Output of [14C]adenine nucleotides and their derivatives from cerebral tissues. Tetrodotoxine-resistant and calcium ion-requiring components

1. Neocortical tissues, exposed briefly to [(14)C]adenine and containing over 98% of their (14)C as adenine nucleotides, when superfused with glucose-bicarbonate salines released about 0.1% of their (14)C content/min to the superfusate. 2. Addition of unlabelled adenosine to the superfusing fluid increased the (14)C output three- to four-fold; half-maximal increase was given by about 40mum-adenosine, and reasons are adduced for considering the activity of adenosine kinase to be a major factor in conditioning the (14)C output. Adenosine similarly increased the enhanced (14)C output caused by electrical excitation of the superfused tissue; it brought about only a small increase in tissue glycolysis. 3. Output of (14)C from the [(14)C]adenine-labelled tissues was increased when Ca(2+) was omitted from the superfusing fluids, but electrical stimulation did not then liberate more (14)C. Nevertheless, such tissues still responded to electrical stimulation by increased glycolysis, and their (14)C output again became susceptible to increase by electrical stimulation when Ca(2+) was restored. 4. The six-fold increase in tissue glycolysis caused by electrical excitation was almost completely inhibited by tetrodotoxin at 0.1mum and above, but this was associated with about 50% inhibition only in the output of (14)C from tissues preincubated with [(14)C]adenine. The (14)C-labelled compounds of which output was most inhibited by tetrodotoxin were adenosine, inosine and hypoxanthine whereas output in a nucleotide fraction was little affected.     

ACTIONS OF NEUROHUMORAL AGENTS AND CEREBRAL METABOLITES ON OUTPUT OF ADENINE DERIVATIVES FROM SUPERFUSED TISSUES OF THE BRAIN

—Adenine nucleotides of guinea-pig neocortical tissues were labelled by prior incubation with [¹⁴C]adenine and excess of adenine was then removed by superfusion with precursor-free media. During continued superfusion labelled adenine derivatives were released at a stable rate of about 0·05 per cent of the tissue ¹⁴C/min and this rate was increased about five-fold by electrical stimulation. Various compounds, including some known to increase the cyclic AMP content of cerebral tissues, were examined for action on the release of [¹⁴C]adenine derivatives from the tissue and also on the rates of lactate production by the tissue, both before and during electrical excitation. The tissue content of adenine nucleotides following exposure of the tissue to these compounds was also determined. Noradrenaline, γ-aminobutyrate and acetylcholine together with carbamoylcholine at the concentrations examined were without effect on the release of ¹⁴C compounds from the tissue. Also, noradrenaline and γ-aminobutyrate caused no alteration in lactate production but brought about some decrease in the adenylate energy charge of the tissue. Histamine, 100 μm , brought about a small but consistent increase (35 per cent) both in release of ¹⁴C-compounds and lactate output, while reducing the adenylate energy charge of the tissues. l -Glutamate at 5 mm decreased the tissue adenylate energy charge to a greater extent than did histamine; it increased the release of ¹⁴C-compounds seven to eight-fold and similarly increased the tissues' rates of lactate production. Lower concentrations of glutamate had smaller effects. In those cerebral tissues whose cyclic AMP content is increased by l -glutamate, the increase is probably brought about by intermediation of released adenosine.     

Adenine derivatives as neurohumoral agents in the brain. The quantities liberated on excitation of superfused cerebral tissues

Adenine nucleotides of guinea-pig neocortical tissues were labelled by incubation with [(14)C]adenine and excess of adenine was then removed by superfusion with precursor-free medium. Adenine derivatives released from the tissue during continued superfusion, including a period of electrical stimulation of the tissue, were collected by adsorption and examined after elution and concentration. The stimulation greatly increased the (14)C output, and material collected during and just after stimulation had a u.v. spectrum which indicated adenosine to be a major component. The additional presence of inosine and hypoxanthine was shown by chromatography and adenosine was identified also by using adenosine deaminase. Total adenine derivatives released from the tissue during a 10min period of stimulation were obtained as hypoxanthine, after deamination and hydrolysis of adenosine and inosine, and amounted to 159nmol/g of tissue. This corresponded to the release of approx. 7pmol/g of tissue per applied stimulus. The hypoxanthine sample derived from superfusate hypoxanthine, inosine and adenosine was of similar specific radioactivity to the sample of inosine separated chromatographically, and each was of higher specific radioactivity than the adenine nucleotides obtained by cold-acid extraction of the tissue.     

Interaction between adenosine generated endogenously in neocortical tissues,and homocysteine and its thiolactone

[¹⁴C]Adenine derivatives in normal guinea pig or rat neocortical tissues maintained by superfusion included ATP, ADP and AMP collectively forming some 98% of the acid-extracted ¹⁴C; adenosine, inosine and hypoxanthine each at less than 0.5% and S-adenosylhomocysteine at about 0.1%. l-Homocysteine and/or its thiolactone increased only a little the S-adenosylhomocysteine. The superfusion fluid carried from the tissue per minute about 0.1% of its acid-extractable [¹⁴C]adenine derivatives. Electrical stimulation of the superfused tissue increased 10-fold its output of [¹⁴C]adenine derivatives and diminished the 5′-nucleotides in the tissue to 94% of the acid-extractable [¹⁴C]adenine derivatives, the remainder being adenosine, inosine and hypoxanthine with little change in S-adenosylhomocysteine. Homocysteine in the superfusion fluids now caused large increases in tissue S-adenosylhomocysteine, which became the preponderant non-nucleotide ¹⁴C-derivative when homocysteine was 0.1 mM or greater. The total [¹⁴C]adenine conversion to non-nucleotide derivatives then increased and the 5′-nucleotides fell to 88% of the total. It is concluded that concentration relationships observed in the action of homocysteine make it feasible that convulsive conditions and mental changes associated with administered homocysteine and with homocystinuria are due to cerebral adenosine concentrations being diminished through formation of S-adenosylhomocysteine. Adenosine is preponderantly depressant in cerebral actions; effects of the S-adenosylhomocysteine produced may also be relevant.     

The metabolism of adenine derivatives in different parts of the brain of the rat, and their release from hypothalamic preparations on excitation.

Five enzymes concerned with the metabolism of adenine derivatives were assayed in seven regions of the rat brain. A region which included the hypothalamus had the highest AMP deaminase and adenosine deaminase activities, while its 5'-nucleotidase activities were relatively low. The enzymes named and also the uptake of [14C]adenine by incubated tissue samples were more active with hypothalamic than with neocortical tissues. On superfusion with glucose-bicarbonate saline after assimilating [14C]adenine, the hypothalamic tissues released about 0.2 per cent of their 14C content per minute. This release was increased fourfold with electrical excitation but the presence of 0.25 muM tetrodotoxin prevented most of this increase. The compounds released during superfusion and electrical stimulation were preponderantly hypoxanthine, inosine, and adenosine, with only small amounts of adenine nucleotides. The output of all these compounds increased during the period of stimulation and also the proportion of adenine nucleotides increased when stimulation was carried out in the presence of tetrodotoxin. The output of the nucleotides and adenosine increased more promptly when stimulated than did that of the other compounds named. The results are discussed in terms of the metabolic roles of the enzymes concerned. and in relation to whether the enzymes are acting on intracellular or extracellular substrates.     

Adenosine in cerebral homeostatic role: appraisal through actions of homocysteine, colchicine, and dipyridamole

Mammalian neocortical tissues were incubated in [14C]adenine-containing fluids and their newly-synthesized adenine derivatives examined after periods of superfusion. Increased [K+] released adenine derivatives from the tissues, a release diminished by homocysteine. Homocysteine acted also to diminish the tissue content of adenosine plus its metabolites hypoxanthine and inosine, while increasing that of S-adenosylhomocysteine. Hypoxia also increased the tissue content and the output of adenosine plus its metabolites, and again homocysteine augmented the S-adenosylhomocysteine. Glutamic acid also increased tissue content and output of adenosine and derivatives, an action diminished by homocysteine and associated with augmented S-adenosylhomocysteine. Colchicine or dipyridamole did not prevent augmentation of S-adenosylhomocysteine by the reagents described; the sequence from adenosine phosphates to S-adenosylhomocysteine is concluded to be intracellular and not to involve extracellular formation of precursor adenosine. Adenosine displayed properties consistent with its being involved in two distinct categories of homeostasis, and also with its exerting an inhibitory tone in normal cerebral systems.     

The metabolism of adenine derivatives in different parts of the brain of the rat,and their release from hypothalamic preparations on excitation

Five enzymes concerned with the metabolism of adenine derivatives were assayed in seven regions of the rat brain. A region which included the hypothalamus had the highest AMP deaminase and adenosine deaminase activities, while its 5'-nucleotidase activities were relatively low. The enzymes named and also the uptake of [¹⁴C]adenine by incubated tissue samples were more active with hypothalamic than with neocortical tissues. On superfusion with glucose-bicarbonate saline after assimilating [¹⁴C]adenine, the hypothalamic tissues released about 0.2% of their ¹⁴C content per minute. This release was increased fourfold with electrical excitation but the presence of 0.25 μUM tetrodotoxin prevented most of this increase. The compounds released during superfusion and electrical stimulation were preponderantly hypoxanthine, inosine, and adenosine, with only small amounts of adenine nucleotides. The output of all these compounds increased during the period of stimulation and also the proportion of adenine nucleotides increased when stimulation was carried out in the presence of tetrodotoxin. The output of the nucleotides and adenosine increased more promptly when stimulated than did that of the other compounds named. The results are discussed in terms of the metabolic roles of the enzymes concerned, and in relation to whether the enzymes are acting on intracellular or extracellular substrates.     

REDISTRIBUTION OF ADENINE DERIVATIVES AMONG SUBCELLULAR FRACTIONS FROM GUINEA PIG NEOCORTICAL TISSUES, ON INCUBATION IN VITRO

Abstract— After the brief in vitro exposure of guinea-pig neocortical tissue to [¹⁴C]adenine, synaptosomal fractions prepared from the incubated tissue contained about 6% of its retained ¹⁴C. On continued incubation and superfusion with or without stimulation, the synaptosomal proportion of the ¹⁴C increased, while the protein and K content of the fraction underwent smaller changes only. Colchicine, 0.5 m m , diminished the synaptosomal enrichment in [¹⁴C]adenine derivatives and also in some cases increased the ¹⁴C effluent from tissues to superfusates. Colchicine also diminished the uptake of adenosine, but not of adenine, to the neocortical tissues. It is concluded that nerve terminal regions receive adenine derivatives from other tissue components as part of their normal metabolism, and that much of this can arrive by extracellular fluids; transport cytoplasmically is not excluded.     

5'-adenine mononucleotides in preparations from guinea pig cerebral cortex. Accumulation, translocation and release.

Isolated nerve terminals (synaptosome beds) were prepared from the neocortex of guinea pig and their ability to accumulate and release adenine nucleotides was studied. Synaptosome beds prelabelled with [14C]adenosine released newly synthesized [14C] adenine derivatives on superfusion. Electrical stimulation and K+ depolarization gave augmented output of both [14C] adenine derivatives and lactate from the preparations. Action of metabolic inhibitors on this output was examined. During incubation and superfusion, the synaptosomes displayed glycolysis and synthesis of ATP. Supply of adenine derivatives to the nerve terminals also occurred by translocation from other parts of the tissue.     

UPTAKE AND RELEASE OF [14C]ADENINE DERIVATIVES AT BEDS OF MAMMALIAN CORTICAL SYNAPTOSOMES IN A SUPERFUSION SYSTEM

(1) Synaptosomal fractions from guinea pig neocortical dispersions prepared in sucrose solutions were deposited from saline media as ‘beds’ on nylon bolting cloth. When incubated with 0.5–10 μm -[¹⁴C]adenine or adenosine in glucose bicarbonate salines, uptake of ¹⁴C from adenosine proceeded at about four times the rate of uptake of [¹⁴C]adenine. This contrasted with the relative uptake of the two compounds to neocortical tissue slices or to beds made from mitochondrial fractions, where uptake was similar with the two precursors. Uptake of both precursors to synaptosome beds was much greater than uptake of inosine. (2) Synaptosome beds, [¹⁴C]adenosine-loaded, contained 88 per cent of the ¹⁴C as 5′-adenine nucleotides, the remainder being present as cyclic AMP, inosine, hypoxanthine and adenosine. When superfused, the ¹⁴C output consisted mainly of adenosine, inosine and hypoxanthine, with some 7 per cent of 5′-nucleotides and 4 per cent of cyclic AMP. (3) Electrical pulses and the addition of 50 mm -KCl each increased the efflux of ¹⁴C from superfused [¹⁴C]adenosine-loaded beds. The superfusates issuing after excitation contained the same ¹⁴C-labelled compounds as issued before, with a small increase in the proportional yield of adenosine. The additional output of ¹⁴C following electrical pulses was diminished by about 50 per cent by 0.5 μm -tetrodotoxin while that following KCl was not affected; it was however prevented when the superfusing fluids were free of Ca²⁺.     

Purine ribonucleotide biosynthesis, interconversion and catabolism in mouse brain in vitro

The relative rates of the synthetic, interconversion and catabolic reactions of purine metabolism in chopped mouse cerebrum were studied. The rates of incorporation of [(14)C]adenine and [(14)C]hypoxanthine into purine ribonucleotides were much less than the potential activities of adenine phosphoribosyltransferase and hypoxanthine phosphoribosyltransferase, and the rates of incorporation were stimulated by the addition of guanosine to the incubation mixture. The availability of ribose phosphates may be a limiting factor for the formation of ribonucleotides from purine bases. The rate of incorporation of [(14)C]adenosine into purine ribonucleotides was at least seven- to eight-fold higher than that of adenine. The radioactivity in adenine ribonucleotides synthesized from adenine and hypoxanthine was about 100- and ten-fold respectively higher than that in the radioactive guanine ribonucleotides. The conversion of inosinate into guanine ribonucleotides was probably limited by the amount of inosinate available, and the conversion of adenine ribonucleotides into guanine ribonucleotides was probably limited by the activity of adenylate deaminase. The rate of catabolism of [(14)C]adenosine was low in comparison with its rate of utilization for ribonucleotide synthesis. A fraction of the [(14)C]hypoxanthine was catabolized to xanthine and urate. [(14)C]Guanine was completely converted into xanthine, mostly by the guanine deaminase that was released during incubation of chopped mouse cerebrum.     

Experimental modulation of PRPP availability for ribonucleotide synthesis from hypoxanthine in human skin fibroblast cultures

The intracellular concentration of the cosubstrate 5-phosphoribosyl 1-pyrophosphate (PRPP) may be rate-limiting for the reactions, catalysed by hypoxanthine phosphoribosyltransferase, by which mammalian cells convert the purine bases hypoxanthine, xanthine, and guanine to their ribonucleotide derivatives. The rate of conversion of [¹⁴C]hypoxanthine to radioactive phosphorylated products by intact human diploid skin fibroblasts was measured in the presence of compounds previously reported to alter PRPP concentration in a variety of cell types Methylene blue, previously reported to increase PRPP concentration in a variety of cultured cells including skin fibroblasts, increased product formation from hypoxanthine, with maximum effect following 60 min preincubation with 0.4 mM. Incubation with adenine, orotic acid, allopurinol, or adenosine has been shown to decrease PRPP concentration. Of these compounds, only adenine and adenosine decreased the rate of ribonucleotide synthesis from hypoxanthine in cultured skin fibroblasts. This decrease probably resulted from decreased PRPP synthesis rather than increased PRPP utilization. The reaction products isolated from cells following incubation with either [¹⁴C]adenine or [¹⁴C]adenosine included adenosine monophosphate and adenosine diphosphate, both inhibitors of PRPP synthetase.     

Experimental modulation of PRPP availability for ribonucleotide synthesis from hypoxanthine in human skin febroblast cultures

The intracellular concentration of the cosubstrate 5-phosphoribosyl 1-pyrophosphate (PRPP) may be rate-limiting for the reactions, catalysed by hypoxanthine phosphoribosyltransferase, by which mammalian cells convert the purine bases hypoxanthine, xanthine, and guanine to their ribonucleotide derivatives. The rate of conversion of [¹⁴C]hypoxanthine to radioactive phosphorylated products by intact human diploid skin fibroblasts was measured in the presence of compounds previously reported to alter PRPP concentration in a variety of cell types Methylene blue, previously reported to increase PRPP concentration in a variety of cultured cells including skin fibroblasts, increased product formation from hypoxanthine, with maximum effect following 60 min preincubation with 0.4 mM. Incubation with adenine, orotic acid, allopurinol, or adenosine has been shown to decrease PRPP concentration. Of these compounds, only adenine and adenosine decreased the rate of ribonucleotide synthesis from hypoxanthine in cultured skin fibroblasts. This decrease probably resulted from decreased PRPP synthesis rather than increased PRPP utilization. The reaction products isolated from cells following incubation with either [¹⁴C]adenine or [¹⁴C]adenosine included adenosine monophosphate and adenosine diphosphate, both inhibitors of PRPP synthetase.     

Intracellular formation of analogs of cyclic AMP. Studies with brain slices labeled with radioactive derivatives of adenine and adenosine.

A variety of radioactive analogs of adenine and adenosine were incubated with guinea pig cerebral cortical slices. Neither 1,N6-etheno[14C] adenosine nor 1,N6-etheno[14C] adenine were significantly incorporated into intracellular nucleotides. 2-chloro[8-3H] adenine was incorporated, but at a very low rate and conclusive evidence for the formation of intracellular radioactive 2-chloro-cyclic AMP was not obtained. N6-Benzyl[14C] adenosine was converted only to intracellular monophosphates and significant formation of radioactive N6-benzylcyclic AMP was not detected during a subsequent incubation. 2'-Deoxy-[8-14C] adenosine was converted to both intracellular radioactive 2'-deoxy-adenine nucleotides and radioactive adenine nucleotides. Stimulation of these labeled slices with a variety of agents resulted in formation of both radioactive 2'-deoxycyclic AMP and cyclic AMP. Investigation of the effect of various other compounds on uptake of adenine or adenosine suggested that certain other adenosine analogs might serve as precursors of abnormal cyclic nucleotides in intact cells.     

Adenine mononucleotides and their metabolites liberated from and applied to isolated tissues of the mammalian brain

Preincubation with [¹⁴C] adenine labeled the nucleotide fraction of isolated cerebral tissues, which subsequently released 0.18% of their¹⁴C content per minute, a proportion increased threefold by electrical excitation. Of the¹⁴C released, 2–3% was as 5-adenine nucleotides and about 2% as cyclic adenosine 35-monophosphate (cAMP). Among the 5-nucleotides AMP greatly preponderated, and ATP and ADP were detected. When added to (unlabeled) incubating neocortical tissue, ATP and AMP yielded adenosine as the major product, with smaller quantities of inosine and hypoxanthine, to effluent fluids. cAMP so added yielded 5-nucleotides and the other compounds named; adenosine yielded mainly inosine and hypoxanthine. Results from these reactions and others in which theophylline was included led to the conclusion that an appreciable proportion of the effluent [¹⁴C] adenosine, inosine, and hypoxanthine derived from cAMP.     

SUBCELLULAR LOCALIZATION OF [14C]ADENINE DERIVATIVES NEWLY-FORMED IN CEREBRAL TISSUES AND THE EFFECTS OF ELECTRICAL EXCITATION

Abstract— Guinea pig neocortical tissues were incubated with [¹⁴C]adenine, dispersed in cold isotonic sucrose and subcellular fractions prepared by centrifugation. Some 98 per cent of the assimilated ¹⁴C was found as acid-soluble nucleotides in the incubated tissues. In primary fractions obtained by differential centrifugation, about 60 per cent of the [¹⁴C]-nucleotides were in supernatant fractions, in distinction to ATP of which the greatest molar quantity (61 per cent of that in the dispersion) was in the crude mitochondrial fraction. When the crude mitochondrial fraction was separated by density gradient centrifugation, most ¹⁴C was found in synaptosomal fractions and about 85 per cent of this ¹⁴C was adenine nucleotides.
Electrical stimulation of incubating tissues immediately prior to their dispersion and centrifugation greatly diminished the proportion of ¹⁴C subsequently found in nucleotides (collectively) in the supernatant fraction, and increased their inosine and hypoxanthine. Stimulation increased the tissue's cyclic AMP but a preferential localization for this was not established. Results are tentatively interpreted in terms of liberation of an adenine derivative on excitation, and its action or reuptake at a tissue component different from that from which it was liberated. Fractionation of tissues which had been incubated with both [¹⁴C]-adenine and [³H]adenosine suggested that of the two compounds, more adenosine was taken up by synaptic regions in preference to other cellular regions of the tissue.     

The mechanism of thyrotrophin action in relation to lipid metabolism in thyroid tissue

1. The effects of thyrotrophin in vitro on the incorporation of [(14)C]-glucose, -glycerol, -palmitate and -oleate into the lipids of thyroid tissue were examined. 2. Thyrotrophin increased the incorporation of these (14)C-labelled precursors into phosphatidylinositol specifically. 3. Thyrotrophin also increased the proportion of (14)C radioactivity from labelled glucose, glycerol, palmitate and oleate incorporated into the 1,2-diglycerides. 4. The addition of thyrotrophin to thyroid slices for 10min., after 2hr. of prelabelling with [(14)C]glycerol, also increased the proportion of (14)C radioactivity incorporated into the 1,2-diglyceride fraction. 5. After incubation of thyroid tissue with [1-(14)C]palmitate, thyrotrophin caused a two- to three-fold increase in the specific radioactivity of palmitate isolated from phosphatidylinositol and 1,2-diglycerides. In contrast, the specific radioactivity of palmitate isolated from the choline and ethanolamine phosphoglycerides, 1,3-diglycerides and triglycerides was not increased by thyrotrophin.     

Intracellular formation of analogues of cyclic AMP: Studies with brain slices labeled with radioactive derivatives of adenine and adenosine

A variety of radioactive analogs of adenine and adenosine were incubated with guinea pig cerebral cortical slices. Neither 1,N⁶-ethano[¹⁴C]adenosine nor 1,N⁶-ethanol[¹⁴C]adenine were significantly incorporated into intracellular nucleotides. 2-chloro[8-³H]adenine was incorporated, but at a very low rate and conclusive evidence for the formation of intracellular radioactive 2-chlorocyclic AMP was not obtained. N⁶-Benzyl[¹⁴C]adenosine was converted only to intracellular monophosphates and significant formation of radioactive N⁶-benzylcyclic AMP was not detected during a subsequent incubation. 2′-Deoxy-[8-¹⁴C] adenosine was converted to both intracellular radioactive 2′-deoxyadenine nucleotides and radioactive adenine nucleotides. Stimulation of these labeled slices with a variety of agents resulted in formation of both radioactive 2′-deoxycyclic AMP and cyclic AMP. Investigation of the effect of various other compounds on uptake of adenine or adenosine suggested that certain other adenosine analogs might serve as precursors of abnormal cyclic nucleotides in intact cells.     

Adenosine as a constituent of the brain and of isolated cerebral tissues, and its relationship to the generation of adenosine 3′:5′-cyclic monophosphate

1. Adenosine was determined in rapidly frozen rat and guinea-pig brain and in guinea-pig cerebral tissues after incubation in vitro. Adenosine concentrations were approx. 2nmol/g wet wt. in frozen tissue, diminished at room temperature, and returned to 2nmol/g on incubation in oxygenated glucose/salines. 2. Superfusion with noradrenaline then increased the tissue's adenosine concentration 2.5-fold, and hypoxia caused an 8-fold increase. 3. Electrical stimulation alone or in the presence of noradrenaline or histamine increased the tissue's adenosine and cyclic AMP, but adenosine concentrations reached their peak later and were maintained for longer than those of cyclic AMP. 4. Superfusion with l-glutamate with and without electrical excitation raised adenosine concentrations to 15-34nmol/g. The increases in cyclic AMP on electrical stimulation, superfusion with glutamate or a combination of these treatments were diminished by addition of adenosine deaminase or theophylline. 5. It is concluded that adenosine can be produced endogenously in cerebral systems, in sufficient concentrations to accelerate an adenosine-activated adenylate cyclase, and by this route can contribute to the cerebral actions of electrical stimulation and of the neurohumoral agents. In certain instances cyclic AMP as substrate contributes to an increase in adenosine.     

Metabolic fate of extracellular NAD in human skin fibroblasts

Extracellular NAD is degraded to pyridine and purine metabolites by different types of surface-located enzymes which are expressed differently on the plasmamembrane of various human cells and tissues. In a previous report, we demonstrated that NAD-glycohydrolase, nucleotide pyrophosphatase and 5'-nucleotidase are located on the outer surface of human skin fibroblasts. Nucleotide pyrophosphatase cleaves NAD to nicotinamide mononucleotide and AMP, and 5'-nucleotidase hydrolyses AMP to adenosine. Cells incubated with NAD, produce nicotinamide, nicotinamide mononucleotide, hypoxanthine and adenine. The absence of ADPribose and adenosine in the extracellular compartment could be due to further catabolism and/or uptake of these products. To clarify the fate of the purine moiety of exogenous NAD, we investigated uptake of the products of NAD hydrolysis using U-[(14)C]-adenine-NAD. ATP was found to be the main labeled intracellular product of exogenous NAD catabolism; ADP, AMP, inosine and adenosine were also detected but in small quantities. Addition of ADPribose or adenosine to the incubation medium decreased uptake of radioactive purine, which, on the contrary, was unaffected by addition of inosine. ADPribose strongly inhibited the activity of ecto-NAD-hydrolyzing enzymes, whereas adenosine did not. Radioactive uptake by purine drastically dropped in fibroblasts incubated with (14)C-NAD and dipyridamole, an inhibitor of adenosine transport. Partial inhibition of [(14)C]-NAD uptake observed in fibroblasts depleted of ATP showed that the transport system requires ATP to some extent. All these findings suggest that adenosine is the purine form taken up by cells, and this hypothesis was confirmed incubating cultured fibroblasts with (14)C-adenosine and analyzing nucleoside uptake and intracellular metabolism under different experimental conditions. Fibroblasts incubated with [(14)C]-adenosine yield the same radioactive products as with [(14)C]-NAD; the absence of inhibition of [(14)C]-adenosine uptake by ADPribose in the presence of alpha-beta methyleneADP, an inhibitor of 5' nucleotidase, demonstrates that ADPribose coming from NAD via NAD-glycohydrolase is finally catabolised to adenosine. These results confirm that adenosine is the NAD hydrolysis product incorporated by cells and further metabolized to ATP, and that adenosine transport is partially ATP dependent.