Triphosphate- and Deoxyadenosine 5'- Triphosphate-Activated Nucleotidase from Human Malignant Lymphocytes1

Home , AMP deaminase, Deoxyadenosine, Deoxyguanosine

[CANCER RESEARCH 42, 4321-4324, November 1982] 0008-5472/82/0042-OOOOS02.00 Characterization of an Adenosine S'-Triphosphate- and Deoxyadenosine 5'- Triphosphate-activated Nucleotidase from Human Malignant Lymphocytes1

Dennis A. Carson and D. Bruce Wasson

Department of Clinical Research, Scripps Clinic and Research Foundation, La Jolla, California 92037

ABSTRACT dATP (5, 14, 19, 23). Subsequently, ATP levels, and indeed the total intracellular pools of adenine nucleotides, slowly de The kinetic properties of a soluble, magnesium-dependent cline (1,5,19). The individual enzymes potentially catalyzing 5'-nucleotidase from human malignant lymphocytes have been nucleotide degradation in lymphocytes with elevated dATP determined. The partially purified enzyme is distinct from levels have not been characterized. Recently, an ATP- and plasma membrane-associated 5'-nucleotidase and is free of dATP-stimulated IMP-dephosphorylating activity was de nonspecific phosphatase activity. Among purine ribonucleo- scribed in crude extracts of a human T-lymphoblastoid cell line tides, it reacted efficiently with inosine 5'-monophosphate and guanosine 5'-monophosphate and to a lesser degree with (1). deoxyguanosine 5'-monophosphate. Adenosine 5'-monophos- In the present investigations, we have partially purified from phate and deoxyadenosine 5'-monophosphate were 30-fold malignant human lymphocytes a soluble nucleotidase with properties analogous to the rat and chicken liver enzymes. The less efficient substrates. Increasing concentrations of adenosine S'-triphosphate and deoxyadenosine 5'-triphosphate from kinetics of the enzyme has been determined, with special 0 to 3 rriM enhanced 5'-nucleotidase activity up to 7-fold. reference to the effects of adenine deoxynucleotides as sub Guanosine 5'-triphosphate and deoxyguanosine 5'-triphos- strates and activators. phate were much less effective enzyme activators, while uridine S'-triphosphate was without effect. Inorganic phosphate in MATERIALS AND METHODS hibited dephosphorylating activity in both adenosine 5'-triphos- Enzyme Purification. Splenic tissue, largely replaced by a well- phate-supplemented and unsupplemented buffer. The activa differentiated lymphocytic lymphoma, was removed during surgery and tion of this 5'-nucleotidase by deoxyadenosine 5'-triphosphate, then frozen at -70Â° for 3 weeks. 5'-Nucleotidase was purified from combined with the relative inability of the enzyme to dephos- the rapidly thawed specimen, following the method developed by Itoh phorylate deoxyadenosine 5'-monophosphate, conceivably (11), except that the second phosphocellulose column was eluted with may contribute to the adenine nucleotide degradation induced a linear gradient of 200 to 800 mw NaCI instead of with ATP and the by deoxyadenosine in normal and malignant lymphocytes. low-ionic strength precipitation step was omitted. When stored as described by Itoh (11), the enzyme was stable at -70Â° for at least 1 month. INTRODUCTION 5'-Nucleotidase Assay. 5'-Nucleotidase activity was determined radiochemically using [8-14C]IMP as substrate and by measuring the Intracellular nucleotide degradation in human cells is highly release of inorganic phosphate from various nucleoside 5'-monophos- regulated (10). However, the exact enzymes catalyzing the phates, exactly as described earlier (4). Unless stated otherwise, the dephosphorylation of purine 5'-monophosphates have not buffer was 100 mw imidazole-HCI (pH 6.5), 50 rnw MgCI2, 500 mw been well characterized. The most abundant human 5'-nucle- NaCI, 0.1% bovine serum albumin, containing varying concentrations otidase [5'-ribonucleotide phosphohydrolase (EC 3.1.3.5)] is of nucleoside 5'-monophosphate, and effectors as indicated. The re on the external surface of the plasma membrane and probably actions were initiated by the addition of 1 to 5 HQenzyme protein and, after 15 to 45 min at 37Â°, were terminated by heating to 100Â° for 1 plays no role in intracellular nucleotide degradation (2, 5, 18, 19). min. For the radiochemical assay, inosine was separated from nucleo Recently, several investigators have described a novel 5'- tides by chromatography on polyethyleneimine-cellulose jn methanol: water (1:1) in the presence of appropriate standards (4). The nucleo- nucleotidase in the cytoplasm of rat liver and chicken liver that is distinct from the plasma membrane enzyme (11 -13, 24-26). sides were visualized under UV, cut out, and counted in a liquid scintillation spectrometer. Enzyme activities are expressed as nmol The activity of the cytosolic enzyme was enhanced by ATP at product per min per mg protein and were linear with protein concentra concentrations that inhibited plasma membrane 5'-nucleotid- tion and with time for the data reported. Protein content was determined ase. Several lines of evidence suggested that in rat liver the by the method of Lowry ef al. (18), using bovine serum albumin as a ATP-stimulated cytoplasmic nucleotidase was a predominant standard. enzyme catalyzing the dephosphorylation of purine nucleotides Materials. All common nucleosides and nucleotides were purchased (24-26). from Sigma Chemical Co. (St. Louis, Mo.). The [8-14C]IMP (50 mCi/ In the presence of an inhibitor of adenosine deaminase, mmol) came from Amersham/Searle Corp. (Arlington Heights, III.) and was purified by high-performance liquid chromatography. ATP and nondividing human lymphocytes exposed to /Â¿Mconcentrations GTP were purified immediately before use by DEAE-cellulose chro of deoxyadenosine in vitro or in vivo progressively accumulate matography, using a 100 to 500 ITIMNaCI gradient in 5 mw potassium phosphate, pH 5.0, and contained less than 2 nmol orthophosphate ' Supported by Grants GM 23200 and CA 31497 from the USPHS. per /imol nucleotide. Other materials were of the highest grade com Received April 26. 1982; accepted July 21, 1982. mercially available.

NOVEMBER 1982 4321

RESULTS

Enzyme Purification. Because the IMP-dephosphorylating activity in the crude lymphocyte extract (5.0 nmol/min/mg protein) represented a mixture of nucleotidases and phosphatases, it was not possible to determine the exact purification dGMP factor or yield of the 5'-nucleotidase. However, the estimated cytosolic nucleotidase activity in lymphocytes ranged from 0.4 to 1.0 nmol/min/mg protein. The final enzyme preparation was free of detectable ÃŸ-glycerophosphate- and ATP-dephosphorylating activities (<0.5 nmol dephosphorylated/min/mg pro AMP tein at a substrate concentration of 40 ITIM)(4). It also lacked measurable adenosine deaminase or purine nucleoside phos- phorylase (3). At an IMP concentration of 3 mM in the standard 4 5 buffer lacking ATP, the specific activity of the partially purified Substrate(mM) enzyme was 1.4 /imol/min/mg protein. The total yield from Chart 2. Relation of substrate concentration to enzyme velocity. The reaction 14.2 starting material was 2.7 mg protein, containing 3.78 conditions were the same as in Chart 1 except that varying concentrations of jumol IMP-dephosphorylating activity per min. IMP, dIMP, QMP, dGMP, and AMP were used. Effect of pH and MgCI2. When assayed in imidazole-HCI buffer with IMP as substrate, the pH optimum of the enzyme was 6.4 (Chart 1). At neutral pH, 5'-nucleotidase activity was approximately 50% maximal. The enzyme activity was not inhibited by 10 mM tartrate or 5 mM jS-glycerophosphate. Ten mM fluoride inhibited enzyme activity by 65%. In the absence of added MgCI2, no IMP- dephosphorylating activity was measurable. Il Substrate Specificity. The lymphocyte 5'-nucleotidase dephosphorylated all 8 purine and pyrimidine 5'-monophosphates studied (Table 1). Over the range of concentrations tested,

0.4

10 20 30 40 50 Substrate (mM) Chart 3. Relation of substrate concentration to enzyme velocity and effect of ATP. The reaction conditions were as described in Chart 1 except that varying concentrations of AMP and dAMP were used, insert, effect of 3 mM ATP upon 0.1 the rate of the reactions.

substrate-velocity plots were hyperbolic with the preferred 5.5 6 6.5 7.5 substrates IMP, dIMP, GMP, dGMP (Chart 2), and CMP (not pH shown). Linear regression analysis of Eadie-Hofstee plots of Chart 1. Effect of pH on 5'-nucleotidase activity. Enzyme activity was assayed the data by the method of least squares yielded r* values in 100 mM imidazole-HCI containing 50 IÃ•IMMgCI.. 0.1% bovine serum albumin, 2 mM IMP, and 1.5 Â«gprotein in a total volume of 100 Â»IProduct formation was >0.90 in each case. determined after 45 min at 37" by the inorganic phosphate method. With the inefficient substrates AMP and dAMP (Chart 3), Table 1 substrate-velocity plots were approximately linear at concen Substrate specificity of the Â¿'-nucleotidase trations up to 12 mM. Maximal enzyme activity was approached for each nucleotide, product formation was determined by the release of only at nucleotide concentrations of 30 to 50 mM, far above inorganic phosphate at substrate concentrations from 0.1 to 100 mM. The any achievable physiological range. With the latter 2 sub reaction buffer lacked ATP or inorganic phosphate. The V,,,,. values for the respective nucleotides are shown relative to IMP, for which the specific activity strates, the nucleotide concentration at which dephosphorylat- was 1.4 ;imol/min/ mg protein. The S â€¢¿isthe substrate concentration at which ing activity was half-maximal (S0 5) O 5) was used to estimate the velocity was 0.5 V,.,,. and is reported for the inefficient substrates AMP, dAMP, and dCMP, which did not follow Michaelis-Menten kinetics. The ratios of overall substrate efficiency by the ratio Vmax/S05. For the other V,â€žâ€ž/Kâ€žâ€žorV,,,,,/S.,... shown in parentheses, yielded estimates of the overall substrates, the ratio Vmax/Kmwas used (16). As shown in Table efficiency of the enzyme toward the individual substrates and are compared to 1, the enzyme reacted most efficiently with IMP, dIMP, and IMP. GMP; dGMP was dephosphorylated 5-fold less efficiently. or Sos When compared to IMP and dIMP, AMP and dAMP were 30- SubstrateIMPdIMPAMPdAMPGMPdGMPCMPdCMPRelativeVâ„¢,1.001.060.910.901.050.720.520.57Km(mM)0.570.62>15>150.642.155.98>15Relativeefficiency10.98<0.03<0.030.940.190.05<0.002(1.75)(1.71)Â«0.06)Â«0.06)(1 fold less efficient substrates. Effect of ATP, dATP, GTP, dGTP, and Inorganic Phos phate. With 200 juM IMP as substrate, increasing concentra tions of ATP or dATP from 0 to 3 mM increased dephosphory- .64)(0.33)(0.087)Â«0.04) lating activity up to 7-fold (Chart 4). GTP and dGTP were poor but detectable activators of the enzyme, while UTP barely augmented IMP dephosphorylation. ATP and dATP also en-

4322 CANCER RESEARCH VOL. 42

0.6 developed by Itoh (11) for the rat liver cytosolic nucleotidase effectively separated the human enzyme from other dephos- 0.5 phorylating activities. Among purine nucleotides, the preferred substrates for the 0.4 enzyme were IMP, dIMP, GMP, and dGMP. When compared to IMP, both AMP and dAMP were 30-fold less efficient sub 0.3 strates. At optimal pH and magnesium concentrations, the net

GTP dephosphorylating activity of the enzyme was modulated pri 0.2 dorp marily by the relative concentrations of substrate, ATP, and inorganic phosphate. 0.1 Previous kinetic analyses of mammalian cytosolic nucleoti dases did not examine in detail the effects of dATP and dGTP on IMP dephosphorylation, although dATP was reported to 0123 stimulate the chicken liver enzyme (12). We found that dATP NucleosidetriphosphateImMl was as effective as ATP in enhancing IMP dephosphorylation Chart 4. Activation of 5'-nucleotidase by ATP, dATP, GTP. dGTP, and UTP. The reaction mixtures contained 200 /IM [8-'4C]IMP (0.03 jiCi) and 0.5 fig protein by the 5'-nucleotidase from human malignant lymphocytes. in 100 /il of the standard buffer. After 20 min. inosine was separated from IMP by Indeed, even 100 /Ã•MdATP significantly augmented IMP-de thin-layer chromatography. Point, mean of 2 experiments, each performed in phosphorylating activity in a buffer containing 3 rriM ATP and duplicate: bars, S.E. 2 mM phosphate. By comparison, GTP and dGTP were far less 0.1 potent activators of the enzyme. As noted earlier, although dATP activated the nucleotidase, 0.6. dAMP was an inefficient substrate for the enzyme. In these 2 aspects, the kinetics of the nucleotidase strongly resemble I 0.4 human erythrocyte adenylate deaminase (1, 17). The latter enzyme also reacts inefficiently with dAMP but is stimulated ÃŽ0.2 markedly by both ATP and dATP. The 2 enzymes, adenylate deaminase and 5'-nucleotidase, are probably pivotal in regu 01234 lating the overall degradation of adenine ribonucleotides in PhosphateImMl human cells (1, 6, 10, 22, 24-26). The activation of both Chart 5. Inhibition of 5'-nucleotidase by inorganic phosphate. The reactions enzymes by dATP, combined with their relative inability to contained 200 fiM [8-'4C]IMP (0.03 /iCi) and 1.5 jig protein in a standard reaction metabolize dAMP, could contribute to adenine nucleotide deg buffer containing 3 ITIM ATP (â€¢)or lacking ATP (O). Product formation was assayed radiochemically. radation in adenosine deaminase-inhibited human lymphocytes exposed to deoxyadenosine (1,5,19). Under such conditions, hanced the rate of dephosphorylation of AMP and dAMP (Chart Â¡ntracellular dATP levels may reach nearly 30% of the total 3, inset). However, even in a buffer containing 3 ITIMATP, the ATP content (3). rate of dephosphorylation of AMP and dAMP was an order of In preliminary experiments, we have also purified and char magnitude less than the rate for IMP or GMP. acterized an ATP-stimulated nucleotidase from extracts of nor Inorganic phosphate at concentrations from 0 to 4 mM sub mal resting human peripheral blood lymphocytes. However, no stantially reduced IMP-dephosphorylating activity. The inhibi detectable ATP-regulated nucleotidase has been recovered tion was significant in both ATP-supplemented and unsupple- from human erythrocytes. The lack of an ATP- or dATP-acti mented buffers (Chart 5). vated nucleotidase in human erythrocytes may explain why Chart 6 shows the effects of increasing concentrations of dATP, ATP, and dGTP upon enzyme activity in a buffer con taining 0.2 rriM IMP, 3 HIM ATP, 0.5 rriM GTP, and 2 ITIM inorganic phosphate. Notably, as little as 100 JUMdATP in creased IMP dephosphorylation significantly. Those concentra tions of dATP are achieved routinely in normal or malignant lymphocytes incubated with deoxyadenosine and the adenosine deaminase inhibitor deoxycoformycin, both in vitro and in vivo (4, 19). ÃŽ .28

DISCUSSION

This study presents evidence demonstrating that human malignant lymphocytes contain a soluble, magnesium-depend ent nucleotidase whose activity is enhanced by ATP and is inhibited by inorganic phosphate. Our earlier attempts to char 0 100 200 400 600 800 1000 acterize cytosolic nucleotidases in crude extracts of human NucleosideTriphosphateluMi tissues were hampered by interference from the more abundant Chart 6. Effect of dATP, ATP. and dGTP upon 5'-nucleotidase activity in a plasma membrane-associated 5'-nucleotidase, as well as from buffer containing 3 mM ATP, 0.5 mM GTP, 2 mM phosphate, and 0.2 mM IMP. nonspecific phosphatases. Fortunately, the purification method Point, mean for 2 experiments, each performed in duplicate; bars, S.E.

NOVEMBER 1982 4323

dATP rises to higher levels in erythrocytes than lymphocytes, 6. Chapman, A. G . and Atkinson, D. E. Stabilization of adenylate energy before ATP degradation ultimately ensues (4, 19, 23). charge by the adenylate deaminase reaction. J. Biol. Chem., 248: 8309- Since neither AMP deaminase nor the ATP-stimulated 5'- 8312, 1973. 7. Fox, R. M., Tripp, E. H., Paddington, S. K., and Tattersall, M. H. N. Sensitivity nucleotidase effectively degrade dAMP, other dAMP-dephos- of human null lymphocytes to deoxynucleosides. Cancer Res., 40: 3383- 3388, 1980. phorylating enzyme(s) must exist in the cytoplasm of human 8. Fritzson, P. Kinetic analysis of a rat-liver nucleotidase activated by deoxyri- cells. In support of this notion, the cumulative evidence of bonucleic acid and constituents. Eur. J. Biochem., 38: 408-415. 1973. several laboratories indicates that mammalian cells actually 9. Fritzson, P. Regulation of nucleotidase activities in animal tissues. Adv. contain at least 3 separate nucleotidases: (a) the soluble nu- Enzyme Regul., 76:43-61, 1978. 10. Henderson, J. F., Bagnara, A. S., Crabtree. G. W., Lomax, C. A., Shantz. G. cleotidase described herein that reacts preferentially with IMP D., and Snyder, F. F. Regulation of enzymes of purine metabolism in intact and GMP and is stimulated by ATP (11-13, 24-26); (b) a tumor cells. Adv. Enzyme Regul., ÃŽ3:37-64, 1975. 11. Itoh, R. Purification and properties of cytosol 5'-nucleotidase from rat liver. soluble nucleotidase that reacts preferentially with deoxynucle- Biochim. Biophys. Acta, 657: 402-410, 1981. otides, including dAMP, independent of ATP levels (4, 8, 9); 12. Itoh, R., Mitsui, A., and Tsushima, K. 5'-Nucleotidase from chicken liver. and (c) a plasma membrane-associated nucleotidase that effi Biochim. Biophys. Acta, 746: 151-159, 1967. 13. Itoh, R., Usami, C., Nishino, T.. and Tsushima, K. Kinetic properties of ciently dephosphorylates AMP and dAMP and which is in cytosol 5'-nucleotidase from chicken liver. Biochim. Biophys. Acta, 526: hibited by ATP (2, 4). Most workers now agree that the plasma 154-162, 1978. 14. Kefford, P. F., and Fox, R. M. Purine deoxyribonucleoside toxicity in non- membrane enzyme plays no significant role in intracellular dividing human lymphoid cells. Cancer Res., 42: 324-330, 1982. purine nucleotide degradation (4, 7, 20, 21). The markedly 15. Koshland, D. E., Jr., Nemethy. G.. and Filmer. D. Comparison of experimental different substrate specificities of the 2 described intracellular binding data and theoretical models in proteins containing subunits. Bio chemistry, 5: 365-385, 1966. nucleotidases is consistent with the view that lymphocytes 16. Krenitsky, T. A., Tuttle, J. V.. Koszalka, G. W., Chen, I. S., Beacham, C. M., regulate purine ribonucleotide and deoxyribonucleotide catab- Ill, Rideout. J. L., and Elion, G. B. Deoxycytidine kinase from calf thymus: substrate and inhibitor specificity. J. Biol. Chem., 25Ã•: 4055-4060, 1976. olism via different mechanisms. Indeed, considering the diver 17. (Jan, C-H., and Harkness, D. R. The kinetic properties of adenylate deami gent functions of the 2 classes of purine nucleotides for cell nase from human erythrocytes. Biochim. Biophys. Acta, 341: 27-40, 1974. growth and metabolism, this conclusion is not unexpected. 18. Lowry, O., Rosebrough, N. J., Farr, A., and Randall. R. J. Protein measure ment with the Folin phenol reagent. J. Biol. Chem., 793: 265-271, 1951. 19. Mitchell, B. S., Koller, C. A., and Heyn, R. Inhibition of adenosine deaminase results in cytotoxicity to T lymphoblasts in vivo. Blood, 56: 556-559, 1980. REFERENCES 20. Newby, A. C. Role of adenosine deaminase ecto-(5'-nucleotidase) and ecto- (nonspecific phosphatase) in cyanide-induced adenosine monophosphate 1. Bagnara. A. S., and Hershfield. M. S. Mechanism of deoxyadenosine-in- catabolism in rat polymorphonuclear leucocytes. Biochem. J., 786: 907- duced catabolism of adenine ribonucleotides in adenosine deaminase in 918, 1980. hibited human T lymphoblastoid cells. Proc. Nati. Acad. Sei. U. S. A., 79: 21. Newby, A. C., and Holmquist, C. A. Adenosine production inside rat poly 2673-2677, 1982. morphonuclear leucocytes. Biochem. J., 200: 399-403, 1981. 2. Burger, R. M., and Lowenstein, J. M. 5'-Nucleotidase from smooth muscle 22. Sauer, L. A. Control of adenosine monophosphate catabolism in mouse of small intestine and from brain: inhibition by nucleotides. Biochemistry, 4: ascites tumor cells. Cancer Res., 38: 1057-1063, 1978. 2362-2366, 1975. 23. Siaw, M. F. E., Mitchell, B. S., Koller, C. A., Coleman, M. H., and Hutton, J. 3. Carson, D. A., Kaye, J., and Seegmiller, J. E. Differential sensitivity of human J. ATP depletion as a consequence of adenosine deaminase inhibition in leukemic T cell lines and B cell lines to growth inhibition by deoxyadenosine. man. Proc. Nati. Acad. Sei. U. S. A., 77: 6157-6161, 1980. J. Immunol., 121: 1726-1731, 1978. 24. Van den Berghe, G., Bronfman, M., Vannester. R., and Heis, H-G. The 4. Carson, D. A., Kaye, J., and Wasson, D. B. The potential importance of mechanism of adenosine triphosphate depletion in the liver after a load of soluble deoxynucleotidase activity in mediating deoxyadenosine toxicity in fructose. Biochem. J., 762: 601-609, 1977. human lymphoblasts. J. Immunol., 726: 348-352, 1981. 25. Van den Berghe, G., Van Pottelsberghe, C., and Hers, H-G. A kinetic study 5. Carson, D. A., Lakow. E., Wasson, D. B., and Kamatani, N. Possible of the soluble 5'-nucleotidase of rat liver. Biochem. J., 762:611 -616,1977. metabolic basis for the different immunodeficient states associated with 26. Vincent. M-F., Van den Berghe, G.. and Heis, H-G. The pathway of adenine genetic deficiencies of adenosine deaminase and purine nucleoside phos- nucleotide degradation and its control in isolated rat hepatocytes subjected phorylase. Proc. Nati. Acad. Sei. U. S. A., 12: 3848-3852, 1982. to anoxia. Biochem. J., 202: 117-123, 1982.

4324 CANCER RESEARCH VOL. 42

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1982 American Association for Cancer Research. Characterization of an Adenosine 5′-Triphosphate- and Deoxyadenosine 5 ′-Triphosphate-activated Nucleotidase from Human Malignant Lymphocytes

Dennis A. Carson and D. Bruce Wasson

Cancer Res 1982;42:4321-4324.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/42/11/4321

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/42/11/4321. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.