New derivatives of ursolic acid through the biotransformation by Bacillus megaterium CGMCC 1.1741 as inhibitors on nitric oxide production

Article abstract of DOI:10.1016/j.bmcl.2017.03.076

Microbial transformation of ursolic acid (1) by Bacillus megaterium CGMCC 1.1741 was investigated and yielded five metabolites identified as 3-oxo-urs-12-en-28-oic acid (2); 1β,11α-dihydroxy-3-oxo-urs-12-en-28-oic acid (3); 1β-hydroxy-3-oxo-urs-12-en-28, 13-lactoe (4); 1β,3β, 11α-trihydroxyurs-12-en-28-oic acid (5) and 1β,11α-dihydroxy-3-oxo-urs-12-en-28-O-β-D-glucopyranoside (6). Metabolites 3, 4, 5 and 6 were new natural products. Their nitric oxide (NO) production inhibitory activity was assessed in lipopolysaccharide (LPS) – stimulated RAW 264.7 cells. Compounds 3 and 4 exhibited significant activities with the IC₅₀ values of 1.243 and 1.711?μM, respectively. A primary structure-activity relationship was also discussed.

Full text of DOI:10.1016/j.bmcl.2017.03.076

Accepted Manuscript
New derivatives of ursolic acid through the biotransformation by Bacillus meg-
aterium CGMCC 1.1741 as inhibitors on nitric oxide production
Chao Zhang, Shao-Hua Xu, Bai-Ling Ma, Wei-wei Wang, Bo-Yang Yu, Jian
Zhang
PII:
S0960-894X(17)30324-4
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.03.076
BMCL 24828
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
16 January 2017
11 March 2017
24 March 2017
Please cite this article as: Zhang, C., Xu, S-H., Ma, B-L., Wang, W-w., Yu, B-Y., Zhang, J., New derivatives of
ursolic acid through the biotransformation by Bacillus megaterium CGMCC 1.1741 as inhibitors on nitric oxide
production, Bioorganic & Medicinal Chemistry Letters (2017), doi: http://dx.doi.org/10.1016/j.bmcl.2017.03.076
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

New derivatives of ursolic acid through the biotransformation by
Bacillus megaterium CGMCC 1.1741 as inhibitors on nitric oxide
production
a
Chao Zhang , Shao-Hua Xu , Bai-Ling Ma , Wei-wei Wang , Bo-Yang Yu , Jian
a
b
b
b
†
†
Zhanga,*
a State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing, Jiangsu 210009, China
b Jiangsu Key Laboratory of TCM Evaluation and Translational Research, China Pharmaceutical University,
Nanjing, Jiangsu 211198, China
†
These authors contributed equally to this work.
* Corresponding author:
Corresponding author:
Jian Zhang Ph.D
State Key Laboratory of Natural Medicines,
China Pharmaceutical University,
24# Tong Jia Xiang Street,
Nan Jing, 210009, China
E-mail: 1020071849@cpu.edu.cn
Tel: 86-25-86185157
Fax: 86-25-86185158

Abstract
Microbial transformation of ursolic acid (1) by Bacillus megaterium CGMCC
1.1741 was investigated and yielded five metabolites identified as
3-oxo-urs-12-en-28-oic acid (2); 1β, 11α-dihydroxy-3-oxo-urs-12-en-28-oic acid (3);
1β-hydroxy-3-oxo-urs-12-en-28, 13-lactoe (4); 1β, 3β, 11α-trihydroxyurs-12-en-28-oic
acid (5) and 1β, 11α-dihydroxy-3-oxo-urs-12-en-28-O-β- D- glucopyranoside (6).
Metabolites 3, 4, 5 and 6 were new natural products. Their nitric oxide (NO) production
inhibitory activity was assessed in lipopolysaccharide (LPS) - stimulated RAW 264.7
cells. Compounds 3 and 4 exhibited significant activities with the IC50 values of 1.243
and 1.711 µM, respectively. A primary structure-activity relationship was also
discussed.
Keyword
Ursolic acid; Nitric oxide production; Microbial transformation; Bacillus megaterium
CGMCC 1.1741;
Nitric oxide (NO), a potent pro-inflammatory mediator, naturally generated from
L-arginine by the action of NO synthase (NOS), is involved in regulating numerous
physiological processes including neurotransmission and events of the immune
system.1 Nitric oxide (NO), superoxide (O2-) and peroxynitrite (ONOO-) may be
generated in excess during the host response against viral and antibacterial infections
and contribute to some pathogenesis by promoting oxidative stress, tissue injury and
cancer.2 Over-production of NO by inducible nitric oxide synthase (iNOS) generally
destroy functional normal tissues during acute and chronic inflammation.3 It has been
shown that NO can have pro-inflammatory (immunostimulatory, antiapoptotic ) or anti-
inflammatory ( immunosuppressive , pro–apoptotic ) activities.4 Therefore, appropriate
management of mediators may represent a therapeutic strategy to reduce inflammation.
Currently, although many medicines are applied in clinic to the treatment of
inflammation, more attention has been focused on the research on new substances of
natural bioactive constituents which have demonstrated effectiveness in the described

fields of application and low degree of toxicity for humans. So, the exploration on
natural derived products with NO inhibitory properties and their structure-activity
relationship studies can provide more choices for drug development programs.
Ursolic acid (UA, 1), a naturally occurring ursane-type pentacyclic triterpene, is
widely distributed in dietary and medicinal herbs in the form of free acid or aglycones
for triterpenoid saponins. It shows a broad range of biological functions, including
anti-neoplastic, anti-angiogenic, anti-HIV, anti-oxidative, anti-inflammatory.5
Derivatives of UA with more functional groups such as tormentic acid6 have generally
exhibited enhanced biological activities than UA itself. However, chemical
functionalization of the un-activated UA skeleton has proven to be quite challenging
which requires laborious protecting group chemistry, co-factor supplement and tedious
downstream process.7-12 On the other hand, microbial biotransformation has been
provided as a valuable and attractive alternative to functionalize molecules in a regio-
and stereo-selective manner that is otherwise difficult to chemically synthesize. Several
accounts have been published in the literature to microbially biotransform UA to afford
unique derivatives that exhibited significantly enhanced pharmacological activities
and/or reduced toxicities.13-16 In our continuous research interest in microbial
biotransformation of triterpenoids,7 herein we wish to communicate our further
investigation on UA by Bacillus megaterium CGMCC 1.1741 as well as the inhibitory
activity of the isolated metabolites on the LPS-induced NO release in RAW 264.7 cells.
Preparative scale biotransformation, isolation and identification of products were
completed routinely.7 Microbial transformation of ursolic acid (1, 300mg) with Bacillus
megaterium CGMCC 1.1741 resulted one less polar metabolite 2 (18.7 mg), four more
polar metabolites, compounds 3 (40.5 mg), 4 (15.1mg), 5 (80.6 mg) and 6 (25.7 mg)
(Fig. 1). Their 13C-NMR spectroscopic data are given in Table 1. (Detailed
experimental procedures see Supplementary data)

HO
OH
COOH
COOH
O
O
2 (6.23%)
(13.50%)
3
O
O
OH
COOH
Bacillus megaterium
CGMCC 1.1741
HO
O
1
(5.03%)
4
HO
OH
O
HO
OH
O
COOH
Glc
O
HO
(8.57%)
6
5 (26.87%)
Fig. 1 Biotransformation of ursolic acid (1) by Bacillus megaterium CGMCC 1.1741
Compound 2 was isolated as white powder. The HR-ESI-MS spectrum displayed
an [M-H]- ion at m/z 453.3366, indicating a 2 amu mass decrease to that of ursolic acid
(1). In the 1H-NMR spectrum, the characteristic signal of δH 3.45 ppm (H-3)
disappeared. And in the 13C-NMR spectrum, one new carbonyl carbon signal showed at
δC 216.57 ppm. The H-2 at δH 2.62 (d, J = 11.3 Hz, 2H) only showed the coupling with
H-1. Based on all the observations and comparison of the spectroscopic data with those
reported in the literature via chemical synthesis11 or microbial biotransformation16,
compound 2 was characterized as 3-oxo-urs-12-en-28-oic acid.
Compound 3 was isolated as white powder. The HR-ESI-MS spectrum displayed
an [M-H]- ion at m/z 485.3268 (calcd for C30H45O5, 485.3272), which indicated an
increment of 32 amu as compared to compound 2. The 1H-NMR spectrum indicated the
presence of two new proton signals at δH 4.28 (t, J = 7.1 Hz, 1H) and 4.61 (dd, J = 8.7,
3.4 Hz, 1H) ppm, while the DEPT spectrum showed the presence of two new carbon
signals at δC 79.27 and 67.34 ppm and the disappearance of two methylene signals
compared with 2, confirming that 3 was a dihydroxylated metabolite of 2. In the HSQC

experiment, the new proton signals at 4.28 ppm showed correlations with the new
methine carbon signal at δC 79.27 ppm, and in the HMBC experiment (Fig. 2), the new
CH signal show long-range 1H-13C correlations with C-2, C-9 and C-25, confirming
C-1 of the one site of hydroxylation. The another new proton signal at δH 4.61 ppm
showed correlations with the new carbon signal at δC 67.34 ppm, and in the HMBC
experiment, the new proton signal show long-range 1H-13C correlations with C-9, C-10,
C-12 and C-13, confirming C-11 of the other site of hydroxylation. The carbon signal at
δC 215.68 ppm was assigned to C-3 since it showed long-range correlations with H-1,
H-2, H-23 and H-24. The relative stereochemistry of the hydroxy groups at C-1 and
C-11 were assigned as β and α, based on their coupling constants (7.1 Hz for H-1, 3.4
and 8.7 Hz for H-11) , strong NOE at H-9 and H-5 given by irradiation of H-1 and
strong NOE at H-25 given by irradiation of H-11 (Fig. 2). Based on all the evidences,
metabolite 3 was characterized as 1β, 11α-dihydroxy-3-oxo-urs-12-en-28-oic acid and
it is a new compound.
Fig. 2 Key HMBC and NOESY correlations for compound 3-6
Compound 4 was isolated as white powder. The HR-ESI-MS spectrum displayed
an [M-COOH]- ion at m/z 513.3200 (calcd. for C31H45O6, 513.3222). The DEPT

spectrum showed the presence of one new methine carbon signal at δC 79.27 ppm (δH
4.16 ppm) and in the HMBC experiment (Fig. 2), the new proton signal showed the
long-range 1H-13C correlations with C-3, C-9, C-25, confirming C-1 of the one site of
hydroxylation. The 1H-NMR showed that the signal of 1 at δH 5.46 ppm (s, 1H, H-12)
disappeared, and two new methene signals were finally assigned to double-bond
protons at δH 7.52 ppm (d, J = 10.6 Hz, 1H, H-11) and δH 5.67 ppm (dd, J = 10.5, 2.9 Hz,
1H, H-12), and the long-range correlations could be observed from H-12 to C-9, C-13,
C-14 and C-18 and from H-11 to C-8 and C-13. The above evidence suggested the
double bond of 4 had shifted from C-12 (13) to C-11 (12). The C-13 signal showed at δC
89.68 ppm that speculated it was the other site of hydroxylation. Although two
additional oxygen atoms were introduced at C-1 and C-13 for compound 4, its
molecular weight was determined to be 468 (C30H44O4) rather than the expected 486
(C30H48O5). This result indicated a further condensation reaction in 4 to eliminate one
molecule of water (H2O) that could take place between 28-COOH and 13-OH due to
their relatively close distance. And the 13C-NMR of 4 showed the signal of C-28 shifted
from 180.29 ppm (compound 2) to 179.77 ppm. This evidence confirmed the presence
of a 13, 28-lactone ring. The relative stereochemistry of the hydroxy groups at C-1 was
assigned as β, based on the strong NOE at H-9 given by irradiation of H-1 (Fig. 2). The
HR-ESI-MS spectrum data also conformed with the structural changes. Based on the
above evidences, metabolite 4 was characterized as 1β-hydroxy-3-oxo-urs-12-en-28,
13-lactone and it is also a new natural product. Interestingly, an endophytic fungus was
reported to catalyze the reaction of ursolic acid (1) to form a 13, 28-lactone ring.14
Compound 5 was isolated as white powder. The molecular formula of 5 was
determined to be C30H48O5 by its HR-ESI-MS at m/z 511.3392 [M + Na]+ (calcd. for
C30H48O5Na, 511.3394), which indicated two oxygen atoms added to compound 1.
When compared to ursolic acid (1), the 1H-NMR showed two new proton signals at δH
3.97 (dd, J = 11.4, 4.4 Hz, 1H) and 4.69(dd, J = 7.9, 3.7 Hz, 1H) ppm that showed
correlations with the new signals at δc 78.17, 66.93 ppm, indicating the 5 was a
dihydroxylated metabolite of 1. The NMR signals for C-2, C-10 and C-25 were shifted
by +9.7, +7.9, -1.8 ppm, respectively, indicating the presence of a 1-OH group. H-1 (δH

3.97 ppm) had long-range correlations with C-2, C-9 and C-25 and showed NOE
enhancements with H-5, H-9, H-2 and H-3, indicating that the 1-OH group should be
located in the β-configuration. The other proton (δH 4.69 ppm) showed long-range
correlations with C-9, C-12, confirming C-11 of the one site of hydroxylation. The
H-11(δH 4.69 ppm) had NOE enhancements (Fig. 2) with H-9, H-12 and H-25,
indicating that 11-OH should be located in the α-configuration. Based on the above
evidence, compound 5 was identified as 1β, 3β, 11α-trihydroxyurs-12-en-28-oic acid.
Compound 6 was isolated as white powder. The HR-ESI-MS spectrum displayed
an [M+Na]+ ion at m/z 671.3754 (calcd for C36H56O10Na, 671.3766) , indicating a 194
amu mass which speculated one hexose unit and two oxygen atoms added to compound
2. The 1H-NMR spectrum displayed signal of one sugar at δH 6.28 ppm, which
correlated to the anomeric carbon at δc 96.24 ppm in the HSQC spectrum and the
13C-NMR spectrum showed the presence of six new signals at δ 96.24, 74.49, 79.67,
71.70, 79.35 and 62.67 ppm, which are characteristic of β-D-glucose. The glucuronic
acid was connected to C-28 of the aglycone, which was deduced from the HMBC
correlation between Glu H-1’(δH 6.28 ppm) and C-28 (δc 176.5 ppm) (Fig. 2). Besides,
the DEPT showed the presence of two new CH signal at δc 79.32 and 67.41 ppm that
were assigned to C-1 and C-11, respectively, based on the HMBC data (from H-1 to
C-2,C-3 and C-9; from H-11 to C-9, C-10, C-12 and C-13). The H-1 had NOE
enhancements with H-5 and H-9, indicating that 1-OH should be located in the
β-configuration, the H-11 had NOE enhancements with H-9 and H-25, indicating that
11-OH should be located in the α-configuration. Based on the above evidence,
compound 6 was identified as 1β,
11α-dihydroxy-3-oxo-urs-12-en-28-o-β-D-glucopyranoside.
As above, Bacillus megaterium CGMCC 1.1741 showed remarkable capability on
the structural modification of UA, almost all the fed substrate was consumed during the
incubation, and several types of reactions were observed in the biotransformation. It
was worthy to note that the formation of compound 4 which might go through the
double bond transferred from C12-C13 to C11-C12 and the lactone formation between

C-28 and C-13.We found that the structure which contained the lactone of pentacyclic
triterpene always existed in medicine plants such as Ilex Kudingcha17 and Rubus
parvifolius18, also could be generated by using microorganism fermentation.13, 19 The
enzyme which guided the synthesis of these metabolites have the potential to reflect the
biogenetic synthesis way. The introduction of hydroxyl groups to the skeleton is one of
the key steps in the structural modification of UA which can provide functional groups
for the follow- up reactions and made it possible to modify its structure through further
derivatization. 13, 15 The sugars attached to skeletons often dictate key pharmacological
properties and/or molecular mechanism of action, combined with hydroxylation and
glycosylation that will offer us a practical method to produce rare natural triterpenoids.7,
20
All the isolated metabolites (2-6, Fig. 1), along with the parent compound UA(1),
were tested for their inhibitory effects on the NO production in LPS-stimulated RAW
264.7 cells with quercetin as the positive control (Detailed experimental procedures see
Supplementary data). The cell viability was examined by the MTT assay to determine
the non-toxic concentration range of each test compound, which was employed in the
NO production assay. As can be seen in Table 2, all five metabolites showed enhanced
activities when compared to the parent UA with compound 3 the most active (IC50 of
1.24 µM) and 6 the least active (IC50 > 60 µM), whereas parent UA displayed no
inhibitory activity. Due to the limited number of compounds tested, thorough SAR
analysis is not possible. However, a couple of points might be worth noting. Firstly, C-3
ketone contributed positively against the NO production, 2>1 and 3>5. Secondly,
glycosylation of the C-28 carboxyl group drastically decreased the activity while
13,28-lactone ring formation appeared to have no significant impact.

Table 2. Inhibitory effect of compounds 1–6 on LPS-induced NO production in RAW 264.7 cells a
Compound
IC50 (µM) b
Stimulation
4.94
Compound
IC50 (µM)b
1
2
5
6
40.09
＞60
3
4
1.24
1.71
quercetin
4.41
a :quercetin was used as a positive control. Values are an average of two separate experiments.
b: Concentration necessary for 50% inhibition (IC50)
In conclusion, microbial transformation of ursolic acid by Bacillus megaterium
CGMCC 1.1741 was explored and yielded one known and four new metabolites. Four
types of reactions were involved in the biotransformation, oxidization, hydroxylation,
condensation and glycosylation. Therefore, the green and efficient culture could be
applied for the synthetic chemistry of UA and the unique catalytic capability and
enzymatic system of Bacillus megaterium CGMCC 1.1741 deserve further exploration.
The inhibitory activities of all compounds on LPS-induced NO production in RAW
264.7 cells revealed that the UA derivatives are one type of promising potential
candidates for the treatment of inflammatory diseases caused by excessive production
of NO in future. The active guided synthesis and the structural design are undergoing.
Acknowledgements
This work was supported by National Nature Science Foundation of China (NSFC
NO. 21302052) and the “Program for New Century Excellent Talents in University”
awarded to Prof. Jian Zhang. Thanks also give to the ‘111 Project’ from the Ministry of
Education of China, and the Fundamental Research Funds for Central Universities
(JKZ2011017), and the scientific and innovation research of college graduate in Jangsu
province(CXLX11_0788)

Table 1. 13C-NMR spectral data of the compounds 1-6 (recorded in C5D5N, 125MHz)
Carbon
1
1
2
3
4
5
6
39.13
28.15
78.17
39.39
55.87
18.82
33.62
40.02
48.09
37.46
23.66
125.68
139.30
42.54
28.73
24.95
48.09
53.60
39.52
39.43
31.11
37.33
28.83
16.56
15.69
17.48
23.93
179.84
17.52
21.41
39.75
34.80
216.57
47.56
55.80
20.27
33.37
40.33
47.80
37.27
24.24
125.86
139.78
43.04
29.12
25.35
48.54
54.07
39.95
39.89
31.57
37.89
27.20
22.07
15.63
17.79
24.17
180.29
17.97
21.86
79.27
44.38
215.68
47.79
51.61
19.76
33.76
43.80
55.75
44.71
67.34
129.99
141.86
43.17
29.27
25.27
48.17
53.14
39.70
39.75
31.31
37.66
27.68
21.52
13.13
19.11
23.94
180.16
17.97
21.69
79.95
45.61
214.41
47.94
51.18
19.22
30.81
42.72
53.95
42.66
137.69
128,90
89.68
42.78
26.47
23.59
46.49
60.80
40.75
38.67
31.44
32.37
27.16
21.05
14.38
19.42
16.54
179.77
18.30
19.56
78.17
37.85
75.95
40.06
53.52
18.77
34.32
44.14
57.3 6
45.34
66.93
129.91
141.93
42.89
29.40
25.27
47.95
52.83
39.76
39.75
31.31
37.73
29.13
16.51
13.85
19.25
24.13
180.43
18.03
21.75
79.32
44.37
215.59
47.76
51.54
19.81
33.76
43.97
55.68
44.70
67.41
130.49
140.96
43.14
29.14
24.97
48.48
52.89
39.45
39.54
31.02
36.99
27.74
21.48
13.24
19.38
23.80
176.54
17.83
21.54
96.24
74.49
79.67
71.70
79.35
62.67
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
1’
2’
3’
4’
5’
6’

References
1. Fang L, Zhang YH, Lehmann J, Wang Y, Ji H, Ding DY. Bioorg Med Chem Lett. 2007,17, 1062.
2. Conforti F, Menichini F. Curr Med Chem. 2011,18,1137.
3. Knowles RG, Moncada S. Biochem J. 1994,298,249.
4. Mayer B, Hemmens B. Trends Biochem Sci. 1997,22,477.
5. Wozniak L, Skapska S, Marszalek K. Molecules. 2015,20,20614.
6. Banno N, Akihisa T, Tokuda H, et al. Bioscience Biotechnology and Biochemistry. 2004,68,85.
7. Zhu Y-Y, Qian L-W, Zhang J, Liu J-H, Yu B-Y. Tetrahedron. 2011,67, 4206.
8. Honda T, Rounds BV, Bore L, et al. J Med Chem. 2000,43,4233.
9. Abodunde OA, LevakaVeera RR, Desai R, Nweke N, Berrou M. J Clin Pharm Ther. 2013,38,243.
10. Solet JM, Bistermiel F, Galons H, Spagnoli R, Guignard J. Phytochemistry. 1993,33,817.
11. Fu HJ, Zhou YR, Bao BH, et al. J Med Chem. 2014,57,4692.
12. Huang LR, Luo H, Li QJ, et al. Eur J Med Chem. 2015,95,64.
13. Huang FX, Yang WZ, Ye F, et al. Phytochemistry. 2012,82,56.
14. Fu SB, Yang JS, Cui JL, Feng X, Sun DA. Chem Pharm Bull (Tokyo). 2011,59,1180.
15. Fu SB, Yang JS, Cui JL, Sun DA. Fitoterapia. 2013,86,123.
16. Ibrahim A, Khalifa SI, Khafagi I, et al. Chem Pharm Bull (Tokyo). 2008,56,1253.
17. Che Y-Y, Li N, Zhang L, Tu P-F. Chin J Nat Med. 2011,9,22.
18. Zhang X, Zhu ZX, Wang J, et al. Chin J Nat Med. 2016,14,377.
19. Guo N, Zhao Y, Fang WS. Planta Med. 2010,76,1904.
20. Feng X, Zou ZM, Fu SB, Sun LZ, Su ZH, Sun DA. J Mol Catal B-Enzym. 2010,66,219.

New derivatives of ursolic acid through the biotransformation by
Bacillus megaterium CGMCC 1.1741 as inhibitors on nitric oxide
production
a
Chao Zhang , Shao-Hua Xu , Bai-Ling Ma , Wei-wei Wang , Bo-Yang Yu , Jian
a
b
b
b
†
†
Zhanga,*
a State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing, Jiangsu 210009, China
b Jiangsu Key Laboratory of TCM Evaluation and Translational Research, China Pharmaceutical University,
Nanjing, Jiangsu 211198, China
Graphical abstract
Microbial transformation of ursolic acid by Bacillus megaterium CGMCC 1.1741 was
explored and yielded one known metabolite and four new metabolites. All compounds were tested
for their inhibitory effects on the NO production in LPS-stimulated RAW 264.7 cells.