ap,A), the Artemiu Np,Nase had a very strong preference for Ap,A with no activity towards Ap2A and 8% or less when n > 3 (Table 2). However with Gp,,G, both
Specific
Crude supernatant 50-100% Ammonium sulphate fraction Q-Sepharose fraction Ultrogei AcA44 fraction Mono-Q fraction Bio-Gel HPHT fraction (concentrated)
100
Fig. 5. Inhibition of Np,Nase activity by divalent cations. Np,Nase activity was determined with Ap,A as substrate in the presence of 2 mM Mg2+ and different concentrations of Zn*+ (@), Ni2+ (D) or Co*+ (0) ions using method 3 in order to avoid any effect on auxiliary enzymes. 63.6 mU of enzyme were used in each assay.
Table I. Purification scheme for the Np,Nase from Artemin cysts
Fraction
80
[Divalent ion] (j&i)
[ Divalent ion I fmM)
temperature
60
Volume (ml)
Protein (m8)
activity
Activity (units)
(units/m~)
Yield (%f
670
7640
271
0.035
100
146 31.5 162 36 0.6
4220 246 79 12 0.97
418 93 42 9.1 7.3
0.1 0.38 0.53 0.81 7.5
154 34 is 3.6 2.7
The figures shown in this table refer to one particular purification and are typical of those obtained on at least five occasions. Activities of each pooled fraction were determined using assay method 2 after dialysis of a sample to remove salts.
569
Artemia dinucleoside triphosphatase Table
2. Substrate
specificity Np,Nase
of the Arfemia
1985; Sillero et al., 1977) and E. coli (Hurtado et al., 1987) enzymes and 1.2 PM for the lupin Np,Nase (Jakubowski and Guranowski, 1983).
% Relative hvdrolvsis
Substrate
0
AP,A AP,A APIA AP,A AP,A GP~G GP,G Go, G
100 8 9 1.4 0 71 30 33
G;;G
Substrate specificity was determined by assay method 4 (h.p.l.c.) with 106mU enzyme and 5OOpM of each substrate. Preliminary experiments showed this concentration to be saturating for all substrates tested. Rates were determined by integration of the product peak areas after adjusting for the hypochromicity of the substrates and are expressed relative to that for Ap,A.
Gp,G and Gp,G were roughly half as effective as substrates as Gp, G which itself was 70% as effective as Ap,A. Where similar determinations have been made with the rat and lupin enzymes, there is good agreement (Costas et al., 1985; Jakubowski and Guranowski, 1983) although the marked lack of chain length specificity with Gp,G has not been examined or observed before. With regard to its mechanism of action, for each substrate examined by h.p.l.c., the Artemia enzyme always produced an NMP as one of the products, e.g. Gp,G gave GMP+GDP, GpIG gave GMP+GTP and Gp,G gave GMP + Gp., . The K,,, of the Np,Nase for Ap,A was determined to be 15 PM by assay method 1, while for Gp,G it was 1OpM (assay method 4). These compare with values for Ap,A of 12 PM for the rat (Costas et al.,
120L100
80 60 4020 -
0’
0
’
I
20
.
I
40
a
i
60
I 80
Developmental time (h) Fig. 6. Np,Nase activity during pre-emergence and early larval development of Artemiu. Crude extracts were prepared from decapsulated cysts or larvae at various developmental stages by homogenising 1 g portions in I ml 10 mM potassium phosphate buffer, pH 7.5, 10% (v/v) glycerol, 5 mM 2-mercaptoethanol, I mM EDTA, 1mM phenylmethylsulphonyl fluoride, 100 ng/ml soybean trypsin inhibitor with a Potter-Elvehjem homogeniser. After high-speed centrifugation and dialysis, 200 ~1 samples were layered on 4.8 ml 520% (w/v) sucrose gradients in homogenising buffer and centrifuged at 182,OOOg,, for 21 hr at 4°C. Portions (0.5 ~1) of the crude extracts (0) and each gradient fraction were assayed for Np,Nase activity using method 2. For each gradient, the activities of those fractions that contained Np,Nase were summed and the total activity obtained plotted against time of development (a).
Np,Nuse activity during embryonic and early larval development We have previously demonstrated a rapid, 3-fold increase in Np,Nase activity during pre-emergence development (t&l6 hr) of Artemiu embryos which is followed by a decline during subsequent larval growth (16-72 hr) (Prescott et al., 1989b). This fluctuation is only observed when the pyrophosphohydrolase activity sedimenting with an M, of 17,000 is separated from inhibitory or modulating factors present in crude cell homogenates by sucrose density gradient centrifugation; otherwise a very gradual 2-fold increase in enzyme activity up to 72 hr of development is seen. An analogous study of the activity sedimenting with an A4, between 100,000 and 120,000 (Np,Nase) showed that, in contrast to the Np,Nase, the activity in crude extracts and sucrose gradient fractions behaved in a similar manner, both rising slightly upon hatching then declining to about 2&30% of maximum between 40 and 72 hr (Fig. 6). Although the inclusion of EDTA, PMSF and soybean trypsin inhibitor in the homogenising buffers was found to inhibit most of the endogenous neutral protease activity, the loss of activity at the later times may still be the result of limited, non-specific proteolysis due to the increasing amounts of proteases that appear as the larval gut differentiates. Thus, in contrast to Np,Nase, there appears to be little change in the activity of the Np,Nase during pre-emergence development (i.e. during the first 24 hr). DISCUSSION
Since it has proved difficult to purify the Artemiu Np,Nase to homogeneity, it is not possible to be certain about its subunit structure. The lupin (Jakubowski and Guranowski, 1983), rat (Costas et al., 1984, 1985) and E. cofi (Hurtado et al., 1987) enzymes have been shown to be monomers of h4, 41,000, 3@35,000 and 36,000 respectively. In this respect, therefore, the Artemiu enzyme appears to represent a new size class of Np, Nase. We have never observed any Np, Nase activity within the 30-50 kDa region. The Artemiu and lupin enzymes do, however, appear to have a similar mechanism of action. The Artemiu Np,Nase always produces an NMP as one of the products. Originally, the Iupin enzyme was reported always to produce an NDP product (Jakubowski and Guranowski, 1983). Thus, Ap,A, Ap, A and Ap,A gave ADP + ADP, ADP + ATP and ADP + Ap, respectively. However, more recent studies performed with highly purified preparations of the lupin Np,Nase have revealed that the enzyme degrades Ap,A to AMP and ATP and not to two molecules of ADP (Guranowski and Sillero, 1992). It appears, therefore, that the original observation was incorrect and was probably due to the activity of contaminating adenylate kinase and ATPase activities in the Np,Nase preparation (Guranowski and Sillero, 1992). We have previously shown that the Artemiu Np,Nase and other asymmetrically cleaving
MARK F’RFXOTT et al
570
Np, Nases always yield an NTP as one of the products, while the symmetrically cleaving Np,Nases always generate an NDP (Prescott et al., 1989). This, coupled with the cleavage data for both enzymes with a range of phosphonate analogues of Ap,A, has led to the proposal that the asymmetrically cleaving Np,Nases normally accommodate a “pppN” moiety in the active site, while the symmetrically cleaving Np,Nases bind a “ppN”, though this distinction is lost with certain phosphonate substrates (McLennan et al., 1989). By analogy, it may be argued that the lupin and Artemia Np, Nases accommodate a “pN” moiety in their active site, thus completing the “family” of specificities among the Np,Nases. The products of hydrolysis of the rat or E. coli enzymes have not been determined with substrates containing more than three phosphate groups (Costas et al., 1985; Hurtado et al., 1987). A further similarity between the lupin and rat enzymes is their inhibition by low concentrations of Zn2+ ions. A possible role for Zn2+ in the modulation of cellular Ap, A and Ap,A levels has been suggested as a result of its stimulation of the synthesis of these nucleotides by certain aminoacyl-tRNA synthetases (Blanquet et al., 1983). Since the Artemiu Np,Nase is unaffected by concentrations of Zn2+ ions up to 100 PM (Prescott et al., 1989b), Zn2+ ions may have an effect on the relative intracellular concentrations of Np,N and Np,N nucleotides in Artemiu through differential effects on their degradation. We have already suggested that the ratio of Ap,A to Ap,A within the cell may be more important than their absolute concentrations (Miller and McLennan, 1986). What is the likely physiological role of the Artemiu Np,Nase? In other cells, Np,Nase activities are presumably involved in the regulation of Ap,N levels under normal conditions and after oxidative stress. In Artemiu, however, there is the additional role of the catabolism of the yolk platelet Gp,G store to be considered. The enzyme itself appears to have no particular preference for Gp,G or Ap, A: the ratio of V,,,/K,,, (Ap,A) to V,,,/Km (Gp,G) is 0.94. On the other hand, substrate specificity in uivo may be dictated by compartmentalisation of both the enzyme and its potential substrates: Gp,G is in the yolk platelets, while Ap,A appears to be cytosolic (Miller and McLennan, 1986). Although the Np,Nase studied here was isolated from the soluble fraction, Sillero and co-workers have reported that at least half of the rat liver Np,Nase is particulate, most of this being associated with the mitochondrial fraction (Costas et al., 1986; Garcia-Agundez et al., 1991). Finally, the
rather relaxed phosphate chain length specificity of the Artemiu Np,Nase with guanine dinucleotides suggests that this enzyme could also be important in the catabolism of Gp,G. The products of hydrolysis, GMP and GTP, are the same as those produced by the asymmetrical-Np,Nase which we and others have characterised (Prescott et al., 1989b; Vallejo et al., 1976). Further work will be required to clarify the role(s) of this enzyme in dinucleoside oligophosphate metabolism in Artemia. SUMMARY
A Pt,P’-bis(S-nucleosidyl)triphosphate pyrophosphohydrolase, Np,Nase, (E.C. 3.6.1.29) has been
partially purified from Artemiu embryos. This is the first time that such an activity has been reported in Artemiu. The Np,Nase has a native M, of 115,000 and has maximum activity at pH 7.0 in the presence of 1 mM Mg2+ or 0.1 mM Mn2+ ions. Ca2+ ions also stimulate activity while Zn2+, Co2+ and Ni2+ ions are inhibitory (IC,, = 8, 35 and 50 PM respectively). It hasaK,forAp,Aof15pMandforGp,GoflOpM and preferentially hydrolyses substrates of the form Np,N: relative rates of hydrolysis are Ap,A (V,,= 1.0) Gp,G (V,,,=O.71), ApIA (?‘,,,=0.08), Ap,A (VCe,= 0.09), GplG (V,,, = 0.3) and Gp,G (V,,, = 0.33); an NMP is always one of the products. The activity of the Np,Nase remains constant during pre-emergence development of encysted embryos but decreases slightly after hatching. Acknowledgements-This
work was supported by grants to A.G.McL from the Science and Engineering Research Council, the Wellcome Trust and the North West Cancer Research Fund. Note added in proof: After submission of this paper, Brevet et al. (Brevet A., Chen J., Fromant M., Blanquet S.
and Plateau P. (1991) J. Bact. 173, 5275-5279) reported the partial purification of an Np, Nase of M, 55,000 from yeast which has properties very similar to those reported here. The Artemia enzyme may be a dimeric form of this activity.
REFERENCES
Baker J. C. and Jacobson M. K. (1986) Alteration of adenyl dinucleotide metabolism by environmental stress. Proc. natn. Acad. Sci. U.S.A. 83, 235G2352. Baker J. C. and Ames B. N. (1988) Alterations in levels of 5’-adenyl dinucleotides following DNA damage in normal human fibroblasts and fibroblasts derived from patients with xeroderma pigmentosum. Mufat. Res 208, 87-93. Blanquet S., Plateau P. and Brevet A. (1983) The role of zinc in 5’,5’-diadenosine tetraphosphate production bv aminoacvl-transfer RNA synthetases. Molec. cell. B>ochem. 52,. 3-l 1. Bochner B. R.. Lee P. C.. Wilson S. W.. Cutler C. W. and Ames B. N. (1984) AppppA and related adenylylated nucleotides are synthesized as a consequence of oxidation stress. Cell 37, 2255232. Costas M. J., Montero J. M., Cameselle J. C., Sillero M. A. and Sillero A. (1984) Dinucleosidetriphosphatase from rat brain. Int. J. Biochem. 16. 757-762. Costas M. J., Cameselle J. C~, Giinther Sillero M. A. and Sillero A. (1985) Occurrence of dinucleosidetriohosnhatase in the cytosol and the particulate fractions ‘from*rat liver. Int. J: Biochem. 17, $03-907. Costas M. J.. Cameselle J. C. and Sillero A. (1986) Mitochondrial location of rat liver dinucleoside triphosphatase. J. biol. Chem. 261, 2064-2067. Garcia-Agtindez J. A., Cameselle J. C., Costas M. J., Gunther Sillero M. A. and Sillero A. (1991) Particulate diadenosine 5’,5’“-Pi, P’-triphosphate hydrolases in rat brain-two specific dinucleoside triphosphatases and two phosphodiesterase I-like hydrolases. Biochim. biophys. Acfa 1073, 402409.
Garrison P. N., Roberson G. M., Culver C. A. and Barnes L. D. (1982) Diadenosine 5’,5”‘-Pi, P4-tetraphosphate pvrophosphohydrolase from Physarum polycepha-
iurn. Substrate specificity. Biochemistry 21, 6129-6133. Garrison P. N.. Mathis S. A. and Barnes L. D. (1986) In uiuo levels of diadenosine tetraphosphate and adenosine tetraphospho-guanosine in Physarum polycephalum during the
Artemia dinucleoside triphosphatase
cell cycle and oxidative stress. Molec. cell. Biol. 6, 1179-l 186. Gilson G., Ebel J. P. and Remy P. (1988) Is Ap.,A involved in DNA repair processes? Expl Cell Res. 177, 143-153. Grummt F. (1988) Diadenosine tetraphosphate as a putative intracellular signal of eukaryotic cell cycle control. Modern Cell Bio,l. 6, 2964.
Guranowski A. and Sillero A. (1992) Enzymes cleaving dinucleoside polyphosphates. In Ap,A and Other Dinucleoside Polyphosphates (Edited by McLennan A. G.). CRC Press, Boca Raton, Fla. In press. Hurtado C., Ruiz A., Sillero A. and Sillero M. A. (1987) Specific magnesium-dependent diadenosine 5’,5”‘-P’,P’triphosphate pyrophosphohydrolase in Escherichia coli. J. Bact. 169, 1718-1723.
Jakubowski H. and Guranowski A. (1983) Enzymes hydrolyzing ApppA and/or AppppA in higher plants. Purification and some properties of diadenosine triphosphatase, diadenosine tetraphosphatase, and phosphodiesterase from yellow lupin (Lupinus luteus) seeds. J. biol. Chem. 258, 9982-9989.
Krystal G. (1987) A silver-binding assay for measuring nanogram amounts of protein in solution. Analyt. Biothem. 167, 8696.
McLennan A. G., Taylor G. E., Prescott M. and Blackburn G. M. (1989) Recognition of fib’-substituted and c@, a’P’-disubstituted phosphonate analogues of bis(5’adenosyl) tetraphosphate by the bis (5’-nucleosidyl)-tetraphosphate pyrophosphohydrolases from Artemia embryos and Escherichia coli. Biochemistry 28, 386883875.
Miller D. and McLennan A. G. (1986) Changes in intracellular levels of Ap,A and Ap,A in cysts and larvae of Artemia do not correlate with changes in protein synthesis after heat-shock. Nucl. Acids Res. 14, 603 I-6040.
Montero J. M., Garcia-Agindez J. A., Costas M. J., Cameselle J. C., Gunther Sillero M. A. and Sillero A. (1986) Hydrolytic activities on diadenosine 5’,5”‘-P’,P3triphosphate in the 27,000 x g supernatants and precipitates from rat brain, muscle, kidney and liver. Cienc. Biol. Portugal 11, 1-8. Ng K. E. and Orgel L. E. (1987) The action of a water-
soluble carbodiimide
on adenosine-5’-polyphosphates.
Nucl. Acids Res. 15, 3573-3580.
Ogilvie A. and Jakob P. (1983) Diadenosine 5’,5”‘-P1,Pstriphosphate in eukaryotic cells: identification and quantitation. Analyt. Biochem. 134, 382-392.
571
Prescott M. and McLennan A. G. (1990) Synthesis and applications of I-azido photoaffinity analogues of P’,P’-bis(S-adenosyl) triphosphate and P’,P4-bis(Sadenosyl)tetraphosphate. Analyt. Biochem. 184, 33&337. Prescott M.. Milne A. D. and McLennan A. G. (1989a) Enzymes of dinucleoside oligophosphate metabohsm in Artemia cysts and larvae. In Cell and Molecular Biology of Artemia Development (Edited by Warner A. H., MacRae T. H. and Bagshaw J. C.). NATO ASI Series A, Lz$ Sciences, Vol. 174, pp. 223-243. Plenum Press, New York. Prescott M., Milne A. D. and McLennan A. G. (1989b) Characterization of the bis(S-nucleosidyl) tetraphosphate pyrophosphohydrolase from encysted embryos of the brine shrimp Artemia. Biochem. J. 259, 831-838. Sillero A. and Gunther Sillero M. A. (1987) Interconversion of purine nucleotides in Artemia: a review. In Artemia Research and its Applications (Edited by Decleir W., Moens L., Slegers H., Sorgeloos P. and Jaspers E.), Vol. 2, pp. 289-307. Universa Press, Wetteren, Belgium. Sillero M. A. G., Villalba R., Moreno A., Quintanilla M., Lobaton C. D. and Sillero A. (1977) Dinucleosidetriphosphatase from rat liver. Purification and properties. Eur. J. Biochem. 76, 331-337.
Smith P. K., Krohn R. I., Hermanson G. T., Mallia A. K., Gartner F. H., Provenzano M. D., Fujimoto E. K., Goeke N. M., Olson B. J. and Klenk D. C. (1985) Measurement of protein using bicinchoninic acid. Analyt. Biochem. 150, 78&793. Vallejo C. G., Lobaton C. D., Quintanilla M., Sillero A. and Sillero M. A. G. (1976) Dinucleosidetetraphosphatase in rat liver and Artemia salina. Biochem. biophys. Acta 438, 304309.
Warner A. H. (1979) Studies on the biosynthesis and function of dinucleoside polyphosphates in Artemia embryos In Regulation of Macromolecular Synthesis by Low Molecular Weight Mediators (Edited by Koch 6. and Richter D.), pp. 161-177. Academic Press, New York. Warner A. H. (1992) Diguanosine and related nonadenylated polyphosphates. In Ap,A and Other Dinucleoside Polyphosphates (Edited by McLennan A. G.). CRC Press, Boca Raton, Florida. In press. Warner A. H. and Finamore F. J. (1965) Isolation, purification and characterization of P’ P’-diguanosine-5’triphosphate from brine shrimp eggs. Biochim. biophys. Acta 108, 525-530. Zamecnik P. (1983) Diadenosine 5’,5”‘-P’,P4-tetraphosphate (A-4A): its role in cellular metabolism. Analyt. Biochem. 134, l-10.
0020-71 IX/92 $5.00 + 0.00 Copyright Q 1992 Pergamon Press plc
Inr. J. Biochem.
CHARACTERISATION OF A ~~~(5‘-NUCLEOSIDYL) TRIPHOSPHATE PYROPHOSPHOHYDROLASE FROM ENCYSTED EMBRYOS OF THE BRINE SHRIMP ARTEMIA MARK PRESCOTT,N. Department
MATTHEWH,THORNE,ANDREWD. MILNEand ALEXANDER G. MCLENNAN*
of Biochemistry,
University of Liverpool, P.O. Box 147, Liverpool L69 3BX, U.K. [Tel. (051) 794-4369; Fax (051) 7944349) (Receiued I7 July 1991)
Abstract- 1. A P’,P’-bis(S-nucleosidyl)triphosphate pyrophosphohydrolase (Np,Nase) has been partially purified from Artemiu embryos. 2. The Np,Nase has a native M, of 115,000 and preferentially hydrolyses substrates of the form Np,N. Relative rates of hydrolysis are Ap,A (V, = I.O), Gp,G (V,, = 0.71), AplA (V,,, = 0.08), ApJA (V,, = O.OY),Gp,G (V,, = 0.3) and Gp,G (V,,, = 0.33). An NMP is always one of the products. 3. The K,,, values for Ap,A and Gp,G are 15 and 10 PM respectively. 4. Mg*+, Mn’+ and Ca2+ Ions all stimulate the activity, while Zn2+, Coz+ and Ni2+ ions are inhibitory. 5. The activity of the Np,Nase remains constant during pre-emergence development of encysted embryos but decreases slightly after hatching.
INTRODUCTION Bis(S’-nucleosidyl)oligophosphates of genera1 structure Np,N have been detected at submicromolar to micromolar concentrations in all organisms studied so far. Of the tetraphosphates, the most widely studied has been ~~~(S’-adenosyl)tetraphosphate, Ap,A. This ubiquitous nucleotide has been ascribed roles in DNA replication (Grummt, 1988; Zamecnik, 1983) and, along with other adenylylated tetraphosphates, e.g. Ap,G, in the response of cells to heat shock, oxidative stress and DNA damage (Baker and Ames, 1988; Baker et al., 1988; Bochner et al., 1984; Garrison et al., 1986; Gilson et al., 1988). Gp,G, on the other hand, is found at millimolar levels in the yolk platelets of eggs and embryos of a few branchiopod crustacea where it serves as a source of purines for development and as a potential source of phosphate bond energy (Sillero et al., 1987; Warner, 1979; Warner, 1992). All cells possess specific hydrolases or phosphorylases whose functions appear to be to regulate the catabolism of these compounds (Guranowski and Sillero, 1992). Less is known about the possible functions of the related bis(5-nucleosidyl) triphosphates (Np,N). Adenine-containing members of this family, e.g. Ap,A, Ap,G, also respond to metabolic stress in prokaryotes and, to a lesser extent, in eukaryotes with an increase in their concentrations (Bochner et al.,
1984; Miller and McLennan, 1986), but no specific roles have yet been suggested for them. Again, Gp,G is found in appreciable quantities in the yolk platelets of embryos of the brine shrimp Artemia and in other branchiopod crustacea, where it is presumed to have functions similar to those of Gp,G (Warner, 1992; Warner and Finamore, 1965). One approach to understanding the biological roles of these nucleotides is to study the enzymes responsible for their synthesis and degradation. Hydrolytic enzymes displaying a high degree of specificity for ~~~(5-nucleos~dyl)triphosphates have been isolated from rat liver (Costas ef ai., 1985, 1986; Sillero et al., 1977) and other rat tissues (Costas et ai., 1984; Garcia-AgGndez, 1991; Montero et al., 1986), lupin seeds (Jakubowski and Guranowski, 1983) and Escherichia coli (Hurtado et al., 1987). The properties of these enzymes have recently been reviewed (Guranowski and Sillero, 1992). Here, we describe the pu~fication and properties of a ~~~(5-nucleosidyl) triphosphate pyrophosphohydrolase (Np, Nase) from encysted embryos of the brine shrimp Artemia. This enzyme was originally reported to be absent from Artemiu (Sillero et al., 1977). Some of these results have been reported in a preliminary form (Prescott et al., 1989a). MATERIALSAND METHODS Materials
‘To whom all correspondence should be addressed. ~~~re~j~r~ff~~: Ap,A, Pj,P3-~~~(S’-adenosyl)triphosphate; P’,~j-~i~(S-guanosyl)triphos~hate~ &,N, GP, G, P’,P”-bis(S’-nudeosidvl)oligophosphate; other 6is(Snucleosid$)oligophosphateH are abbreviated in a similar fashion; Np, Nase, P’,P’-bis(S’-nucleosidyl)triphosphatepyrophosphohydrolase (EC. 3.6.1.29); Np,Nase, P’,P“-bis(S’-nucleosidyl)tetraphosphate pyrophosphohydrolase (EC 3.6.1.17).
Great Salt Lake Artemia cysts were from the Sanders Brine Shrimp Co., Ogden, Utah, U.S.A. and were obtained in 1980. Calf intestine alkaline phosphatase. pyruvate kinase and molecular weight calibration proteins were from Boehringer. All dinucleoside oligophosphates were from Sigma. [P-32P]Ap,A was prepared by a two-step method involving the phosphorylation of AMP to v-32P]ADP using [Y-)~P]ATP (Amersham) and adenylate kinase (Boehringer) and the condensation of the labelled ADP with AMP using a water-soluble carbodiimide (Ng and Orgel, 1987; Prescott 565
MARK PRESCOTT et al.
566
60
80
100
120
140
Fraction number Fig. 1. Ultrogel AcA44 chromatography of a Q-Sepharose eluate containing both Np,Nase (0) and Np,Nase (0) activities. Approximately 30 ml of a concentrated Q-Sepharose eluate (Prescott et al., 1989b) were applied to a 5 x 95 cm column of Ultrogel AcA44 in 50 mM potassium phosphate buffer, pH 7.5, 10% (v/v) glycerol, 5 mM 2-mercaptoethanol, 1 mM EDTA and the column eluted at 80 ml/hr with the same buffer. Fractions (12.4 ml) were collected and 12.5 ~1 portions assayed for both activities using the luminescence assay (method 1). For molecular weight determinations (m), the column was calibrated with (1) rabbit muscle aldolase (ISS,OOO), (2) bovine serum albumin (68,000), (3) ovalbumin (45,000), (4) chymotrypsinogen A (25,000) and (5) cytochrome c (12,500). Absorbance at 280 nm (...).
and McLennan, 1990). The final product was purified on a Mono-Q anion-exchange column. The carbodiimide condensation procedure was also used to prepare [3H]Ap,A. Enzyme assays Several different assay procedures were employed for the detection of Np,Nase. They were employed under different circumstances depending on the substrate or when either speed, lack of auxiliary enzymes or precise quantitation was most important. One unit of Np,Nase activity is defined as the amount that degrades 1 nmol of AplA per min at 30°C. Np,Nase activity was assayed as previously described (Prescott et al., 1989b). Method l-Luminescence assay. This was based on a previous procedure (Ogilvie and Jakob, 1983) and contained 23 mM Hepes-KOH, pH 7.8, 4 mM acetate, 100 PM Ap,A, 2 mM phosphoenolpyruvate, 2 units pyruvate kinase and 25~1 ATP-monitoring reagent (LKB) in a total volume of 125 ~1. The increase in luminescence was monitored over a 2min period at 25-C with an LKB 1250 Luminometer. Method 2-Radiolabel assays (DEAE). These contained 20mM bisTris+HCI, pH 7.0, lOO~g/ml BSA, 1 mM Mg acetate, 5 units alkaline phosphatase and 100 PM [‘H]Ap,A (40 mCi/mmol) in a total volume of 50 ~1. After incubation for 10 min at 3O’C, assays were chilled on ice and 300 ~1 of a 25% (v/v) suspension of DEAE-Sephacel in IOmM Tris-HCl, pH 7.5, were then added followed by centrifugation at 16,000g for 5 min. 200~1 samples of the supernatants were added to 4 ml Optiphase-MP scintillant (LKB) and the radioactivity due to [‘Hladenosine released from the products by the alkaline phosphatase was determined. Method 3-Radiolabel assays (t.i.c.). Np,Nase was assayed for 15 min at 30’C in a total volume of 25 ~1 containing 20 mM bisTris+HCl, pH 7.0, 100 pg/ml BSA, 1 mM Mg acetate and [/?-‘2P]ApjA (20 mCi/mmol). Five ~1 samples were mixed with 0.2~1 each of IOmM ADP and AplA markers and spotted on to a PEI-cellulose plate which was then run its full length in 1.6 M LiCI. The ADP spots were cut out and counted directly in Optiphase MP scintillant. Method I-H.p.l.c. assays. Np,Nase was incubated for 10min at 30°C with 100 or 500pM nucleotide substrates in 20 mM bisTris-HCl, pH 7.0, 100 pg/ml BSA, 1 mM Mg acetate in a total volume of 50~1. Seventy ~1 of 50mM
ammonium phosphate, pH 5.2, were added and 50 ~1 of the mixture injected on to a 4.6 x 250mm Partisil lo-SAX column. The column was developed isocratically for 10 min with 5% buffer B followed by a 44min gradient from 5
to 80% buffer B where buffer A = 50 mM ammonium phosphate, pH 5.2, and buffer B = 1M ammonium phosphate,
pH 5.7 (Garrison
et al., 1982).
Protein determination This was assayed by the bicinchoninic acid (Smith et al., 1985) and silver-binding (Krystal, 1987) methods.
RESULTS Pkjication
of Np,Nase
from
Artemia
cysts
During the early stages of purification (extraction, ammonium sulphate precipitation and chromatography on Q-Sepharose), the Np,Nase co-purified with the Np,Nase activity which we and others have previously described. Therefore the details of purification of the Np,Nase from 250 g dry cysts up to and including the running of the Ultrogel AcA-44 column are exactly as described for the Np,Nase (Prescott et al., 1989b). Fractions eluting from Ultrogel AcA-44 and which contained the Np,Nase activity separated from Np,Nase were pooled and dialysed overnight against 2 x 10 vol. Mono-Q loading buffer (20 mM Tris-acetate, pH 8.0, 10% glycerol, 1 mM 2-mercaptoethanol). This fraction was then applied at 4 ml/min to an 8 ml Mono-Q HRlOjlO column and the column eluted with a 160 ml gradient of (M.75 M sodium acetate in loading buffer. Fractions containing the major peak of Np,Nase activity were dialysed overnight against 50 vol 10mM potassium phosphate, pH 6.8, 10 PM CaCl, and applied to a 100 x 7.8 mm Bio-Gel HPHT column (Bio-Rad) at 0.7 ml/min. After eluting unbound protein, the column was eluted with a 31.5 ml gradient of l&500 mM potassium phosphate, pH 6.8, containing 10 p M CaCl, . The active fractions were pooled and
Artemia dinucleoside triphosphatase
s
J
600
1.L
500
1.0
._ 3E
400
Z 0.8 ‘:
!Z B
300
0.6 fB
-2
200
0.4 9
i
100
0.2 4B
&
0
0.0 0
10
20
30
40
1
B 0.5 g g E 0
50
Fraction number Fig. 2. Chromatography of Np,Nase on Mono-Q. Activity eluting from the Ultrogel AcA44 column was chromatographed on an 8 ml Mono-Q HRlO/lO column as described in the Experimental section. Three ml fractions were collected and 12.5 ~1 portions assayed for Np,Nase activity (0) with the luminescence assay (method 1). Fractions 17-24, which comprised 87% of the eluted activity, were pooled. Na acetate gradient (-); absorbance at 280 nm (. .). diluted with water to reduce the phosphate concentration of approx 100 mM and then concentrated to
0.55ml with an Amicon C-10 Centricon. Glycerol was added to 10% (v/v) and 2-mercaptoethanol to 1 mM and the final preparation stored at 4°C. The final yield was 2.7% with a purification factor of 215-fold over the crude extract. This low figure reflects the substantial loss of activity during purification. The final Bio-Gel fraction was used for all the studies described below, unless otherwise stated. Throughout the purification, fractions were assayed immediately with the luminescence assay (method 1). Pooled fractions were subsequently quantitated with the radiolabel assay (method 2). The relatively mild extraction conditions employed appeared to avoid the extraction of non-specific phosphodiesterases which would otherwise interfere with the quantitation of the enzymes at early states of the purification. Np,Nase activity co-eluted from Q-Sepharose with the Np,Nase but could not be assayed conveniently with the luminescence assay owing to the presence of endogenous ADP-binding proteins, hence its elution profile is not shown. The
two pyrophosphohydrolases were, however, completely separated by chromatography on Ultrogel AcA-44 (Fig. 1). As previously shown for the Artemia Np,Nase activity, the Np,Nase exhibited a degree of heterogeneity when purified on Mono-Q and displayed one major and one minor peak (Fig. 2). The major peak had a higher affinity for Bio-Gel HPHT and coincided with a major protein peak (Fig. 3). Unfortunately, despite repeated attempts, further purification of the enzyme proved impossible because of the instability of the preparation at this stage. A similar instability has been reported for the other eukaryotic Np,Nases (Guranowski and Sillero, 1992). Molecular mass Calibration of the Ultrogel AcA44 column with standard proteins gave a native molecular mass of 115,000 for the Np,Nase (Fig. 1). Since the final preparation was not homogeneous it has not been possible accurately to assign particular polypeptides to this enzyme. However, analysis of the peak
200
0.8 0.6
100
0.4 0.2 c 0 Fraction number Fig. 3. Chromatography
of Np,Nase on Bio-Gel HPHT. Activity eluting from Mono-Q was chromatographed on a Bio-Gel HPHT column as described in Materials and Methods. OSml fractions were collected and 12.5 ~1 portions assayed for Np,Nase activity (0) with the luminescence assay (method 1). Fractions 23-33, which comprised 92% of the eluted activity, were pooled. Potassium phosphate gradient (-); absorbance at 280 nm (. . .).
MARK PRESCWTet al.
568
0’
Of
0.0
1.0
0.5
2.0
1.5
I
0
20
40
Fig. 4. Activation of Np,Nase activity by added divalent cations. Np,Nase activity was determined with Ap,A as substrate in the presence of different concentrations of Mg*+ (a), Mn2+ (0) and Ca*+ (8) ions using method 3 in order to avoid any effect on auxiliary enzymes. 42.4 mU of enzyme were used in each assay.
Bio-Gel HPHT fractions by SDS-polyacrylamide gel electrophoresis showed at least two major polypeptides within the M, range 5060,000 (not shown). The enzyme may therefore be a homo- or heterodimer, but any firmer conclusion would be premature. and pH opt~~la
Although the dependence on temperature was not very critical within the range 2545’C, optimal activity of the Np,Nase was observed at 30°C and it was most active at neutral pH in the absence of any auxiliary enzymes in the assay (assay methods 3 and 4, not shown). All measurements were made at a constant ionic strength of 0.1. In these respects, this enzyme is very similar to other dinucleoside oligophosphate pyrophosphohydrolases that have been described. It was found to be particularly unstable in the presence of Tris buffers, and quickly lost activity at 4°C when stored in Tris-HCI. Eflect
ofdivalent cations
The activity of the Np,Nase was stimulated equally by Mg2+ ions at 0.5 to 2 mM and MnZ+ ions at 0.1 mM, while 1 mM Ca’+. in contrast to its inhibitory effect on the Artemia Np,Nase (Prescott et al., 1989b), was 45% as effective (Fig. 4). Identical properties have been reported for the rat enzyme (Costas et al., 1985; Sillero et al., 1977) while the lupin Np,Nase shows little activity with Mn*+ and none
120
with Ca*+ (Jakubowski and Guranowski, 1983). Interestingly, 14% activity was displayed in the absence of added divalent cations. This activity was abolished by pre-incubation with 1 mM EDTA but only the basal level (12% maximum) could be restored by the addition of excess Mg*+ suggesting the involvement of a different endogenous ion in the activity of this enzyme. Zn *+ ions are powerful inhibitors of the lupin and rat liver (Costas et al., 1985; Jakubowski and Guranowski, 1983). Np,Nases and were also found to inhibit the Artemia enzyme with an I& of 8 PM in the presence of 2 mM Mg*+ (Fig. 5). Co*+ and Ni*+ ions are also effective inhibitors of both the lupin (Jakubowski and Guranowski, 1983) and Artemia enzymes, with IC, values of 35 and 50 pM respectively for the Artemia enzyme (Fig. 5). Substrate speczjicity and products
The enzyme had an absolute specificity for dinucleotide substrates of the form Np,N (n > 2). Using the h.p.1.c. assay, no hydrolysis was detected with the following substrates: AMP, ADP, ATP, GMP, GDP, GTP, UMP, TTP, CAMP, NAD+, FAD, p-nitrophenyl phosphate and bis-fp-nitrophenyl)pyrophosphate. With saturating concentrations of diadenosine substrates (Ap,A), the Artemiu Np,Nase had a very strong preference for Ap,A with no activity towards Ap2A and 8% or less when n > 3 (Table 2). However with Gp,,G, both
Specific
Crude supernatant 50-100% Ammonium sulphate fraction Q-Sepharose fraction Ultrogei AcA44 fraction Mono-Q fraction Bio-Gel HPHT fraction (concentrated)
100
Fig. 5. Inhibition of Np,Nase activity by divalent cations. Np,Nase activity was determined with Ap,A as substrate in the presence of 2 mM Mg2+ and different concentrations of Zn*+ (@), Ni2+ (D) or Co*+ (0) ions using method 3 in order to avoid any effect on auxiliary enzymes. 63.6 mU of enzyme were used in each assay.
Table I. Purification scheme for the Np,Nase from Artemin cysts
Fraction
80
[Divalent ion] (j&i)
[ Divalent ion I fmM)
temperature
60
Volume (ml)
Protein (m8)
activity
Activity (units)
(units/m~)
Yield (%f
670
7640
271
0.035
100
146 31.5 162 36 0.6
4220 246 79 12 0.97
418 93 42 9.1 7.3
0.1 0.38 0.53 0.81 7.5
154 34 is 3.6 2.7
The figures shown in this table refer to one particular purification and are typical of those obtained on at least five occasions. Activities of each pooled fraction were determined using assay method 2 after dialysis of a sample to remove salts.
569
Artemia dinucleoside triphosphatase Table
2. Substrate
specificity Np,Nase
of the Arfemia
1985; Sillero et al., 1977) and E. coli (Hurtado et al., 1987) enzymes and 1.2 PM for the lupin Np,Nase (Jakubowski and Guranowski, 1983).
% Relative hvdrolvsis
Substrate
0
AP,A AP,A APIA AP,A AP,A GP~G GP,G Go, G
100 8 9 1.4 0 71 30 33
G;;G
Substrate specificity was determined by assay method 4 (h.p.l.c.) with 106mU enzyme and 5OOpM of each substrate. Preliminary experiments showed this concentration to be saturating for all substrates tested. Rates were determined by integration of the product peak areas after adjusting for the hypochromicity of the substrates and are expressed relative to that for Ap,A.
Gp,G and Gp,G were roughly half as effective as substrates as Gp, G which itself was 70% as effective as Ap,A. Where similar determinations have been made with the rat and lupin enzymes, there is good agreement (Costas et al., 1985; Jakubowski and Guranowski, 1983) although the marked lack of chain length specificity with Gp,G has not been examined or observed before. With regard to its mechanism of action, for each substrate examined by h.p.l.c., the Artemia enzyme always produced an NMP as one of the products, e.g. Gp,G gave GMP+GDP, GpIG gave GMP+GTP and Gp,G gave GMP + Gp., . The K,,, of the Np,Nase for Ap,A was determined to be 15 PM by assay method 1, while for Gp,G it was 1OpM (assay method 4). These compare with values for Ap,A of 12 PM for the rat (Costas et al.,
120L100
80 60 4020 -
0’
0
’
I
20
.
I
40
a
i
60
I 80
Developmental time (h) Fig. 6. Np,Nase activity during pre-emergence and early larval development of Artemiu. Crude extracts were prepared from decapsulated cysts or larvae at various developmental stages by homogenising 1 g portions in I ml 10 mM potassium phosphate buffer, pH 7.5, 10% (v/v) glycerol, 5 mM 2-mercaptoethanol, I mM EDTA, 1mM phenylmethylsulphonyl fluoride, 100 ng/ml soybean trypsin inhibitor with a Potter-Elvehjem homogeniser. After high-speed centrifugation and dialysis, 200 ~1 samples were layered on 4.8 ml 520% (w/v) sucrose gradients in homogenising buffer and centrifuged at 182,OOOg,, for 21 hr at 4°C. Portions (0.5 ~1) of the crude extracts (0) and each gradient fraction were assayed for Np,Nase activity using method 2. For each gradient, the activities of those fractions that contained Np,Nase were summed and the total activity obtained plotted against time of development (a).
Np,Nuse activity during embryonic and early larval development We have previously demonstrated a rapid, 3-fold increase in Np,Nase activity during pre-emergence development (t&l6 hr) of Artemiu embryos which is followed by a decline during subsequent larval growth (16-72 hr) (Prescott et al., 1989b). This fluctuation is only observed when the pyrophosphohydrolase activity sedimenting with an M, of 17,000 is separated from inhibitory or modulating factors present in crude cell homogenates by sucrose density gradient centrifugation; otherwise a very gradual 2-fold increase in enzyme activity up to 72 hr of development is seen. An analogous study of the activity sedimenting with an A4, between 100,000 and 120,000 (Np,Nase) showed that, in contrast to the Np,Nase, the activity in crude extracts and sucrose gradient fractions behaved in a similar manner, both rising slightly upon hatching then declining to about 2&30% of maximum between 40 and 72 hr (Fig. 6). Although the inclusion of EDTA, PMSF and soybean trypsin inhibitor in the homogenising buffers was found to inhibit most of the endogenous neutral protease activity, the loss of activity at the later times may still be the result of limited, non-specific proteolysis due to the increasing amounts of proteases that appear as the larval gut differentiates. Thus, in contrast to Np,Nase, there appears to be little change in the activity of the Np,Nase during pre-emergence development (i.e. during the first 24 hr). DISCUSSION
Since it has proved difficult to purify the Artemiu Np,Nase to homogeneity, it is not possible to be certain about its subunit structure. The lupin (Jakubowski and Guranowski, 1983), rat (Costas et al., 1984, 1985) and E. cofi (Hurtado et al., 1987) enzymes have been shown to be monomers of h4, 41,000, 3@35,000 and 36,000 respectively. In this respect, therefore, the Artemiu enzyme appears to represent a new size class of Np, Nase. We have never observed any Np, Nase activity within the 30-50 kDa region. The Artemiu and lupin enzymes do, however, appear to have a similar mechanism of action. The Artemiu Np,Nase always produces an NMP as one of the products. Originally, the Iupin enzyme was reported always to produce an NDP product (Jakubowski and Guranowski, 1983). Thus, Ap,A, Ap, A and Ap,A gave ADP + ADP, ADP + ATP and ADP + Ap, respectively. However, more recent studies performed with highly purified preparations of the lupin Np,Nase have revealed that the enzyme degrades Ap,A to AMP and ATP and not to two molecules of ADP (Guranowski and Sillero, 1992). It appears, therefore, that the original observation was incorrect and was probably due to the activity of contaminating adenylate kinase and ATPase activities in the Np,Nase preparation (Guranowski and Sillero, 1992). We have previously shown that the Artemiu Np,Nase and other asymmetrically cleaving
MARK F’RFXOTT et al
570
Np, Nases always yield an NTP as one of the products, while the symmetrically cleaving Np,Nases always generate an NDP (Prescott et al., 1989). This, coupled with the cleavage data for both enzymes with a range of phosphonate analogues of Ap,A, has led to the proposal that the asymmetrically cleaving Np,Nases normally accommodate a “pppN” moiety in the active site, while the symmetrically cleaving Np,Nases bind a “ppN”, though this distinction is lost with certain phosphonate substrates (McLennan et al., 1989). By analogy, it may be argued that the lupin and Artemia Np, Nases accommodate a “pN” moiety in their active site, thus completing the “family” of specificities among the Np,Nases. The products of hydrolysis of the rat or E. coli enzymes have not been determined with substrates containing more than three phosphate groups (Costas et al., 1985; Hurtado et al., 1987). A further similarity between the lupin and rat enzymes is their inhibition by low concentrations of Zn2+ ions. A possible role for Zn2+ in the modulation of cellular Ap, A and Ap,A levels has been suggested as a result of its stimulation of the synthesis of these nucleotides by certain aminoacyl-tRNA synthetases (Blanquet et al., 1983). Since the Artemiu Np,Nase is unaffected by concentrations of Zn2+ ions up to 100 PM (Prescott et al., 1989b), Zn2+ ions may have an effect on the relative intracellular concentrations of Np,N and Np,N nucleotides in Artemiu through differential effects on their degradation. We have already suggested that the ratio of Ap,A to Ap,A within the cell may be more important than their absolute concentrations (Miller and McLennan, 1986). What is the likely physiological role of the Artemiu Np,Nase? In other cells, Np,Nase activities are presumably involved in the regulation of Ap,N levels under normal conditions and after oxidative stress. In Artemiu, however, there is the additional role of the catabolism of the yolk platelet Gp,G store to be considered. The enzyme itself appears to have no particular preference for Gp,G or Ap, A: the ratio of V,,,/K,,, (Ap,A) to V,,,/Km (Gp,G) is 0.94. On the other hand, substrate specificity in uivo may be dictated by compartmentalisation of both the enzyme and its potential substrates: Gp,G is in the yolk platelets, while Ap,A appears to be cytosolic (Miller and McLennan, 1986). Although the Np,Nase studied here was isolated from the soluble fraction, Sillero and co-workers have reported that at least half of the rat liver Np,Nase is particulate, most of this being associated with the mitochondrial fraction (Costas et al., 1986; Garcia-Agundez et al., 1991). Finally, the
rather relaxed phosphate chain length specificity of the Artemiu Np,Nase with guanine dinucleotides suggests that this enzyme could also be important in the catabolism of Gp,G. The products of hydrolysis, GMP and GTP, are the same as those produced by the asymmetrical-Np,Nase which we and others have characterised (Prescott et al., 1989b; Vallejo et al., 1976). Further work will be required to clarify the role(s) of this enzyme in dinucleoside oligophosphate metabolism in Artemia. SUMMARY
A Pt,P’-bis(S-nucleosidyl)triphosphate pyrophosphohydrolase, Np,Nase, (E.C. 3.6.1.29) has been
partially purified from Artemiu embryos. This is the first time that such an activity has been reported in Artemiu. The Np,Nase has a native M, of 115,000 and has maximum activity at pH 7.0 in the presence of 1 mM Mg2+ or 0.1 mM Mn2+ ions. Ca2+ ions also stimulate activity while Zn2+, Co2+ and Ni2+ ions are inhibitory (IC,, = 8, 35 and 50 PM respectively). It hasaK,forAp,Aof15pMandforGp,GoflOpM and preferentially hydrolyses substrates of the form Np,N: relative rates of hydrolysis are Ap,A (V,,= 1.0) Gp,G (V,,,=O.71), ApIA (?‘,,,=0.08), Ap,A (VCe,= 0.09), GplG (V,,, = 0.3) and Gp,G (V,,, = 0.33); an NMP is always one of the products. The activity of the Np,Nase remains constant during pre-emergence development of encysted embryos but decreases slightly after hatching. Acknowledgements-This
work was supported by grants to A.G.McL from the Science and Engineering Research Council, the Wellcome Trust and the North West Cancer Research Fund. Note added in proof: After submission of this paper, Brevet et al. (Brevet A., Chen J., Fromant M., Blanquet S.
and Plateau P. (1991) J. Bact. 173, 5275-5279) reported the partial purification of an Np, Nase of M, 55,000 from yeast which has properties very similar to those reported here. The Artemia enzyme may be a dimeric form of this activity.
REFERENCES
Baker J. C. and Jacobson M. K. (1986) Alteration of adenyl dinucleotide metabolism by environmental stress. Proc. natn. Acad. Sci. U.S.A. 83, 235G2352. Baker J. C. and Ames B. N. (1988) Alterations in levels of 5’-adenyl dinucleotides following DNA damage in normal human fibroblasts and fibroblasts derived from patients with xeroderma pigmentosum. Mufat. Res 208, 87-93. Blanquet S., Plateau P. and Brevet A. (1983) The role of zinc in 5’,5’-diadenosine tetraphosphate production bv aminoacvl-transfer RNA synthetases. Molec. cell. B>ochem. 52,. 3-l 1. Bochner B. R.. Lee P. C.. Wilson S. W.. Cutler C. W. and Ames B. N. (1984) AppppA and related adenylylated nucleotides are synthesized as a consequence of oxidation stress. Cell 37, 2255232. Costas M. J., Montero J. M., Cameselle J. C., Sillero M. A. and Sillero A. (1984) Dinucleosidetriphosphatase from rat brain. Int. J. Biochem. 16. 757-762. Costas M. J., Cameselle J. C~, Giinther Sillero M. A. and Sillero A. (1985) Occurrence of dinucleosidetriohosnhatase in the cytosol and the particulate fractions ‘from*rat liver. Int. J: Biochem. 17, $03-907. Costas M. J.. Cameselle J. C. and Sillero A. (1986) Mitochondrial location of rat liver dinucleoside triphosphatase. J. biol. Chem. 261, 2064-2067. Garcia-Agtindez J. A., Cameselle J. C., Costas M. J., Gunther Sillero M. A. and Sillero A. (1991) Particulate diadenosine 5’,5’“-Pi, P’-triphosphate hydrolases in rat brain-two specific dinucleoside triphosphatases and two phosphodiesterase I-like hydrolases. Biochim. biophys. Acfa 1073, 402409.
Garrison P. N., Roberson G. M., Culver C. A. and Barnes L. D. (1982) Diadenosine 5’,5”‘-Pi, P4-tetraphosphate pvrophosphohydrolase from Physarum polycepha-
iurn. Substrate specificity. Biochemistry 21, 6129-6133. Garrison P. N.. Mathis S. A. and Barnes L. D. (1986) In uiuo levels of diadenosine tetraphosphate and adenosine tetraphospho-guanosine in Physarum polycephalum during the
Artemia dinucleoside triphosphatase
cell cycle and oxidative stress. Molec. cell. Biol. 6, 1179-l 186. Gilson G., Ebel J. P. and Remy P. (1988) Is Ap.,A involved in DNA repair processes? Expl Cell Res. 177, 143-153. Grummt F. (1988) Diadenosine tetraphosphate as a putative intracellular signal of eukaryotic cell cycle control. Modern Cell Bio,l. 6, 2964.
Guranowski A. and Sillero A. (1992) Enzymes cleaving dinucleoside polyphosphates. In Ap,A and Other Dinucleoside Polyphosphates (Edited by McLennan A. G.). CRC Press, Boca Raton, Fla. In press. Hurtado C., Ruiz A., Sillero A. and Sillero M. A. (1987) Specific magnesium-dependent diadenosine 5’,5”‘-P’,P’triphosphate pyrophosphohydrolase in Escherichia coli. J. Bact. 169, 1718-1723.
Jakubowski H. and Guranowski A. (1983) Enzymes hydrolyzing ApppA and/or AppppA in higher plants. Purification and some properties of diadenosine triphosphatase, diadenosine tetraphosphatase, and phosphodiesterase from yellow lupin (Lupinus luteus) seeds. J. biol. Chem. 258, 9982-9989.
Krystal G. (1987) A silver-binding assay for measuring nanogram amounts of protein in solution. Analyt. Biothem. 167, 8696.
McLennan A. G., Taylor G. E., Prescott M. and Blackburn G. M. (1989) Recognition of fib’-substituted and c@, a’P’-disubstituted phosphonate analogues of bis(5’adenosyl) tetraphosphate by the bis (5’-nucleosidyl)-tetraphosphate pyrophosphohydrolases from Artemia embryos and Escherichia coli. Biochemistry 28, 386883875.
Miller D. and McLennan A. G. (1986) Changes in intracellular levels of Ap,A and Ap,A in cysts and larvae of Artemia do not correlate with changes in protein synthesis after heat-shock. Nucl. Acids Res. 14, 603 I-6040.
Montero J. M., Garcia-Agindez J. A., Costas M. J., Cameselle J. C., Gunther Sillero M. A. and Sillero A. (1986) Hydrolytic activities on diadenosine 5’,5”‘-P’,P3triphosphate in the 27,000 x g supernatants and precipitates from rat brain, muscle, kidney and liver. Cienc. Biol. Portugal 11, 1-8. Ng K. E. and Orgel L. E. (1987) The action of a water-
soluble carbodiimide
on adenosine-5’-polyphosphates.
Nucl. Acids Res. 15, 3573-3580.
Ogilvie A. and Jakob P. (1983) Diadenosine 5’,5”‘-P1,Pstriphosphate in eukaryotic cells: identification and quantitation. Analyt. Biochem. 134, 382-392.
571
Prescott M. and McLennan A. G. (1990) Synthesis and applications of I-azido photoaffinity analogues of P’,P’-bis(S-adenosyl) triphosphate and P’,P4-bis(Sadenosyl)tetraphosphate. Analyt. Biochem. 184, 33&337. Prescott M.. Milne A. D. and McLennan A. G. (1989a) Enzymes of dinucleoside oligophosphate metabohsm in Artemia cysts and larvae. In Cell and Molecular Biology of Artemia Development (Edited by Warner A. H., MacRae T. H. and Bagshaw J. C.). NATO ASI Series A, Lz$ Sciences, Vol. 174, pp. 223-243. Plenum Press, New York. Prescott M., Milne A. D. and McLennan A. G. (1989b) Characterization of the bis(S-nucleosidyl) tetraphosphate pyrophosphohydrolase from encysted embryos of the brine shrimp Artemia. Biochem. J. 259, 831-838. Sillero A. and Gunther Sillero M. A. (1987) Interconversion of purine nucleotides in Artemia: a review. In Artemia Research and its Applications (Edited by Decleir W., Moens L., Slegers H., Sorgeloos P. and Jaspers E.), Vol. 2, pp. 289-307. Universa Press, Wetteren, Belgium. Sillero M. A. G., Villalba R., Moreno A., Quintanilla M., Lobaton C. D. and Sillero A. (1977) Dinucleosidetriphosphatase from rat liver. Purification and properties. Eur. J. Biochem. 76, 331-337.
Smith P. K., Krohn R. I., Hermanson G. T., Mallia A. K., Gartner F. H., Provenzano M. D., Fujimoto E. K., Goeke N. M., Olson B. J. and Klenk D. C. (1985) Measurement of protein using bicinchoninic acid. Analyt. Biochem. 150, 78&793. Vallejo C. G., Lobaton C. D., Quintanilla M., Sillero A. and Sillero M. A. G. (1976) Dinucleosidetetraphosphatase in rat liver and Artemia salina. Biochem. biophys. Acta 438, 304309.
Warner A. H. (1979) Studies on the biosynthesis and function of dinucleoside polyphosphates in Artemia embryos In Regulation of Macromolecular Synthesis by Low Molecular Weight Mediators (Edited by Koch 6. and Richter D.), pp. 161-177. Academic Press, New York. Warner A. H. (1992) Diguanosine and related nonadenylated polyphosphates. In Ap,A and Other Dinucleoside Polyphosphates (Edited by McLennan A. G.). CRC Press, Boca Raton, Florida. In press. Warner A. H. and Finamore F. J. (1965) Isolation, purification and characterization of P’ P’-diguanosine-5’triphosphate from brine shrimp eggs. Biochim. biophys. Acta 108, 525-530. Zamecnik P. (1983) Diadenosine 5’,5”‘-P’,P4-tetraphosphate (A-4A): its role in cellular metabolism. Analyt. Biochem. 134, l-10.
