- Email: [email protected]
Modulation of macrophage superoxide release by purine metabolism
Life Sciences, Vol. 32, pp. 1359-1362 Printed in the U.S.A.
Pergamon Pres~
MODULATION OF MACROPHAGE SUPEROXIDE RELEASE BY PURINE METABOLISM George L. Tritsch* and Paul W. Niswander Department of Surgical Oncology, RoswelI Park Memorial Institute, New York State Department of Health Buffalo, N.Y. 14263 (Received in final form December 13, 1982) Summa ry Metabolic flux through the purine salvage pathway appears to modulate superoxide secretion by elicited macrophages. Exogenous adenosine, the first substrate of this pathway, stimulates superoxide secretion, and Allopurinol, a specific inhibitor of xanthine oxidase, inhibits superoxide secretion. The effects of these agents are additive since it was possible for each to neutralize the effects of the other when given in combination. In these experiments, the purine salvage pathway was responsible for over ten times the superoxide production attributable to the NADPH oxidase system. During macrophage membrane perturbation or phagocytosis, adenosine deaminase (ADA), the first enzyme of the purine salvage pathway, increases in activity in direct proportion to superoxide (OF) secretion (1,2,3). These findings suggested that the basis for this correlation was that ADA regulated the metabolic flux through xanthine oxidase which produces OF during conversion of its substrates into uric acid. Because the macrophage exercises broad influence on the regulation of normal immune function (4), we further proposed that our finding accounts for the need of adequate ADA activity for normal immune function, and may form the basis of the association of ADA deficiency with immunodeficiency (5). Herein we extend these observations by demonstrating the effects of inhibition and stimulation of metabolic flux through the purine salvage pathway with exogenous Allopurinol (Hydroxypyrazolo(3,4-D)pyrimidine) and adenosine, respectively. We used Tuftsin (L-Threonyl-L-Lysyl-L-Prolyl-L-Arginine) as the acute membrane perturbant because of its normal endogenous occurrence within the Fc ~ortion of its parent cytophilic gamma globulin leukokinin (6). Because O~ production by the NADPH oxidase system has been studied extensively (7), we also evaluated this enzyme activity in our experiments. Under our experimental conditions , ~he data indicate that xanthine oxidase provides over ten times the O~ attributable to oxidases for reduced pyridine nucleotides, and that ADA provides the substrate for the reaction, i.e., hypoxanthine. M a t e r i a l s and Methods We have chosen to use p e r i t o n e a l exudate c e l l s (PEC) e l l i c i t e d w i t h t h i o g l y c o ] l a t e in mice. In view of the m e t a b o l i c d i f f e r e n c e s between resident macrophages, c e l l s a c t i v a t e d by i n j e c t i o n w i t h various organisms, and c e l l s e l l i c i t e d w i t h caseinate or o t h e r agents ( 8 , 9 ) , the f i n d i n g s herein presented need not be expected, a p r i o r i , to apply to these o t h e r systems. * To whom correspondence should be addressed 0024-3205/83/121359-04503.00/0 Copyright (c) 1983 Pergamon Press Ltd.
1360
Superoxide Release and Purine Metabolism
Vol. 32, No. 12, 1983
Thioglycollate elicited PEC were lavaged from ICR Swiss mice of the Roswell Park colony 5 days after i.p. injection of 3 ml brewer's thioglycollate broth (Becton Dickinson & Co., Cockeysville, ND) as previously described (1,2). Four tubes were prepared (see Table I), each of which contained sufficient balanced salt solution (2) to make a final volume of 3.0 ml, 0.30 ml of the lavaged PEC (10.1 x 106 PEC/ml), i.e., 105 PEC/ml during the incubation, 0.30 ml of 1.2 mM cytochrome c (Type IEI from horse heart, Sigma Chemical Co., St. Louis, MO), and the reaction was started by the addition of 0.45 ml of 2.5 x IO-~M Tuftsin (Sigma; used without further purification). This resulted in a concentration of 375 nM Tuftsin during the incubation, a level producing near maximum O~ secretion with Tuftsin with I0 s PEC/ml (10), i.e., 18 nmole O~/min/IO 6 cells, a number in good agreement with that observed in Karnovsky's laboratory (7) for phagocytosis (133±35 nmole O~ per minute per 107 cells). Samples of 1.0 ml were removed after, O, 3 and 6 minutes at 37 ° , immediately centrifuged at 8,000 x g in an Eppendorf ultracentrifuge, and the supe~natant and residue chilled at 0 °. The absorbance of the supernatant was determined at 550 nm with the zero time supernatant in the reference beam, and O~ produced calculated from the concentration of cytochrome c reduced with the equation AE=2.1xlO~M-~cm -I. The sedimented cells were washed once with Ca ++ and Mg ++ free balanced salt solution (2), suspended in 0.20 ml of 0.25 M sucrose adjusted to ph 7.3, and frozen. After freezing and thawing twice, 0.1 ml was used for assay of ADA activity and 0.I ml for assay of NADPH oxidase activity. To demonstrate the effects of adenosine and Allopurinol, the PEC were incubated for 15 minutes at 37 ° before Tuftsin addition with 0.30 ml of I mM adenosine, 0.30 ml of 0.30 mM Allopurinol, and a combination of the two, respectively (tubes 2, 3 and 4, respectively, in Table I). The choice of the 15 minute incubation and the adenosine concentration were arbitrary, but the Allopurinol level selected was based on the work of Tubaro et al (11). Enzyme Assays - In the spectrophotometric assays to be described, the substrates show appreciable absorption at the wave lengths of the assay. Thus, low substrate concentrations were selected (about equal to Km) so that relatively small rates of change in absorbance could be detected. Had high substrate concentrations been used to assure saturation of the enzymes with substrate, low levels of enzyme activity may have been missed. To certify linearity of the reaction, it was desired to use assays based on continuous recording of changes in substrate or product concentrations, rather than analyses of discrete samples. Selection of pH and buffer was based on the cited literature. ADA - The assay of Kalckar (12) as described by Kaplan (13) was used with the modification suggested by Beutler (14) for use of Tris instead of phosphate buffer because our crude cell lysates are expected to contain purine nucleoside phosphorylase activity, which, in the presence of phosphate would convert, by phosphorolysis, inosine as quickly as it is formed by ADA into hypoxanthine and ribose-l-phosphate. The decrease in absorbance at 265 nm which accompanies conversion of adenosine into inosine is recorded in a Cary model 14 spectrophotometer in quartz cells of I cm light path and maintained at 37 ° . The O.1 ml aliquot of the PEC-sucrose mixture is added to 0.65 ml H20, equilibrated at 37 ° and rapidly added to a mixture of O.1 ml of 1.0 mM adenosine and 0.15 ml of 1.O6 M Tris - HCl buffer of pH 7.3. Micromole adenosine deaminated per minute per 106 cells is calculated by use of AE=8,0OO M-Icm -I (12). NADPH Oxidase - The following was adapted from the literature (15-17) for estimating the rate of change in absorbance at 340 nm which accompanies oxidation of NADPH: Phosphate buffer, 50 nM, pH 5.5 which is O.17 mM in MnCI2 and 50 uM in NADPH, and the 0.1 ml PEC-sucrose mixture, all in a 1.O ml total volume, were incubated at 37 ° in a spectrophotometer cuvet. Micromole NADPH oxidized per minute per 106 cells is calculated by use of the equation AE=6,200 M-icm -I
Vol. 32, No. 12, 1983
Superoxide Release and Purine Metabolism
1361
Table I Effects of Adenosine and Allopurinol on Superoxide Production Exp. No.
Addition During Preincubation
0~
ADA**
Intracellular NADPH Oxidase***
1.59
4.3
<0.32
Extracellular*
I
none
2
Adenosine
3.70
3.8
<0.32
3
Allopurinol
0.70
5.7
<0.32
4
Adenosine + Allopurinol
1.40
4.2
<0.32
* ** ***
nmole O~/min/106 cells nmole adenosine turned over per min per 106 cells nmole NADPH turned over per min per 10 ~ cells
Each mixture was 3.0 ml in total volume, contained 101,000 PEC/ml, and was 120 uM in cytochrome c. After 15 min preincubation at 37 ° , the reactions were started by the addition of Tuftsin as described in the text. (15). Because of the considerable evidence (18,19) that in the presence of Mn ++ some nonenzymatic NADPH oxidation occurs via an O~ dependent free radical chain reaction, the values herein reported for NADPH oxidase activity are probably greater than the true in vivo activity. Results Table I shows the results of each experiment. Omission of Tuftsin (data not shown) resulted in no OF release. Incubation with adenosine for 15 minutes before Tuftsin stimulation resulted in more than doubling of O~ release (1.59 to 3.70). Allopurinol approximately halved O~ release (1.59 to 0.7), and the particular combination of adenosine and Allopurinol used essentially neutralized the effects of each when used individually. ADA activities are shown in the table; differences between values were negligible, especially when compared to changes in 0~. No NADPH oxidase activity could be detected in any of the samples; more than I0 s ceils are generally used to detect this enzyme activity (15-17). When l~sates of 107 cells were used, activity could readily be demonstrated. Thus O~ production attributable to NADPH oxidase could be no more than 10 percent of that attributable to ADA. Discussion Our findings demonstrate metabolic modulation of O~ production by macrophages. Of the large number of possibilities, we have selected adenosine, the first substrate of the purine salvage pathway, to stimulate metabolic flux through this pathway, and Allopurinol, a recognized agent used in clinical medicine to selectively inhibit xanthine oxidase. The effects of these agents were additive since it was possible for each to neutralize the effects of the other when given in combination. These results are qualitatively similar to those obtained in vivo with Allopurinol and xanthlne in experiments where phagocytosis was the end point (11). Combinations of stimulants and inhibitors were not tested in these experiments, and it was not possible to compare relative contributions of NADPH oxidases and purine salvage pathway enzymes to O~ production. The importance of the purine salvage pathway in O~ production is furthermore implied by the findings of increased uric acid excretion during
1362
Superoxide Release and Purine Metaboli3m
Vol. 32, No. 12, 1983
phagocytosis (20) and the increased ADA titer in caseinate-elicited macrophages (21) and during macrophage chemotaxis (22). The experiments herein described demonstrate that the purine salvage pathway is responsible for over ten times the 03 production attributable to the NADPH oxidase system. Thus, oxidases for reduced pyrimidine nucleotides are not the key enzymes in this context in these cells. The ability of the xanthine oxidase reactions to provide most of the OF in these cells appears to be related to ADA activity regulation of substrate flux towards hypoxanthine formation. The sensitivity of the enzymes of the purine salvage pathway to inhibition by a number of agents, especially many of those used in chemotherapy of neoplastic disease, suggests that impairment of macrophage function by these agents may be part of the basis for the immunosuppressive side effects of several of such therapies. References I. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
G.L. TRITSCH and P.W. NISWANDER, J. Medicine 11393-399 (1980). G.L. TRITSCH and P.W. NISWANDER, Immunol. Commu---n. 10 I-8 (1981). G.L. TRITSCH and P.W. NISWANDER, Biochem. Med. 26 I--~r5-190 (1981). A.S. ROSENTHAL, N. Eng]. J. Med. 3031153-1156 -(-]-980). E.R. GIBLETT, J.E. ANDERSON, F. COHEN, B. POLLARA and H.J. MEUWISSEN, Lancet ii 1067-1069 (1972). V.A. NAJJAR, Adv. Enzymol. 41129-178 (1974). J.A. BADWEY and M.L. KARNOVSK---Y, Ann. Rev. Biochem. 4_99695-726 (1980) M.L. KARNOVSKY, J.K. LAZDINS, J. Immunol. 121809-813 (1978). M.L. KARNOVSKY, J.K. LAZDINS, D. DR~TH and A. HARPER, Ann. N.Y. Acad. Sci. 256266-274 (1975). G.L. TRITSCH and P.W. NISWANDER, Molec. Cell. Biochem. in press. E. TUBARO, B. LOTTI, C. SANTIANGELI and G. CAVALLO, Biochem. Pharmacol. 293018-3020 (1980). H.M. KALCKAR, J. Biol. Chem. 167445-459 (1954). N.O. KJ~PLAN, Methods in Enzymol. 2473-474 (1955). E. BEUTLER, Red Cell Metabolism, A Manual of Biochem. Methods, 2nd ed., p. 86, Grune & Stratton, New York (1975). G.Y.N. IYER and J.H. QUASTEL, Canad. J. Biochem. Physiol. 41427-434 (1963). P. PATRIARCA, R. CRAMER, S. MONCALVO, F. ROSSI and ROMEO, Arch. Biochem. Biophys. 145255-262 (1971). K. TAKANAKA and P.J. O'BRIEN, Arch. Biochem. Biophys. 169436-442 (1975). J.T. CURNUTTE, M.L. KARNOVSKY and B.M. BABIOR, J. Clin. Invest. 5710591067 (1976)~ P. PATRIARCA, P. DRI, K. KAKINUMA, F. TEDESCO and F. ROSSI, Biochim. Biophys. Acta 385380-386 (1975). T. CHAN, Proc. Natl. Acad. Sci. USA 76925-929 (1979). R.J. SOBERMAN and M.L. KARNOVSKY, J. Expl. Med. 152241-246 (1980). R. SNYDERMAN, M.C. PIKE, D.G. FISCHER and H.S. KOREN, J. Immunol. 119 2060-2066 (1977).
Pergamon Pres~
MODULATION OF MACROPHAGE SUPEROXIDE RELEASE BY PURINE METABOLISM George L. Tritsch* and Paul W. Niswander Department of Surgical Oncology, RoswelI Park Memorial Institute, New York State Department of Health Buffalo, N.Y. 14263 (Received in final form December 13, 1982) Summa ry Metabolic flux through the purine salvage pathway appears to modulate superoxide secretion by elicited macrophages. Exogenous adenosine, the first substrate of this pathway, stimulates superoxide secretion, and Allopurinol, a specific inhibitor of xanthine oxidase, inhibits superoxide secretion. The effects of these agents are additive since it was possible for each to neutralize the effects of the other when given in combination. In these experiments, the purine salvage pathway was responsible for over ten times the superoxide production attributable to the NADPH oxidase system. During macrophage membrane perturbation or phagocytosis, adenosine deaminase (ADA), the first enzyme of the purine salvage pathway, increases in activity in direct proportion to superoxide (OF) secretion (1,2,3). These findings suggested that the basis for this correlation was that ADA regulated the metabolic flux through xanthine oxidase which produces OF during conversion of its substrates into uric acid. Because the macrophage exercises broad influence on the regulation of normal immune function (4), we further proposed that our finding accounts for the need of adequate ADA activity for normal immune function, and may form the basis of the association of ADA deficiency with immunodeficiency (5). Herein we extend these observations by demonstrating the effects of inhibition and stimulation of metabolic flux through the purine salvage pathway with exogenous Allopurinol (Hydroxypyrazolo(3,4-D)pyrimidine) and adenosine, respectively. We used Tuftsin (L-Threonyl-L-Lysyl-L-Prolyl-L-Arginine) as the acute membrane perturbant because of its normal endogenous occurrence within the Fc ~ortion of its parent cytophilic gamma globulin leukokinin (6). Because O~ production by the NADPH oxidase system has been studied extensively (7), we also evaluated this enzyme activity in our experiments. Under our experimental conditions , ~he data indicate that xanthine oxidase provides over ten times the O~ attributable to oxidases for reduced pyridine nucleotides, and that ADA provides the substrate for the reaction, i.e., hypoxanthine. M a t e r i a l s and Methods We have chosen to use p e r i t o n e a l exudate c e l l s (PEC) e l l i c i t e d w i t h t h i o g l y c o ] l a t e in mice. In view of the m e t a b o l i c d i f f e r e n c e s between resident macrophages, c e l l s a c t i v a t e d by i n j e c t i o n w i t h various organisms, and c e l l s e l l i c i t e d w i t h caseinate or o t h e r agents ( 8 , 9 ) , the f i n d i n g s herein presented need not be expected, a p r i o r i , to apply to these o t h e r systems. * To whom correspondence should be addressed 0024-3205/83/121359-04503.00/0 Copyright (c) 1983 Pergamon Press Ltd.
1360
Superoxide Release and Purine Metabolism
Vol. 32, No. 12, 1983
Thioglycollate elicited PEC were lavaged from ICR Swiss mice of the Roswell Park colony 5 days after i.p. injection of 3 ml brewer's thioglycollate broth (Becton Dickinson & Co., Cockeysville, ND) as previously described (1,2). Four tubes were prepared (see Table I), each of which contained sufficient balanced salt solution (2) to make a final volume of 3.0 ml, 0.30 ml of the lavaged PEC (10.1 x 106 PEC/ml), i.e., 105 PEC/ml during the incubation, 0.30 ml of 1.2 mM cytochrome c (Type IEI from horse heart, Sigma Chemical Co., St. Louis, MO), and the reaction was started by the addition of 0.45 ml of 2.5 x IO-~M Tuftsin (Sigma; used without further purification). This resulted in a concentration of 375 nM Tuftsin during the incubation, a level producing near maximum O~ secretion with Tuftsin with I0 s PEC/ml (10), i.e., 18 nmole O~/min/IO 6 cells, a number in good agreement with that observed in Karnovsky's laboratory (7) for phagocytosis (133±35 nmole O~ per minute per 107 cells). Samples of 1.0 ml were removed after, O, 3 and 6 minutes at 37 ° , immediately centrifuged at 8,000 x g in an Eppendorf ultracentrifuge, and the supe~natant and residue chilled at 0 °. The absorbance of the supernatant was determined at 550 nm with the zero time supernatant in the reference beam, and O~ produced calculated from the concentration of cytochrome c reduced with the equation AE=2.1xlO~M-~cm -I. The sedimented cells were washed once with Ca ++ and Mg ++ free balanced salt solution (2), suspended in 0.20 ml of 0.25 M sucrose adjusted to ph 7.3, and frozen. After freezing and thawing twice, 0.1 ml was used for assay of ADA activity and 0.I ml for assay of NADPH oxidase activity. To demonstrate the effects of adenosine and Allopurinol, the PEC were incubated for 15 minutes at 37 ° before Tuftsin addition with 0.30 ml of I mM adenosine, 0.30 ml of 0.30 mM Allopurinol, and a combination of the two, respectively (tubes 2, 3 and 4, respectively, in Table I). The choice of the 15 minute incubation and the adenosine concentration were arbitrary, but the Allopurinol level selected was based on the work of Tubaro et al (11). Enzyme Assays - In the spectrophotometric assays to be described, the substrates show appreciable absorption at the wave lengths of the assay. Thus, low substrate concentrations were selected (about equal to Km) so that relatively small rates of change in absorbance could be detected. Had high substrate concentrations been used to assure saturation of the enzymes with substrate, low levels of enzyme activity may have been missed. To certify linearity of the reaction, it was desired to use assays based on continuous recording of changes in substrate or product concentrations, rather than analyses of discrete samples. Selection of pH and buffer was based on the cited literature. ADA - The assay of Kalckar (12) as described by Kaplan (13) was used with the modification suggested by Beutler (14) for use of Tris instead of phosphate buffer because our crude cell lysates are expected to contain purine nucleoside phosphorylase activity, which, in the presence of phosphate would convert, by phosphorolysis, inosine as quickly as it is formed by ADA into hypoxanthine and ribose-l-phosphate. The decrease in absorbance at 265 nm which accompanies conversion of adenosine into inosine is recorded in a Cary model 14 spectrophotometer in quartz cells of I cm light path and maintained at 37 ° . The O.1 ml aliquot of the PEC-sucrose mixture is added to 0.65 ml H20, equilibrated at 37 ° and rapidly added to a mixture of O.1 ml of 1.0 mM adenosine and 0.15 ml of 1.O6 M Tris - HCl buffer of pH 7.3. Micromole adenosine deaminated per minute per 106 cells is calculated by use of AE=8,0OO M-Icm -I (12). NADPH Oxidase - The following was adapted from the literature (15-17) for estimating the rate of change in absorbance at 340 nm which accompanies oxidation of NADPH: Phosphate buffer, 50 nM, pH 5.5 which is O.17 mM in MnCI2 and 50 uM in NADPH, and the 0.1 ml PEC-sucrose mixture, all in a 1.O ml total volume, were incubated at 37 ° in a spectrophotometer cuvet. Micromole NADPH oxidized per minute per 106 cells is calculated by use of the equation AE=6,200 M-icm -I
Vol. 32, No. 12, 1983
Superoxide Release and Purine Metabolism
1361
Table I Effects of Adenosine and Allopurinol on Superoxide Production Exp. No.
Addition During Preincubation
0~
ADA**
Intracellular NADPH Oxidase***
1.59
4.3
<0.32
Extracellular*
I
none
2
Adenosine
3.70
3.8
<0.32
3
Allopurinol
0.70
5.7
<0.32
4
Adenosine + Allopurinol
1.40
4.2
<0.32
* ** ***
nmole O~/min/106 cells nmole adenosine turned over per min per 106 cells nmole NADPH turned over per min per 10 ~ cells
Each mixture was 3.0 ml in total volume, contained 101,000 PEC/ml, and was 120 uM in cytochrome c. After 15 min preincubation at 37 ° , the reactions were started by the addition of Tuftsin as described in the text. (15). Because of the considerable evidence (18,19) that in the presence of Mn ++ some nonenzymatic NADPH oxidation occurs via an O~ dependent free radical chain reaction, the values herein reported for NADPH oxidase activity are probably greater than the true in vivo activity. Results Table I shows the results of each experiment. Omission of Tuftsin (data not shown) resulted in no OF release. Incubation with adenosine for 15 minutes before Tuftsin stimulation resulted in more than doubling of O~ release (1.59 to 3.70). Allopurinol approximately halved O~ release (1.59 to 0.7), and the particular combination of adenosine and Allopurinol used essentially neutralized the effects of each when used individually. ADA activities are shown in the table; differences between values were negligible, especially when compared to changes in 0~. No NADPH oxidase activity could be detected in any of the samples; more than I0 s ceils are generally used to detect this enzyme activity (15-17). When l~sates of 107 cells were used, activity could readily be demonstrated. Thus O~ production attributable to NADPH oxidase could be no more than 10 percent of that attributable to ADA. Discussion Our findings demonstrate metabolic modulation of O~ production by macrophages. Of the large number of possibilities, we have selected adenosine, the first substrate of the purine salvage pathway, to stimulate metabolic flux through this pathway, and Allopurinol, a recognized agent used in clinical medicine to selectively inhibit xanthine oxidase. The effects of these agents were additive since it was possible for each to neutralize the effects of the other when given in combination. These results are qualitatively similar to those obtained in vivo with Allopurinol and xanthlne in experiments where phagocytosis was the end point (11). Combinations of stimulants and inhibitors were not tested in these experiments, and it was not possible to compare relative contributions of NADPH oxidases and purine salvage pathway enzymes to O~ production. The importance of the purine salvage pathway in O~ production is furthermore implied by the findings of increased uric acid excretion during
1362
Superoxide Release and Purine Metaboli3m
Vol. 32, No. 12, 1983
phagocytosis (20) and the increased ADA titer in caseinate-elicited macrophages (21) and during macrophage chemotaxis (22). The experiments herein described demonstrate that the purine salvage pathway is responsible for over ten times the 03 production attributable to the NADPH oxidase system. Thus, oxidases for reduced pyrimidine nucleotides are not the key enzymes in this context in these cells. The ability of the xanthine oxidase reactions to provide most of the OF in these cells appears to be related to ADA activity regulation of substrate flux towards hypoxanthine formation. The sensitivity of the enzymes of the purine salvage pathway to inhibition by a number of agents, especially many of those used in chemotherapy of neoplastic disease, suggests that impairment of macrophage function by these agents may be part of the basis for the immunosuppressive side effects of several of such therapies. References I. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
G.L. TRITSCH and P.W. NISWANDER, J. Medicine 11393-399 (1980). G.L. TRITSCH and P.W. NISWANDER, Immunol. Commu---n. 10 I-8 (1981). G.L. TRITSCH and P.W. NISWANDER, Biochem. Med. 26 I--~r5-190 (1981). A.S. ROSENTHAL, N. Eng]. J. Med. 3031153-1156 -(-]-980). E.R. GIBLETT, J.E. ANDERSON, F. COHEN, B. POLLARA and H.J. MEUWISSEN, Lancet ii 1067-1069 (1972). V.A. NAJJAR, Adv. Enzymol. 41129-178 (1974). J.A. BADWEY and M.L. KARNOVSK---Y, Ann. Rev. Biochem. 4_99695-726 (1980) M.L. KARNOVSKY, J.K. LAZDINS, J. Immunol. 121809-813 (1978). M.L. KARNOVSKY, J.K. LAZDINS, D. DR~TH and A. HARPER, Ann. N.Y. Acad. Sci. 256266-274 (1975). G.L. TRITSCH and P.W. NISWANDER, Molec. Cell. Biochem. in press. E. TUBARO, B. LOTTI, C. SANTIANGELI and G. CAVALLO, Biochem. Pharmacol. 293018-3020 (1980). H.M. KALCKAR, J. Biol. Chem. 167445-459 (1954). N.O. KJ~PLAN, Methods in Enzymol. 2473-474 (1955). E. BEUTLER, Red Cell Metabolism, A Manual of Biochem. Methods, 2nd ed., p. 86, Grune & Stratton, New York (1975). G.Y.N. IYER and J.H. QUASTEL, Canad. J. Biochem. Physiol. 41427-434 (1963). P. PATRIARCA, R. CRAMER, S. MONCALVO, F. ROSSI and ROMEO, Arch. Biochem. Biophys. 145255-262 (1971). K. TAKANAKA and P.J. O'BRIEN, Arch. Biochem. Biophys. 169436-442 (1975). J.T. CURNUTTE, M.L. KARNOVSKY and B.M. BABIOR, J. Clin. Invest. 5710591067 (1976)~ P. PATRIARCA, P. DRI, K. KAKINUMA, F. TEDESCO and F. ROSSI, Biochim. Biophys. Acta 385380-386 (1975). T. CHAN, Proc. Natl. Acad. Sci. USA 76925-929 (1979). R.J. SOBERMAN and M.L. KARNOVSKY, J. Expl. Med. 152241-246 (1980). R. SNYDERMAN, M.C. PIKE, D.G. FISCHER and H.S. KOREN, J. Immunol. 119 2060-2066 (1977).