Determination of the transglycosidation activity of NAD+ glycohydrolases with 4-(2′-alkyl-sulfanyl-vinyl)-pyridine derivatives generating chromophoric NAD+ analogs

Article abstract of DOI:10.1016/S0045-2068(03)00041-5

The base exchange of nicotinamide with pyridine derivatives 1a-5a, catalyzed by pig brain NAD⁺ glycohydrolase and ADP-ribosyl cyclase from Aplysia californica, generated the corresponding NAD⁺ analogs 1b-5b. These analogs exhibited a high absorbance band in the visible region. The transglycosidation rate was determined by monitoring the absorbance increase. Among the tested derivatives, (E)-4-[2-(methylsulfanyl)-vinyl]-pyridine 1a was the most suitable substrate for pig brain NAD⁺ glycohydrolase while 4-[1,3]-dithiolan-2-ylidenemethyl-pyridine 3a was the most efficient for ADP-ribosyl cyclase from A. californica.

Full text of DOI:10.1016/S0045-2068(03)00041-5

BIOORGANIC
CHEMISTRY
Bioorganic Chemistry 31 (2003) 288–305
www.elsevier.com/locate/bioorg
Determination of the transglycosidation
activity of NADþ glycohydrolases with
4-(20-alkyl-sulfanyl-vinyl)-pyridine derivatives
generating chromophoric NADþ analogs
*
Karine Pacaud,Denis Tritsch, Alain Burger,
and Jean-Franßcois Biellmann
Laboratoire de Chimie Organique Biologique, UMR7123, Faculteꢀ de Chimie,
Universiteꢀ Louis Pasteur, 1 rue Blaise Pascal, FR-67008 Strasbourg, Cedex, France
Received 16 January 2003
Abstract
The base exchange of nicotinamide with pyridine derivatives 1a–5a,catalyzed by pig brain
NADþ glycohydrolase and ADP-ribosyl cyclase from Aplysia californica,generated the corre-
sponding NADþ analogs 1b–5b. These analogs exhibited a high absorbance band in the visible
region. The transglycosidation rate was determined by monitoring the absorbance increase.
Among the tested derivatives,( E)-4-[2-(methylsulfanyl)-vinyl]-pyridine 1a was the most suit-
able substrate for pig brain NADþ glycohydrolase while 4-[1,3]-dithiolan-2-ylidenemethyl-pyr-
idine 3a was the most efficient for ADP–ribosyl cyclase from A. californica.
Ó 2003 Elsevier Science (USA). All rights reserved.
Keywords: 4-(20-Alkyl-sulfanyl-vinyl)-pyridine; NADþ analogs; NADþ glycohydrolase; ADP-ribosyl
cyclase; NADase; Transglycosidation; Push–pull double bond
* Corresponding author. Fax: +33-3-90-24-13-45.
E-mail address: dtritsch@chimie.u-strasbg.fr (D. Tritsch).
0045-2068/03/$ - see front matter Ó 2003 Elsevier Science (USA). All rights reserved.
doi:10.1016/S0045-2068(03)00041-5

K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
289
Scheme 1. Multicatalytic activity of NADþ glycohydrolases.
1. Introduction
NAD(P)þ glycohydrolases (NADases;1 EC 3.2.2.5 and EC 3.2.2.6) constitute a
heterogeneous family of enzymes. They catalyze the hydrolytic cleavage of the N-gly-
cosidic bond of b-NADþ leading to the formation of an ADP-ribosyl oxocarbenium
ion. Then this intermediate could evolve according to three competing pathways
(Scheme 1). The reaction between ribosyl C-10 and water gives ADP–ribose. Intra-
molecular cyclization between ribosyl C-10 and N-1 of the adenine ring produces cy-
þ
clic ADP–ribose (cADPR). In the presence of pyridine derivatives,NAD analogs
are obtained via a transglycosidation reaction [1,2]. In fact NADases have multicat-
alytic activities and the yield in each product depends on the enzyme and the reaction
conditions. Mammalian ‘‘classical’’ NADases transform NADþ essentially into
ADP–ribose. Only small proportions of cADP–ribose (less than 2% of the reaction
products) are detected [3,4]. On the other hand ADP–ribosyl cyclase from Aplysia
californica produces almost exclusively cADPR [5,6]. Moreover in the presence of
nicotinic acid,this enzyme catalyzes the exchange of the nicotinamide group of
1
Abbreviations used: NADase(s),NAD(P) þ glycohydrolase(s); c-ADPR,cyclic adenosine diphosphate
ribose; NAAD(P)þ,nicotinic acid adenine dinucleotide (phosphate).

290
K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
NADðPÞþ with nicotinic acid to give nicotinic acid adenine dinucleotide (phosphate)
[7]. The interest in NADases is growing since the dþiscovery that cADPR and nico-
tinic acid adenine dinucleotide phosphate (NAADP ) have intracellular Ca2þ-mobi-
lizing properties (for a review,see [8]).
4-(20-Alkyl-sulfanyl-vinyl)-pyridinium salts absorb strongly in the near visible re-
gion whereas the corresponding (20-alkyl-sulfanyl-vinyl)-pyridines do not [9]. These
spectral properties were at the origin of the development of new specific cysteine re-
agents [9]. As the enzyme exchanges nicotinamide with a large variety of pyridine
derivatives [10,12] we developed a continuous spectrophotometric assay allowing
the determination of the transglycosidation rate of NADases. It is based on the base
exchange between nicotinamide and (20-alkyl-sulfany0l-vinyl)-pyridines generating
NADþ analogs absorbing,as the corresponding 4-(2 -alkyl-sulfanyl-vinyl)-pyridi-
nium salts,in the visible region. The enzymatic activity could be determined by fol-
lowing the absorbance increase due to the formation of these NADþ analogs.
Scheme 2. Transglycosidation reaction of NADase with pyridine derivatives 1a–5a.

K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
291
Pyridine derivatives 1a–5a were synthesized and tested on pig brain NADase and
ADP-ribosyl cyclase from A. californica. The kinetic constants of the transglycosida-
tion reaction were determined and the generated NADþ analogs 1b–5b characterized
(Scheme 2).
2. Materials and methods
Thin-layer chromatography was performed on silica analytical Si 60 F254 plates
(Merck),flash chromatography on silica gel Si 60 230–400 mesh (Merck) pretreated
with 1% triethylamine in the elution solvent.1H and 13C NMR spectra were recorded
either on a Bruker WP-200 SY,on a Bruker AM 400 or on a Bruker ARX 500 spec-
trometer. The chemical shifts (d) are reported in parts per million (ppm) relative to
tetramethylsilane. The letters s,d,t and m indicate the multiplicity of the signals: sin-
glet,doublet,triplet and multiplet,respectively. The coupling constants ( J) are given
in hertz (Hz). The mass spectra were recorded on a LKB 9000S apparatus by elec-
tronic impact (EI,70 eV). Spectrophotometric measures were performed on an Uvi-
kon 933 spectrophotometer (Kontron instruments). The molar extinction coefficients
(e) are expressed in MÀ1 cmÀ1. The melting points (mp) were determined with a Rei-
chert hot stage microscope and were not corrected. Elemental analyses were per-
formed by the Strasbourg analytical service of the CNRS.
Chemicals. Unless otherwise stated,all reagents were obtained from commercial
suppliers and were used without purification. Anhydrous tetrahydrofuran (THF)
was freshly distilled before use from sodium benzophenone ketyl.
Synthesis of (E)-4-[2-(methylsulfanyl)-vinyl]-pyridine (1a) and (Z)-4-[2-(meth-
yl-sulfanyl)-vinyl]-pyridine (2a). The two isomers were synthesized,as described,
from isonicotinaldehyde and methylthiomethyl-triphenylphosphonium chloride [9].
Synthesis of 4-[1,3]-dithiolan-2-ylidenemethyl-pyridine (3a). A 1.1 M solution of
butyl lithium in hexane (10.8 mL; 11.9 mmol) was added to a solution of di-isopro-
pylamine (1.65 mL; 11.9 mmol) in dry THF (19 mL) at À65°C. The mixture was stir-
red for 20 min and the temperature kept below À20°C. 4-Methylpyridine (0.525 mL;
5.4 mmol) was added,and the resulting red solution was stirred at À20°C for 1 h.
CS2 (0.711 mL; 11.9 mmol) was then added dropwise during 10 min at À20°C. The
resulting suspension was stirred for 1 h. After addition of dibromoethane (1 mL;
11.9 mmol) the mixture was stirred at room temperature for 4 h. Water saturated
Et2O (3 mL) was added cautiously,and the solvents were removed. The residue
was dissolved in CHCl3 (3 mL) and the solution was washed successively with
0.5 M H2SO4 (3 mL),H 2O (3 mL),and brine (3 mL). The aqueous layer was basified
with 4 M NaOH and extracted with CHCl3 (3 mL). The combined organic layers
were washed with brine,dried over Na SO4 and concentrated. Compound 3a was
2
purified by flash chromatography on silica (eluent: ethyl acetate) and was obtained
1
as a yellow powder (650 mg; yield 65%). mp 91–93 °C. H NMR (CDCl3; 200 MHz)
d: 8.51 (2H,dd, J1 ¼ 4:7; J2 ¼ 1:7); 7.27 (2H,dd, J1 ¼ 4:7; J2 ¼ 1:7); 6.55 (1H,s);
3.64–3.58 (2H,m); 3.41–3.35 (2H,m). 13C NMR (CDCl3; 50 MHz) d: 149.8; 145.7;
144.2; 121.5; 113.1; 40.36; 35.9. UV(CH3OH) kmax 325 nm (e 24,800), kmax 262 nm

292
K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
(e 3400); UV(100 mM potassium phosphate buffer,pH 8.0) kmax 324 nm (e 14,200),
kmax 264 nm (e 9700). MS (EI) 195,167,135. Anal. Calcd for C9H9NS2: C,55.35;
H,4.64; N,7.17. Found: C,55.46; H,4.56; N,6.89.
Synthesis of 4-(2,2-bis-methylsulfanyl-vinyl)-pyridine (4a). Compound 4a was
prepared essentially as described above for compound 3a. Iodomethane (0.84 mL;
13.5 mmol) was added instead of dibromoethane. The compound was purified by
flash chromatography on silica gel (eluent: ether/hexane; 1:1) and was obtained as
1
a red oil (698 mg; yield 66%). H NMR (CDCl3; 200 MHz) d: 8.57 (2H,m); 7.48
(2H,dd, J3À2 ¼ 4:8; J3À6 ¼ 1:5); 6.51 (1H,s); 2.47 (3H,s); 2.44 (3H,s). 13C NMR
(CDCl3; 50 MHz) d: 149.7; 143.6; 142.18; 124.3; 123.18; 40.36; 17.3. UV(CH3OH)
kmax 320 nm (e 13,250), kmax 262 nm (e 3400); UV(100 mM potassium phosphate buf-
fer,pH 8.0) kmax 316 nm (e 11,800), kmax 264 nm (e 9700). MS (EI) 197; 182; 167; 150;
135.
Synthesis of ethyl N-(1-pyridin-4-yl-ethylidene)-hydrazinecarboxylate (6). A mix-
ture of 4-acetylpyridine (0.55 mL; 5 mmol),carbethoxyhydrazine (521 mg; 5 mmol),
methanol (4 mL),and acetic acid (12.5 lL) was refluxed 1 h and left overnight at
room temperature. Compound 6 was precipitated by addition of ether and obtained
1
in a 80% yield (831 mg) as a white powder. mp 181–183 °C. H NMR (CDCl3;
200 MHz) d: 8.64 (2H,dd, J1 ¼ 4:6; J2 ¼ 1:7); 7.98 (1H,s); 7.64 (2H,dd,
J1 ¼ 4:6; J2 ¼ 1:7); 4.34 (2H,q); 2.20 (3H,s); 1.37 (3H,t).
13C NMR (CDCl3;
50 MHz) d: 154.0; 150.1; 145.3; 145.1; 120.36; 62.33; 14.52; 12.22. UV(CH3OH)
max 276 nm (e 15,900). MS (EI) 207; 162; 134; 119; 106. Anal. Calcd for C10H13N3O2:
C,57.96; H,6.32; N,20.28. Found: C,58.10; H,6.25; N,20.03.
k
Synthesis of 4-[1,2,3]thiadiazol-4-yl-pyridine (7). Compound 6 (621 mg; 3 mmol)
was added to cold thionyl chloride (1.5 mL) at 0 °C. The mixture was heated at 60 °C
for 1 h,then excess thionyl chloride was removed in vacuo. Aqueous sodium carbon-
ate was added to the residue and the aqueous layer extracted 3-fold with ethyl ace-
tate. The organic layers were dried over Na2SO4,and the solvent was removed.
Compound 7 was purified by flash chromatography (eluent: ethyl acetate). It was ob-
1
tained as white crystals (431 mg; yield 88%). mp 119–120 °C. H NMR (CDCl3;
200 MHz) d: 8.87 (1H,s); 8.77 (2H,dd,
J2À3 ¼ 4:4; J2À5 ¼ 1:6); 7.94 (2H,dd,
J3À2 ¼ 4:4; J3À6 ¼ 1:6). 13C NMR (CDCl3; 50 MHz) d: 149.8; 145.7; 144.2; 121.5;
113.1; 40.36; 35.9. UV(CH3OH) kmax 270 nm (e 6350), kmax 244 nm (e 12,800). MS
(EI) 163; 135. Anal. Calcd for C7H5N3S: C,51.52; H,3.09; N,25.75. Found: C,
51.56; H,2.86; N,25.85.
Synthesis of 4-(methylsulfanylethynyl)-pyridine (5a). A 1.4 M phenyl lithium so-
lution in benzene/ether (7:3) (0.7 mL,1.0 mmol) was added dropwise on a stirred and
precooled ()78 °C) suspension of compound 7 (150 mg; 0.92 mmol) in anhydrous
THF (1.4 mL) at a such rate that the temperature did not exceed )60 °C. The reac-
tion medium was stirred at )60 °C for further 30 min. Dimethyl sulfate (94 lL;
1 mmol) was added in one portion. The reaction was allowed to warm to 0 °C.
Ice-water (5 mL) was added. The organic layer was separated and the aqueous layer
extracted with ether (2 Â 10 mL). The combined organic layers were dried (Na2SO4),
and the solvents removed in vacuo. Pyridine derivative 5a was purified by flash chro-
matography (eluent: ether/hexane 1:1) and was obtained as a colorless oil (112 mg;

K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
293
1
yield 82%). H NMR (CDCl3; 200 MHz) d: 8.51 (2H,dd, J1 ¼ 4:5; J2 ¼ 1:6); 7.20
(2H,dd, J1 ¼ 4:5; J2 ¼ 1:6); 2.50 (3H,s). 13C NMR (CDCl3; 50 MHz) d: 149.67;
131.54; 124.65; 90.03; 87.82; 19.3. UV(100 mM potassium phosphate buffer,pH
8.0) kmax 284 nm (e 17,500). MS (EI) 149; 134.
Determination of the pKa of compounds 1a–5a. Pyridine derivatives 1a–5a (30 lM)
were dissolved in a 100 mM potassium phosphate buffer at pH values between 3.0
and 8.0 (15–20 points were undertaken for each compound). UV/visible spectra were
recorded between 300 and 500 nm to determine the absorbance of the pyridinium
form. The absorbance values were plotted against the pH and the pKa was taken
at the inflexion point of the graph.
Synthesis of (E) 1-methyl-4(2-methylsulfanyl-vinyl)pyridinium iodide. Compound
1a was methylated as described by Holler et al. [9]. The 1H NMR spectrum of (E) 1-
methyl-4(2-methylsulfanyl-vinyl)pyridinium iodide was taken in a 10 mM deuterated
1
potassium phosphate buffer. H NMR (200 MHz) d: 8.25 (d, J ¼ 6:8,2H); 7.8 (d,
J ¼ 15:4,1H); 7.6 (d, J ¼ 6:8,2H); 6.29 (d, J ¼ 15:4,1H); 4.09 (s,3H); 2.44 (s,3H).
Solubilization of pig brain NADase. Pig brain NADase was solubilized from an ac-
etone powder,purchased from Sigma Chemical,according to the procedure of
Windmueller and Kaplan [13]. Acetone powder (111 mg) was washed 2-fold with
water to eliminate soluble contaminants. The powder was suspended in water
(2 mL) under constant stirring for 10 min. The suspension was centrifuged at
20,000 rpm for 30 min with a Sigma 3K30 centrifuge. The supernatant was discarded.
The pellet was suspended in a 100 mM potassium phosphate,pH 7.2 buffer (3 mL)
under stirring at 37 °C. Trypsin (0.5 mg) was added to the suspension. After incuba-
tion for 40 min,the proteolysis was stopped by addition of a 0.01 M phenylmethyl-
sulfonyl fluoride solution in isopropanol (30 lL). After 5 min at 37 °C,the suspension
was chilled in an ice-bath and stirred for 30 min,and then centrifuged at 200, 00 rpm
for 60 min. The supernatant containing soluble NADase was used for the kinetic
studies.
The concentration of protein was determined according to the Bradford method
using bovine serum albumin as standard [14].
Determination of the rate of NADþ hydrolysis for pig brain NADase with KCN.
The hydrolysis (1 mL reaction volume) was performed at 25 °C in a 100 mM potas-
sium phosphate buffer,pH 8.0 with NAD þ (0.45 mM) and NADase (32 lg). Aliquots
(100 lL) were withdrawn at different times and added to a 1 M potassium cyanide
solution (900 lL). Absorbance at 327 nm was measured to determine the amount
of unhydrolyzed NADþ using a molar extinction coefficient of 5900 MÀ1 cmÀ1 for
the NAD–cyanide adduct [15]. The Km of NADþ and the Vmax were determined by
varying the concentration of NADþ from 11 to 84 lM.
Determination of the kinetic constants of the transglycosidation reaction with com-
pounds 1a–5a for pig brain NADase. Assays were carried out in a 100 mM potassium
phosphate buffer,pH 8.0 at 25 °C (final volume 1 mL). The pyridine derivatives were
dissolved in acetone and concentrated stock solutions in buffer were prepared. These
solutions were diluted to the appropriate concentration with buffer. The concentra-
tion of acetone was maintained constant in the assays (<5% in volume). NADase
from pig brain (38 lg) was added to initiate the enzymatic reaction. Except for

294
K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
compounds 1a and 2a (k ¼ 380 nm, e ¼ 21,500 MÀ1 cmÀ1),the transglycosidation re-
action was followed by monitoring the increase of the absorbance at the kmax of the
corresponding NADþ analogs during 10 min (Table 1).
The Km of NADþ was determined at a fixed concentration of compound 1a
(3 mM) and the concentration of NADþ was varied from 0.015 to 0.45 mM. The ki-
netic constants of compounds 1a–5a were established at a fixed concentration of
NADþ (0.45 mM). The concentration of pyridine derivatives was varied from
0.045 to 1.35 mM for compound 1a,from 0.090 to 0.9 mM for compound 2a,from
0.13 to 1.3 mM for compound 3a,from 0.45 to 2.25 mM for compound 4a,and
from 0.62 to 3.1 mM for compound 5a. The reactions were initiated by the addition
of NADase (26–95 lg depending on the substrate).
Preparation of the NADþ analogs 1b–5b. They were synthesized enzymatically.
The same procedure was used for each analog. The preparation of analog 1b is given.
Pyridine derivative 1a (80 mg) and NADþ (80 mg) were incubated in a 10 mM Tris/
HCl buffer,pH 8.0 (final volume 80 mL) with NADase from pig brain (20 mg). The
reaction was followed by measuring the absorbance at 374 nm. The reaction was
stopped when the absorbance reached a plateau. The excess of compound 1a was ex-
tracted with ethyl acetate. The aqueous layer was applied on a Dowex 1 Â 2
(HCOOÀ form) column (1 Â 5 cm) equilibrated with water. The column was washed
with water to eliminate residual compound 1a. A gradient of formic acid (0–0.4 M)
(2 Â 50 mL) was applied to elute the retained compounds. The analog was detected
by its absorbance at 374 nm. Fractions having the highest 374 nm/252 nm absorbance
ratio were combined and lyophilized.
The spectrophotometric properties of the different NADþ analogs are given in
Table 1. The yields of the isolated analogs were about 40–60%. NMR data of the
NADþ analogs 1b–5b are given below.
1
Analogs 1b and 2b. H NMR (D2O; 400 MHz) d: 8.73 (2H,d, J ¼ 7; Z form);
8.65 (2H,d, J ¼ 7; E form); 8.43 (1H,s, E and Z forms); 8.21 (1H,s, E and Z forms);
7.89 (1H,d, J ¼ 15, E form); 7.78 (2H,d, J ¼ 7, Z form); 7.58 (2H,d, J ¼ 7:5, E
form); 7.4 (1H,d, J ¼ 11, Z form); 6.39 (1H,d, J ¼ 11, Z form); 6.18 (1H,d,
J ¼ 15, E form); 6.11 (1H,d, J ¼ 6:5, E form); 6.09 (1H,d, J ¼ 6, Z form); 5.98
Table 1
Spectrophotometric properties of the NADþ analogs 1b–5b
Compounds
kmax (nm)
e (MÀ1 cmÀ1
)
kmax (nm)
e (MÀ1cmÀ1
)
1b
2b
3b
4b
5b
252a
13,500a
374a
22,300a
b
b
b
b
—
—
—
—
254
255
259
13,300
14,700
14,500
405
405
354
26,500
23,500
17,300
Note. The NADþ analogs were dissolved in 100 mM potassium phosphate buffer,pH 8.0. The UV/
visible spectra were recorded between 240 and 480 nm. Measures were carried out in triplicate.
a For a mixture of 90% 1b and 10% 2b due to the isomerization of the methylsulfanyl-vinyl group.
b Could not be determined due to the isomerization of 2b to 1b.

K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
295
(1H,d, J ¼ 5, E and Z forms); 4.5-3 (m, E and Z forms); 2.72 (3H,s, Z form); 2.56
(3H,s, E form). 13C NMR (D2O; 100.6 MHz) d: 153.6; 150.4; 148.4; 146.6; 145.9;
142.2; 139.5; 122.0; 118.4; 117.4; 98.6; 87.7; 86.4; 84.4; 77.3; 74.6; 70.7; 70.6; 65.5;
65.1; 13.6.
Analog 3b. 1H NMR (D2O; 400 MHz) d: 8.35 (2H,d, J ¼ 7:1); 8.21 (1H,s); 8.06
(1H,s); 7.34 (2H,d, J ¼ 7:1); 6.30 (1H,s); 5. 85 (1H,d, J ¼ 6); 5.67 (1H,d, J ¼ 5);
3.61 (2H,m); 3.33 (2H,m). 13C NMR (D2O; 100.6 MHz) d: 154.9; 152.6; 151.9;
149.6; 139.2; 138.4; 122.4; 118.0; 109.2; 100.2; 97.4; 84.3; 83.8; 76.9; 73.9; 70.3;
69.1; 68.3; 65.4; 65.0; 41.5; 36.1.
1
Analog 4b. H NMR (D2O; 400 MHz) d: 8.61 (2H,d, J ¼ 6:8); 8.39 (2H,d,
J ¼ 6:8); 8.18 (1H,s); 8.04 (1H,s); 5. 83 (2H,m); 5.69 (1H,m); 4.4–4 (m); 2.43
(3H,s); 2.35 (3H,s). 13C NMR (D2O; 100.6 MHz) d: 152.8; 151.5,149.5; 148.6;
140.7; 138.8; 123.9; 119.0; 113.1; 102.4; 98.5; 87.0; 86.3; 84.3; 77.2; 74.4; 70.7; 70.6;
65.7; 65.3; 16.7; 15.7.
Analog 5b. 1H NMR (D2O; 400 MHz) d: 8.75 (2H,d, J ¼ 6:8); 8.47 (1H,s); 8.28
(1H,s); 7.67 (2H,d, J ¼ 6:8); 6.04 (1H,d, J ¼ 5:6); 5.89 (1H,d, J ¼ 5:2); 2.51 (3H,
s). 13C NMR (D2O; 100.6 MHz) d: 151.3; 149.2; 147.1; 141.6; 141.4 139.4; 139.4;
127.1; 118.3; 105.1 98.9; 90.5; 87.4; 86.4; 84.0; 77.4; 70.6; 70.3; 65.2; 64.9; 18.8.
Isomerization of NADþ analogs (1b) and (2b). The transglycosidation reaction
with compounds 1a and 2a werþe followed by 1H NMR (500 MHz). The pyridine de-
rivative 1a or 2a (1 mg),NAD (1.5 mg) and NADase from pig brain (3 mg) were
incubated at 37 °C in a 10 mM deuterated potassium phosphate buffer (final volume
500 lL). The deuterated buffer was obtained from a 100 mM potassium phosphate
buffer,pH 8.0 (10 mL) by five successive ‘‘lyophilization/dissolution in D O’’ steps.
2
Finally the residue was dissolved in 10 mL of D2O to obtain a stock solution of
100 mM deuterated potassium phosphate buffer. NADase was dialyzed against
10 mM deuterated buffer by a series of concentration/dilution steps by ultracentrifu-
gation using a Centricon YM-10 unit (Millipore). The pyridine derivatives were dis-
solved in a minimum of deuterated acetone (20 lL). 1H NMR spectra were recorded
after 7,60,120 and 240 min and 24 h incubation. The signal of H 2O was suppressed
by presaturation. The ratio of the products resulting from the enzymatic reaction
was determined by integration of typical signals: for NADþ d: 9.45 (1H,s,N2);
9.28 (1H,d, J ¼ 6:5,N6); 8.95 (1H,dt, J4À5 ¼ 8; J4À2 ¼ 3; J4À6 ¼ 1:5,N4); 8.54
(1H,s,A8);08.31 (1H,dd,
J5À4 ¼ 8; J5À6 ¼ 6:5,N5); 8.26 (1Hs,,A2); 6.20 (1H,d,
0
J ¼ 5:5,N 1); 6.14 (1H,d, J ¼ 6,A 1) [16]; for free nicotinamide d ¼ 9:03 (1H,m,
N2); 8.08 (1H,m,N6) [17],anomeric protons for
(1H,d, J ¼ 2:5) and d ¼ 5:43 (1H,d, J ¼ 4),respectively [18],vinylic protons for
a- and b-ADP–ribose d ¼ 5:31
NADþ analogs 1b (d: 6.18 and 7.89) and 2b (d: 6.39 and 7.4).
Hydrolysis of NADþ analogs 1b and 5b with pig brain NADase. The analogs
(40 lM) were incubated at 25 °C in a 0.1 M sodium phosphate buffer,pH 8.0 in
the presence of pig brain NADase (50 lg). The enzymatic reaction was followed
by monitoring the absorbance at 374 and 354 nm for compounds 1b and 5b,respec-
tively,during 30 min.
Kinetic studies with ADP-ribosyl cyclase from A. californica. ADP-ribosyl cyclase
from A. californica was obtained from Dr. F. Schuber [2]. Samples were also

296
K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
purchased from Sigma. Except otherwise indicated the enzymatic reactions were per-
formed in a 100 mM potassium phosphate buffer,pH 8.0 at 25 °C.
The rate of hydrolysis of NADþ was measured according to Kaplan [15]. The re-
action medium (600 L) contained NADþ (0.45 mM) and ADP-ribosyl cyclase
(0.37 lg). Aliquots of 100 lL were withdrawn at fixed time to determine the concen-
tration of unhydrolyzed substrate.
The kinetic constants for the transglycosidation reaction with pyridines 1a–5a
were determined as described above. The reactions were performed at saturating
concentration of NADþ: 1 mM for assays with compounds 1a and 2a,0.6 mM for
assays with pyridine 3a,2 mM for assays with compounds 4a and 5a. The concentra-
tion of pyridine derivatives was varied from 80 to 500 lM for compound 1a,from 8
to 65 lM for compound 2a,from 14 to 210 lM for compound 3a,from 80 to 560 lM
for compound 4a,from 190 to 1900 lM for compound 5a. The reactions were initi-
ated by the addition of enzyme (0.2 lg) and monitored during 2 min.
3. Results
3.1. Syntheses of pyridine derivatives (1a–5a)
Compounds 1a and 2a were prepared according to Holler et al. [9]. The two iso-
mers were easily distinguished by the 1H NMR coupling constants of the vinylic pro-
tons: 15.5 Hz for the (E) isomer and 11 Hz for the (Z) isomer.
The method described by Goodwin et al. [19] for the preparation of pyridines
substituted at position 2 with ketene dithioacetal groups was applied to synthesize
compounds 3a and 4a. Compound 4a was unstable and thus,prepared and purified
before use.
The synthesis of 1-alkynylthioethers via 1,2,3-thiadiazole by Raap and Micetich
[20] was efficiently extended to the preparation of pyridine 5a. The compound was
obtained in 3 steps from 4-acetyl-pyridine (Scheme 3).
The pKa of the pyridine derivatives 1a–5a was measured to determine the optimal
pH for the transglycosidation reaction. As the enzymatic substrate is the unproto-
nated pyridine,the enzymatic assays were performed at 2 pH units above the p Ka
in order to have high concentration of unprotonated pyridine. The protonated
and unprotonated forms absorbing at different wavelengths,their p KaÕs were deter-
mined spectrophotometrically (Table 2). The pKa values of the pyridine nitrogen
group were significantly higher than that of nicotinamide (pKa ¼ 3:33) or pyridine
(pKa ¼ 5:23) [11],due to the electron donating effect of the alkylsulfanyl-vinyl group.
3.2. Transglycosidation reaction with pyridine derivatives 1a–5a catalyzed by pig brain
NADase: structure of the transglycosidation products
The reaction with pyridine derivatives 1a and 2a was extensively studied. The
incubation of both compounds with NADþ and NADase led to the formation
of an absorption band in the visible region,suggesting that the enzyme exchanged

K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
297
Scheme 3. Synthesis of 4-(methylsulfanylethynyl)pyridine 5a.
Table 2
pKa of compounds 1a–5a
Compounds
kmax Unprotonated form
kmax Protonated form
pKa
1a
2a
3a
4a
5a
304
310
325
316
284
355
358
377
374
335
6.0
5.8
6.5
6.2
4.9
Note. Pyridine derivatives 1a–5a (30 lM) were dissolved in a 100 mM potassium phosphate buffer at
pH between 3 and 8. Spectra were recorded between 300 and 500 nm to determine the absorbance of the
pyridinium form. The absorbance values were plotted against the pH and the pKa was taken at the
inflexion point of the graph.
nicotinamide with both pyridine derivatives 1a and 2a. The transglycosidation reac-
tion was performed on a large scale to confirm the structure of the NADþ analogs 1b
and 2b. The spectroscopic analysis of the isolated products showed that both gave a
mixture of analogs 1b and 2b in a same 9/1 ratio. Clearly a cis–trans isomerization of
the methylsulfanyl-vinyl group took place in the reaction medium and/or during the
1
purification of the analogs. Following the transglycosidation reaction by H NMR
solved this question (Figs. 1 and 2). Integration of typical signals allowed quantify-
ing of the different molecules present in the reaction medium (Tables 3 and 4). The
assignments of the peaks for NADþ,free nicotinamide and ADP–ribose were based
on published spectra [16–18]. The NMR spectra clearly showed that the isomeriza-
tion occurred in the reaction medium. This fact was particularly perceptible with
pyridine 2a where the analogs 2b/1b ratio decreased during the incubation time.

298
K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
Fig. 1. Transglycosidation reaction of pig brain NADase with pyridine 1a. Reaction conditions are given
in Table 3. Partial 1H NMR (500 MHz) spectra after incubation of 7 minþ(A) and 360 min (B) at 37 °C are
shown. Characteristic peaks of NADþ (M),free nicotinamide ( ),NAD analogs 1b (s) and 2b (j) are
annotated.
After 24 h,the 9/1 ratio in favor of analog 1b was reached. The stability of pyridine
derivatives 1a and 2a in the deuterated buffer was checked at 37 °C. No isomerization
was detected after an incubation of 24 h. So the formation of the mixture of NADþ
analogs 1b and 2b was not due to a preliminary isomerization of the pyridine deriv-
atives. The catalytic event did not seem to be responsible for the isomerization. In-
deed the alkylation of pyridine 1a with iodomethane gave also a mixture of (E) and
1
(Z) isomers in the same 9/1 ratio as evidenced by H NMR in deuterated buffer.
Likely the isomerization took place after the analogs dissociated from the enzyme.

K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
299
Fig. 2. Transglycosidation reaction of pig brain NADase with pyridine 2a. Reaction conditions are given
in Table 4. Partial 1H NMR (500 MHz) spectra after incubation of 7 minþ(A) and 360 min (B) at 37 °C are
shown. Characteristic peaks of NADþ (M),free nicotinamide ( ),NAD analogs 1b (s) and 2b (j) are
annotated
þ
In the two experiments,NAD disappeared at almost the same rate. On the other
hand the efficiency of pyridine exchange,illustrated by the ratio of analog/ADP–ri-
bose concentrations,was more important with compound 1a (87/13) than with com-
pound 2a (69/31). Yost and Anderson reported for NADase from Bungarus fasciatus
venom that the efficiency of the transglycosidation increased linearly with respect to
the concentration of the pyridine derivatives [12]. In our case,as the reactions were
performed at the same concentration of pyridine derivative,the higher efficiency
with compound 1a could be linked to the different affinity of the enzyme for the

300
K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
Table 3
1H NMR study of the transglycosidation with compound 1a
Reaction time (min)
7
60
120
240
360
Disappeared NADþ (%)
Analog 1b (%)
Analog 2b (%)
7
43
89
62
81
6
85
81
6
90
78
9
Traces
ND
ND
11
ND
a- and b-ADP–ribose (%)
13
13
13
Note. Pyridine derivative 1a (1 mg),NAD þ (1.5 mg) and NADase from pig brain (3 mg) were incubated
in a 10 mM deuterated potassium phosphate buffer (final volume 600 lL) at 37 °C. 1H NMR (500 MHz)
spectra were recorded at indicated time. The proportion of disappeared NADþ,analogs 1b and 2b and
ADP–ribose was calculated from the integration of characteristic peaks: for NADþ d: 9.45 (1H,s,N2);
9.28 (1H,d, J ¼ 6:5,N6); 8.95 (1H,dt, J4–5 ¼ 8; J4–2 ¼ 3; J4–6 ¼ 1:5,N4); 8.54 (1H,s,A8); 8.31 (1H,dd,
J5–4 ¼ 8; J5–6 ¼ 6:5,N5); 8.26 (1H,s,A2); 6.20 (1H,d, J ¼ 5:5,N 01); 6.14 (1H,d, J ¼ 6,A 01) [16]; for free
nicotinamide d ¼ 9:03 (1H,m,N2); 8.08 (1H,m,N6) [17],anomeric protons for
a- and b-ADP–ribose
þ
d ¼ 5:31 (1H,d, J ¼ 2:5) and d ¼ 5:43 (1H,d, J ¼ 4),respectively [18],vinylic protons for NAD analogs
1b (d: 6.18 and 7.89) and 2b (d: 6.39 and 7.4).
ND,not detected.
Table 4
1H NMR study of the transglycosidation with compound 2a
Reaction time (min)
7
60
120
240
360
Disappeared NADþ (%)
Analog 1b
Analog 2b
6
42
26
46
28
57
28
43
29
78
37
32
31
87
46
26
28
ND
Traces
ND
a- and b-ADP–ribose
Note. Pyridine derivative 2a (1 mg),NAD þ (1.5 mg) and NADase from pig brain (3 mg) were incubated
in a 10 mM deuterated potassium phosphate buffer (final volume 600 lL) at 37 °C. 1H NMR (500 MHz)
spectra were recorded at indicated time. The proportion of disappeared NADþ,analogs 1b and 2b and
ADP–ribose was calculated from the integration of characteristic peaks as described in Table 2.
ND,not detected.
Table 5
Kinetic constants of the transglycosidation reaction for pig brain NADase
Substrates
Km (lM)
Vmax (lM/min/mg prot)
V )
max=Km (minÀ1 mg protÀ1
1a
2a
300
760
1460
950
470
34b
77a
45a
67
0.257
0.059
0.046
0.040
0.038
—
3a
4a
38
18
77b
5a
NADþ
Note. The reaction conditions are given in Section 2. Except for compounds 1a and 1b (see text),the
transglycosidation reactions were followed at the kmax of the generated NADþ analogs (Table 4).
a Determined at 380 nm, e ¼ 21,500 MÀ1 cmÀ1 for a mixture of analogs 1b and 2b (9/1).
b Determined with compound 1a as the fixed substrate.

K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
301
two isomers. Indeed,taking the Km values in Table 5 into account,the enzymatic re-
actions were performed at a concentration corresponding to 35-fold the Km with pyr-
idine 1a and to 15-fold the Km with pyridine 2a.
Due to the fact that transglycosylation reaction with either reagent 1a or 2a fur-
nished a mixture of same ratio of (E) and (Z) isomers 1b and 2b and that pure NADþ
products 1b and 2b could not be isolated,the kmax and e spectrophotometric data of
the two products could not be determined accurately. The absorption bands of the
2 isomers may have different intensity because (E) 1-methyl-4-[2-(methylsulfanyl)-
vinyl]pyridinium iodide has an 1.5-fold higher e than the (Z)-isomer [9]. Neverthe-
less,the e of the isolated NADþ analog was determined from the 9/1 mixture of
the two isomers 1b and 2b (Table 1).
The problem encountered with derivatives 1a and 2a led us to synthesize and to
test compounds 3a–5a. All the 3 derivatives were substrate of the NADase. The
1H and 13C NMR spectra of the isolated NADþ analogs 3b–5b agreed with the
anticipated structures.
3.3. Transglycosidation reaction with pyridine derivatives 1a–5a catalyzed by pig brain
NADase: kinetic studies
The transglycosidation reaction may satisfy several criteria to be suitable as an
enzymatic assay: linearity of the absorbance increase in function of time,saturat-
ing kinetics,proportionality between the rate and enzyme concentration. Assays
were performed with compound 1a. The reaction was followed at a slightly higher
wavelength (380 nm) than the kmax to minimize the ground level of absorbance. So
the reaction could be followed for a longer time and with high concentrations of
compound 1a without saturation of the spectrophotometer. The reaction was gen-
erally followed over 10 min to determine the rate of the enzymatic reaction,but
the increase of the absorbance was linear for at least 1 h (DDO380 ¼ 0:4). In the
enzymatic reaction the pyridine derivatives competed with water producing
NADþ analogs and ADP–ribose,respectively. The linearity of the absorbance in-
crease indicated that the product ratio NADþ analog/ADP–ribose was constant
during the reaction time and that the enzyme remained stable. The rate of the
transglycosidation was also proportional to the concentration of protein (data
not shown).
The kinetic constants (Km and Vmax) of pyridines 1a–5a are given in Table 5. The
transglycosidation rates with pyridines 1a and 2a were calculated by using the molar
extinction coefficient obtained with the 9/1 mixture of compounds 1b and 2b. How-
ever as the equilibrium was plainly in favor of analog 1b and was reached rapidly,the
transglycosidation rate for compound 1a could reasonably be considered as close to
the true rate. On the other hand,the rate calculated with this e value for the base
exchange with compound 2a would be underestimated,the equilibrium being
reached more slowly.
The maximal rate for the transglycosidation (Vmax ¼ 77lM/min/mg prot with
compound 1a) was lower than the NADþ disappearance rate determined by the
KCN method (Vmax ¼ 197lM/min/mg prot with NADþ as the substrate).

302
K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
Table 6
Kinetic constants of the transglycosidation reaction for ADP-ribosyl cyclase of Aplysia californica
Substrates
Km (lM)
Vmax (mM/min/mg prot)
V )
max=Km (minÀ1 mg protÀ1
1a
2a
3a
4a
5a
110
37
116a
45a
78
1055
1216
1733
552
45
58
32
94
290
324
Note. The reaction conditions are given in Section 2. Except for compounds 1a and 2a (see text),the
transglycosidation reactions were followed at the kmax of the generated NADþ analogs (Table 4).
a Determined at 380 nm, e ¼ 21,500 MÀ1 cmÀ1 for a mixture of analogs 1b and 2b (9/1).
3.4. Transglycosidation reaction with pyridine derivatives 1a–5a, catalyzed by ADP-
ribosyl cyclase from A. californica
The kinetic constants of the cyclase for compounds 1a–5a are given in Table 6.
Pyridine 3a seemed to be the most efficient substrate. The NADþ hydrolysis rate
measured with the KCN method (2652 mM/min/mg prot) at saturating concentra-
tion of NADþ was about 34-fold higher than that of the transglycosidation with
compound 3a.
4. Discussion
The base exchange of nicotinamide with pyridine derivatives was considered as a
noteworthy property of NADases to synthesize NADþ analogs [10]. As pointed out
for the calf spleen NADase [21],the transglycosidation reaction may occur according
to a ping-pong Bi Bi mechanism. After cleavage of the glycosidic nicotinamide–ri-
bose bond and release of nicotinamide,pyridine derivatives would compete with
water for the nucleophilic attack of the ADP-ribosyl oxocarbenium ion intermediate.
The corresponding NADþ analogs are generated with retention of the b configura-
tion [10]. All NADases do not perform this reaction. The enzymes catalyzing the
transglycosidation are easily revealed by the fact that their activity is inhibited by
nicotinamide [15].
Only few extensive studies of the transglycosidation reaction catalyzed by NAD-
ases are reported. The reason could be the difficulty to perform the enzymatic as-
says,based on the separation of the different reaction products by high
performance liquid chromatography [7,12]. Yost and Anderson described some as-
pects on the translycosidation of venom NADase such as substrate specificity of the
base exchange [11,12,22]. For CD38, radiolabelled nicotinamide was used to study
the kinetic mechanism of base exchange [1]. The development of a specific assay
for the transglycosidation reaction would be helpful. With our test the progress
of the reaction is monitored continuously by following the absorbance increase
due to the formation of the NADþ analogs and the rate is calculated instanta-
neously from the slope.

K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
303
All five pyridine derivatives tested were substrate of the enzyme in the transglycos-
idation reaction. The pyridine derivatives had relatively high pKa values (Table 2).
Therefore,as pointed out by Tarnus and Schuber [23] which found a 6000-fold lower
hydrolysis rate for a NADþ analog with a pyridine derivative having a pKa of nearly
6.0 with regard to NADþ,the hydrolysis of the NAD þ analogs by pig brain NADase
was insignificant. The high pKa values required performing the reactions at pH 8,
slightly higher than the usual,pH 7.2. The enzyme has a broad pH optimum (pH
5 to 8.5) and the influence on the rate would be negligible (<10%) [13].
The isomerization of NADþ analogs 1b and 2b pointed out by the NMR analysis
of the enzymatic reaction could be attributed to the push–pull nature of the double
bond. As reported for the (Z) to (E) isomerization of (Z) 1-methyl-4-[2-(methylsulfa-
nyl)-vinyl]pyridinium iodide,the electron transfer from the electron donating meth-
ylsulfur group to the electron withdrawing pyridinium group lowers the
isomerization barrier [24]. While the isomerization in DMSO-d6 of the (Z) 1-meth-
yl-4-[2-(methylsulfanyl)-vinyl]pyridinium iodide to the (E) isomer was rapid and
quantitative,the isomerization of the analogs 1b and 2b in the buffered medium
was slower and gave a 9/1 mixture in favor of analog 1b. The nature of the solvent
and of the N-substituent (a ribofuranosyl group instead of a methyl group) might be
at the origin of these differences.
The observed saturating kinetics confirmed that a specific ternary complex be-
tween the enzyme,the ADP-ribosyl oxocarbenium ion intermediate and the pyridine
derivative takes place before the nucleophilic attack. The transglycosidation reaction
did not result from a bimolecular reaction between the ADP-ribosyl oxocarbenium
ion intermediate and the pyridine derivative. With pig brain NADase the Vmax=Km
ratio indicates that compound 1a was the most efficient (Table 5). Compound 1a
had the lowest Km,about 2- to 3-fold lower than its isomer 2a. Any of the pyridines
3a–5a synthesized to avoid the problem of the cis–trans isomerization of the meth-
ylsulfanyl-vinyl group,can be used as an alternative. However compounds 3a and
4a were poorly soluble in buffer and had high Km values. So they were not suitable
for kinetic studies because the enzymatic reaction could not be performed in optimal
conditions of substrate concentration. The transglycosidation rate with compound
5a was 4-fold slower than the exchange with compound 1a.
The transglycosidation assay was applied to ADP-ribosyl cyclase from A. cal-
ifornica. This multifunctional enzyme catalyzes not only the formation of cyclic-
ADP–ribose but also the exchange of nicotinamide of NADðPÞþ with nicotinic
þ
acid [7]. As cyclic ADP–ribose,the NADP analog,nicotinic acid adenine dinu-
cleotide phosphate (NAADPþ) is implicated in the Ca2þ release mechanism while
no activity for the NADþ analog NAADþ has been detected [7,8]. The base ex-
change reaction is dominant over the cyclization at acidic pH. At pH 7.0 with
NADþ as the substrate,even in the presence of nicotinic acid,only the synthesis
of cyclic ADP–ribose is observed [7]. With compounds 1a–5a the transglycosida-
tion reaction occurred even at pH 8.0. Our assay would make it possible to deter-
mine the factors privileging one reaction or the other and to show if the base
exchange between nicotinamide and nicotinic acid is likely to occur in vivo and
is not an in vitro reaction.

304
K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
The comparison of the kinetic parameters of pig brain NADase and ADP-ribosyl
cyclase showed some differences with regard to their affinity for the pyridine deriva-
tives. The Km values of the cyclase were lower than those of the NADase and the two
enzymes showed a different specificity for the pyridine derivatives. If the NADase
seemed to better recognize compounds 1a and 5a,the cyclase preferred compounds
2a–4a. This reflects a different nicotinamide binding site,the recognition site for the
pyridine derivatives.
The hydrolysis of NADþ (measured with the KCN method) was faster (about 2.5-
and 34-fold for pig brain NADase and ADP-ribosyl cyclase,respectively) than the
transglycosidation. As nicotinamide,the pyridine derivatives would inhibit the
NADþ cleavage. In the case of calf spleen NADase,a 3-fold decrease of the reaction
rate was observed in the presence of nicotinamide. The base exchange with [14C]nic-
otinamide was about 10-fold slower than the NADþ cleavage [21].
5. Conclusion
We developed a continuous spectrophotometric assay allowing the determination
of the transglycosidation reaction of NADases. The assay is very sensitive owing to
the high absorption coefficient of the NADþ analogs. More the latter absorb in the
near visible so that the measurements would not be hindered by the absorption of pro-
teins and NADðPÞþ. Optimal assay conditions have been determined for the two en-
zymes and are the following: ½NADþꢀ ¼ 0:45 mM and ½compound 1aꢀ ¼ 3 mM for
the pig brain NADase; ½NADþꢀ ¼ 0:6 mM and ½compound 3aꢀ ¼ 1 mM for the
ADP-ribosyl cyclase. The use of our assay as a common method to measure the
enzymatic of NADases could also be envisaged.
Acknowledgments
The authors are indebted to Dr. F. Schuber and Dr. H. Muller-Steffner for the gift
of ADP-ribosyl cyclase from Aplysia californica and helpful discussions. We are
grateful to R. Graff,M. Martigneaux,and J.D. Sauer for the NMR measurements.
References
[1] A.A. Sauve,C. Munshi,H.C. Lee,V.L. Schramm,Biochemistry 37 (1998) 13239–13249.
[2] C. Cakir-Kiefer,H. Muller-Steffner,F. Schuber,Biochem. J. 349 (2000) 203–210.
[3] H.C. Lee,R. Aarhus,Biochim. Biophys. Acta 1164 (1993) 68–74.
[4] G. Orsomando,V. Polzonetti,P. Natalini,Comp. Biochem. Physiol. B: Biochem. Mol. Biol. 126B
(2000) 89–98.
[5] M.R. Hellmich,F. Strumwasser,Cell Regul. 2 (1991) 193–202.
[6] H.C. Lee,R. Aarhus,Cell Regul. 2 (1991) 203–209.
[7] R. Aarhus,R.M. Graeff,D.M. Dickey,T.F. Walseth,H.C. Lee,J. Biol. Chem. 270 (1995) 30327–
30333.

K. Pacaud et al. / Bioorganic Chemistry 31 (2003) 288–305
[8] H.C. Lee,Physiol. Rev. 77 (1997) 1133–1164.
305
[9] M. Holler,H.-S. Sin,A. James,A. Burger,D. Tritsch,J.-F. Biellmann,Chem. Eur. J. 6 (2000) 2053–
2062.
[10] B.M. Anderson,in: J. Everse,B. Anderson,K. You (Eds.),The Pyridine Nucleotide Coenzymes,
Academic Press,New York,1982,pp. 91–133.
[11] D.A. Yost,B.M. Anderson,J. Biol. Chem. 257 (1982) 767–772.
[12] D.A. Yost,B.M. Anderson,J. Biol. Chem. 258 (1983) 3075–3080.
[13] H.G. Windmueller,N.O. Kaplan,Biochim. Biophys. Acta 56 (1962) 388–391.
[14] M.M. Bradford,Anal. Biochem. 72 (1976) 248–254.
[15] N.O. Kaplan,Methods Enzymol. 2 (1955) 660–663.
[16] N.J. Oppenheimer,in: J. Everse,B. Anderson,K. You (Eds.),The Pyridine Nucleotide Coenzymes,
Academic Press,New York,1982,pp. 51–89.
[17] J.L. Czeisler,D.P. Hollis,Biochemistry 14 (1975) 2781–2785.
[18] J. Moss,N.J. Oppenheimer,R.E. West Jr,S.J. Stanley,Biochemistry 25 (1986) 5408–5414.
[19] J.A. Goodwin,I.M.Y. Kwok,B.J. Wakefield,Synthesis 11 (1990) 991–992.
[20] A. Raap,R.G. Micetich,Can. J. Chem. 46 (1968) 1057–1063.
[21] F. Schuber,P. Travo,M. Pascal,Eur. J. Biochem. 69 (1976) 593–602.
[22] D.A. Yost,B.M. Anderson,J. Biol.Chem. 256 (1981) 3647–3653.
[23] C. Tarnus,F. Schuber,Bioorg. Chem. 15 (1987) 31–42.
[24] H.-S. Sin,M. Holler,A. Burger,J.-F. Biellmann,Tetrahedron Lett. 38 (1997) 3585–3586.