Studies on the Biosynthesis of Nicotinamide COMPARATIVE IN VIVO STUDIES ON NICOTINIC ACID, NICOTINAMIDE, AND QUINOLINIC ACID AS PRECURSORS OF NICOTINAMIDE ADENINE DINUCLEOTIDE*

In order to investigate the biosynthesis of nicotinamide adenine dinucleotide in vivo and to evaluate comparatively the quantitative significance of three precursors, nicotinic acid, nicotinamide, and quinolinic acid, the radioactive substrates were injected directly into the portal vein of mice, and the radioactive compounds in the liver were analyzed. When administered in small doses, nicotinic acid was a much better precursor of NAD than nicotinamide. The incorporation of nicotinic acid-4C into NAD proceeded almost linearly up to 10 min, thereafter NAD-14C gradually decreased to about one-half by 10 hours. In contrast, nicotinamide-' 4 C injected in this way was not utilized significantly for the synthesis de novo of NAD during the 1st hour after the injection. Quinolinic acid14C hardly penetrated into the liver cells. On the contrary, when administered in large quantities, nicotinamide was a much better precursor of NAD in the liver than nicotinic acid. With a large dose of nicotinamide-' 4 C as substrate, the total radioactivity in the liver decreased rapidly during the 1st hour and then increased up to almost 8 hours after the injection. During the initial phase, the incorporation of 14C into NAD was almost insignificant, but NAD-' 4C in the liver started to increase about 1 hour after the injection with the concomitant increase of the total radioactivity in the liver. Analyses of the distribution of 14C in various organs and tissues indicated that a large portion of nicotinamide4 C was first excreted from liver, accumulated in the gastrointestinal tract, deamidated to nicotinic acid, reabsorbed into the liver, and served as precursor to NAD over a prolonged period of time.

It has been well established that the biosynthesis of nicotinamide adenine dinucleotide occurs from three different precursors, nicotinic acid, nicotinamide, and tryptophan, in mammalian tissues. The biosynthetic pathway from nicotinic acid has been studied by Preiss and Handler, who found that nicotinic acid ribonucleotide and deamido-NAD are intermediates in this synthesis (1)(2)(3). In a series of previous reports from this laboratory, quinolinic acid, which is produced from tryptophan via 3-hydroxyanthranilic acid, was shown to be converted directly to nicotinic acid ribonucleotide in the presence of 5-phosphoriboxyl-l-pyrophosphate in rat liver (4)(5)(6). Free nicotinic acid was excluded as an intermediate in this conversion.
The 5-phosphoribosyl-l-pyrophosphate-dependent formation of nicotinamide ribonucleotide from nicotinamide was described also by Preiss and Handler (7). Recently an enzyme has been partially purified from mammalian liver by Greengard and his co-workers which catalyzes the deamidation of nicotinamide to nicotinic acid (8)(9)(10). The affinity of these enzymes for nicotinamide, however, are considerably low, and high concentrations of the substrate are required for these reactions. The enzymic formation of nicotinamide ribonucleoside from nicotinamide and ribose 1-phosphate was described by Rowen and Kornberg (11), but the equilibrium of this reaction favors the phosphorolysis of the ribonucleoside rather than its synthesis. All previous studies in vitro thus indicated that nicotinic acid and tryptophan are better precursors of NAD than nicotinamide in the mammalian liver.
In marked contrast to these observations in vitro, however, studies in vivo by Kaplan and his co-workers indicated that the intraperitoneal injection of nicotinamide causes a striking rise in the liver NAD level in mice and rats (12)(13)(14)(15)(16)(17). Moreover, nicotinic acid has been shown to give a much smaller rise in NAD than does nicotinamide (13). These contradictory observations in studies in vivo and in vitro, mentioned above, have not yet been reasonably explained, and the biosynthetic pathway of NAD from nicotinamide under physiological conditions has so far remained obscure.
, radioactivity; [1, optical density at 260 mu. Recoveries were almost quantitative. Nicotinamide ribonucleoside and N'-methylnicotinamide were not absorbed on the column. The following abbreviations are used in the figures: Namide, nicotinamide; Nacid, nicotinic acid; NaR, nicotinic acid ribonucleoside; NMN, nicotinamide ribonucleotide; NaMN, nicotinic acid ribonucleotide; d-NAD, deamido-NAD; QA, quinolinic acid. of these precursors by the liver as well as the biosynthetic pathway of NAD in vivo from these precursors has been investigated.

EXPERIMENTAL PROCEDURE
Materials-Nicotinamide-7-1 4 C (10.7 mC per mM), nicotinic acid-7-14 C (10.7 mC per mM) and uniformly labeled aniline sulfate-' 4 C (12.7 mC per mM) were purchased from Radiochemical Centre, Amersham, England. Nicotinamide-14C and nicotinic acid-' 4 C were purified by paper chromatography by consecutive runs in 1 M ammonium acetate-ethyl alcohol (3:7, pH 5.1) and in n-butyl alcohol saturated with 15% ammonia. Quinolinic acid-2,3,7,8-14 C (720,000 cpm per guM) was prepared from uniformly labeled aniline sulfate-14 C according to the method of Gholson, Ueda, Ogasawara, and Henderson (18) except that free quinolinic acid-l 4 C was obtained from copper quinolinate-l4C by the use of a Chelex-100 column before use. Nicotinic acid ribonucleotide-1 4 C and deamido-NAD-' 4 C were prepared from nicotinic acid-7-4 C, ATP, and ribose 5-phosphate with crude extracts of human erythrocytes by the method of Preiss and Handler (3). Nicotinic acid ribonucleoside-' 4 C and nicotinamide ribonucleoside were obtained from the respective ribonucleotides with prostatic phosphomonoesterase which was kindly donated by Drs. Shimono and Sugino. The ribonucleotides and ribonucleosides of nicotinic acid and nicotinamide thus prepared were isolated by chromatography on a column of Dowex 1-X2 (formate) and purified by paper chromatography on Whatman No. 3 filter paper with isobutyric acid-ammonia-water (66: 1.7:33, pH 3.8) (5). Nicotinamide ribonucleotide was purchased from Sigma. Nicotinic acid, nicotinamide, quinolinic acid, and other chemicals were obtained from commercial sources.
Methods-Experiments were carried out with 2-to 3-month-old dd-mice, 25 to 30 g in weight, which had been raised on an ad libitum diet of CLEA laboratory chow. After laparotomy under a slight ether anesthesia, a 0.1-ml solution containing 70 to 80 mgtmoles of radioactive compounds was injected into the portal vein via the superior mesenteric vein during a period of 20 sec. At various time intervals after the termination of the injection, namely, 20 sec, 1 min, several minutes or hours, the livers were quickly removed and immediately frozen in an alcohol-Dry Ice mixture. The liver was then homogenized with 3.3 ml of 3% perchloric acid in a glass homogenizer and centrifuged at 3,000 rpm for 10 min. The precipitate was washed twice with 2-ml portions of 3% perchloric acid. The supernatant solution and the washings were combined, and the pH was adjusted to 6 with 5 N KOH at 0°. Potassium perchlorate was removed by centrifugation, and an aliquot (0.05 to 0.2 ml) of the supernatant solution was placed on an aluminum disk, dried, and the radioactivity was determined by the use of a Nuclear-Chicago gas flow counter. The values obtained were then corrected for selfabsorption by comparison with a known amount of nicotinic acid-' 4 C or nicotinamide-' 4 C treated in the same manner. A portion of the neutralized extract was placed on a column of Dowex 1-X2 (formate) (200 to 400 mesh, diameter 0.8 cm, length 40 cm). Elution was carried out by the application of a formic acid concentration gradient initially with 250 ml of water in the mixing chamber, into which 150 ml of 0.01 N formic acid, 250 ml of 0.25 N formic acid, and 400 ml of 2 N formic acid were introduced from the reservoir in that order. Fractions (10 ml) were collected at the rate of about 35 ml per hour, and the radioactivity and ultraviolet absorption at 260 mgz were determined. A typical elution pattern of various compounds is presented in Fig. 1. The fractions containing radioactive products were combined and concentrated under reduced pressure, and then chromatographed on Whatman No. 3 paper with the authentic samples in three different solvent systems: 1 M ammonium acetate-ethyl alcohol (3:7, pH 5.1), isobutyric acid-ammoniawater (66:1.7:33, pH 3.8), and n-butyl alcohol saturated with 15% NH 4 OH (5). Spots were detected under ultraviolet light with a Mineralight lamp, and the radioactivity on paper chromatograms was determined by directly counting paper strips with a Packard Tri-Carb liquid scintillation spectrometer (19). Toluene containing 0.25% 2,5-diphenyloxazole and 0.015% 1,4-bis-2'-(5'-phenyloxazolyl)benzene was used as the scintillator. NAD-14C was further identified as nicotinamide-' 4 C or as nicotinamide ribonucleotide-' 4 C by paper chromatography after digestion either with spleen NAD nucleosidase or with snake venom phosphodiesterase, respectively. Deamido-NAD-' 4 C was converted to nicotinic acid ribonucleotide-' 4 C by the action of snake venom phosphodiesterase and identified by paper chromatography (5). Nicotinic acid ribonucleotide-14 C was identified as nicotinic acid ribonucleoside-' 4 C by paper chromatography after digestion with prostatic phosphomonoesterase (5).
NAD and NADP were determined spectrophotometrically by measuring the increase in absorbance at 340 mtu in the presence of either alcohol and alcohol dehydrogenase or glucose 6-phosphate and glucose 6-phosphate dehydrogenase, respectively. Quinolinic acid was determined by measuring the absorbance at 268 mi in 0.1 M potassium phosphate buffer, pH 6.8 ( = 3.4 X 103) (20). Nicotinic acid and nicotinamide were determined at 261 mit in 0.1 N HCl (: nicotinic acid = 4.36 X 103, nicotinamide = .5.25 X 103) (21,22). All spectrophotometric measurements were carried out with a Shimadzu spectrophotometer, type QR 50. Absorption spectra were measured with a Cary model 15 recording spectrophotometer.
tions of 78 mgmoles each of either nicotinic acid-' 4 C or nicotinamide-' 4 C containing 4 X 105 cpm into the portal vein. The livers of three mice were removed at each of the following time intervals: 20 sec, 1 min, 3 min, 10 min, 1 hour, 4 hours, and 8 hours after the injection. The livers were extracted with 3 % perchloric acid, and the radioactivity was determined as described under "Methods." As shown in Fig. 2, approximately 40 to 50% of the total radioactivity of nicotinic acid or nicotinamide injected was found in the liver 20 sec after the injection. When nicotinamide-' 4 C was injected, however, a large portion of the radioactivity disappeared rapidly from the liver during the initial 3 main, and less than 10% of the total radioactivity was found in the liver for the next several hours. After 8 hours, the radioactivity in the liver appeared to increase slightly. In contrast, the radioactivity of injected nicotinic acid remained in the liver in much larger quantities. More than 20% of the total radioactivity was recovered from the liver even after 4 hours. When 80 mamoles of quinolinic acid-14 C containing 6 X 104 cpm were injected into the portal vein of mice by a similar procedure, less than 10% of the radioactivity was recovered in the liver even 20 see after the injection. This result suggests that quinolinic acid hardly penetrated into the liver cells. Although the blood contaminated the liver extract under the conditions employed, the radioactivity recovered in the blood was not significant as compared with that in the liver, irrespective of the substrate injected and the time of analysis.

Fate of Nicotinic Acid-" 4 C in Liver-When nicotinic acid-' 4 C
(78 mpmoles) was employed as substrate, the radioactivity was taken up rapidly by the liver and subsequently disappeared gradually as mentioned above (Fig. 2). Upon analysis of the liver extracts by chromatography on a Dowex 1 (formate) column, nicotinic acid-' 4 C in the liver disappeared rapidly (Fig.  3). After 3 min, less than 5% of the total radioactivity injected was recovered as nicotinic acid-' 4 C, and by 10 min it disappeared almost completely. The rapid disappearance of nicotinic acid-' 4 C was accompanied by the appearance of nicotinic acid ribonucleotide-'4C and deamido-NAD-' 4 C. The radioactivity in nicotinic acid ribonucleotide was maximal after 1 min and then decreased to about one-fifth after 3 min. The radioactivity of deamido-NAD was maximal after 1 min and then decreased to about one-half after 3 min. After 10 min, these radioactive Radioactivity in the liver after intraportal injection of 14C-labeled nicotinic acid, nicotinamide, and quinolinic acid in small doses. Experimental procedures are described in the text. Each value for nicotinic acid and nicotinamide represents the average of three mice, and the values for quinolinic acid are the mean of five mice. Since the conversion of quinolinic acid-2,3,7,8-' 4 C to NAD is accompanied by the release of 1 mole of 1 4 CO2 from each mole of quinolinic acid utilized, the values for quinolinic acid are corrected for the loss of ' 4 C by multiplying the radioactivity recovered as nicotinic acid ribonucleotide, deamido-NAD, NAD, and nicotinic acid by . 0, nicotinic acid-' 4 C; X---X, nicotinamide-4C; A ----E /, quinolinic acid-4C. compounds disappeared almost completely. Radioactivity in NAD, in contrast, increased with time reaching a maximum at about 10 min and then decreased gradually. When the NAD curve was extrapolated, the half-life of NAD-' 4 C under the specified conditions was calculated to be about 10 hours. These results are consistent with the interpretation that nicotinic acid-' 4 C is converted to NAD via nicotinic acid ribonucleotide and deamido-NAD. In addition to these three products, a small amount of radioactive nicotinamide was detected during the 1to 8-hour period after the injection, presumably due to the degradation of NAD-' 4 C in the liver. Fate of Nicotinamide-' 4 C in Liver-Nicotinamide-' 4 C (78 mumoles) was injected and radioactive products in the liver extracts were analyzed as mentioned above. Nicotinamide-' 4 C in the liver disappeared rapidly as did nicotinic acid-' 4 C. In marked contrast to the results obtained with nicotinic acid-1 4 C, however, a small but definite incorporation of nicotinamide-' 4 C into NAD was observed immediately after the injection, but no increase of NAD-' 4 C occurred during a 10-min period (Fig. 4). After several hours, NAD-' 4 C in the liver started to increase slightly. Any other radioactive nicotinic acid derivatives did not accumulate to any appreciable extent; only a small quantity of nicotinamide ribonucleotide was detected 1 min after the injection.

Effect of Nonradioactive Nicotinic Acid on Incorporation of
Nicotinamide-' 4 C into NAD-The above results indicated that nicotinamide was a much poorer precursor of liver NAD than nicotinic acid when a small dose was injected into the portal vein. However, a small amount of nicotinamide-'4C was incorporated into liver NAD immediately after the injection. In order to , nicotinamide-' 4 C was given without nicotinic acid; Ol, nicotinamide-l 4 C was injected with 50 mumoles of nicotinic acid; M, nicotinamide-4 C was given with 500 mromoles of nicotinic acid.

4-
3. test the possibility that nicotinamide-14 C was deamidated in the liver in situ to nicotinic acid-14 C prior to incorporating into NAD during the 1-min period after the injection, 78 mjumoles of nicotinamide-14 C containing 4 X 105 cpm were injected simultaneously with 50 or 500 mumoles of unlabeled nicotinic acid into the portal vein. The livers were removed 1 min later, and the radioactive products in the liver were analyzed as described above. Since separate experiments showed that nicotinamide in the given doses does not inhibit the NAD biosynthesis from nicotinic acid, more than 4.5 mymoles of nicotinic acid, 2.5 mremoles of nicotinic acid ribonucleotide, and 7.5 mrmoles of deamido-NAD should be present in the liver following injection of 50 mumoles of nicotinic acid, under the specified conditions, as would be anticipated from the experimental results shown in Fig. 3. Accordingly, if nicotinamide-1 4 C is converted to NAD by way of nicotinic acid, nicotinic acid ribonueleotide, and deamido-NAD during the 1 min period after the injection, radioactivity of nicotinamide injected should be trapped to an appreciable extent in these compounds, and the incorporation of radioactivity into NAD should be decreased. However, as shown in Fig. 5, unlabeled nicotinic acid had very little effect on the uptake of nicotinamide-1 4 C by the liver as well as on the incorporation of the radioactivity into NAD. Radioactive nicotinic acid ribonucleotide and deamido-NAD were not detected to any appreciable extent, and only a trace of radioactive nicotinic acid was found in the liver. The results are consistent with the interpretation that nicotinamide-' 4 C injected intraportally is incorporated into liver NAD without being deamidated to nicotinic acid immediately after the injection, assuming that nicotinic acid and nicotinamide are not located in different compartments of the liver cells.

Fate of Large Doses of Nicotinamide-' 4 C in Liver-The intraperitoneal injections of large doses of nicotinamide (82 to 164
Moles per mouse) were shown to produce about a 10-fold increase in liver NAD in 8 to 12 hours, which confirms the previous reports by Kaplan and his co-workers (12)(13)(14)(15)(16)(17). Injections of nicotinic acid in large quantities gave a much smaller rise in NAD than did injections of nicotinamide. When small doses were employed, however, nicotinic acid was shown to be more effective than nicotinamide in stimulating the increase in the liver NAD. Similar results were obtained when these compounds were injected into the portal vein. In order to elucidate the discrepancy, a large quantity of nicotinamide-' 4 C (75 Amoles, 4 105 cpm) was injected directly into the portal veins of mice. The livers were removed and analyzed 20 see, 1 min, 3 min, 10 min, 30 min, 1 hour, 2 hours, 4 hours, 8 hours, 13 hours, and 16 hours after the injection. As shown in Fig. 6A, approximately 14% of the total radioactivity injected was recovered from the liver 20 sec after the injection. The radioactivity in the liver disappeared rapidly, and less than 4% of the radioactivity remained in the liver after 1 hour. However, the total radioactivity in the liver began to increase about 1 to 2 hours after the injection, and the increase continued up to 8 hours. In terms of specific radioactive compounds in the liver, the rapid decrease of the total radioactivity during the initial phase indicates the excretion of nicotinamide-14 C from the liver, and during this period, the formation of NAD-1 4 C was almost insignificant. The increase of NAD-14 C in the latter phase was almost parallel with the increase of total radioactivity, which indicates that nicotinamide-14C excreted from the liver returned to the liver probably in a different form and was utilized as the precursor of NAD. Since the above results merely represent the uptake of radioactivity in NAD, the net increase of the amount of NAD in the liver was also determined. As shown in Fig. 6B, the increase of the radioactivity in NAD was associated with an increase in the amount of NAD. In addition to NAD, small amounts of radioactive nicotinic acid, deamido-NAD, NADP, and a trace of nicotinic acid ribonucleotide were detected 1 to 13 hours after the injection (Fig. 6A).

Effect of Prior Administration of Nicotinamide on Synthesis of NAD from Nicotinic Acid-
The above results indicate that different mechanisms may exist for the rapid incorporation of nicotinamide-4 C into NAD during the 1st min and for the slow elevation of NAD level during 8 hours following the injection of nicotinamide-' 4 C. When nicotinamide-4 C was injected into the portal vein, some of the nicotinamide-' 4 C was immediately incorporated into the liver NAD, presumably by an exchange reaction catalyzed by NAD nucleosidase, but the major portion was rapidly excreted or passed through the liver without being utilized by the cell (Figs. 4 and 6). Therefore, the possibility must be considered that nicotinamide not utilized by the liver cells immediately after the injection may well be transported to the other tissues, stored, and converted to a different form, e.g. nicotinic acid which then returns to the liver and is synthesized to NAD over a prolonged period of time. In order to see if nicotinamide is converted to NAD by way of nicotinic acid, 82 tumoles of nonradioactive nicotinamide were injected intraperitoneally into mice, and 4, 8, and 24 hours after the injection, 60 mpzmoles of nicotinic acid-' 4 C containing 3.1 X 105 cpm were injected into the portal vein. Mice were killed 1 min later, and the livers were analyzed. If nicotinamide is converted to nicotinic acid prior to being synthesized to liver NAD, nicotinic acid-14 C should be diluted or trapped by unlabeled nicotinic acid, nicotinic acid ribonucleotide, or deamido-NAD. When nicotinic acid-1 4 C was injected without prior administration of nicotinamide, about 35% (109,000 cpm) of the radioactivity injected was found in the liver 1 min after the injection. As shown in Fig.  7, approximately 20% of the radioactivity found in the liver was recovered as nicotinic acid, 10% as nicotinic acid ribonucleotide, 49% as deamido-NAD, and 19% as NAD. At 4 and 8 hours after the injection of nicotinamide when the liver NAD was increasing markedly, almost the same amount (30.6% and 31.5%, respectively) of radioactivity was found in the liver. However, the majority of radioactivity was recovered as nicotinic acid, with a lesser amount in nicotinic acid ribonucleotide. Less than 10% was found in the deamido-NAD fraction, and only a negligible amount of radioactivity was incorporated into NAD. At 24 hours after the nicotinamide injection when the level of NAD had returned to normal, nicotinic acid-" 4 C was again incorporated into deamido-NAD and NAD in a similar pattern to that obtained without prior administration of nicotinamide, indicating that most, if not all, nicotinamide was synthesized to NAD by way of nicotinic acid.
Accumulation of Nicotinamide-' 4 C in Gastrointestinal Tract-Survey of the distribution of radioactivity after intraperitoneal injection of nicotinamide-1 4 C (82 ,moles, 4 X 105 cpm) showed that a large portion of the radioactivity injected was excreted into the urine; this is in good agreement with the finding by Petrack, Greengard, and Kalinsky. However, as shown in Fig.  8, as much as 20% of the radioactivity was found in the gastrointestinal tract 1 hour after the injection, whereas only about 3% was recovered in the liver, and less than 2% was found in other tissues. The radioactivity in the gastrointestinal tract decreased to about one-third after 4 hours with the concomitant increase in the radioactivity in the liver NAD. Preliminary analysis for the precise distribution of radioactivity in the gastrointestinal tract showed that more than 90 % of the radioactivity was located in the contents rather than the stomach and intestinal wall. 1 Personal communication. Each value represents the average of four mice. Total radioactivity recovered in the liver was 109,000 cpm when nicotinic acid-4 C was injected without prior administration of non-radioactive nicotinamide and 95,000 cpm, 97,500 cpm, and 95,000 cpm when nicotinic acid-4 C was injected 4, 8, and 24 hours, respectively, after the injection of nicotinamide. Stippled bars denote radioactivity recovered as nicotinic acid; hatched bars denote nicotinic acid ribonucleotide; plain bars denote deamido-NAD; solid bars denote NAD. Nicotinamide-14 C (82 molese, 4 X 105 cpm) was dissolved in 0.2 ml of water and injected intraperitoneally to mice. The mice were killed at 1 and 4 hours after the injection. The abdominal cavity was washed three times with NaCI solution, the various organs were removed, extracted with 3% perchloric acid, and the radioactivity was determined as described under "Methods." The radioactivity found in the third washings of the abdominal cavity was less than 500 cpm. Each value represents the average of three mice. 1, radioactivity recovered 1 hour after injection; l, after 4 hours.
When nicotinamide-14 C (41 moles, 8 X 105 cpm) was injected intraportally, accumulation of radioactivity was more pronounced; as much as 50% of the radioactivity injected was found in the gastrointestinal tract 1 hour after the injection, and about 6% was recovered in the liver. Analysis of the radioactive materials found in the gastrointestinal tract at various intervals showed that the radioactive nicotinic acid increased with concomitant decrease of nicotinamide-1 4 C during the 4 hours after the injection (Fig. 9A). Since only a little radioactivity was found in the feces during the entire period of time, nicotinic acid into the portal veins of mice, and radioactive compounds in the liver were analyzed as described tinder "Methods." Radioactivity found in the gastrointestinal tract 1 hour after the injection was analyzed by chromatography on a column of Dowex 1-X2 (formate), and the radioactivity found in the gastrointestinal tract 4 and 8 hours after the injection were analyzed by paper chromatography on Whatman No. 3 paper with n-butyl alcohol saturated with 15% ammonia as solvent. Each value represents the average of three mice. A, radioactive compounds found in the gastrointestinal tract after injection of nicotinamide-14C. B, radioactive compounds found in the gastrointestinal tract following administration of nicotinic acid-' 4 C. C, radioactive compounds found in the liver after administration of nicotinic acid-14C. * 0, total radioactivity; X---X, radioactivity recovered as nicotinic acid-'4C; O--O, NAD-' 4 C; *-·---, nicotinamide-14 C; --, sum of nicotinic acid ribonucleotide, deamido-NAD, and NAD. produced from nicotinamide may presumably be reabsorbed and utilized as precursor to NAD.
On the other hand, when nicotinic acid-14 C (82 molese, 4 105 cpm) was injected intraperitoneally, a larger portion of it than nicotinamide was excreted into the urine in confirmation of the previous results of Petrack, Greengard, and Kalinsky.l Radioactivity recovered in the gastrointestinal tract 1 hour after the injection was less than 10 % of the total radioactivity injected and that in the liver was about 4.5%. When nicotinic acid-14 C (41 ,moles, 8 X 105 cpm) was injected into the portal vein, about 27 % of the total radioactivity was found in the gastrointestinal tract, and about 7% was recovered in the liver 1 hour after the injection (Fig. 9, B and C). Radioactivity in the gastrointestinal tract as well as in the liver decreased gradually to a very low level during 8 hours. In terms of the specific radioactive compounds in the liver as well as in the gastrointestinal tract, the disappearance of nicotinic acid-14 C was almost parallel with the decrease of the total radioactivity. The formation of NAD-' 4 C in the liver was insignificant even after 8 hours, and neither nicotinic acid ribonucleotide-4C nor deamido-NAD-14C was detected to any appreciable extent in the liver, indicating that nicotinic acid supplied to the liver in large quantities does not serve as a good precursor to NAD.
Gastrointestinal Nicotinamide Deamidase-Available evidence indicated that nicotinamide deamidase is very active in the gastrointestinal tract. The enzyme has been detected in stomach mucosa, stomach contents, and cecum contents of rat and was partially purified from mucosal membrane of the bovine rumen. The IK, value of the enzyme for nicotinamide was approximately 6 X 10-6 M.2 Details of the experimental results on the properties as well as on the role of this deamidase during the biosynthesis of NAD from nicotinamide will be described in a following paper in this series. DISCUSSION The method employed in the present study involves the direct injection of 14 C-labeled substrates into the portal vein, followed by the analysis of the radioactive intermediates in the liver. The pulse-labeling technique in vivo has enabled us to follow the NAD biosynthesis from three different precursors in the liver under various conditions. This method may be comparable to perfusion in situ but is more simple, convenient, and reproducible. Only by the use of such techniques at several levels of organization, from the intact organism to isolated enzymes, have we been able to understand somewhat better the uniqueness of different organs and their interrelations in the over-all physiological transformations of nicotinic acid and nicotinamide.
When nicotinic acid-' 4 C was injected into the portal vein in small doses, nicotinic acid was converted rapidly and efficiently to NAD via nicotinic acid ribonucleotide and deamido-NAD. Within 1 min after the injection, more than 70% of the total radioactivity in the liver was recovered as nicotinic acid ribonucleotide, deamido-NAD, and NAD, and the radioactivity in NAD reached a maximum in about 10 min. On the other hand, when nicotinamide-14 C was similarly injected, the incorporation of radioactivity into NAD was much lower than that with nicotinic acid-14 C as substrate. Nicotinic acid, nicotinic acid ribonucleotide, and deamido-NAD did not accumulate in any detectable amount during the entire period after the injection. This suggests that the rapid incorporation of nicotinamide-4 C into liver NAD was probably due to the exchange reaction catalyzed by NAD nucleosidase (23) and that nicotinamide could not be utilized significantly by the liver cells for synthesis de novo of NAD under the specified conditions employed.
When a large dose of nicotinamide was injected, however, the liver NAD level increased markedly for a prolonged period of time. In contrast, injection of nicotinic acid in a large dose gave a much smaller rise in liver NAD than did injection of nicotinamide. This discrepancy has sometimes been attributed, at least in part, to the permeability difference of nicotinic acid and nicotinamide. Indeed, experiments with several ascites tumor cells by Dietrich and Ahuja demonstrated that at pH above 7.4, nicotinamide was more permeable to these cells than nicotinic acid (24). Our results of experiments in vivo, however, indicated that under the conditions employed, there was no significant difference of uptake by the liver cells between nicotinic acid and nicotinamide. These compounds were absorbed almost to the same extent by the liver cells regardless of the size of the dosage. Moreover, a large dose of nicotinic acid injected intraportally with nicotinamide-14 C did not affect the uptake of nicotinamide-14 C by the liver and vice versa.
When nicotinamide-14 C was administered in a large quantity and the radioactive compounds in the liver were analyzed, quite contrary to our expectation, nicotinamide was fairly quickly excreted from the liver; the total radioactivity in the liver decreased rather rapidly but started to increase after several hours. During the initial phase, the formation of NAD-' 4 C was almost negligible. The increase of the total radioactivity during the latter phase was accompanied by the formation of NAD-14 C. Small amounts of radioactive nicotinic acid, deamido-NAD, and NADP were also formed in this period. These results thus suggest that when a large dose of nicotinamide-' 4 C was injected into the portal vein, a large portion of it was excreted from the liver without being utilized as precursor of NAD, but it was later reabsorbed by the liver and synthesized to NAD. Although it is difficult to determine experimentally whether the nicotinamide-' 4 C returns to the liver in its original form or is converted to nicotinic acid or some other derivatives of nicotinamide, there is no reason to conceive that nicotinamide returns to liver as unchanged nicotinamide, since it is hard to believe that nicotinamide taken up by the liver shortly after the injection and that was later reabsorbed behaves entirely differently. A rather large portion of nicotinamide-' 4 C was found in the gastrointestinal tract 1 hour after the injection. Since an active nicotinamide deamidase has been detected in rat and bovine gastrointestinal tract, and since the enzyme partially purified from bovine rumen has a low Km, the diphasic nature of the curve shown in Fig. 6A may be best explained by the assumption that a large portion of nicotinamide-1 4 C administered to mice was excreted from the liver, stored in the gastrointestinal tract where it was deamidated to nicotinic acid, reabsorbed into the liver cells, and synthesized to NAD, although the role of nicotinamide deamidase in the liver cannot be rigorously excluded. This assumption is consistent with the fact that a small dose of nicotinic acid-' 4 C injected into the portal vein is synthesized efficiently to liver NAD (Fig. 3) and that most, if not all, of nicotinamide-'4C injected in large quantities is converted to NAD by way of nicotinic acid, nicotinic acid ribonucleotide, and deamido-NAD (Fig. 7). On the other hand, when a large dose of nicotinic acid-14 C was injected, nicotinic acid-'4C was absorbed by the liver cells almost to the same extent as nicotinamide 1 hour after the injection, but failed to serve as a good precursor to NAD. The rapid excretion of nicotinic acid-' 4 C into the urine may account, at least in part, for the fact that nicotinic acid in large quantities is much less effective than nicotinamide as a precursor to liver NAD, but it does not explain the fact that nicotinic acid supplied to the liver in small and large quantities behaves quite differently. Although high concentrations of nicotinic acid do not inhibit the nicotinic acid ribonucleotide pyrophosphorylase of beef liver in vitro (25), nicotinic acid in large quantities may exert some inhibitory effect on the biosynthesis of NAD from nicotinic acid in the liver. The mechanism of the inhibition should be explained by further investigations in vitro. Tryptophan may be a good precursor of NAD in the liver (26)(27)(28). However, quinolinic acid, a direct precursor of NAD, hardly penetrated the liver cell, presumably due to its strong polarity, and was a poorer precursor of NAD. The quantitative relationship of tryptophan and nicotinic acid in the biosynthesis of NAD in mammalian liver must be elucidated by further investigations.