Nucleoside Salvage Pathway for NAD Biosynthesis in - Journal of

Vol. 152, No. 3

JOURNAL OF BACTERIOLOGY, Dec. 1982, p. 1111-1116

0021-9193/82/121111-06S02.000/ Copyright C 1982, American Society for Microbiology

Nucleoside Salvage Pathway for NAD Biosynthesis in Salmonella typhimurium GSEPING LIU,1 JOHN FOSTER,2 PURITA MANLAPAZ-RAMOS,1 AND BALDOMERO M. OLIVERAl* Department ofBiology, University of Utah, Salt Lake City, Utah 841121 and Department of Microbiology, Marshall University, Huntington, West Virginia 257012 Received 8 June 1982/Accepted 31 August 1982

A previously undescribed nucleoside salvage pathway for NAD biosynthesis is defined in Salmonella typhimurium. Since neither nicotinamide nor nicotinic acid is an intermediate in this pathway, this second pyridine nucleotide salvage pathway is distinct from the classical Preiss-Handler pathway. The evidence indicates that the pathway is from nicotinamide ribonucleoside to nicotinamide mononucleotide (NMN) and then to nicotinic acid mononucleotide, followed by nicotinic acid adenine dinucleotide and NAD. The utilization of exogenous NMN for NAD biosynthesis has been reexamined, and in vivo evidence is provided that the intact NMN molecule traverses the membrane.

In the enteric bacteria Escherichia coli and Salmonella typhimurium, there are two wellestablished pathways for the synthesis of the pyridine nucleotides NAD and NADP (1, 2, 5, 13, 15, 16). The bacteria can synthesize these nucleotides de novo, starting with dihydroxyacetone phosphate and aspartate. In addition, however, a pathway exists which permits the salvage of the free bases (nicotinamide [Nm] and nicotinic acid [Na]). This salvage pathway, first elucidated in erythrocytes by Preiss and Handler (13), is shown below: Nm -Na -NaMN -NaAD -* NAD NADP where NaMN is nicotinic acid mononucleotide, and NaAD is nicotinic acid adenine dinucleotide. Recently, it was shown that S. typhimurium double mutants defective in both the de novo and the salvage pathway for biosynthesis of NAD can use nicotinamide mononucleotide (NMN) as a precursor for NAD (4). This result indicates the presence of a pathway permitting utilization of the nucleotide without conversion to the free base. In this report, we establish the presence of a second salvage pathway for pyridine nucleotide synthesis and define the metabolic steps involved. The results show that in addition to utilization of NMN, nicotinamide ribonucleoside can be a precursor for NAD. In vivo evidence is also provided that NMN uptake involves the intact phosphorylated molecule traversing the cellular membranes.

MATERIALS AND METHODS Materials. [3H]nicotinamide and [14C]NAD were purchased from Amersham Searle Corp. [14C]nicotinamide ribonucleoside and [14C]NMN were prepared from [14C]carbonyl NAD by treatment with venom phosphodiesterase and bacterial alkaline phosphatase (for the nicotinamide ribose). NAD and NMN were separated by chromatography on DEAE-Sephadex as previously described (8). Nicotinamide ribonucleoside was treated with bacterial alkaline phosphatase (BAPF from Worthington Diagnostics) and either purified on the paper chromatography system of Witholt (17) or used unpurified. All chemicals used were reagent grade. Bacterial strains and culture conditions. S. typhimurium LT2 and the mutants JF63 (nadB51 pncAl5 trpA49) and JF76 (pnuAl7 nadAS6 pncAl5 trpA49) have been previously described (10). The high-uptake mutant JF63A was obtained by growing S. typhimurium JF63 in 10' M NMN; the culture was plated, and a colony was isolated. This was shown to still retain the nadA and pncA markers. S. typhimurium TT6586 (pncBl85:: TnIO nadB32 purC7 proA46 ilv40S rfa461 Strrfla-56 Fim-) was used for some experiments. Enzymatic reactions. Reactions with alkaline phosphatase were carried out in a 0.2-mi mixture containing 0.05 M Tris (pH 7.6). Alkaline phosphatase (Worthington BAP-F; 5 FLl) was added, and the reaction mixture was incubated for 15 min at 37°C. A second equal aliquot of the phosphatase was added, and the rest was incubated at 37°C for an additional 15 min. Venom phosphodiesterase digestion (8) was carried out in a 0.5-ml reaction mixture containing 45 ,umol of Tris buffer (pH 8.6) and 0.45 LmoI of MgSO4 and incubated for 1 h at 37°C. The reaction was quenched by boiling in water for 3 min. Chromatography. Two chromatography systems were used routinely. The first involved DEAE paper (Whatman DE 81), with 0.25 M ammonium bicarbon-

1111

1112

J. BACTERIOL.

LIU ET AL.

ate as the developing solvent. Generally, these chromatograms were run for 4 to 5 h. The second paper chromatography system for analysis of pyridine nucleotides was the citrate ethanol system of Witholt (17). Other methods. Absorbance at 595 nm was measured in a Cary model 15 spectrophotometer. Cell number was determined with a Coulter Counter (model F). Chromatograms were analyzed by cutting 1-cm strips and determining radioactivity on a Beckman LS200 scintillation counter with a toluene-based scintillation fluid. Preparation of [3H, 32PJNMN. E. coli cells were grown either in [3H]nicotinic acid or 'pi as previously described (8). Cells were harvested and lysed, and the NAD was purified. [3H]NMN and [32P]NMN were prepared from the purified NAD by treatment with venom phosphodiesterase as described above. In vivo labeling protocol with [3H, 32P]NMN. A culture containing M9 minimal salts, plus 0.2% glucose and tryptophan, containing NMN (final concentration, 10-4 M) labeled with both [32p] in the phosphate moiety and [3H] in the nicotinamide moiety was inoculated with a culture of S. typhimurium J63A. The cells were allowed to grow for 8 h (with the growth of the culture being monitored at approximately hourly intervals); under these conditions, the generation time was 45 min. At t-he end of 8 h, the absorbance of the culture was 2.5 at 595rnm. A 1.5-ml amount of the cell culture was harvested and centrifuged, and the supernatant was saved. The cells were washed in nutrient broth, centrifuged again, and resuspended in 0.33 M HCI (0.2 ml); the lysed cells were left in an ice bucket overnight. The lysate was centrifuged, and the acid extract, along with the culture medium and a control culture medium (in which no bacteria had been introduced), was analyzed by chromatography on 3MM paper as described above. NADP, NAD, NMN, nicotinamide and nicotinic acid were added as markers. Another aliquot of the solutions was analyzed on DEAE paper, with 0.25 M ammonium bicarbonate as the developing solvent (results not shown). A portion of the acid extract was also neutralized and treated with alcohol dehydrogenase as described above on the DEAE paper system, with NAD and NADH as markers.

RESULTS Nicotinamide ribonucleoside, a precursor for NAD. In S. typhimurium JF63, a pncA nadB strain, both the de novo and the Preiss-Handler salvage pathways are defective: the pncA and nadB mutations cause blocks in nicotinamide deamidation and quinolinate synthesis, respectively. However, NMN was found to be a precursor for NAD in JF63. This result indicated that NMN is converted to NAD without nicotinamide as an intermediate (10). We tested whether nicotinamide ribonucleoside would also serve as an NAD precursor in these strains (Fig. 1). It is clear that JF63 cells can grow more rapidly with nicotinamide ribonucleoside than NMN under the experimental conditions used. Much higher concentrations of NMN are required to give the same cell density in the period tested (13 h). These results indicate that the nucleoside can

also be converted to NAD without first being converted to nicotinamide. To test for the possibility that deamidation took place at the nucleoside level, followed by glycosidic cleavage to nicotinic acid, we determined whether the ribonucleoside would support growth in nadB pncB strains (pncB mutants are unable to convert nicotinic acid to the corresponding nucleotide). This strain (TT6586) can grow in the presence of nicotinamide ribonucleoside, indicating that neither nicotinic acid nor nicotinamide is an obligatory intermediate in the utilization of nicotinamide riboside for the synthesis of NAD. The data above suggest that both the ribonucleoside and the ribonucleotide of nicotinamide can be utilized by a salvage pathway distinct from the Preiss-Handler pathway. A genetic locus, pnuA, affects NMN uptake: mutants in this locus are unable to grow on NMN (10). We tested whether the pnuA mutant (JF76) also fails to grow on nicotinamide ribonucleoside. The pnuA mutant grows normally on the nucleoside, but, as previously reported, fails to grow on the nucleotide (10-4 M). Conversion of nicotinamide ribonucleoside to NMN in vitro. To serve as a precursor for NAD, nicotinamide ribonucleoside might initially be converted to the mononucleotide. We have tested Salmonella lysates for their ability to convert the nucleoside to the nucleotide. Since nicotinamide ribonucleoside is also cleaved by Salmonella spp. to nicotinamide, the possibility that extracts form NMN with the free base as an intermediate must be considered, i.e., ribose- 1 -P

Nm-R

_Nm7\MN

PRPP Pi P-Pi where P is phosphate, Nm is nicotinamide, Pi is inorganic pyrophosphate, R is ribose, and PRPP is phosphoribosyl pyrophosphate. To evaluate whether any NMN could be formed by direct phosphorylation, the bacterial extract was incubated with [14C]nicotinamide ribonucleoside and ATP (no PRPP), and [3H]nicotinamide was added to the incubation mixture. As shown in Fig. 2, only the [14C]nicotinamide ribonucleoside was converted to NMN; there was no measurable conversion of [3H]nicotinamide to the mononucleotide. (Even when PRPP is included in the reaction mixture, there is no detectable production of [3H]NMN. However, [3H]nicotinic acid mononucleotide, which moves slightly behind NMN in the chromatographic system used, is

NAD BIOSYNTHESIS IN S. TYPHIMURIUM

VOL. 152, 1982

1113

I.0

0.8 L) CYr 0.6F

/

X

/

0.2 0

/

/

0.41

AF I

I

..I..I

-

I

a

-A-,A I

50

10

5

I

(

a....I

100

PM)

FIG. 1. Growth of nadB pncA strains of S. typhimurium on nicotinamide ribose and NMN. Media containing E minimal salts, 0.2% glucose, and 1 ,ug of tryptophan per ml, and the amount of NMN (triangles) or nicotinamide ribose (circles) indicated were inoculated either with S. typhimurium JF63 (nadA pncA) (closed symbols) or the high-uptake mutant, JF63A (open symbols). Cultures (2.5 ml) were inoculated with 0.05 ml of a single colony resuspended in 1 ml and then incubated at 37°C with a rotatory shaker (110 rpm). After 13 h, the absorbance at 595 nm (A595) was determined. Absorbance was also read after 46 h (not shown). An increase in cell density was generally observed between 13 and 46 h (for JF63A growing in 7.5 ,uM NMN, the absorbance increased from 0.01 to 0.74 in this period).

produced (results not shown). In the presence of PRPP, the [3H]nictinic acid mononucleotide and the barely detectable levels of [14C]nicotinic acid mononucleotide are presumably formed by the Preiss-Handler pathway.) We conclude that the NMN was produced by direct phosphorylation of the nucleoside. This experiment demonstrates the presence of a kinase activity in crude extracts of Salmonella which converts nicotinamide ribonucleoside to NMN. These results suggest the following pathway for the entry of nicotinamide ribonucleoside into the pyridine nucleotide pool of Salmonella: Nm-ribose

--

NMN -3

NaMN

--

NaAD

NAD It has previously been shown that NMN can be deamidated to nicotinic acid mononucleotide in E. coli, Salmonella spp., and other bacteria (3, 6-9, 11). The remaining steps proposed are well-established steps common to both the de novo and Preiss-Handler Pathway (5, 13). NMN uptake. A spontaneous mutant was isolated which grew more rapidly on NMN (here designated pnuB). In Fig. 1, the rate of growth of the high-efficiency-uptake mutant is compared in nicotinamide ribonucleoside and in NMN to

the original JF63 strain. The JF63 culture grows very slowly in 1i-0 M NMN, whereas the pnuB mutant uses 1i-0 M NMN efficiently. Both cultures grew at comparable rates in nicotinic acid and in nicotinamide ribonucleoside. We have taken advantage of the pnuB mutant, capable of relatively efficient NMN uptake, to distinguish between two alternative schemes for the utilization of NMN: I

NMN

NMN--

Nm-R

NmR

NaMNN---*NaAD--oNAD

'~celt membrane

'-.cell

membrane

NaAD NAD

To distinguish between these alternatives, a double-label experiment was performed. The

1114

LIU ET AL.

J. BACTERIOL.

i0

E

0~

I

2 NAD

I

NMN

0

O

20

0

30

40

cm FROM ORIGIN FIG. 2. S. typhimurium LT2 was grown in Penassay broth to an absorbance at 595 nm of 0.43. Cells were harvested by centrifugation by using an RC2B refrigerated centrifuge for 5 min at 7,000 rpm and lysed on cellophane disks (108 cells per cm2) by the method of Schaller et al. (14). The top panel shows a control incubation without bacterial extract; the lower panel is the experimental incubation. The cellophane disks were incubated for 15 min at 37°C on 0.05-ml drops containing 1 mM ATP, 20 ,M [14C]nicotinamide ribose (4.8 x 107 cpm/,Lmol), 0.52 ,uM [3H]nicotinamide (1.573 x 10"° cpm/,umol), 0.12 M MOPS (morpholine propanesulfonic acid) buffer (pH 7.4), 0.1 M KCI, 5 mM Mg2+, and 10-4 M EDTA (pH 7.2). After incubation, the drops were analyzed by chromatography on 3MM paper according to the method of Witholt (17). NAD, NMN, and nicotinamide were spotted as visual markers. After development, the nicotinamide marker was detected in fractions 32 to 36 and 33 to 37 in the control and experimental chromatograms, respectively. The [14C]nicotinamide ribose had been previously run with authentic nicotinamide ribose in the same chromatographic system, and coincidence was demonstrated. The paper was analyzed for radioactivity as described in the text. Closed circles, 14C radioactivity; open circles, 3H.

first mechanism predicts that the original phosphate on NMN will be incorporated into NAD; the second mechanism predicts that the original phosphate moiety will be lost. A culture was fed NMN labeled in both the nicotinamide (with 3H) and the phosphate (with 32p) moieties. The experimental results are shown in Fig. 3. Both labels were taken up by the cells. When a chromatographic analysis was made of the intracellular radioactivity, both the

3H and 32p migrated with the NAD marker in two different chromatographic systems. Further, when the NAD was reduced to NADH (thereby altering its chromatographic behavior) both 3H and 32p showed altered mobility at the NADH position. We therefore conclude that the phosphate moiety of NMN was efficiently incorporated into the NAD molecule, indicating that NMN had been transported intact across the membrane. If phosphate hydrolysis had oc-

NAD BIOSYNTHESIS IN S. TYPHIMURIUM

VOL. 152, 1982

Q~L

SUPERNATANT

10

1

il7

I0

o~~~~~~~~~~N Na

NADP NAD 1.0

NMN

Nm

1.62

I CELLS

0.5

-

k1.29

1.88$

1115

nucleotide and nicotinamide ribonucleoside are exogenous sources of the pyridine moiety of NAD and can be incorporated by a route distinct from the classical Preiss-Handler pathway. However, NMN may not be significant source of the pyridine ring under normal in vivo conditions. The concentrations of NMN required for rapid enough uptake to support growth in a nadstrain are unphysiological and unlikely to be encountered in vivo. In contrast, nicotinamide ribonucleoside can support growth at micromolar concentrations; thus, nucleoside uptake could contribute significantly to NAD biosynthesis. The data presented define the following salvage pathway: Nm-ribose -* NMN -* NaMN -* NaAD NAD All enzymes except nicotinamide ribonucleoside kinase have previously been detected in Salmonella spp. (15). The kinase has previously been described in eucaryotic cells (12), but this is the first evidence for its occurrence in a procaryote. The presence of a second salvage pathway in Salmonella spp. distinct from the Preiss-Handler pathway, is useful for the experimental manipulation of pyridine nucleotide metabolism in bacteria. This salvage pathway allows an alternative source of the pyridine ring if both the de novo pathway and the Preiss-Handler pathway are blocked by mutation. In addition, analogs of both the nucleoside and the free base can now be used to select for mutants in the geneti-

1.49

10

0

20

30

FROM ORIGIN FIG. 3. Intracellular fate of 13H, 32P]NMN in exponentially growing S. typhimurium JF63A. A culture of S. typhimurium JF63A was labeled with [3H, 32P]NMN, and the radioactivity was analyzed as described in the text. The results of chromatography on cm

3MM paper of the culture supernatant, the harvested cells, and a control medium (with no bacteria) are shown. Open circles, 32P radioactivity; closed circles, 3H. The numbers indicate the 32P/3H ratio for that fraction. Na, Nicotinic acid; Nm, nicotinamide.

Aspartate

+NA NAD

Dihydroxyacetone phosphate nadB

NaAD

nadA

CELL

Qa

-a

Cellular

Envelope

MEDIUM Qa

curred, the 32p label would have dispersed into all the phosphorylated intracellular metabolites and only a minor fraction retained in the NAD pool. These results show that, although NMN is a phosphorylated compound, transport occurs across the cell membrane. DISCUSSION The data above demonstrate that in the bacterium S. typhimurium, both nicotinamide mono-

I

nadC Na

I

NaMN -.NMN

.. Nm_-NmR :9 . ......

Nm NmR NMN Na FIG. 4. Biosynthetic pathways for NAD in S. typhimurium. Metabolic steps which have been demonstrated and known genetic loci affecting these metabolic steps are shown. The five specific precursors that can be fed to S. typhimurium for the pyridine ring of NAD and NADP are included. Some of the steps may occur in the periplasmic space or on the inner membrane. Abbreviations: Qa, quinolinate; Na, nicotinate; Nm, nicotinamide; NmR, nicotinamide ribonucleoside; NaMN, nicotinic acid mononucleotide; NaAD, nicotinic acid adenine dinucleotide.

1116

LIU ET AL.

cally undefined metabolic steps of NAD biosynthesis. Our results demonstrate that NMN is one of the few phosphorylated compounds that can traverse the cellular membrane. In the experiments shown in Fig. 3, exogenously added double-labeled 13H, 32P]NMN was incorporated essentially intact into NAD; indeed, the 3H/32P ratio remained relatively unchanged, even after 8 h of growth. This is additional evidence for the extreme stability of the NMN moiety of NAD in vivo in enteric bacteria. In E. coli, it has previously been shown that, although there is readily measurable NAD breakdown, the NMN moiety is almost perfectly recycled (8). The same stability must be true in this S. typhimurium strain under the growth conditions used. Two types of mutations have been isolated which affect NMN uptake, the pnuA mutants, which are blocked in NMN uptake (10), and the pnuB mutants, which utilize NMN more efficiently than does the wild type. Neither pnuA nor pnuB significantly affects nicotinamide ribose or nicotinic acid utilization. Our present prejudice is that the pnuA and pnuB loci are not specific for an NMN uptake system, but probably represent more generalized membrane mutations. Changes in the membrane could clearly change the rate of penetration by a phosphorylated compound such as NMN. Preliminary results of transduction experiments with a TnWO closely linked to pnuA indicate that pnuA and pnuB are distinct loci and suggest that the pnuB phenotype (accelerated NMN uptake) is expressed in apnuA pnuB double mutant. A genetic analysis of all pnu loci is in progress, and the results will be presented elsewhere (J. Foster, submitted for publication). A summary of the presently known pathways for NAD biosynthesis in S. typhimurium is given in Fig. 4, including the metabolic steps that have been elucidated in this paper for nucleoside and nucleotide salvage. LITERATURE CITED 1. Andreoll, A. J., T. W. Ideda, T. Nishizuka, and 0. HayaislI. 1963. Quinolinic acid: a precursor to nicotinamide adenine dinucleotide in Escherichia coli. Biochem. Biophys. Res. Commun. 12:92-97. 2. Chandler, J. L. R., R. K. Gholson, and T. A. Scott. 1970.

J. BACTERIOL. Studies on the de novo biosynthesis of nicotinamide adenine dinucleotide in Escherichia coli. I. Labeling patterns from precursors. Biochim. Biophys. Acta 222:523526. 3. Foster, J. W., and A. M. Baskowsky-Foster. 1980. Pyridine nucleotide cycle of Salmonella typhimurium. In vivo recycling of nicotinamide adenine dinucleotide. J. Bacteriol. 142:1032-1035. 4. Foster, J. W., D. M. Kinney, and A. G. Most. 1979. Pyridine nucleotide cycle of Salmonella typhimurium: isolation and characterization of pWcA, pncB, and pncC mutants and utilization of exogenous nicotinamide adenine dinucleotide. J. Bacteriol. 137:1165-1175. 5. Foster, J. W., and A. G. Moat. 1980. Nicotinamide adenine dinucleotide biosynthesis and pyridine nucleotide cycle metabolism in microbial systems. Microbiol. Rev. 44:83-105. 6. Fredmann, H. C. 1971. Preparation of nicotinic acid mononucleotide from nicotinamide mononucleotide by enzymatic deamidation. Methods Enzymol. 18B:192-197. 7. Fyfe, S. A., and H. C. Frledmann. 1969. Vitamin B-12 biosynthesis enzyme studies on the formation of the aglycosidic nucleotide precursor. J. Biol. Chem. 244:16591666. 8. Hillyard, D., M. Rechsteiner, P. Manlapaz-Ramos, J. S. Imperial, L. J. Cruz, and B. M. Olhvera. 1981. The pyridine nucleotide cycle. Studies in E. coli and the human cell line D98/AH2. J. Biol. Chem. 256:8491-8497. 9. Imal, T. 1973. Purification and properties of nicotinamide mononucleotide amidohydrolase from Azotobacter vinelandii. J. Biochem. (Tokyo) 73:139-153. 10. Kinney, D. M., J. W. Foster, and A. G. Moat. 1979. Pyridine nucleotide cycle of Salmonella typhimurium: in vitro demonstration of nicotinamide mononucleotide deamidase and characterization of pnuA mutants defective in nicotinamide mononucleotide transport. J. Bacteriol. 140:607-611. 11. Manlapaz-Fernandez, P., and B. M. Olivera. 1973. Pyridine nucleotide metabolism in Escherichia coli. IV. Turnover. J. Biol. Chem. 248:5067-5073. 12. Nlshizuka, Y., and 0. Hayaishl. 1971. Mammalian pyridine ribonucleotide phosphokinase. Methods Enzymol. 18B:141-144. 13. Prelss, J., and P. Handler. 1958. Biosynthesis of diphosphopyridine nucleotides. I. Identification of the intermediates. J. Biol. Chem. 194:269-278. 14. Schaller, H., B. Otto, V. Nusslein, J. Huf, R. Herrmann, and F. Bonhoeffer. 1972. Deoxyribonucleic acid replication in vitro. J. Mol. Biol. 63:183-200. 15. Suzuki, N., J. Carsdon, G. Griffith, and R. K. Gholson. 1973. Studies on the de novo biosynthesis of nicotinamide adenine dinucleotide in Escherichia coli. V. Properties of the quinolinic acid synthetase system. Biochim. Biophys. Acta 304:309-315. 16. Wicks, F. D., S. Sakakibara, R. K. Gholson, and T. A. Scott. 1977. The mode of condensation of aspartic acid and dihydroxyacetone phosphate in quinolinate synthesis in Escherichia coli. Biochim. Biophys. Acta S00:213-216. 17. Wltholt, B. 1971. A bioautographic procedure for detecting TPN, DPN, NMN, and NR. Methods Enzymol. 18B:813-816.