2 N. H. Hughes, G. W. Kenner and A. Todd, J. Chem. Soc., 1957, 3733;
L. J. Haynes, N. H. Hughes, G. W. Kenner and A. Todd, J. Chem. Soc.,
1957, 3727.
bromide salt 12 as a crystalline solid in 80% isolated yield. It is
noteworthy that, contrary to the previous report,4 we found both
11 and 12 to be air-stable and non-hygroscopic solids.
3 While b-NMN is commercially available, its high price ($508 per gram,
Sigma 1997 catalog) and a lack of supply made large scale preparation
of NAD+ difficult. For other preparations of b-NMN, see D. R. Walt,
V. M. Rios-Mercadillo, J. Auge and G. M. Whitesides, J. Am. Chem.
Soc., 1980, 102, 7806; D. R. Walt, M. A. Findeis, V. M. Rois-
Mercadillo, J. Auge and G. M. Whitesides, J. Am. Chem. Soc., 1984,
106, 234; G. M. Whitesides and D. R. Walt, US Pat. 4,411,995-A, 1983;
R. Jeck, P. Heik and C. Woenckhaus, FEBS Lett., 1974, 42, 161.
4 I. A. Mikhailopulo, T. I. Pricota, V. A. Timoshchuk and A. A. Akhrem,
Synthesis, 1981, 387.
The phosphorylation of the nicotinamide triol 12 has been
reported in the literature using 1 equiv. of POCl3 in PO(OMe)3
with variable yield (35–64%).4,5 Accordingly, we reexamined
the reaction conditions and observed that reaction was very
slow below 25 °C, while at > 5 °C a chlorination reaction was
the favored pathway.6,7 Thus the triol 12 was treated with 4.0
equiv. of POCl3 in PO(OMe)3 between 25 and 0 °C for 7 h to
give the desired phosphorylated product in 90–92% yield
(Scheme 1). Upon neutralization of the reaction mixture, the
crude b-NMN 5 was separated from the reaction mixture by
addition of MeCN–Et2O (1:3). This crude product was further
purified using resin chromatography to give the b-NMN in
> 97% purity in 80% isolated yield.8,9 Use of excess of POCl3
(4 equiv.) at controlled temperature proved far superior to the
more conventional stoichiometric amount of the reagent.
With b-NMN in hand, our attention turned to development of
a practical method for pyrophosphate bond formation en route
to NAD+.10 Conventional ways to make the pyrophosphate
bonds include carbodiimide coupling,2,11 the Michelson proce-
dure,12 and the Khorana–Moffatt procedure.13 This last method,
involving the coupling of a suitable salt of a sugar phosphate
and a nucleotide phosphoromorpholidate, has been widely used
for the synthesis of various sugar nucleotides.13 However, the
reaction between b-NMN and AMP-morpholidate13b (13) was
attempted and gave NAD+ in < 5% yield.14 Our initial attempts
using the AMP-imidazolidate,15 triazolidate, or tetrazolidate, or
to react AMP with b-NMN-imidazolidate16 under the modified
Khorana–Moffat conditions gave only low conversion to NAD+
(formamide, 16 h to 6 days). We reasoned that this low
reactivity of AMP-amidates might be due to the fact that b-
NMN is an inner salt and is thus less nucleophilic. Under these
conditions, a second dissociation of the phosphoric acid in the
b-NMN, pKa ~ 7, is probably not occurring to any significant
extent. Accordingly, we attempted to activate the reaction using
Lewis acid additives. The coupling between b-NMN and AMP-
imidazolidate in the presence of MnCl2–4H2O in formamide
gave the desired NAD+ in 25% yield after 16 h.16 Further fine-
tuning of the reaction conditions improved conversion up to
78% (HPLC assay yield). Typically, the reaction was carried
out using 1.0 equiv. of AMP-morpholidate (13) with 1.1 equiv.
of b-NMN (5) in 0.2 M solution of MnCl2 (1.5 equiv.)17 and
MgSO4 (2 equiv.) in formamide at room temperature for 16 h.
Upon completion, the crude NAD+ was precipitated from the
reaction mixture with MeCN, and further purified by resin
chromatography (Sephadex QAE A-25, elution with 0.25 M aq.
(NH4)HCO3 solution; then Amberite XAD-16 with water–
MeOH gradient at 5 °C)18 followed by freeze-drying to give the
NAD+ as its ammonium salt with > 99% purity in 58% isolated
yield.
5 M. Yoshikawa, T. Kato and T. Takenishi, Tetrahedron Lett., 1967,
5065; K. Imai, S. Fujii, K. Takanohashi, Y. Furukawa, T. Masuda and
M. Honjo, J. Org. Chem., 1969, 34, 1547.
6 In considering an efficient method for the synthesis of the pyr-
ophosphate, we attempted to prepare the intermediate dichlorophos-
phate (phosphorodichloridate). Dichloridate was then treated in situ
with AMP monosodium salt and triazole. However, no pyrophosphate
bond formation was observed by 31P NMR spectroscopy.
7 In an attempt to eliminate the bis-adduct formation, the reaction was
also carried out using a 1:1 mixture of POCl3–H2O [in situ generation
of (HO)POCl2]; see M. Yoshikawa, T. Kato and T. Takenishi, Bull.
Chem. Soc. Jpn., 1969, 3505.
8 Chromatography on Dowex resin (two successive columns, 1X2
formate, 50WX8 H+ form) followed by subsequent lyophilization of the
aqueous fractions provided b-NMN in 80% yield, 97% purity (120 g
scale). The two resin column isolation process proved to be operation-
ally simple and the Dowex 50WX8 resin (H+ form) served the dual
purpose of purification and, more importantly, pH adjustment to provide
the desired inner salt form of b-NMN.
9 The crude b-NMN was initially purified by a resin-free isolation [i.e.
precipitation followed by treatment with activated carbon (Ecorsob-C,
P-502-H)]. Although the process is simple, the yield for the next
coupling reaction with AMP-morpholidate under the optimized condi-
tion was only moderate (it still required pH adjustment using H+ cation
resin). Alternatively, a direct resin chromatography using Dowex (50W
X 8-100) resin column was developed (97% purity, eluted with 5%
formic acid in water). However, this procedure required high resin loads
with the product eluting in a large volume of water.
10 For examples of general pyrophosphate bond formation, see: A. F.
Cook, M. J. Holman and A. L. Nussbaum, J. Am. Chem. Soc., 1969, 91,
1522; K. Furusawa, M. Sekine and T. Hata, J. Chem. Soc., Perkin
Trans. 1, 1976, 1711; F. Cramer and H. Schaller, Chem. Ber., 1961, 94,
1634; V. J. Davisson, A. B. Woodside, T. R. Neal, K. E. Stremler, M.
Muehlbacher and C. D. Poulter, J. Org. Chem., 1986, 51, 4768; J. R.
Falck, K. K. Reddy, J. Ye, M. Saady, C. Mioskowski, S. B. Shears, Z.
Tan and S. Safrany, J. Am. Chem. Soc., 1995, 117, 12172; V. Wittmann
and C.-H. Wong, J. Org. Chem., 1997, 62, 2144.
11 V. C. Bailey, J. K. Sethi, A. Galione and B. V. L. Potter, Chem.
Commun., 1997, 695.
12 A. M. Michelson, Biochim. Biophys. Acta, 1968, 91, 1; H. Kim and B. E.
Haley, J. Biol. Chem., 1990, 265, 3636.
13 (a) J. G. Moffat and H. G. Khorana, J. Am. Chem. Soc., 1959, 81, 1265;
(b) J. G. Moffat and H. G. Khorana, J. Am. Chem. Soc.,, 1961, 83, 649;
(c) E. S. Simon, S. Grabowski and G. M. Whitesides, J. Org. Chem.,
1990, 55, 1834.
The practical synthesis of NAD+ outlined above has a number
of noteworthy steps. The mild deacetylization procedure
allowed isolation of the crystalline air stable nicotinamide
nucleoside. Phosphorylation conditions were developed to
provide b-NMN in 80% isolated yield. A new catalyzed
pyrophosphate formation between b-NMN and AMP-morpholi-
date using a combination of MnCl2 and MgSO4 gave NAD+ in
78% yield. An improved resin isolation process provided NAD+
in 58% isolated yield.
14 L. M. Mel’nikova and V. M. Berezovkii, Zh. Obshch. Khim., 1970, 40,
918.
15 R. Lohrmann and L. E. Orgel, Tetrahedron, 1978, 34, 853.
16 B. C. F. Chu and L. E. Orgel, Biochim. Biophys. Acta, 1984, 782, 103;
M. Shimazu, K. Shinozuka and H. Sawai, Tetrahedron Lett., 1990, 31,
235.
17 For a successful high yielding coupling reaction, reagents and the
solvents were dried. A MnCl2 solution was prepared by dissolution of
the commercial tetrahydrate in formamide to give a 0.2 m stock solution.
The solution was dried over 4 Å molecular sieves for several days prior
to use. Typically the solution had 0.5 m H2O content and was used for
the reaction successfully. Use of commercial anhydrous MnCl2 proved
inferior.
We are grateful to Dr R. Czaja, Mr J. Bergan and Mr Bob
Reamer for technical assistance.
18 Purification of NAD+ using resin chromatography at room temperature
resulted in partial decomposition of purified NAD+ (5% purity loss).
Therefore, the chromatography was carried out at 5 °C (110 g scale,
> 99% purity).
Notes and references
1 The Merck Index, 6259, 11th edn., ed. S. Budavari, Merck, Rahway, NJ,
1989.
Communication 8/09930H
730
Chem. Commun., 1999, 729–730