Efficient synthesis of Guanfu base G via highly regioselective lipase-catalyzed acylation in non-aqueous phase

Article abstract of DOI:10.1016/j.bmcl.2012.03.066

Lipase-catalyzed acylation of Guanfu alcohol-amine (GFAA) with vinyl acetate (VA) was performed in non-aqueous system for the preparation of Guanfu base G (GFG), a plant-originated alkaloid with significant antiarrhythmic activity. Among the eight lipases from different origins, Novozym 435 was found to be the best biocatalyst. The most suitable molecular sieve amount, substrate concentration, molar ratio of VA to GFAA, enzyme amount and reaction temperature were proved to be 40 mg/mL, 6 μmol/mL, 10:1, 2 mg/mL and 50 °C, respectively. A maximum GFG yield of 37.4% was achieved under the selected conditions with methanol served as the optimal reaction medium. The structure of the acetylated product was elucidated by ¹H NMR and ¹³C NMR analysis.

Full text of DOI:10.1016/j.bmcl.2012.03.066

Bioorganic & Medicinal Chemistry Letters 22 (2012) 3100–3103
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Efficient synthesis of Guanfu base G via highly regioselective
lipase-catalyzed acylation in non-aqueous phase
Ye Xia a, Ji-Hua Liu a, , Jing-Han Liu c, Bo-Yang Yu b,
⇑
⇑
a Department of Complex Prescription of Traditional Chinese Medicine, China Pharmaceutical University, Long Mian Road 639, Nanjing 211198, PR China
b Key Laboratory of Modern Chinese Medicines, China Pharmaceutical University, Nanjing 211198, PR China
c Department of Natural Medicinal Chemistry, China Pharmaceutical University, Nanjing 210009, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Lipase-catalyzed acylation of Guanfu alcohol-amine (GFAA) with vinyl acetate (VA) was performed in
non-aqueous system for the preparation of Guanfu base G (GFG), a plant-originated alkaloid with signif-
icant antiarrhythmic activity. Among the eight lipases from different origins, Novozym 435 was found to
be the best biocatalyst. The most suitable molecular sieve amount, substrate concentration, molar ratio of
Received 25 December 2011
Revised 8 March 2012
Accepted 16 March 2012
Available online 21 March 2012
VA to GFAA, enzyme amount and reaction temperature were proved to be 40 mg/mL, 6 lmol/mL, 10:1,
2 mg/mL and 50 °C, respectively. A maximum GFG yield of 37.4% was achieved under the selected con-
ditions with methanol served as the optimal reaction medium. The structure of the acetylated product
was elucidated by 1H NMR and 13C NMR analysis.
Keywords:
Guanfu base G
Vinyl acetate
Lipases
Ó 2012 Elsevier Ltd. All rights reserved.
Acylation
Guanfu base A (GFA), a naturally occurring diterpene alkaloid
originally isolated from the root of Aconitum coreanum (Lévl.)
Rapaics, has been proved to exhibit significant antiarrhythmic
activity.1–3 In August 2005, hydrochloric GFA received the certifi-
cate of State Category I New Drug issued by Chinese Health Minis-
try.4 To date, GFA has been extensively used for the therapeutic
treatment of arrhythmia in clinical practice. It is more appealing
that Guanfu base G (GFG), an acetyl derivative of GFA, showed
improved antiarrhythmic activity compared with GFA in various
animal models.5,6 In addition, GFG could antagonize the ventricular
fibrillation induced by electrical stimulation in isolated guinea pig
hearts, which could not be achieved by GFA.7 According to HPLC
analysis, the content of GFA and GFG accounted for only 0.124%
and 0.012% in Aconitum coreanum, respectively.8 Moreover, the
natural resource of Aconitum coreanum was extremely limited,
with only a small-scale distribution in the Northeast of China. Cur-
rently, GFA was mainly prepared by traditional extraction process.
Since the content of GFG was much lower than that of GFA, it was
not feasible to produce GFG by means of extraction and isolation.
During the preparation of GFA, Guanfu alcohol-amine (GFAA)
was obtained in large quantity, which was found to be an abundant
and inactive component in Aconitum coreanum. Since the content
of GFAA was up to 1% in plants, it would be of great significance
if GFAA was successfully converted to GFG.
In our laboratory, chemical methods have been applied for the
structural modification of GFAA. However, conventional chemical
routes for the selective acylation of GFAA resulted in a poor yield
of GFG, let alone the numerous by-products and the environmental
concerns of the process. In order to overcome the above shortcom-
ings, enzymatic catalysis was introduced in present study for the
acylation of GFAA to produce GFG. Due to the remarkable selectiv-
ity and specificity, enzymatic catalysis has been frequently used in
the field of food, energy and medicine for several decades,9–11 and
has been demonstrated to be an effective and practical tool in the
development of green chemistry.12,13 So far, however, there has
been no report on the enzymatic synthesis of GFG. We herein pres-
ent the selective acylation of GFAA by a lipase to produce GFG for
the first time. Since the source of lipase and reaction conditions are
responsible for dramatic differences in the reaction outcome,
several crucial parameters were also systematically investigated
with a view to obtain the desired product in high yield. The
lipase-catalyzed acylation of GFAA could be a promising protocol
for the efficient synthesis of GFG, a potential antiarrhythmic candi-
date with considerably improved activity than GFA.
Amongst the numerous enzymes in related researches, lipases
have been successfully employed in the selective acylation of
various target molecules.14 In order to determine the optimal
biocatalyst, eight commercially available lipases from different ori-
gins were screened for their efficiency in the acetylation of GFAA
with vinyl acetate (VA). An overall reaction procedure was outlined
in Scheme 1. In a typical experiment, 5 mL of anhydrous solution
containing 0.028 mmol GFAA, 0.14 mmol VA and enzyme was
incubated at 30 °C with continuous shaking of 200 rpm for 48 h.
⇑
Corresponding authors. Tel.: +86 025 86185157; fax: +86 025 86185158.
E-mail addresses: jihualiu88@163.com (J.-H. Liu), boyangyu59@163.com
(B.-Y. Yu).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.03.066

Y. Xia et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3100–3103
3101
25 26
CH3COO
HO
13
12
13
12
23 24
17
17
CH3COO
HO
20
11
11
CH3COOCH=CH2
CH3CHO
16
15
16
15
21 22
20
OH
1
OH
1
CH3COO
HO
14
14
9
9
6
8
7
2
3
8
7
10
10
2
3
N
N
Lipases
Organic solvents
5
5
4
4
6
19
19
18
18
Scheme 1. Lipase-catalyzed acylation of GFAA with VA non-aqueous medium.
Aliquots were withdrawn periodically from the reaction mixture
and filtered to remove the enzymes. The filtrates were then evap-
orated to dryness and redissolved with methanol for HPLC analy-
sis.15 As shown in Table 1, considerable conversion (15.3%) was
achieved by Candida antarctica lipase B (CAL-B, Novozym 435) with
GFG as the sole product. Slight activities were also observed for li-
pases from Burkholderia cepacia (Lipase PS) and Pseudomonas fluo-
rescens (Lipase AK), while the rest exhibited no catalytic activity for
the synthesis of GFG. According to the above results, Novozym 435
was selected for the subsequent condition optimization process.
Since enzyme activity depends largely on reaction medium, the
acylation of GFAA was performed in nine organic solvents with dif-
ferent polarities as listed in Table 2. The result indicated that this
acylation reaction could only take place in methanol, ethanol and
n-butanol, which might be attributed to the fact that GFAA enjoyed
better solubility in alcohol solvents. In addition, the initial rate and
GFG yield were closely related to the polarity of the solvents. The
most favorable result was obtained in methanol with the lowest
logP value of À0.76, whereas both initial rate and GFG yield
decreased along with the increase of hydrophobicity.
It is well established that a minimal amount of water is neces-
sary for the enzyme to ensure its optimal conformation and then to
become optimally active. However, an excess of water decreases
the enzyme catalytic activity from both kinetic and thermody-
namic points of view.16 Among the numerous water removal pro-
tocols, molecular sieve was frequently used for the adsorptive
control of water in non-aqueous lipase-catalyzed reaction.17,18
The effect of concentration of 4 Å molecular sieve on the yield of
GFG was then investigated. As described in Fig. 1, the highest initial
rate and substrate conversion was achieved at 40 mg/mL, while
further increase in the amount of molecular sieve led to an obvious
decline in the yield of GFG, which was ascribed to the loss of essen-
tial water.
followed by a decrease with further increase of GFAA concentra-
tion. This might be partially due to the fact that all active sites of
the enzymes were saturated, thus additional substrate could not
lead to a greater rate of product formation. Moreover, at high sub-
strate levels, most of the GFAA remained undissolved in methanol,
which would be the main reason for the lower GFG yield at higher
concentration, since the yield was calculated by the molar ratio of
GFG to initial GFAA.
It has been reported that in the enzymatic acylation with vinyl
esters served as acyl donor, an excessive amount of acyl donor is
necessary due to the presence of a side reaction, that is, the hydro-
lysis of vinyl esters.19 Therefore, an appropriate molar ratio of VA
to GFAA was crucial for the effective synthesis of GFG. As shown
in Fig. 3, an enhanced substrate conversion was achieved with
Table 2
Solvent effects on the Novozym 435-catalyzed acetylation of GFAA with VA
Solvent
LogP
V0 (mmol LÀ1 hÀ1
)
GFG (%) Regioselectivity (%)
DMF
À1.00 nd
À0.76 12.4
À0.34 nd
À0.30 7.1
À0.24 nd
0.35 nd
0.80 2.6
1.30 nd
4.00 nd
nd
19.6
nd
15.3
nd
nd
8.1
nd
nd
nd
100
nd
100
nd
nd
100
nd
nd
Methanol
Acetonitrile
Ethanol
Acetone
t-Butanol
n-Butanol
Cyclohexanone
Hexane
Except for biocatalyst category and reaction medium, the effect
of substrate concentration on the lipase-catalyzed reaction was
also investigated. As seen in Fig. 2, GFG yield reached the maxi-
mum when the substrate concentration rose to
6 lmol/mL,
Table 1
Acylation of GFAA with VA catalyzed by different lipases in ethanol
Enzyme
V0 (mmol LÀ1 hÀ1
)
GFGb (%)
Regioselectivity (%)
Novozym 435
Lipase AK
Lipase PS
Lipase G
PPL
7.1
3.4
1.64
nda
nd
15.3
7.4
4.8
nd
100
100
100
nd
nd
nd
RNL
nd
nd
nd
Lipase A
Lipase M
nd
nd
nd
nd
nd
nd
a
Not determined.
Conversion determined by HPLC.
Figure 1. Effect of 4 Å molecular sieve concentration on the acylation of GFAA with
b
VA catalyzed by Novozym 435 in methanol (aConversion determined by HPLC).

3102
Y. Xia et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3100–3103
Figure 4. Effect of enzyme amount on the acylation of GFAA with VA catalyzed by
Novozym 435 in methanol.
Figure 2. Effect of substrate concentration on the acylation of GFAA with VA
catalyzed by Novozym 435 in methanol.
Preparative scale acylation of GFAA (200 mg) by Novozym 435
was carried out under the above optimized conditions. The solu-
tion was pooled, filtered, and evaporated to dryness under reduced
pressure. The residue (336 mg) was subjected to silica gel column
chromatography eluting with chloroform–methanol (70:1, v/v),
followed by further purification on Sephadex LH-20 column
chromatography (methanol) to afford GFG (47 mg). The purified
compound was structurally characterized by 1H NMR and 13C
NMR analysis and was confirmed to be GFG undoubtedly.
In conclusion, an efficient and practical approach for the syn-
thesis of GFG through the lipase-catalyzed acylation of the inactive
GFAA was proposed in this Letter. A maximum GFG yield of 37.4%
was achieved under the optimized conditions, which provided a
feasible solution for overcoming the deficiency of GFG from natural
origin and thus contributed to the conservation of the wild medic-
inal plant resources. Furthermore, this non-aqueous enzymatic
reaction showed outstanding regioselectivity with GFG as the sole
the ratio rising from 1:1 to 10:1, above which further increase of
molar ratio had a marginal effect on the yield of GFG.
The influence of enzyme amount on the acetylation of GFAA
was also evaluated with the amount of Novozym 435 ranging from
1 to 3 mg/mL. As indicated in Fig. 4, GFG yield increased with the
increase of catalyst amount and reached a maximum value of
23.1%. However, there was no appreciable increase in either initial
rate or substrate conversion when the amount of Novozym 435
was beyond 2 mg/mL, suggesting that the enzyme amount was
excessive for the acetylation of GFAA.
Temperature was associated with the solubility of substrate and
the stability of enzyme, thus had profound influence on the acety-
lation of GFAA.20 As illustrated in Fig. 5, initial rate and substrate
conversion increased rapidly when the temperature rose from 30
to 50 °C with 50 °C being optimal. When the temperature went
up sequentially to 60 °C, an apparent decrease in GFG yield was
observed due to the loss of lipase activity.
Figure 3. Effect of molar ratio of VA to GFAA on the acylation of GFAA catalyzed by
Figure 5. Effect of temperature on the acylation of GFAA with VA catalyzed by
Novozym 435 in methanol.
Novozym 435 in methanol.

Y. Xia et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3100–3103
4. See the announcement: Chin. J. Nat. Med. 2006, 4, 76.
3103
product, thus simplified the subsequent purification of GFG as well
as the recovery of remaining GFAA, which could be reused for
further reaction. Over the last few decades, lipases have been
known as an emerging and versatile biocatalyst capable of catalyz-
ing various acylation reactions. However, this was the first success-
ful lipase-catalyzed conversion of GFAA to GFG that could never be
accomplished by routine chemical methods, expanding the
application of lipases into the field of structural modification of
alkaloids. Hopefully, the present research will provide a new
insight into the exploration of new antiarrhythmia agents.
5. Hou, D. H.; Wu, L. Y.; Xu, W. R.; Liu, J. H. J. Chin. Pharm. Univ. 1987, 18, 268.
6. Zhang, L. Y.; Ji, H. F.; Xia, L. W.; Liu, J. H.; Xing, Y. W. J. Chin. Pharm. Univ. 1993,
24, 242.
7. Hou, D. H.; Wang, Q. J.; Zhang, L. Y.; Li, Y. M.; Xu, W. R.; Wu, L. Y.; Liu, J. H. Chin.
J. Pharmacol. Toxicol. 1988, 2, 183.
8. Determination of GFA and GFG content in Aconitum coreanum (Lévl.) Rapaics:
1 g of dried and pulverized herbal materials were ultrasonic extracted with
50 mL of 1% HCl solution for 30 min. Five milliliter of sample was then removed
and centrifuged at 4000 rpm for 30 min. After filtrated through a 0.45 lm
membrane, an aliquot of 10
the method in Ref. 14.
lL solution was subjected to HPLC analysis using
9. Li, Z. J.; Cai, L.; Qi, Q. S.; Styslinger, T. J.; Zhao, G. H.; Wang, P. G. Bioorg. Med.
Chem. Lett. 2011, 21, 5084.
10. Zhang, L. P.; Sun, S. Z.; Xin, Z.; Sheng, B. Y.; Liu, Q. Fuel 2010, 89, 3960.
11. Li, X. F.; Zong, M. H.; Zhao, G. L. Bioorg. Med. Chem. Lett. 2010, 20, 4125.
12. Poliakoff, M.; Fitzpatrick, J. M.; Farren, T. R.; Anastas, P. T. Science 2002, 297,
807.
13. Paravidino, M.; Hanefeld, U. Green Chem. 2011, 13, 2651.
14. González-Sabín, J.; Morán-Ramallal, R.; Rebolledo, F. Chem. Soc. Rev. 2011, 40,
5321.
Acknowledgments
This work was financially supported by the Priority Academic
Program Development of Jiangsu Higher Education Institutions
(PAPD) and the ‘111 Project’ from the Ministry of Education of Chi-
na and the State Administration of Foreign Expert Affairs of China
(No. 111-2-07).
15. HPLC analysis: HPLC analysis was performed on an Agilent series 1100 HPLC
instrument with
a DAD detector at 205 nm, using a ODS-2 C18 column
(4.6 Â 250, m). The mobile phase was composed of 0.2% SDS buffer
5
l
(containing 0.1% triethylamine, adjusted with phosphoric acid to pH 3.0) and
acetonitrile using a gradient elution of 20–50% acetonitrile at 0–30 min with a
flow rate of 1.0 mL/min. The chromatographic peaks of GFAA and GFG were
obtained at 8.9 and 18.2 min, respectively.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.03. 066.
16. Colombié, S.; Tweddell, R. J.; Condoret, J. S.; Marty, A. Biotechnol. Bioeng. 1998,
60, 362.
17. Li, R.; Jia, C. S.; Yue, L.; Zhang, X. M.; Xia, Q. Y.; Zhao, S. L.; Feng, B.; Zhong, F.;
Chen, W. J. J. Agric. Food. Chem. 2010, 1898, 58.
18. He, W. S.; Jia, C. S.; Ma, Y.; Yang, Y. B.; Zhang, X. M.; Feng, B.; Yue, L. J. Mol. Catal.
B: Enzym. 2010, 67, 60.
References and notes
19. Weber, H. K.; Weber, H.; Kazlauskas, R. J. Tetrahedron: Asymmetry 1999, 102,
635.
20. Habulin, M.; Sabeder, S.; Paljevac, M.; Primozic, M.; Knez, Z. J. Supercrit. Fluids
2007, 43, 199.
1. Wang, F. P.; Chen, Q. H.; Liu, X. Y. Nat. Prod. Rep. 2010, 27, 529.
2. Yang, X. J.; Wang, G. J.; Ling, S. S.; Qiu, N. Y.; Wang, G. G.; Zhu, J.; Liu, J. H. J.
Chromatogr., B 2000, 740, 273.
3. Hou, D. H.; Li, L.; Wang, Q. J.; Liu, J. H. J. Chin. Pharm. Univ. 1981, 17, 68.