Synthesis of 1,2,3-triazole moiety-containing NAD analogs and their potential as redox cofactors

Article abstract of DOI:10.1016/j.tetlet.2011.08.152

Thirteen 1,2,3-triazole moiety-containing NAD analogs were synthesized and evaluated as redox cofactors. Noticeable enzymatic activities were observed when these analogs were employed as cofactors for malic enzyme or alcohol dehydrogenase. Our results ind

Full text of DOI:10.1016/j.tetlet.2011.08.152

Tetrahedron Letters 52 (2011) 5855–5857
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis of 1,2,3-triazole moiety-containing NAD analogs and their potential
as redox cofactors
Shuhua Hou a,b, Wujun Liu a, Debin Ji a,b, Qian Wang a, Zongbao Kent Zhao a,c,
⇑
a Dalian Institute of Chemical Physics, CAS, Dalian 116023, PR China
b Graduate University of Chinese Academy of Sciences, Beijing 100039, PR China
c Dalian National Laboratory for Clean Energy, Dalian 116023, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Thirteen 1,2,3-triazole moiety-containing NAD analogs were synthesized and evaluated as redox cofac-
tors. Noticeable enzymatic activities were observed when these analogs were employed as cofactors
for malic enzyme or alcohol dehydrogenase. Our results indicated that the natural NAD-dependent oxido-
reductases possessed flexible NAD binding pockets.
Received 22 June 2011
Revised 18 August 2011
Accepted 26 August 2011
Available online 1 September 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
NAD analogs
Malic enzyme
Alcohol dehydrogenase
Promiscuity
Nicotinamide adenine dinucleotide (NAD, Fig. 1) has attracted
major interests because of its prominent biological roles as redox
coenzyme, substrate of ADP-ribosyltransferases, deacetylases, and
DNA ligases. When it functions as the coenzyme to mediate an oxi-
dative reaction, the pyridium ring of the nicotinamide mononucle-
otide (NMN) moiety receives a hydride (H–) to form NADH, the
reduced form of NAD. The adenosine monophosphate (AMP) part
of the coenzyme is believed to endorse proper binding and shaping
NAD in the active site. As binding events can lead to protein confor-
mational changes,1 chemical diversifying the structure of NAD
remains attractive in terms of engineering novel redox cofactors.
Over the years, a great variety of NAD analogs have been pre-
pared.2 In terms of diversification of the AMP part, analogs were
synthesized incorporating adenine derivatives, purine derivatives,
triazine dye derivatives3–5 and other heterocycles. However, most
of these compounds were made by laborious procedures, and the
structural diversity remains limited. Considering that NAD is
involved in multifaceted biological processes, synthetic NAD ana-
logs can be valuable tools for chemists and biologists to elucidate
the mechanisms and control these processes. Here we would like
to report the synthesis of a novel series of NAD analogs containing
1,2,3-triazole moiety instead of the adenine structure (Fig. 1) with
the previous reported nucleotide analogs.6 We showed that these
analogs could act as cofactors for the recombinant malic enzyme
(ME)7 from Escherichia coli and the alcohol dehydrogenase (ADH)
from Saccharomyces cerevisiae.
The preparation of NAD analogs was envisioned by coupling of
NMN, prepared from NAD using ZrCl4 as the catalyst following the
literature process,8 and 1,2,3-triazole nucleotide analogs 1a–m,
prepared according to our previous procedure (Scheme 1).5 To
form the pyrophosphate linkage, one can in principle activate
either NMN or the 1,2,3-triazole nucleotide analogs. For the syn-
thesis of NAD, AMP moiety was activated as AMP-morpholidate,
followed by the reaction with NMN in the presence of MnCl2 and
MgSO4 in formamide.9,10 However, we preferred activating NMN
because it may lead to a standardized reaction procedure for a
number of analogs. In the literature, NMN was routinely activated
by carbonyldiimidazole, followed by the coupling reaction for
about 2 days to give the products in low yields.11,12 Activation of
NMN by diphenyl phosphorochloridate was also known, which in-
volved protection of the hydroxyl groups at 20- and 30-positions of
NMN, and took more than 3 days to complete the coupling
reaction.13
The triphenylphosphine/2,20-dipyridyl disulfide (Ph3P/(PyS)2)
redox pair, proposed by Mukaiyama,14 was reported as an effective
mediator for the synthesis of dinucleotide 50-triphosphates.15 We
found that this system was good in promoting the activation of
the phosphate group of NMN in the presence of 1-methylimidazole
in DMSO/DMF. Upon incubation of the mixture of NMN, 1-methyl-
imidazole, (PyS)2, and Ph3P at room temperature for 15 min, the
bis(tri-n-butylammonium) nucleotide analog 1 was added. The
reaction mixture was stirred at room temperature for 20–
420 min. The NAD analog was purified by DEAE-Sepharose FF
⇑
Corresponding author. Address: 457 Zhongshan Rd., Dalian 116023, PR China.
Tel./fax: +86 411 84379211.
E-mail address: zhaozb@dicp.ac.cn (Z.K. Zhao).
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.08.152

5856
S. Hou et al. / Tetrahedron Letters 52 (2011) 5855–5857
O
OH
HO
NH2
O
O
P
O-
O
N
N+
O
O
O
P
OH
N
N
O
3
3
2
2
4
N
C
B
NH2
HO
OH
4
4
1
1
NAD
3
A
9
O
OH
HO
1
2
3
4
5
6
7
8
10
11
12
NH2
O
O
O
Time (min)
N+
O
P
O-
O
O
P
OH
N
N
O
N
Figure 2. Ion chromatography analysis of the reaction mixture of ME catalyzed
reaction. All assays were carried out at 25 °C for 30 min in 50 mM HEPES, pH 7.2, in
the presence of 5 mM L-malate and 2.5 mM MnCl2. A, 1 mM 2c included; B, 1 mM 2c
R
HO
OH
and 0.7
l
M ME included; C, 1 mM NAD and 0.7
l
M ME included. The peaks 1, 2, 3
NAD analogs
ꢀ
and 4 were HCO2 (originated from the process of NAD analog purification),
pyruvate, Clꢀ and
L-malate, respectively.
Figure 1. Structures of NAD and 1,2,3-triazole moiety-containing analogs.
time to that of peak 2 found in line B, suggesting that ME executed
the decarboxylation of -malate to pyruvate using 2c as the cofac-
ꢀ
(HCO2 form) using water and 10 mM NH4HCO3 as the eluents.
L
Compounds 2a–m were obtained in yields of 19–51% (Scheme 1
and Supplementary data). These results suggested that the cou-
pling strategy with the Ph3P/(PyS)2 pair was versatile and power-
ful. Specifically, the reaction time was short and the separation
process was simple. It is likely that this strategy will be generally
applicable for the preparation of other NAD analogs.
tor. When the ME-catalyzed reaction was followed with a UV–vis
spectrophotometer, absorption evolution at 340 nm was observed.
These observations suggested that the replacement of the adenine
moiety of NAD with the 1,2,3-triazole moiety had limited spectro-
photometric effects on the reduced form of these compounds.
Therefore, we were able to measure the enzyme activities spectro-
scopically in the presence of NAD analogs.
We next tested the capacities of these NAD analogs as cofactors
for the recombinant ME. ME catalyzes an oxidative decarboxyl-
The fact that the decarboxylation of L-malate by ME failed to
ation of
L
-malate to yield pyruvate and CO2 with the reduction of
proceed to completion in the presence of 2c and other analogs pre-
cluded us from establishing the quantitative relationship between
the absorption at 340 nm and the concentration of the reduced
cofactor according to the published procedure.16 To measure the
NAD to NADH. We analyzed the incubation mixture of the
L
-malate
and 2c in the presence (Fig. 2, line B), or absence (line A) of ME by
ion chromatography (IC). It was clear that a new peak (peak 2) ap-
peared in line B. The positive control reaction (line C) with ME and
NAD gave the product pyruvate, which had an identical retention
millimolar extinction coefficients
(e340) of the reduced NAD
analogs, we recorded the absorbance changes at 340 nm in the
O
O
NH2
NH2
O
O
PPh3/(PyS)2
-O
P
O
N+
P
O
N
N+
N+
O
O
+O-
O-
NH4
N
N
HO
NMN
OH
HO
OH
R
N
O
Bu3N+
N
O
O
-O P O
O-
OH
N
HO
N
Bu3N+
NH2
O
O
HO
1a - 1m
OH
O
N+
O
P
O-
O
O
P
O-
O
N
N
+
R
NH4
HO
OH
2a - 2m
F3C
H3C
OH
R=
OMe
F
N
b (31%)
i (23%)
d (29%)
O
g (19%)
a (51%)
c (26%)
e (37%)
f (30%)
O
O
O
F
NH
NH
CH3
O-
NH2
m (43%)
l (38%)
h (23%)
k (40%)
j (36%)
Scheme 1. Preparation of NAD analogs. Reagents and conditions: (1) 1-methylimidazole (20 equiv), (PyS)2 (5 equiv), Ph3P (5 equiv), rt, 15 min; (2) 1a–m (2–3 equiv) in DMF,
rt, 20–420 min.

S. Hou et al. / Tetrahedron Letters 52 (2011) 5855–5857
5857
Table 1
NAD and its phosphorylated form, NADP, are the dominant re-
dox cofactors for oxidoreductase and dehydrogenase. Natural en-
zymes may prefer one of these two over the other as its cofactor.
There were examples where enzymes were mutated such that
the mutants had substantial cofactor preference changes compared
to the wild-type enzymes.17,18 It remains to be demonstrated that
enzymes can be engineered to take cofactor analogs with similar
efficiency to NAD. Our results also indicated that the NAD binding
pockets of these oxidoreductases were flexible, which may lead to
cofactor promiscuity. It has been proposed that the low levels of
promiscuity are favored in evolution, such that the natural en-
zymes might be turned into a much more proficient catalyst upon
mutagenesis.19,20 The fact that the wild-type ME and ADH could
facilitate their native redox chemistry using these 1,2,3-triazole
moiety containing NAD analogs as cofactors encouraged us to engi-
neer variant enzymes for better activities using directed evolution
strategies.21
Relative enzymatic activity data in the presence of NAD analogs and millimolar
extinction coefficients of the reduced form of NAD analogs
MEa (%)
ADHb (%)
e340 (mMꢀ1 cmꢀ1
)
c
NAD
2a
2b
2c
2d
2e
2f
2g
2h
2i
100
0.2
0.7
4.5
2.1
1.3
1.6
3.4
1.8
1.3
0.1
0.3
0.7
0.3
100
0.8
0.7
3.9
2.1
1.3
25.3
8.5
3.2
7.9
0.5
2.8
0.4
0.2
6.29
3.60
6.74
4.47
4.49
5.43
4.13
3.42
5.04
5.32
4.79
4.37
5.33
5.30
2j
2k
2l
2m
a
Assays were performed with 5 mM
L-malate, 5 mM MnCl2, 0.6 mM cofactor and
0.02–2.6 M ME in 50 mM HEPES (pH 7.2) at 25 °C.
l
In summary, we synthesized NAD analogs containing substi-
tuted 1,2,3-triazoles. Noticeable enzymatic activities were ob-
served when these analogs were employed as cofactors for ME or
ADH. Our results may inspire further study to develop novel bio-
logical redox systems or to research other NAD-dependent bio-
chemistry using synthetic cofactors.
b
Assays were performed with 0.5 M ethanol,1 mM cofactor and 6.8–780 nM ADH
(Sigma Cat. No. A3263) in 50 mM sodium pyrophosphate (pH 8.8) at 25 °C.
c
Assays were performed with 5 mM
L-malate, 5 mM MnCl2, 1 mM cofactor and
0.7
lM ME in 50 mM HEPES (pH 7.2) at 25 °C for 30 min. All assays were performed
in triplicate.
presence of ME and each synthetic NAD analog after 30 min. The
mixture was heat-shocked at 75 °C for 10 min to stop the reaction,
and the formation of pyruvate was quantified by IC. Based on the
stoichiometry of the ME-catalyzed reaction, the concentration of
each reduced NAD analog should be equal to the concentration
of pyruvate of the respective reaction. Taken together, we were
able to estimate e340 for the reduced form of all NAD analogs and
confirm the activity of ME in the presence of NAD analogs (Table
1). Except for the reduced form of 2b, other analogs gave lower
e340 values compared with that of NADH. It should be noted that
we determined the e340 value of NADH as 6.29 mMꢀ1 cmꢀ1, very
close to 6.22 mMꢀ1 cmꢀ1 used by most researchers, indicating that
the method had a high accuracy and should be applicable to the
analysis of other redox active NAD analogs. Accordingly, we esti-
mated the relative ME activities using these synthetic cofactors
(Supplementary data). It was clear that ME activities with NAD
analogs were less than 5% of that with NAD as the cofactor. Among
these analogs, 2c assumed the highest ME activity. These data also
indicated that the attachment of an aromatic moiety to the 1,2,3-
triazole ring in these analogs 2b–i was beneficial to act as a cofac-
tor for ME. Our results suggested that the NAD binding pocket of
ME was relatively flexible such that it retained a noticeable activity
using a number of structurally diversified NAD analogs as the
cofactor.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2011.08.152.
References and notes
1. James, L. C.; Tawfik, D. S. Trends Biochem. Sci. 2003, 28, 361–368.
2. Woenckhaus, C.; Jeck, R. In Coenzymes and Cofactors Pyridine Nucleotide
Coenzyme; John Wiley & Sons: New York, 1987; Vol. 2A, pp 450–568.
3. Ansell, R. J.; Small, D. A. P.; Lowe, C. R. J. Mol. Catal. B: Enzym. 1999, 6, 111–123.
4. Ansell, R. J.; Small, D. A. P.; Lowe, C. R. J. Mol. Recognit. 1999, 12, 45–56.
5. Ansell, R. J.; Lowe, C. R. Appl. Microbiol. Biotechnol. 1999, 51, 703–710.
6. Hou, S. H.; Liu, W. J.; Ji, D. B.; Zhao, Z. K. Bioorg. Med. Chem. Lett. 2011, 21, 1667–
1669.
7. Wang, J. X.; Tan, H. D.; Zhao, Z. K. Protein Expr. Purif. 2007, 53, 97–103.
8. Liu, R.; Visscher, J. Nucleosides Nucleotides 1994, 13, 1215–1216.
9. Lee, J.; Churchil, H.; Choi, W. B.; Lynch, J. E.; Roberts, F. E.; Volante, R. P.; Reider,
P. J. Chem. Commun. 1999, 729–730.
10. Kennedy, K. J.; Bressi, J. C.; Gelb, M. H. Bioorg. Med. Chem. Lett. 2001, 11, 95–98.
11. Zhang, B.; Wagner, G.; Weber, K.; Garnham, C.; Morgan, A.; Galione, A.; Guse,
A.; Potter, B. J. Med. Chem. 2008, 51, 1623–1636.
12. Dowden, J.; Moreau, C.; Brown, R. S.; Berridge, G.; Galione, A.; Potter, B. V. L.
Angew. Chem., Int. Ed. 2004, 43, 4637–4640.
13. Mort, C. J. W.; Migaud, M. E.; Galione, A.; Potter, B. V. L. Bioorg. Med. Chem.
2004, 12, 475–487.
14. Mukaiyama, T. Phosphorus, Sulfur Silicon Relat. Elem. 1976, 1, 371–387.
15. Abramova, T. V.; Vasileva, S. V.; Serpokrylova, I. Y.; Kless, H.; Silnikov, V. N.
Bioorg. Med. Chem. 2007, 15, 6549–6555.
16. Siegel, J. M.; Montgomery, G. A.; Bock, R. M. Arch. Biochem. Biophys. 1959, 82,
288–299.
17. Woodyer, R.; van der Donk, W. A.; Zhao, H. M. Biochemistry (Mosc.) 2003, 42,
11604–11614.
With e340 values in hand, we further measured the activity of
ADH using these NAD analogs as cofactors by continuously moni-
toring the absorbance at 340 nm. Noticeable ADH activities were
indeed observed with all analogs (Table 1). More interestingly, rel-
ative ADH activity in the presence of the analogs 2f, 2g, and 2i
reached 25.3%, 8.5%, and 7.9%, respectively, of that in the presence
of NAD. The fact that ADH had up to 25.3% relative activity in the
presence of these synthetic NAD analogs indicated that ADH had
a more flexible NAD binding pocket than ME.
18. Ma, C. W.; Zhang, L.; Dai, J. Y.; Xiu, Z. L. J. Biotechnol. 2010, 146, 173–178.
19. Tracewell, C. A.; Arnold, F. H. Curr. Opin. Chem. Biol. 2009, 13, 3–9.
20. Griswold, K. E.; Aiyappan, N. S.; Iverson, B. L.; Georgiou, G. J. Mol. Biol. 2006,
364, 400–410.
21. Turner, N. J. Nat. Chem. Biol. 2009, 9, 567–573.