Cell, Vol. 117, 495–502, May 14, 2004, Copyright 2004 by Cell Press
`
`Discoveries of Nicotinamide Riboside as a Nutrient
`and Conserved NRK Genes Establish a Preiss-Handler
`Independent Route to NADⴙ in Fungi and Humans
`
`Pawel Bieganowski and Charles Brenner*
`Departments of Genetics and Biochemistry
`and the Norris Cotton Cancer Center
`Dartmouth Medical School
`Rubin 733–HB7937
`Lebanon, New Hampshire 03756
`
`Summary
`
`NADⴙ is essential for life in all organisms, both as
`a coenzyme for oxidoreductases and as a source of
`ADPribosyl groups used in various reactions, including
`those that retard aging in experimental systems. Nico-
`tinic acid and nicotinamide were defined as the vitamin
`precursors of NADⴙ in Elvehjem’s classic discoveries
`of the 1930s. The accepted view of eukaryotic NADⴙ
`biosynthesis, that all anabolism flows through nico-
`tinic acid mononucleotide, was challenged experi-
`mentally and revealed that nicotinamide riboside is an
`unanticipated NADⴙ precursor in yeast. Nicotinamide
`riboside kinases from yeast and humans essential for
`this pathway were identified and found to be highly
`specific for phosphorylation of nicotinamide riboside
`and the cancer drug tiazofurin. Nicotinamide riboside
`was discovered as a nutrient in milk, suggesting that
`nicotinamide riboside is a useful compound for eleva-
`tion of NADⴙ levels in humans.
`
`Introduction
`
`In 1938, the pioneering vitamin-hunter Conrad Elvehjem
`and his coworkers put dogs on a synthetic diet supple-
`mented with only the known B vitamins. When the dogs
`were near death and exhibited pellagra-like black
`tongue symptoms, the investigators fed the animals
`small-molecule fractions derived from liver. In this man-
`ner, nicotinic acid and nicotinamide, now collectively
`termed niacin, were identified as the “anti-black tongue
`factor” with essential nutritional activity (Elvehjem et al.,
`1938). Because niacins are the vitamin forms of nicotin-
`amide adenine dinucleotide (NAD⫹), and eukaryotes
`also synthesize NAD⫹ de novo via the kynurenine path-
`way from tryptophan (Krehl et al., 1945; Schutz and
`Feigelson, 1972), niacin supplementation prevents the
`pellagra that can occur in populations with a trypto-
`phan-poor diet. In 1958, Jack Preiss and Philip Handler
`determined that nicotinic acid is phosphoribosylated to
`nicotinic acid mononucleotide (NaMN), which is then
`adenylylated to form nicotinic acid adenine dinucleotide
`(NaAD), which in turn is amidated to form NAD⫹ (Preiss
`and Handler, 1958a, 1958b).
`NAD⫹ was initially characterized as a coenzyme for
`oxidoreductases. Though conversions between NAD⫹,
`NADH, NADP, and NADPH would not be accompanied
`by a loss of total coenzyme, it was discovered that NAD⫹
`is also turned over in cells for unknown purposes
`
`*Correspondence: charles.brenner@dartmouth.edu
`
`(Maayan, 1964). In 2000, it became clear that Sir2 and
`Sir2-related enzymes termed Sirtuins deacetylate lysine
`residues with consumption of an equivalent of NAD⫹
`and that this activity is required for Sir2 function as a
`transcriptional silencer (Imai et al., 2000). It was also
`demonstrated that the Preiss-Handler pathway is re-
`quired for normal silencing activity in vivo (Smith et al.,
`2000). NAD⫹-dependent deacetylation reactions are re-
`quired not only for alterations in gene expression but
`also for repression of ribosomal DNA recombination and
`extension of lifespan in response to calorie restriction
`(Lin et al., 2000, 2002). NAD⫹ is consumed by Sir2 to
`produce a mixture of 2⬘- and 3⬘ O-acetylated ADPribose
`plus nicotinamide and the deacetylated polypeptide
`(Sauve et al., 2001). Additional enzymes, including poly
`(ADPribose) polymerases and cADPribose synthases
`are also NAD⫹-dependent and produce nicotinamide
`and ADPribosyl products (Ziegler, 2000; Burkle, 2001).
`Interest in the noncoenzymatic properties of NAD⫹
`has rekindled interest in NAD⫹ biosynthesis. In the last
`two years, at least five publications have schematized
`NAD⫹ biosynthesis in yeast as shown in Figure 1 (Pa-
`nozzo et al., 2002; Sandmeier et al., 2002; Bitterman et
`al., 2002; Anderson et al., 2003; Gallo et al., 2004). This
`scheme depicts a convergence of the flux to NAD⫹ from
`de novo synthesis, nicotinic acid import, and nicotin-
`amide salvage at NaMN. This is somewhat surprising
`because biochemists from 1950 to the present day have
`been characterizing the Nma1 and Nma2 gene products
`from yeast and their human homologs as NMN adenylyl-
`transferases (Kornberg, 1950; Emanuelli et al., 1999,
`2003; Garavaglia et al., 2002) or as dual specificity en-
`zymes that will use either NaMN or NMN as a substrate
`(Zhou et al., 2002).
`In this study, we show that nicotinamide riboside,
`which was known to be an NAD⫹ precursor in bacteria
`such as Haemophilus influenza (Gingrich and Schlenk,
`1944; Leder and Handler, 1951; Shifrine and Biberstein,
`1960) that lack the enzymes of the de novo and Preiss-
`Handler pathways (Fleischmann et al., 1995), is an NAD⫹
`precursor in a previously unknown but apparently con-
`served eukaryotic NAD⫹ biosynthetic pathway. We iden-
`tify yeast nicotinamide riboside kinase, Nrk1, and both
`human Nrk enzymes and demonstrate their specific
`functions in NAD⫹ metabolism biochemically and genet-
`ically. Their specificity suggests additionally that they
`are the long-sought tiazofurin kinases that perform the
`first step in converting cancer drugs such as tiazofurin
`and benzamide riboside into toxic NAD⫹ analogs
`(Cooney et al., 1983). Finally, we utilize yeast mutants
`of defined genotype to hunt for vitamins in a pathway-
`specific manner and show that milk is a source of nico-
`tinamide riboside.
`
`Results
`
`Recently, we characterized the S. cerevisiae QNS1 gene
`encoding glutamine-dependent NAD⫹ synthetase and
`showed that mutation of either the glutaminase active
`
`THORNE - EXHIBIT 1008
`
`

`

`Cell
`496
`
`site or the NAD⫹ synthetase active site resulted in invia-
`ble cells (Bieganowski et al., 2003). Possession of strains
`containing the qns1 deletion and a plasmid-borne QNS1
`gene allowed us to test whether the canonical de novo,
`import, and salvage pathways for NAD⫹ as depicted in
`Figure 1 (Panozzo et al., 2002; Sandmeier et al., 2002;
`Bitterman et al., 2002; Anderson et al., 2003; Gallo et
`al., 2004) are a complete representation of the metabolic
`pathways to NAD⫹ in S. cerevisiae. This scheme makes
`two specific predictions. First, because nicotinamide is
`deamidated to nicotinic acid before the pyridine ring is
`salvaged to make more NAD⫹, supplementation with
`nicotinamide should not
`rescue qns1 mutants by
`shunting nicotinamide-containing precursors through
`the pathway. Second, because QNS1 is common to the
`three pathways, there should be no NAD⫹ precursor that
`rescues qns1 mutants. As shown in Figure 2A, consis-
`tent with the scheme’s implicit assumptions about nico-
`tinamide metabolism, nicotinamide does not rescue
`qns1 mutants even at 1 or 10 mM. However, apart from
`any intermediate or enzymatic transformation depicted
`in the scheme, nicotinamide riboside, which was charac-
`terized as an NAD⫹ precursor in bacteria such as
`Haemophilus influenza (Gingrich and Schlenk, 1944;
`Leder and Handler, 1951; Shifrine and Biberstein, 1960),
`functions as a vitamin form of NAD⫹ at 10 ␮M.
`Two independent lines of evidence suggest that a
`pathway to NAD⫹ from nicotinamide riboside may exist
`in fungi and other eukaryotes. First, though the recent
`literature (Panozzo et al., 2002; Sandmeier et al., 2002;
`Bitterman et al., 2002; Anderson et al., 2003; Gallo et
`al., 2004) depicts all of the metabolic flux through NaMN,
`the enzymes placed on this scheme as NaMN adenylyl-
`transferases were initially classified as NMN adenylyl-
`transferases (Kornberg, 1950; Emanuelli et al., 1999,
`2003; Garavaglia et al., 2002) or as dual specificity NMN/
`NaMN adenylyltransferases (Zhou et al., 2002). Though
`it is possible that eukaryotic adenylyltransferases were
`misclassified, it is reasonable to ask whether eukaryotes
`have a pathway that produces NMN to feed into these
`enzymes.
`
`Figure 1. Three Known Biosynthetic Routes
`to NAD⫹ in S. cerevisiae
`NAD⫹ is synthesized from the de novo kynur-
`enine pathway that originates with trypto-
`phan using the Bna1-Bna6 gene products, an
`import pathway that originates with nicotinic
`acid, and a salvage pathway that utilizes nico-
`tinamide produced as a function of Sir2-
`related lysine deacetylases. According to
`this scheme, nicotinic acid mononucleotide
`(NaMN) is common to all three pathways and
`Qns1, the glutamine-dependent NAD⫹ syn-
`thetase that converts nicotinic acid adenine
`dinucleotide (NaAD) to nicotinamide adenine
`dinucleotide (NAD⫹), is required for all path-
`ways (Panozzo et al., 2002; Sandmeier et al.,
`2002; Bitterman et al., 2002; Anderson et al.,
`2003; Gallo et al., 2004). Note that Nma1 and
`Nma2 are schematized as NaMN adenylyl-
`transferases, though they were initially char-
`acterized as nicotinamide mononucleotide
`(NMN) adenylyltransferases (Kornberg, 1950;
`Emanuelli et al., 1999, 2003).
`
`Figure 2. Nicotinamide Riboside Allows NAD⫹ Synthetase-Indepen-
`dent Growth Via a Novel Pathway
`(A) qns1 cells transformed with a plasmid carrying the QNS1 and
`URA3 genes were tested for growth on synthetic dextrose complete
`media and 5-fluoroorotic acid (5-FOA). NAD⫹ synthetase mutant
`qns1 is inviable without supplements or with 1 mM nicotinamide
`but is rescued by 10 ␮M nicotinamide riboside.
`(B) Strains with indicated mutations were plated on medium supple-
`mented with 5-FOA and 10 ␮M nicotinamide riboside. Deletion of
`genes for uridine/cytosine kinase, adenosine kinase, and ribokinase
`do not alter the ability of yeast cells to utilize nicotinamide riboside.
`
`

`

`Eukaryotic Nicotinamide Riboside Kinase Pathway
`497
`
`Figure 3. Nicotinamide Riboside and its Analogs, Tiazofurin and
`Benzamide Riboside
`Nicotinamide riboside, first reported as the lowest molecular weight
`“V factor” for Haemophilus influenza (Gingrich and Schlenk, 1944),
`is shown alongside the IMP dehydrogenase prodrugs, tiazofurin
`(Cooney et al., 1983), and benzamide riboside (Krohn et al., 1992).
`NMN and its analogs are produced via phosphorylation of the 5⬘
`hydroxyl groups.
`
`Second, anticancer agents such as tiazofurin (Cooney
`et al., 1983) and benzamide riboside (Krohn et al., 1992)
`have been shown to be metabolized intracellularly to
`NAD⫹ analogs tiazofurin adenine dinucleotide and ben-
`zamide adenine dinucleotide, which inhibit IMP de-
`hydrogenase, the rate-limiting enzyme for guanine nu-
`cleotide biosynthesis. As these compounds can be
`considered analogs of nicotinamide riboside (Figure 3),
`generation of NAD⫹ analogs would necessarily involve
`phosphorylation of the 5⬘ hydroxyl group and subse-
`quent adenylylation of these intermediates. Though an
`NMN/NaMN adenylytransferase is thought to be the en-
`zyme that converts the mononucleotide intermediates
`to NAD⫹ analogs and the structural basis for this has
`
`been established (Zhou et al., 2002), several different
`enzymes including adenosine kinase, 5⬘ nucleotidase
`(Fridland et al., 1986; Saunders et al., 1990) and a spe-
`cific nicotinamide riboside kinase (Saunders et al., 1990)
`have been proposed to be responsible for tiazofurin
`phosphorylation in vivo. Indeed, a putative nicotinamide
`riboside kinase (Nrk) activity was reportedly purified but
`no amino acid sequence information was obtained and,
`as a consequence, no genetic test was ever performed
`to assess its function in nutrient or drug metabolism
`(Sasiak and Saunders, 1996).
`To test whether any of the nucleoside kinases pro-
`posed to phosphorylate tiazofurin are uniquely or collec-
`tively responsible for utilization of nicotinamide riboside,
`we prepared a qns1 deletion strain that was additionally
`deleted for all of the candidate genes for which yeast
`homologs exist, namely adenosine kinase ado1 (Lecoq
`et al., 2001), uridine/cytidine kinase urk1 (Kern, 1990;
`Kurtz et al., 1999), and ribokinase rbk1 (Thierry et al.,
`1990). As shown in Figure 2B, despite these deletions,
`the strain retained the ability to utilize nicotinamide ribo-
`side in an anabolic pathway independent of NAD⫹ syn-
`thetase. Given that mammalian pharmacology provided
`no useful clue to the identity of a putative fungal Nrk,
`we considered whether the gene might have been con-
`served with the Nrk of H. influenza. The Nrk domain of
`H. influenza is encoded by amino acids 225 to 421 of
`the NadR gene product (the amino terminus of which is
`NMN adenylyltransferase). Though this domain is struc-
`turally similar to yeast thymidylate kinase (Singh et al.,
`2002), sensitive sequence searches (not shown) re-
`vealed that bacterial Nrk has no ortholog in yeast. In-
`deed, genomic searches with the Nrk domain of H. influ-
`enza NadR have identified a growing list of bacterial
`genomes predicted to utilize nicotinamide riboside as
`an NAD⫹ precursor (Kurnasov et al., 2002). Thus, had
`fungi possessed NadR Nrk-homologous domains, com-
`parative genomics would have already predicted that
`yeast can salvage nicotinamide riboside.
`To identify the Nrk of S. cerevisiae, we established an
`HPLC assay for the enzymatic activity and utilized a
`biochemical genomics approach to screen for the gene
`encoding this activity (Martzen et al., 1999). Sixty-four
`pools of 90–96 S. cerevisiae open reading frames fused
`to glutathione S-transferase (GST), expressed in S. cere-
`visiae, were purified as GST fusions and screened for
`the ability to convert nicotinamide riboside plus ATP to
`NMN plus ADP. As shown in Figure 4A, whereas most
`pools contained activities that consumed some of the
`input ATP, only pool 37 consumed nicotinamide riboside
`and produced NMN. Examination of the 94 open reading
`frames that were used to generate pool 37 revealed that
`YNL129W encodes a predicted 240 amino acid polypep-
`tide with a 187 amino acid segment containing 23%
`identity with the 501 amino acid yeast uridine/cytidine
`kinase Urk1 and remote similarity with a segment of
`E. coli pantothenate kinase panK (Yun et al., 2000) (Fig-
`ure 4B). Cloning of YNL129W into a bacterial expression
`vector allowed us to test the hypothesis that this homo-
`log of metabolite kinases is the eukaryotic Nrk. Strik-
`ingly, the specific activity of purified YNL129W was
`ⵑ100 times that of pool 37, consistent with the idea that
`all the Nrk activity of pool 37 was encoded by this open
`reading frame. To test genetically whether this gene
`
`

`

`Cell
`498
`
`Figure 4. Cloning and Genetic Validation of Yeast and Human Nicotinamide Riboside Kinases
`(A) HPLC traces of a negative pool (36) and positive pool (37) of GST-ORF fusions. In pool 36, some ATP was converted to ADP and production
`of AMP occurred in several pools including 37. In pool 37, approximately half of the 1 mM ATP was converted to ADP and the 500 ␮M Nr
`peak was almost entirely converted to NMN.
`(B) Amino acid sequence alignment of human Nrk1, human Nrk2, S. cerevisiae Nrk1, S. pombe nrk1, and portions of S. cerevisiae uridine/
`cytidine kinase Urk1 and E. coli pantothenate kinase.
`(C) nrk1 deletion introduced into a qns1 deletion strain blocks the ability to form colonies in the presence of nicotinamide riboside. Expression
`of human NRK1 or human NRK2 cDNAs restores growth to qns1 nrk1 deletion strains, indicating that human Nrk1 and Nrk2 are authentic
`nicotinamide riboside kinases in vivo.
`
`product phosphorylates nicotinamide riboside in vivo,
`we created a deletion of YNL129W in the qns1 back-
`ground and found that nicotinamide riboside rescue of
`the qns1 deletion strain is entirely dependent on this
`gene product (Figure 4C). Having shown biochemically
`and genetically that YNL129W encodes an authentic Nrk
`activity, we named this gene NRK1.
`We ran PSI-BLAST (Altschul et al., 1997) on the pre-
`dicted S. cerevisiae Nrk1 polypeptide and discovered
`the apparent orthologous human protein Nrk1 (locus
`NP_060351) encoded at 9q21.31, encoding a polypep-
`tide of 199 amino acids annotated as an uncharacterized
`protein of the uridine kinase family. In addition, we found
`
`a second human gene product Nrk2 (locus NP_733778)
`that is 57% identical to human Nrk1. Nrk2 is a 230 amino
`acid splice form of what was described as a 186 amino
`acid muscle integrin ␤1 binding protein (ITGB1BP3) en-
`coded at 19p13.3 (Li et al., 1999, 2003). Amino acid
`conservation between S. cerevisiae, S. pombe, and hu-
`man Nrk homologs and similarity with fragments of
`S. cerevisiae Urk1 and E. coli panK is shown in Figure
`4B. As shown in Figure 4C, complementation of the
`failure of qns1 nrk1 to grow on nicotinamide riboside-
`supplemented media was provided by human NRK1 and
`human NRK2 cDNAs expressed from the yeast GAL1
`promoter.
`
`Table 1. Specific Activity (nmole mg⫺1 min⫺1) of human Nrk1, Nrk2, and yeast Nrk1 for Phosphorylation of Nucleoside Substrates
`
`Nicotinamide riboside
`
`Tiazofurin
`
`Uridine
`
`Human Nrk1
`Human Nrk2
`Yeast Nrk1
`
`275 ⫾ 177
`2320 ⫾ 200
`535 ⫾ 600
`
`538 ⫾ 277
`2150 ⫾ 2100
`1129 ⫾ 1344
`
`19.3 ⫾ 1.77
`2220 ⫾ 170
`15.2 ⫾ 3.44
`
`Cytidine
`
`35.5 ⫾ 6.44
`222 ⫾ 88
`82.9 ⫾ 4.44
`
`

`

`Eukaryotic Nicotinamide Riboside Kinase Pathway
`499
`
`Figure 5. Nicotinamide Riboside Is Present in an Acid Whey Vitamin
`Fraction of Cow’s Milk
`Yeast qns1 mutants grow when supplemented with whey in an
`NRK1-dependent manner, indicating that whey is a source of nico-
`tinamide riboside.
`
`Purification of yeast Nrk1 and human Nrk1 and Nrk2
`revealed high specificity for phosphorylation of nicotin-
`amide riboside and tiazofurin (Table 1). In the cases
`of yeast and human Nrk1, the enzymes actually prefer
`tiazofurin to the natural substrate nicotinamide riboside
`by a factor of two and both enzymes retain less than
`7% of their maximal specific activity on uridine and cyti-
`dine. In the case of human Nrk2, the 186 amino acid
`integrin ␤1 binding protein form is devoid of enzymatic
`activity (data not shown). On the other hand, the 230
`amino acid form is essentially equally active on nicotin-
`amide riboside, tiazofurin, and uridine with less than
`10% of corresponding activity on cytidine. Thus, though
`Nrk2 may contribute additionally to formation of uridy-
`late in the tissues in which it is expressed, these data
`demonstrate that fungi and mammals possess specific
`Nrks that function to synthesize NAD⫹ through NMN in
`addition to the well-known pathways through NaMN.
`Identification of Nrk enzymatic activities thus accounts
`for the dual specificity of fungal and mammalian NaMN/
`NMN adenylyltransferases.
`We used the yeast qns1 mutant to screen for natural
`sources of nicotinamide riboside and, as shown in Fig-
`ure 5, we found it in a vitamin fraction of cow’s milk.
`Unlike the original screen for vitamins in protein-depleted
`extracts of liver for reversal of black tongue in starving
`dogs (Elvehjem et al., 1938), this assay is pathway-spe-
`cific in identifying NAD⫹ precursors. As shown in Figures
`1 and 2, because of the qns1 deletion, nicotinic acid
`and nicotinamide do not score positively in this assay.
`Because the factor from milk requires nicotinamide ribo-
`
`side kinase for growth, the nutrient is clearly nicotin-
`amide riboside and not NMN or NAD⫹.
`
`Discussion
`
`A revised metabolic scheme for NAD⫹, incorporating
`Nrk1 homologs and the nicotinamide riboside salvage
`pathway is shown in Figure 6. A little appreciated dif-
`ference between humans and yeasts concerns the or-
`ganisms’ uses of nicotinamide and nicotinic acid, the
`compounds coidentified as anti black tongue factor (El-
`vehjem et al., 1938). Humans encode a homolog of the
`Haemophilus ducreyi nadV gene, termed pre-B-cell col-
`ony enhancing factor, that may convert nicotinamide to
`NMN (Rongvaux et al., 2002), which is highly induced
`during lymphocyte activation (Samal et al., 1994). In con-
`trast, S. cerevisiae lacks a homolog of nadV and instead
`has a homolog of the E. coli pncA gene, termed PNC1,
`that converts nicotinamide to nicotinic acid for entry
`into the Preiss-Handler pathway (Ghislain et al., 2002;
`Sandmeier et al., 2002). Though the Preiss-Handler path-
`way is frequently considered a salvage pathway from
`nicotinamide, it technically refers to the steps from nico-
`tinic acid to NAD⫹ (Preiss and Handler, 1958a, 1958b).
`Reports that nicotinamidase had been purified from
`mammalian liver in the 1960s (Petrack et al., 1965) may
`have contributed to the sense that fungal and animal
`NAD⫹ biosynthesis is entirely conserved. However, ani-
`mal genes for nicotinamidase have not been identified
`and there is no compelling evidence that nicotinamide
`and nicotinic acid are utilized as NAD⫹ precursors
`through the same route in mammals. The persistence
`of “niacin” as a mixture of nicotinamide and nicotinic
`acid may attest to the utility of utilizing multiple path-
`ways to generate NAD⫹ and suggests that supplementa-
`tion with nicotinamide riboside as third importable NAD⫹
`precursor may be beneficial for certain conditions.
`Whereas nicotinamide metabolism is not conserved
`between vertebrates and fungi, presence of NRK genes
`suggests that the nicotinamide riboside kinase pathway
`is conserved more broadly in eukaryotes, though it is
`likely to be time and tissue-restricted in animals.
`First reported in 1955, high doses of nicotinic acid are
`effective at reducing cholesterol levels (Altschul et al.,
`1955). Since the initial report, many controlled clinical
`studies have shown that nicotinic acid preparations,
`alone and in combination with HMG co-A reductase
`inhibitors, are effective in controlling low-density lipo-
`
`Figure 6. NAD⫹ Metabolism in Humans and
`Yeast
`A revised scheme for NAD⫹ metabolism in
`which double arrows depict steps common to
`yeast and humans (with yeast gene names).
`Single arrows depict steps unique to humans
`(PBEF, nicotinamide phosphoribosyltransfer-
`ase) and yeast (Pnc1, nicotinamidase). Addi-
`tional gene products are responsible for
`some of the steps. For example, there are
`multiple Sir2-related lysine deacetylases in
`humans and yeast and additional NAD⫹ glycohydrolases in humans along the arrow marked Sir2, and there are two Nrk enzymes in humans
`along the arrow marked Nrk1. The scheme remains incomplete in that there are enzyme and E.C. names for some additional possible steps
`in NAD⫹ biosynthesis, particularly for breakdown of NAD⫹ to nicotinamide riboside (Magni et al., 2004), that have not been genetically validated.
`
`

`

`Cell
`500
`
`protein cholesterol, increasing high-density lipoprotein
`cholesterol, and reducing triglyceride and lipoprotein a
`levels in humans (Pasternak et al., 1996). Though nico-
`tinic acid treatment effects all of the key lipids in the
`desirable direction and has been shown to reduce mor-
`tality in target populations (Pasternak et al., 1996), its use
`is limited because of a side effect of heat and redness
`termed “flushing,” which is significantly effected by the
`nature of formulation (Capuzzi et al., 2000). Thus, it will
`be important to test whether nicotinamide riboside sup-
`plementation is a preferred route to improve lipid profiles
`in humans. Additionally, study of the expression and
`regulation of NAD⫹ biosynthetic enzymes is expected
`to reveal approaches to sensitize tumors to compounds
`such as tiazofurin, to protect normal tissues from the
`toxicity of compounds such as tiazofurin adenine dinu-
`cleotide, and to stratify patients for the most judicious
`use of tiazofurin chemotherapy.
`
`Experimental Procedures
`
`S. cerevisiae Strains
`Yeast diploid strain BY165, heterozygous for qns1 deletion and hap-
`loid BY165-1d carrying a chromosomal deletion of qns1 gene, trans-
`formed with plasmid pB175 containing QNS1 and URA3 were de-
`scribed previously (Bieganowski et al., 2003). Genetic deletions were
`introduced by direct transformation with PCR products (Brachmann
`et al., 1998) generated from primers listed in Supplemental Data
`available at http://www.cell.com/cgi/content/full/117/4/495/DC1.
`Plating on media containing 5-fluoroorotic acid (Boeke et al., 1987)
`was after growth for 24 hr on complete media. The ado1 disruption
`cassette was constructed by PCR with primers 7041 and 7044 and
`plasmid pRS413 as a template. Yeast strain BY165 was transformed
`with this PCR product, and homologous recombination in histidine
`prototrophic transformants was confirmed by PCR with primers
`7042 and 7043. This strain was transformed with plasmid pB175 and
`subjected to sporulation and tetrad dissection. One of the resulting
`haploids carrying qns1 and ado1 deletions and plasmid was se-
`lected for further experiments and named BY237. The urk1 deletion
`was introduced into strain BY237 by transformation with the product
`of the PCR amplification that used pRS415 as a template and prim-
`ers 7051 and 7052. Disruption was confirmed by PCR with primers
`7053 and 7054, and the resulting strain was named BY247. The rbk1
`disruption cassette was constructed by PCR with primers 7063 and
`7065 and plasmid pRS411 as a template. Disruption was introduced
`into strain BY242 by transformation with the product of this reaction
`and confirmed by PCR with primers 7062 and 7064. The resulting
`strain, carrying deletions of qns1, ado1, urk1, and rbk1 genes was
`named BY252. A qns1 yeast strain carrying disruption of the nrk1
`locus was made by transformation of strain BY165-1d with an
`nrk1⌬::HIS3 cassette generated by PCR with primers 4750 and 4751
`and plasmid pRS413 as template. Correct integration of the HIS3
`marker into the nrk1 locus was confirmed by PCR with primers 4752
`and 4753.
`
`Nicotinamide Riboside and Whey Preparations
`NAD⫹ (Sigma) concentration was determined by conversion to
`NADH with alcohol dehydrogenase using an absorption coefficient
`(340 nm) of 6200 cm⫺1 M⫺1. The concentration of NMN was deter-
`mined by converting NAD⫹ to NMN plus AMP with rattlesnake venom
`NAD⫹ pyrophosphatase (E.C. 3.6.9.1, Sigma). Using 15,400 cm⫺1
`M⫺1 as the absorption coefficient for AMP at 259 nm, we used
`relative peak areas to calculate the absorption coefficient (259 nm)
`of NMN to be 4740 cm⫺1 M⫺1. To prepare nicotinamide riboside,
`120 ␮mol NMN (Sigma, concentration corrected by absorption) was
`treated with 1250 units of calf intestinal alkaline phosphatase
`(Sigma) for 1 hr at 37⬚C in 1 ml 100 mM NaCl, 20 mM Tris [pH 8.0],
`5 mM MgCl2. After hydrolysis of NMN to nicotinamide riboside was
`verified by HPLC, phosphatase was removed by centrifuging the
`reaction through a 5000 Da filter (Millipore). A whey vitamin fraction
`
`of commercial nonfat cow’s milk was prepared by adjusting the pH
`to 4 with HCl, stirring at 55⬚C for 10 min, removal of denatured
`casein by centrifugation, and passage through a 5000 Da filter. In
`yeast media, nicotinamide riboside was used at 10 ␮M and whey
`vitamin fraction at 50% by volume.
`
`Yeast GST-ORF Library
`Preparation of the fusion protein library was as described (Martzen
`et al., 1999; Phizicky et al., 2002) at a 500 ml culture scale for each
`of the 64 pools of 90–96 protein constructs. 10% of each pool
`preparation was assayed for Nrk activity in overnight incubations.
`
`Nicotinamide Riboside Phosphorylation Assays
`Reactions (0.2 ml), containing 100 mM NaCl, 20 mM Na HEPES [pH
`7.2], 5 mM ␤-mercaptoethanol, 1 mM ATP, 5 mM MgCl2, and 500 ␮M
`nicotinamide riboside or alternate nucleoside were incubated at
`30⬚C and terminated by addition of EDTA to 20 mM and heating for
`2 min at 100⬚C. Specific activity assays, containing 50 ng to 6 ␮g
`enzyme depending on the enzyme and substrate, were incubated
`for 30 min at 30⬚C to maintain initial rate conditions. Reaction prod-
`ucts were analyzed by HPLC on a strong anion exchange column
`with a 10 mM to 750 mM gradient of KPO4 [pH 2.6].
`
`NRK Gene and cDNA Cloning and Enzyme Purification
`The S. cerevisiae NRK1 gene was amplified from total yeast DNA
`with primers 7448 and 7449. The amplified DNA fragment was cloned
`in vector pSGA04 (Ghosh and Lowenstein, 1997) for E. coli expres-
`sion using restriction sites for NdeI and XhoI included in primer
`sequences and the resulting plasmid was named pB446. Samples
`of cDNA made from human lymphocytes and spleen were used as
`a template for amplification of human NRK1 performed with primers
`4754 and 4755. Plasmid pB449 was created by cloning of the prod-
`uct of this reaction between restriction sites NcoI and BamHI of
`vector pMR103 (Munson et al., 1994) for E. coli expression. Plasmid
`pB449 was used as a template for PCR reaction with primers 7769
`and 7770. The product of this amplification was cloned between
`BamHI and XhoI sites of vector p425GAL1 (Mumberg et al., 1994)
`and the resulting plasmid carrying human NRK1 gene under GAL1
`promoter control was named pB450. Human NRK2 cDNA was ampli-
`fied with primers 7777 and 7776. The amplified fragment was di-
`gested with NdeI and XhoI enzymes and cloned in plasmid pSGA04
`for E. coli expression (pB457) and into plasmid p425GAL1 for yeast
`expression (pB459). His-tagged enzymes were purified by immobi-
`lized cobalt affinity chromatography (Becton Dickenson).
`
`Acknowledgments
`
`This research was supported by National Cancer Institute grant
`CA77738.
`
`Received: January 22, 2004
`Revised: March 19, 2004
`Accepted: March 19, 2004
`Published: May 13, 2004
`
`References
`
`Altschul, R., Hoffer, A., and Stephen, J.D. (1955). Influence of nico-
`tinic acid on cholesterol
`in man. Arch. Biochem. Biophys. 54,
`558–559.
`Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z.,
`Miller, W., and Lipman, D.J. (1997). Gapped BLAST and PSI-BLAST:
`a new generation of protein database search programs. Nucleic
`Acids Res. 25, 3389–3402.
`Anderson, R.M., Bitterman, K.J., Wood, J.G., Medvedik, O., and
`Sinclair, D.A. (2003). Nicotinamide and PNC1 govern lifespan exten-
`sion by calorie restriction in Saccharomyces cerevisiae. Nature
`423, 181–185.
`Bieganowski, P., Pace, H.C., and Brenner, C. (2003). Eukaryotic
`NAD⫹ synthetase Qns1 contains an essential, obligate intramolecu-
`lar thiol glutamine amidotransferase domain related to nitrilase. J.
`Biol. Chem. 278, 33049–33055.
`
`

`

`Eukaryotic Nicotinamide Riboside Kinase Pathway
`501
`
`Bitterman, K.J., Anderson, R.M., Cohen, H.Y., Latorre-Esteves, M.,
`and Sinclair, D.A. (2002). Inhibition of silencing and accelerated
`aging by nicotinamide, a putative negative regulator of yeast Sir2
`and human SIRT1. J. Biol. Chem. 277, 45099–45107.
`Boeke, J.D., Trueheart, J., Natsoulis, G., and Fink, G.R. (1987).
`5-Fluoroorotic acid as a selective agent in yeast molecular genetics.
`Methods Enzymol. 154, 164–175.
`Brachmann, C.B., Davies, A., Cost, G.J., Caputo, E., Li, J., Hieter,
`P., and Boeke, J.D. (1998). Designer deletion strains derived from
`Saccharomyces cerevisiae S288C: a useful set of strains and plas-
`mids for PCR-mediated gene disruption and other applications.
`Yeast 14, 115–132.
`Burkle, A. (2001). Physiology and pathophysiology of poly(ADP-
`ribosyl)ation. Bioessays 23, 795–806.
`Capuzzi, D.M., Morgan, J.M., Brusco, O.A., Jr., and Intenzo, C.M.
`(2000). Niacin dosing: relationship to benefits and adverse effects.
`Curr. Atheroscler. Rep. 2, 64–71.
`Cooney, D.A., Jayaram, H.N., Glazer, R.I., Kelley, J.A., Marquez, V.E.,
`Gebeyehu, G., Van Cott, A.C., Zwelling, L.A., and Johns, D.G. (1983).
`Studies on the mechanism of action of tiazofurin metabolism to an
`analog of NAD with potent IMP dehydrogenase-inhibitory activity.
`Adv. Enzyme Regul. 21, 271–303.
`Elvehjem, C.A., Madden, R.J., Strong, F.M., and Woolley, D.W.
`(1938). The isolation and identification of the anti-black tongue fac-
`tor. J. Biol. Chem. 123, 137–149.
`Emanuelli, M., Carnevali, F., Lorenzi, M., Raffaelli, N., Amici, A.,
`Ruggieri, S., and Magni, G. (1999). Identification and characterization
`of YLR328W, the Saccharomyces cerevisiae structural gene encod-
`ing NMN adenylyltransferase. Expression and characterization of
`the recombinant enzyme. FEBS Lett. 455, 13–17.
`Emanuelli, M., Amici, A., Carnevali, F., Pierella, F., Raffaelli, N., and
`Magni, G. (2003). Identification and characterization of a second
`NMN adenylyltransferase gene in Saccharomyces cerevisiae. Pro-
`tein Expr. Purif. 27, 357–364.
`Fleischmann, R.D., Adams, M.D., White, O., Clayton, R.A., Kirkness,
`E.F., Kerlavage, A.R., Bult, C.J., Tomb, J.F., Dougherty, B.A., Mer-
`rick, J.M., et al. (1995). Whole-genome random sequencing and
`assembly of Haemophilus influenzae Rd. Science 269, 496–512.
`Fridland, A., Connelly, M.C., and Robbins, T.J. (1986). Tiazofurin
`metabolism in human lymphoblastoid cells: evidence for phosphory-
`lation by adenosine kinase and 5⬘-nucleotidase. Cancer Res. 46,
`532–537.
`Gallo, C.M., Smith, D.L., Jr., and Smith, J.S. (2004). Nicotinamide
`clearance by pnc1 directly regulates sir2-mediated silencing and
`longevity. Mol. Cell. Biol. 24, 1301–1312.
`Garavaglia, S., D’Angelo, I., Emanuelli, M., Carnevali, F., Pierella, F.,
`Magni, G., and Rizzi, M. (2002). Structure of human NMN adenylyl-
`transferase. A key nuclear enzyme for NAD homeostasis. J. Biol.
`Chem. 277, 8524–8530.
`Ghislain, M., Talla, E., and Francois, J.M. (2002). Identification an