Structure–activity relationship and hypoglycemic activity of tricyclic matrines with advantage of treating diabetic nephropathy

Article abstract of DOI:10.1016/j.ejmech.2020.112315

Forty?three tricyclic matrinic derivatives with a unique scaffold were prepared and evaluated for their stimulation effects on glucose consumption in HepG2 cells. The structure?activity relationship was systematically elucidated for the first time. Among them, compound 17a exhibited the most promising potency, and dose-dependently increased glucose consumption in L6 myotubes. It significantly lowered blood glucose, glucosylated haemoglobin and AGE level, and improved glucose tolerance and insulin resistance in KK-Ay mice as well. More importantly, 17a effectively ameliorated diabetic nephropathy (DN), as indicated by the improvement of renal function and pathological changes, and decrease of urinary protein. Furthermore, 17a could induce glycolysis but suppressed aerobic oxidation of glucose, in a similar mechanism to Metform. Our results indicated that in addition to hyperglycemia, 17a may be developed to treat diabetic complication such as DN.

Full text of DOI:10.1016/j.ejmech.2020.112315

Journal Pre-proof
Structure–activity relationship and hypoglycemic activity of tricyclic matrines with
advantage of treating diabetic nephropathy
Sheng Tang, Can Wang, Ying–Hong Li, Tian–Yu Niu, Yuan–Hui Zhang, Yu–Dong
Pang, Yan–Xiang Wang, Wei–Jia Kong, Dan–Qing Song
PII:
S0223-5234(20)30284-1
DOI:
https://doi.org/10.1016/j.ejmech.2020.112315
EJMECH 112315
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 20 January 2020
Revised Date: 27 March 2020
Accepted Date: 6 April 2020
Please cite this article as: S. Tang, C. Wang, Ying–Hong. Li, Tian–Yu. Niu, Yuan–Hui. Zhang, Yu–
Dong. Pang, Yan–Xiang. Wang, Wei–Jia. Kong, Dan–Qing. Song, Structure–activity relationship and
hypoglycemic activity of tricyclic matrines with advantage of treating diabetic nephropathy, European
Journal of Medicinal Chemistry (2020), doi: https://doi.org/10.1016/j.ejmech.2020.112315.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. 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.
© 2020 Published by Elsevier Masson SAS.

Graphic Abstract
17a exhibited the promising activity on glucose consumption, and may be
developed to treat hyperglycemia as well as diabetic nephropathy complication.

Structure–activity relationship and hypoglycemic
activity of tricyclic matrines with advantage of
treating diabetic nephropathy
Sheng Tang†, Can Wang†, Ying–Hong Li, Tian–Yu Niu, Yuan–Hui Zhang, Yu–Dong
Pang, Yan–Xiang Wang, Wei–Jia Kong*, Dan–Qing Song*
Institute of Medicinal Biotechnology, Chinese Academy of Medical Science & Peking
Union Medical College, Beijing 10050, China
*Corresponding authors. E-mail address: songdanqing@imb.pumc.edu.cn (Danqing
Song) and wjkong@ imb.pumc.edu.cn (Weijia Kong).
† These authors made equal contribution to this work.
1

ABSTRACT
Forty−three tricyclic matrinic derivatives with a unique scaffold were prepared
and evaluated for their stimulation effects on glucose consumption in HepG2 cells.
The structure−activity relationship was systematically elucidated for the first time.
Among them, compound 17a exhibited the most promising potency, and
dose-dependently increased glucose consumption in L6 myotubes. It significantly
lowered blood glucose, glucosylated haemoglobin and AGE level, and improved
glucose tolerance and insulin resistance in KK-Ay mice as well. More importantly,
17a effectively ameliorated diabetic nephropathy (DN), as indicated by the
improvement of renal function and pathological changes, and decrease of urinary
protein. Furthermore, 17a could induce glycolysis but suppressed aerobic oxidation of
glucose, in a similar mechanism to Metform. Our results indicated that in addition to
hyperglycemia, 17a may be developed to treat diabetic complication such as DN.
KEYWORDS
Tricyclic matrinic; hypoglycemic; structure−activity relationship; pharmacodynamics;
diabetic nephropathy.
2

1. Introduction
According to the latest report, diabetes mellitus (DM) has become the world’s
third largest patient pool in 2018 and the number of patients will rise to 522 million in
2030 [1]. Complications of DM, such as diabetic nephropathy (DN), diabetic
ophthalmopathy and neuropathy, are the majors’ causes for increased diabetic
morbidity, disability and mortality. And they also bring a threat to the economies of all
countries, especially the developing ones [2]. DN is recognized as the leading cause
of kidney disease and affects approximately 40% of type 1 and type 2 diabetic
patients [3]. Although a lot of hypoglycemic drugs have been successfully applied to
treat patients [4–6], effective treatments for DN are still lacking [6]. Meanwhile,
safety is a big issue for all these drugs, side effects such as acute kidney injury [5,6],
acute pancreatitis [7] and hypoglycemia had been disclosed [8,9]. Therefore, it is
crucial to develop novel anti-diabetic drugs with good safety as well as potential to
benefit DN.
In the past decades, our team has been dedicated to the discovery and
development of innovative drugs from traditional Chinese natural alkaloids, with
novel chemical skeletons and unique biological mechanisms [10–14]. Based on
metform (MET) derived from alkaloid, we screened the library of alkaloid compounds
constructed in our lab to find the lead compounds against hyperglycemia, via a
glucose consumption model in HepG2 cells. Unexpectedly, as depicted in Fig. 1,
matrine (1) and sophoridine (2, 5R-matrine) with an opposite configuration at the
5-position, were identified to promote the glucose consumption in similar degrees,
much better than that of MET. Our previous structure–activity relationship (SAR)
showed 5S-matrinic and 5R-sophoridinic derivatives had different activities, resulting
from the completely different spatial conformation as displayed in Fig.1. The matrinic
3

derivatives with 5S-configuration owned antiviral activities against HCV or CVB3
[15–18], while sophoridinic derivatives with 5R-configuration had anticancer
activities [19–21]. Interestingly, in this study, compounds 1 and 2 showed an
equivalent activity in promoting glucose consumption, indicating that 5S- or
5R-configuration did not affect the activity much, a different SAR from our previous
results. This hinted a plenty of room for structural modification of its kind, and greatly
spurred us to further explore SAR to develop a novel family of hypoglycemic
candidates with a novel mechanism of action.
Fig. 1. Chemical structures, spacial conformation of 1and 2, and tricyclic matrinic scaffold
Thus, a simplified tricyclic matrinic scaffold was constructed, as illustrated in
Fig. 1. Two series of new compounds, tricyclic matrinic derivatives and sophoridinic
derivatives (tricyclic 5R-matrinic derivatives), were designed, synthesized and
evaluated for their stimulating glucose consumption. In our present study, we
described the synthesis of 43 matrinic derivatives, of which 35 were new, SAR
analysis, and in vivo hypoglycemic effects as well as primary mode of action of the
key compound.
2. Results and discussion
2.1. Synthetic routes
4

The semi-synthetic routes of 43 tricyclic matrinic analogues were depicted in
Schemes 1, 2 and 3, taking commercially available 1, 2 or sophocarpine (3) as the
starting materials respectively, which were purchased from Xi'an Tianbao
Biotechnology Co., Ltd. (Shanxi, China) with purities over 95%.
Scheme 1. (a) 5 N NaOH, reflux, 9 h, then 2 N HCl, pH = 5~6; 2 N MeOH/HCl, reflux, 2 h; (b)
substituted benzenesulfonyl chloride, TEA, CH2Cl2, r.t., 4 h; (c) LiAlH4, THF, 0 oC, 30 min; (d) 5 N
NaOH, reflux, 6 h; 10 N HCl, pH = 7~7.5; (e) P2O5/MsOH (1:5), thiosemicarbazide, 70 °C, 8 h; (f)
Acetyl chloride, TEA, THF, r.t., 0.5 h; (g) Phthalimide, PPh3, DIAD, r.t., 12 h; (h) NH2NH2· H2O,
CH3CH2OH, reflux, 2 h; (i) TCDI, DMAP, CH2Cl2, r.t., 2 h; (j) TsCl, TEA, DMAP, CH2Cl2, r.t., 4 h;
(k) LiAlH4, THF, r.t., 1 h; (l) Sodium naphthalenide, THF, 0 oC, 1 h; (m) substituted
benzenesulfonyl chloride or substituted benzene isothiocyanate, CH2Cl2, K2CO3, r.t., 4 h.
As outlined in Scheme 1, the reduction of 5 in anhydrous tetrahydrofuran (THF)
achieved the desired matrinols 6 with lithium tetrahydroaluminum (LiAlH4) as the
reductant [16], while the hydrolysis of 5 in 5 N NaOH attained matrinic acids 7a–c in
5

good yields as reported earlier [16]. Matrinic thiadiazoles 8a–d were obtained by the
cyclization of 7 with thiosemicarbazide or N-substituted thiosemicarbazides in the
presence of phosphorus pentoxide/methanesulfonic acid (1/5) in yields of 63–79%
[22]. Likewise, matrinic oxadiazoles 9a and 9b were gained by the cyclization of 7
and aminourea in good yields. The acylation of 8a in triethylamine (TEA) generated
compound 10 in an 87% yield.
The Mitsunobu reaction of 6a achieved key intermediate 11, then 12 was gained
with hydrazinolysis. Then the reaction of 12 with 1,1'-thiocarbonyl diimidazole
(TCDI) achieved the target compound 13 in an overall yield of 19%. Meanwhile, the
desired matrinanes 17a–l were prepared from 6b. The 11-terminal hydroxyl of 6b was
protected with tosyl (Ts) in the presence of dimethylamino pyridine (DMAP) to form
intermediate 14. Then it underwent the three-step procedure including elimination of
11-terminal OTs by LiAlH4, removal of 12N-Ts by sodium naphthalene, and
12N-benzenesulfonylation or amidation to gain target compounds 17a–l in total yields
of 38–57% [15,16].
Scheme 2. (a) 10% H2SO4, KMnO4, reflux, 2 h; (b) 2 N HCl/MeOH, reflux, 2 h; (c) trifluoro
methyl benzenesulfonyl chloride, TEA, CH2Cl2, 4 h; (d) 3 N HCl, reflux, 1 h; (e) LiAlH4, THF, 0
oC, 30 min; (f) TsCl, TEA, DMAP, CH2Cl2, 4 h; (g) LiAlH4, THF, 1 h.
As depicted in Scheme 2, the key intermediate 19 was acquired via the oxidation
6

and esterification reaction from 3 using potassium permanganate (KMnO4) as an
oxidizing agent in acid condition with an overall yield of 54%. The matrinic acetic
acid 21 was obtained from 19 following 12N benzenesulfonylation and hydrolysis
procedures in high yields. The desired matrinic ethanes 24a and 24b were obtained by
a similar three-step procedure including reduction, hydroxyl protection and removal
of OTs as mentioned above.
Similarly, the desired methyl 12N-benzenesulfonyl sophoridinicates 27a–d,
sophoridinic acids 28a–d, sophoridinols 29a–c and sophoridinanes 31a–d were
achieved from 2, as shown in Scheme 3. All the final products were purified by flash
column chromatography on silica gel with CH2Cl2/CH3OH as the eluents.
Scheme 3. (a) 5 N NaOH, reflux, 9 h, then 2 N HCl, pH = 5~6; (b) 2 N HCl/MeOH, reflux, 2 h; (c)
substituted benzenesulfonyl chloride, TEA, CH2Cl2, 4 h; (d) 5 N NaOH, reflux, 6 h; 10 N HCl, pH
= 7~7.5; (e) LiAlH4, THF, 0 oC, 30 min. (f) TsCl, TEA, DMAP, CH2Cl2, 4 h; (g) LiAlH4, THF, 1 h.
2.2. Biological activity evaluation
2.2.1. SAR analysis for glucose consumption in HepG2 cells
Forty-three matrinic derivatives were gathered and examined for their activities
on glucose consumption in HepG2 cells at concentration of 25 μM, taking MET (5
mM) as the positive control. The activity was expressed as the ratio of Gx to GDMSO
.
The structures and activities of all compounds were listed in Table 1.
7

Initially, the 11-side chain was set as butyric acid, and 12N-benzenesulfonyl with
a good druglike feature [16] was selected and attached by a methyl or trifluoromethyl
group to form compounds 7a–c, respectively. Although none of them showed
enhanced activity compared with the leads, compounds 7b and 7c with a common
trifluoromethyl on the benzene ring gave higher activities than 7a. Similarly, matrinol
6b with a trifluoromethyl on the benzene displayed a higher activity than 6a with a
methyl motif. Next, thiadiazole or oxadiazole motifs as bioisosteres of carboxyl acid
[23] were introduced on 11-terminal site to form the target matrinic thiadiazoles (8a–d
and 10) and oxadiazoles (9a and 9b). Compounds 8a–d and 10 did not display any
potential in promoting glucose consumption, while 9a obviously increased activity
compared with MET as a positive control. These results hinted the advantage of the
electron-withdrawing trifluoromethyl as the substitution.
Next, 12N-trifluoromethylbenzenesulfonyl was retained, 11-terminal group was
changed into isothiocyanate or alkane to generate four compounds 13 and 17a–c,
respectively. Compound 17a induced a significant increase in glucose consumption
with the Gx/GDMSO value of 1.49, while the other three seemed to have minimal
impact, which hinted the importance of a p-CF3 as the substituent and butane as the
11-side chain as well. Thus, 11-butane was retained as a favorable pharmacophore,
and the sulfonyl linker was replaced by amide and a variety of substituents on
benzene ring were selected, by which a series of 12N-benzeneaminoacyl matrinane
compounds 17d–l were invented. Among them, compounds 17i and 17j bearing
o-methyl and m-methyl gave the Gx/GDMSO value of 1.30 and 1.32, respectively,
higher than the leads 1–3, but lower than 17a. And unexpectedly, compounds 17d–l
bearing a CF3 substitutent did not show any positive stimuli on the glucose
consumption, indicating that the structure of linker group might affect the activities.
8

Furthermore, the impact of 11-side chain length was investigated by the
preparation of new matrinic acetic acid (21) and matrinic ethanes (24a and 24b). As
depicted in Table 1, the activity of 21 was comparable to its counterpart 7b, while the
activities of 24a and 24b were even higher than their counterparts 17b and 17c, which
suggested that the shortening of 11-side chain did not reduce the activity. Finally, the
impact of configuration change of 5C atom was confirmed by the evaluation of
methyl sopharidinates (27a–d), sopharidinic acids (28a–c), sopharidinols (29a–d) and
sopharidinanes (31a–d). As anticipated, all of them displayed comparable or higher
activity than their matrinic counterparts, suggesting that the configuration of 5C in the
matrinic scaffold might not affect the activity significantly.
Table 1 The structures and glucose consumption of target compounds in HepG2 cells
a
R1
-
R2
Gx/GDMSO
1.26±0.12
1.22±0.11
1.11±0.13
1.16±0.11
1.04±0.05
0.87±0.07
1.03±0.18
1.05±0.03
R1
R2
Gx/GDMSO
1.30±0.06
1.32±0.11
1.04±0.15
1.14±0.12
1.06±0.09
1.26±0.03
1.32±0.04
-
o-CH3
m-CH3
CH3
CH3
CH3
1
17i
17j
17k
17l
21
-
-
2
-
p CH3
--
-
3
p
p
-
CF3
CH2OH
2,6-di-CH3 CH3
m-CF3 COOH
6a
6b
7a
7b
7c
-CH3 CH2OH
-
p CH3 COOH
o
p
p
-
CF3
CF3
CH3
CF3
CH3
CH3
24a
24b
27a
m-CF3 COOH
-
p
-CF3
COOH
-
COOCH3 1.14±0.03
COOCH3 1.23±0.02
m-CF3
1.04±0.05
o
-
8a
27b
9

m-CF3
m-CF3
0.86±0.09
0.97±0.02
0.79±0.13
1.34±0.15
1.18±0.39
m-CF3
COOCH3 1.21±0.02
COOCH3 1.13±0.04
8b
8c
8d
9a
9b
27c
27d
28a
28b
28c
p
o
-
CF3
-
p CH3
-CF3
COOH
COOH
COOH
1.12±0.02
1.16±0.03
0.99±0.06
m-CF3
m-CF3
-
p CF3
p
-CF3
m-CF3
0.89±0.17
p
o
-
CH3
CH2OH
1.12±0.06
10
29a
-
p CF3
NCS
CH3
CH3
1.02±0.01
1.49±0.01
1.01±0.17
1.03±0.01
1.03±0.12
1.00±0.17
1.08±0.01
-CF3
CH2OH
CH2OH
CH2OH
CH3
1.24±0.04
1.25±0.01
1.06±0.08
1.07±0.03
1.26±0.08
1.20±0.02
1.17±0.12
1.54±0.13
13
29b
29c
29d
31a
31b
31c
31d
p
o
-
CF3
m-CF3
17a
17b
17c
17d
17e
17f
17g
17h
-CF3
p
p
-CF3
-
p CH3 CH3
CF3 CH3
m-CF3 CH3
-CH3
o
-
o-CF3
m-CF3
CH3
CH3
-
p CF3
p-OCF3 CH3
p-F CH3
CH3
p
-CF3
CH3
1.07±0.09 MET -
1.02±0.09
-
a The activity of target compounds was examined at concentration of 25 μM and MET at 5 mM
concentration.
2.2.2. Glucose consumption of 17a in L6 myotubes
The stimulating effect of 17a on glucose consumption was further confirmed in
L6 skeletal muscle cells, taking rosiglitazone (RGZ) as a positive control [24]. As
shown in Fig. 2, 17a significantly increased glucose consumption in a concentration-
dependent manner in L6 myotubes after the treatment of 24 h, with an extent
comparable to RGZ at the same concentration of 20 μM.
10

Fig. 2. Effects of 17a on glucose consumption in L6 myotubes. L6 myotubes were treated with
DMSO, 17a or RGZ at indicated concentrations for 24 h. After treatment, culture supernatants
were harvested for the determination of glucose consumption. Values are mean ± SD of 3
independent experiments. *p<0.05, **p<0.01 vs. that of DMSO treated cells.
2.3. Safety assay of 17a in vitro and in vivo
As compound 17a displayed the highest activity on glucose consumption
stimulation, the cytotoxicity of 17a in HepG2 cells was determined. The growth of
HepG2 cells was not influenced by 17a when its concentration did not exceed 160 μM,
and its CC50 value was about 642 μM. The acute toxicity test of 17a was performed in
Kunming mice by oral administration in a single-dosing experiment at 250, 500, 750
or 1000 mg·kg-1, respectively. The mice were closely monitored for 7 days. The
results showed that the median lethal dose (LD50) value of 17a was over 1000 mg/kg.
Meanwhile, blood samples taken at the end of the experiment were examined for liver
and kidney functions. As shown in Fig. 3, significant abnormality was not found in
glutamic oxalacetic transaminase (GOT), glutamic pyruvic transaminase (GPT), as
well as blood urea nitrogen (BUN) and creatine (CRE), which demonstrated 17a
owned a good oral safety profile.
11

Fig. 3. Effect of 17a on liver and kidney functions. GOT, GPT, BUN and CRE (U/L) levels in the
treatment with 17a.
2.4. Hypoglycemic efficacy 17a in KK-Ay mice
We have demonstrated that 17a owned excellent pharmacokinetics properties
with the maximum concentration (Cmax) of 4.78 μM and AUC of 20.3 μM•h, which
indicated its good stability to metabolism in vivo [16]. Therefore, it was selected to
conduct the hypoglycemic efficacy experiment in KK-Ay mice. Taking RGZ as the
positive control, as shown in Table 2, after 50 days of oral administration, 50 and 100
mg/kg of 17a lowered fasting blood-glucose (FBG) significantly and dose-
dependently. After treatment, 50 or 100mg/kg of 17a reduced the glucosylated
haemoglobin (GHb) level averagely by 19.1% and 22.8%, respectively. In addition,
the serum level of the Advanced Glycation End Product (AGE), which closely
involved in the progression of diabetes mellitus and related complications like
diabetic nephropathy and cardiovascular complications [25], was also greatly reduced
by 17a treatment in KK-Ay mice, comparable to that of RGZ. Our finding suggested
that 17a might ameliorate the development of diabetes mellitus and diabetic
nephropathy, at least in part, through suppressing the production of AGEs.
Furthermore, 17a at 100 mg/kg greatly alleviated hyperinsulinemia and restored
insulin sensitivity (Table 2), which were indicated by a significant increase of the
12

value of homeostasis model assessment-insulin resistance (HOMA-IR) of 46.5,
compared to the value of 137 in KK-Ay mice.
Table 2. Effect of 17a on blood and serum biochemical indexes in KK-Ay mice.
KK-Ay
Measurements
C57BL/6J
Untreated
17a, 50 mg/kg
17a, 100 mg/kg
RGZ, 5 mg/kg
FBG (mM)
Before
5.11±0.45 13.0±2.09*** 13.2±2.51***
5.40±0.31 13.6±1.41*** 9.44±0.98#
13.3±2.85***
8.18±0.99##
6.75±1.36#
1.13±0.24##
128±63.0##
46.5±10.8##
5.55±1.54#
0.82±0.22#
2.39±0.95##
70.4±25.1##
12.4±1.86
13.2±2.06***
7.66±1.28##
6.09±1.04#
1.27±0.40##
89.3±28.4##
30.4±7.61###
8.31±1.95
After
GHb (%)
4.69±0.86 8.74±1.08**
7.07±1.37#
AGE (μg/mL)
Insulin (mU/L)
HOMA-IR
CHO (mM)
LDL-c (mM)
TG (mM)
0.68±0.09 2.85±0.92*** 1.59±0.86#
5.09±1.75 226±15.5*** 177±54.1#
1.22±0.16 137±18.4*** 74.3±15.7##
2.65±0.31 7.38±1.03**
7.05±1.06
0.22±0.06 1.11±0.24*** 1.06±0.26
0.94±0.14 4.01±1.83*** 2.89±1.11#
26.0±7.56 124±35.1*** 88.8±26.9#
1.08±0.27
1.84±0.45##
51.6±15.5##
12.0±1.64
Scr (μM)
BUN (mM)
12.6±1.35 12.8±2.97
12.7±1.81
Values are mean±SD of 8-9 mice in each group.
#
**p<0.01, ***p<0.001 vs. that of C57BL/6J group; p<0.05, ##p<0.01, ###p<0.001 vs. that of
KK-Ay untreated group.
During the experiment, oral glucose tolerance test (OGTT) and insulin tolerance
test (ITT) were performed. As shown in Fig. 4A, in OGTT, the blood glucose levels
reached a peak at 30 min after glucose loading in all groups. Compared to normal
control group, the KK-Ay mice exhibited a significant glucose intolerance, and its
AUCs of OGTT was about 2.52-fold of that of normal control. 17a at 50 or 100
mg/kg/day or RGZ at 5 mg/kg improved glucose tolerance and reduced the AUCs of
the mice significantly. In the ITT experiment, the blood glucose of C57BL/6J mice
declined after insulin injection and reached a minimum at 60 min, and
correspondingly, the blood glucose of KK-Ay mice was almost constant during the
test and its AUC increased by about 5.35-fold of normal control, which indicated the
successful construction of the state of insulin unresponsiveness. To our delight, oral
administration of 17a or RGZ restored insulin responsiveness and reduced the AUC of
13

ITT significantly (Fig. 4B).
Fig. 4. Effect of 17a on OGTT and ITT in KK-Ay mice. Diabetic KK-Ay mice were untreated
or treated with 50 or 100 mg/kg of 17a, or 5 mg/kg of RGZ, respectively; C57BL/6J mice were
used as normal control. After 35 days of treatment, OGTT (A) and ITT (B) were performed, blood
glucose levels were plotted against time and AUCs were calculated. Values are mean ± SD of 8-9
#
mice in each group. **p<0.01, ***p<0.001 vs. that of C57BL/6J mice; p<0.05, ##p<0.01 vs. that
of KK-Ay untreated mice.
As the KK-Ay mice were fed with a lipogenic diet, they developed
hyperlipidemia as compared to the normal control group, as indicated by the high
levels of serum triglyceride (TG), cholesterol (CHO) and low-density lipoprotein
cholesterol (LDL-c) (Table 2). 17a significantly lowered serum TG at the dosages of
50 and 100 mg/kg, and decreased serum CHO and LDL-c levels at 100 mg/kg. For a
comparison, RGZ lowered serum TG greatly, but had no effect on serum CHO and
LDL-c.
The average body weight, body weight gain, and food intake amount of KK-Ay
mice were higher than those of C57BL/6J mice (Table 3). Although 17a did not
influent food intake or the endpoint body weight of KK-Ay mice, it greatly reduced
the extra body weight gain induced by the high fat and high sugar diet after treatment.
For a comparison, the average food intake greatly reduced in the RGZ-treated group.
Although RGZ did not affect the endpoint body weight, it induced a higher body
14

weight gain than non-treated KK-Ay mice group, which might be due to an induction
of preadipocyte differentiation [24,26,27]. Due to the onset of diabetes, KK-Ay mice
drank significantly more water than C57BL/6J mice (p<0.001), and the water intakes
were effectively ameliorated in both 17a and RGZ treatments. Therefore, 17a lowered
blood lipids especially TG and inhibited extra body weight gain in KK-Ay mice,
which suggested that its beneficial effects on both lipid metabolism and obesity.
Table 3. Effects of 17a on food intake, body weight and body weight gain in KK-Ay mice.
KK-Ay
Measurements
C57BL/6J
4.29±0.24
Untreated
17a, 50 mg/kg
17a, 100 mg/kg
5.28±0.46*
RGZ, 5 mg/kg
4.58±0.31#
Food intake (g/day)
5.67±0.55*
5.20±0.44*
Water intake (mL/day) 4.38±0.95
Body weight (g)
21.2±4.17*** 13.8±2.60##
11.9±3.70##
6.10±1.53###
Before
25.6±1.09
29.0±1.53
3.42±0.21
37.9±2.60*
43.5±5.24*
5.61±0.87**
37.2±3.10*
40.7±4.41*
3.52±0.77#
37.8±2.49*
40.3±3.86*
2.53±0.47##
37.9±2.63*
44.7±3.59*
6.82±1.21#
After
Body weight gain (g)
Values are mean±SD of 8-9 mice in each group. *p<0.05, **p<0.01, ***p<0.001 vs. that of
C57BL/6J group; #p<0.05, ##p<0.01 vs. that of KK-Ay untreated group.
2.5. 17a improves pathological changes in KK-Ay mice
Kidney function of the KK-Ay mice was damaged, as indicated by the significant
elevation of serum creatinine (Scr) level, which was restored by the administration of
17a or RGZ effectively (Table 2). Before the end of experiment, 24-h urine samples
of the mice were collected and examined. As shown in Table 4, the 24-h urine amount
of KK-Ay mice was averagely 9.61-fold of C57BL/6J mice, accompanied with high
level of urine glucose. Administration of 17a or RGZ greatly ameliorated polyuria and
reduced urine glucose in KK-Ay mice. In the urine of C57BL/6J mice, only a small
amount of protein existed and no microalbumin (mAlb) was detected. At the
meantime, a large amount of total protein and mAlb were detected in the urine of
KK-Ay mice, which was in accordance with the increase of Scr and indicated severe
renal damage. 17a or RGZ significantly reduced the amounts of 24-h urinary total
15

protein and mAlb (Fig.5).
Fig.5. Effects of 17a and RGZ on urine related parameters in KK-Ay mice. KK-Ay mice were
grouped and treated as described in Fig. 2. Before the end of experiment, 24-h urine samples were
collected for the measurements of volume (A), glucose (B), total protein (C), and mAlb (D) by
kits. Values are mean±SD of 8-9 mice in each group. ***p<0.001 vs. that of C57BL/6J group;
#p<0.05, ##p<0.01 vs. that of KK-Ay untreated group.
The kidneys and pancreases were subjected to pathological examinations, as
showed in Fig. 6 and Fig. 7. In accordance with the assays of renal function and urine,
significant pathological changes were disclosed in the kidney of KK-Ay diabetic mice
(Fig. 6), such as glomerular abnormalities, interstitial fibrosis and tubularatrophy
(IFTA) and inflammatory infiltration. For the pancreas of KK-Ay mice (Fig. 7A), the
volume of islets increased greatly, the structure of islets was largely destructed, the
density of islet cells was greatly reduced, and islet cells were disarranged, cells with
vacuolar degeneration were interspersed in the islets. In addition, apoptotic
characteristic such as nuclear condensation was observed in some islet cells. The
average pancreas pathological score of KK-Ay mice was 3.67 (Fig. 7B), for a
comparison, the values in the treatment of 17a at the dosages of 50, 100 mg/kg and
RGZ at 5 mg/kg were significantly decreased respectively, which demonstrated 17a’s
significant efficacy on improving kidney and pancreas pathological changes.
16

Fig. 6. Effects of 17a on pathological changes of kidney in KK-Ay mice. At the end of
experiment, the mice were sacrificed; their kidneys were harvested for pathological examination
by H&E staining. Representative images were presented in panel A (200×, scale bar = 100 μm), in
which glomerular changes were indicated by a white arrow, tubulointerstital changes were
indicated by a blue arrow, and interstitial inflammation was indicated by an orange arrow. The
pathological changes were scored in panel B. Values are mean ± SD of 8-9 mice in each group.
***p<0.001 vs. that of C57BL/6J mice; #p<0.05, ##p<0.01 vs. that of KK-Ay untreated mice.
Fig. 7. Effects of 17a on pathological changes of pancreas in KK-Ay mice. At the end of
experiment, the mice were sacrificed; their pancreases were harvested for pathological
examination by H&E staining. Representative images were presented in panel (A) (200×, scale bar
= 100 μm), in which vacuolar degeneration was indicated by a white arrow, and nuclear
condensation was indicated by a green arrow. The pathological changes were scored in panel B.
Values are mean ± SD of 8-9 mice in each group. ***p<0.001 vs. that of C57BL/6J mice; #p<0.05,
##p<0.01 vs. that of KK-Ay untreated mice.
DN is a severe complication of diabetes which may cause renal failure of the
17

patients [28]. Our study proved that 17a effectively ameliorated DN in KK-Ay mice,
as indicated by the improvement of renal function, the decrease of urinary protein,
and amelioration of renal pathological changes. Our results indicated that in addition
to hyperglycemia, 17a may be developed to treat diabetic complication like DN.
2.6. Primary mechanism of action of 17a against glucose consumption
In order to further understand the mechanism of action of 17a to stimulate
glucose consumption, we investigated its effects on cellular lactate release and ATP
production in L6 myotubes. As depicted in Fig. 8A, compound 17a greatly enhanced
lactate release in a dose-dependent manner. Accordingly, cellular ATP production was
also greatly reduced by 17a treatment, as displayed in Fig. 8B. The positive control
RGZ also exerted the similar effects on lactate release and ATP production at the
concentration of 20 μM, in agreement with a previous report [29]. Similar results
were obtained in HepG2 cells, as shown in Fig 8B.
Fig. 8A. Effects of 17a on glucose consumption, lactate release and ATP production in L6
myotubes. L6 myotubes were treated with DMSO, 17a or RGZ at indicated concentrations for 24
h. After treatment, culture supernatants were harvested for the determination of glucose
consumption (A) and lactate release (B), and cells were lysed for ATP assay, which was
normalized to protein contents (C). Values are mean ± SD of 3 independent experiments. *p<0.05,
**p<0.01 vs. that of DMSO treated cells.
18

Fig. 8B. Effects of 17a and RGZ on glucose consumption, lactate release, and ATP production in
HepG2 cells. HepG2 cells were treated with 17a or RGZ at indicated concentrations for 24 h.
After treatment, culture supernatants were harvested for determination of glucose consumption (A)
and lactate release (B), cells were lysed for determination of ATP production (C), which was
normalized to protein contents. Value are mean ± SD of 3 or 4 separate experiments. *p<0.05,
**p<0.01, ***p<0.001 vs. that of DMSO.
In cells of metabolic tissues, glucose is metabolized through glycolysis or
aerobic oxidation [30]. In condition of hypoxia, glucose is transformed into lactate
through glycolysis [31]. When there is plenty of oxygen, glucose will be metabolized
thoroughly through Kreb's cycle and mitochondrial respiratory chain and a large
amount of ATP will be produced [32]. In our experiments, 17a stimulated lactate
release and inhibited ATP production in L6 myotubes. Our results proved that 17a
induced glycolysis but suppressed aerobic oxidation of glucose. Therefore, the
hypoglycemic mechanism of 17a was probably similar to MET, which stimulated
glucose consumption through induction of glycolysis [32], as shown in Fig. 9.
19

Fig. 9. Possible mechanisms of 17a to stimulate glucose consumption in skeletal muscle cells.
17a stimulates glucose consumption through inhibition of aerobic oxidation and induction of
glycolysis. The detailed mechanisms are unclear and need investigation.
3. Conclusions
Taken together, 43 tricyclic matrinic analogues with a unique chemical scaffold,
of which 35 were new, were synthesized and evaluated for their potency on
stimulating glucose consumption. SAR study indicated that introduction of a suitable
substituent into both matrinic and sophoridinic scaffold might be beneficial for the
activity, regardless of the configuration change on the 5C atom. In the present study,
we reported for the first time that 17a significantly lowered blood glucose,
glucosylated haemoglobin and AGE level, as well as improved glucose tolerance and
insulin resistance in KK-Ay mice. Importantly, 17a effectively ameliorated DN, as
indicated by the improvement of renal function, decrease of urinary protein and
amelioration of pathological changes. Its hypoglycemic mechanism was probably
similar to MET, which stimulated glucose consumption through induction of
glycolysis. Our results indicated that in addition to hyperglycemia, 17a may be
developed to treat diabetic DN complication. The underly mechanism of 17a to
modulate both glucose and lipid metabolism are actively investigated in our lab.
4. Experimental section
4.1. General
20

All starting materials were commercial products and used without further
purification. Melting point was determined with a MP90 apparatus and uncorrected
(Mettler-Toledo, Greifensee, Switzerland). Specific rotations were obtained at
Rudolph Autopol IV automatic polarimeter, using ethanol as solvent with the
concentration of 0.1 M. (Jasco Products Company, Oklahoma City, USA).1H NMR
and 13C NMR spectra were recorded on a Bruker Avance 400 or III 500 spectrometer
(Varian, San Francisco, USA). HRMS were recorded on an AutospecUitima-TOF
spectrometer (Micromass UK Ltd., Manchester, U.K.). Flash chromatography was
performed on Combiflash Rf 200 (Teledyne, Nebraska, USA) particle size 0.038 mm.
The HPLC analyses were performed on an Agilent 1200 (Agilent, Santa Clara, USA).
The purity of all test compounds was confirmed to be over 95% (Supp. Table 1).
Compounds 6a, 7a–c, 17a, 17b, 24b, 28a and 28c are known [16,20,21].
4.2. Chemistry
4.2.1. 12N-p-Methlybenzenesulfonyl matrinol hydrochloride (6b). To a solution of 4
(20.0 mmol) in CH2Cl2, p-methlybenzenesulfonyl chloride (20.0 mmol) and TEA
(20.0 mmol) were added, and the mixture was stirred for 4 h. Water (50 mL) was
added and the organic phase was separated and dried over Na2SO4, concentrated to
provide 5b. To a solution of 5b (10.0 mmol) in anhydrous THF in an ice bath, LiAlH4
(12.0 mmol) was added portionwise, and the mixture was stirred for 30 min and
quenched with saturated NH4Cl (2 mL). The organic layer was concentrated and
purified by flash column chromatography with CH2Cl2/CH3OH to afford 6b. Yield:
20
65%; white solid; mp 152–153 °C; [α] = 3.41; 1H NMR (600 MHz, DMSO-d6) δ
589
10.25 (br 1H), 7.70 (d, J = 8.2 Hz, 2H), 7.40 (d, J = 8.2 Hz, 2H), 4.00–3.93 (m, 1H),
3.73–3.64 (m, 2H), 3.56–3.53 (m, 1H), 3.23–3.18 (m, 2H), 3.18–3.12 (m, 2H),
2.93–2.85 (m, 2H), 2.39 (s, 3H), 2.29–2.20 (m, 2H), 1.90–1.83 (m, 2H), 1.70–1.52 (m,
21

9H), 1.30–1.21 (m, 1H), 1.17–1.06 (m, 2H), 0.95–0.86 (m, 1H). 13C NMR (126 MHz)
δ 143.0, 138.8, 129.7(2), 126.6(2), 63.0, 60.5, 57.6, 54.6, 54.5, 47.9, 37.9, 34.1, 32.3,
27.9, 24.9, 24.6, 22.1, 21.0, 18.3, 18.2. HRMS: calcd for C22H35O3N2S [M-HCl+H]+:
407.2363, found: 407.2365.
4.2.2. General procedures for the synthesis of compounds 8a–d. Compound 5 (10.0
mmol) was refluxed in 3 N HCl for 1 h, and neutralized with ammonia to pH 6–7. The
residue was suspended in MeOH, and the filtration was concentrated to afford 7. A
mixture of 7 (5.0 mmol), thiosemicarbazide or N-substituted thiosemicarbazide (5.0
mmol), methanesulfonic acid (25.0 mmol) and phosphorus pentoxide (5.0 mmol)
were heated at 70 °C for 9 h. The reaction mixture was poured into ice water (60 mL)
and neutralized with 20% NaOH to pH 8. The solution was extracted with CH2Cl2,
and dried over Na2SO4 and condensed, which was purified by flash column
chromatography with CH2Cl2/CH3OH as the eluent.
4.2.2.1. 2’-Amino-12N-m-trifluoromethlybenzenesulfonyl matrinic thiadiazole (8a).
20
Yield: 47%; white solid; mp 162–163 °C; [α] = -8.93; 1H NMR (500 MHz, CDCl3) δ
589
8.11 (s, 1H), 8.04 (d, J = 7.8 Hz, 1H), 7.77 (d, J = 7.8 Hz, 1H), 7.61 (t, J = 7.8 Hz,
1H), 5.71 (s, 2H), 3.67 (s, 1H), 3.46 (dd, J = 12.7, 6.8 Hz, 1H), 3.21 (t, J = 11.1 Hz,
1H), 2.88–2.80 (m, 2H), 2.47–2.41 (m, 2H), 1.93 (s, 2H), 1.84–1.63 (m, 8H),
1.47–1.20 (m, 7H). 13C NMR (126 MHz) δ 168.4, 161.1, 141.5, 131.2 (q, JCF = 32.8
Hz), 131.0, 129.4, 128.9 (q, JCF = 3.78 Hz), 124.7 (q, JCF = 3.78 Hz), 123.5 (q, JCF
=
273.4 Hz), 62.4, 57.0, 56.4, 56.3, 45.7, 39.7, 33.8, 32.8, 30.3, 28.7, 27.9, 26.0, 20.8,
20.4. HRMS: calcd for C23H31O2N5F3S2[M+H]+: 530.1866, found: 530.1863.
4.2.2.2. 2’-Methylamino-12N-m-trifluoromethlybenzenesulfonyl matrinic thiadiazole
20
hydrochloride (8b). Yield: 33%; white solid; mp 157 °C (dec.); [α] = -5.36; 1H NMR
589
(600 MHz, DMSO-d6) δ 10.38 (br, 1H), 8.18 (d, J = 8.0 Hz, 1H), 8.09–8.03 (m, 2H),
22

7.89 (t, J = 8.0 Hz, 1H), 4.34–4.27 (m, 1H), 3.92–3.80 (m, 2H), 3.64–3.58 (m, 1H),
3.18 (d, J = 11.7 Hz, 2H), 3.01 (s, 3H), 2.94–2.81 (m, 3H), 2.68–2.61 (m, 1H),
2.31–2.23 (m, 2H), 1.92–1.81 (m, 3H), 1.78–1.51 (m, 7H), 1.44–1.35 (m, 1H),
1.27–1.14 (m, 2H). 13C NMR (151 MHz) δ 168.1, 158.0, 142.8, 131.3, 130.7, 129.9 (q,
JCF = 33.2 Hz), 129.7 (q, JCF = 3.02 Hz), 123.3 (q, JCF = 273.3 Hz), 122.9 (q, JCF
=
3.02 Hz), 63.0, 57.7, 54.6, 54.5, 48.8, 38.1, 34.9, 32.0, 28.8, 26.7, 24.7, 24.6, 24.3,
18.3, 18.1. HRMS: calcd for C24H33O2N5F3S2 [M-HCl+H]+: 544.2022, found:
544.2020.
4.2.2.3.2’-Dimethylamino-12N-m-trifluoromethlybenzenesulfonyl matrinic thiadiazole
20
hydrochloride (8c). Yield: 34%; white solid; mp 198 °C (dec.); [α] = 12.16; 1H NMR
589
(600 MHz, DMSO-d6) δ 10.96 (br, 1H), 8.18 (d, J = 8.0 Hz, 1H), 8.10–8.04 (m, 2H),
7.88 (t, J = 8.0 Hz, 1H), 4.38–4.31 (m, 1H), 3.92 (t, J = 13.3 Hz, 1H), 3.87–3.81 (m,
1H), 3.66–3.60 (m, 1H), 3.21–3.15 (m, 7H), 2.96–2.82 (m, 3H), 2.67–2.59 (m, 1H),
2.34–2.28 (m, 1H), 2.28–2.21 (m, 1H), 1.94–1.82 (m, 3H), 1.79–1.51 (m, 7H),
1.43–1.32 (m, 1H), 1.25–1.13 (m, 2H). 13C NMR (151 MHz) δ 170.2, 157.8, 143.1,
131.2, 130.6, 130.0 (q, JCF = 33.2 Hz), 129.7 (q, JCF = 3.02 Hz), 123.3 (q, JCF = 273.3
Hz), 122.9 (q, JCF = 3.02 Hz), 63.0, 57.9, 54.6, 49.0, 45.2, 42.3(2), 38.2, 35.0, 29.0,
26.7, 25.4, 24.7, 24.3, 18.3, 18.0. HRMS: calcd for C25H35O2N5F3S2 [M-HCl+H]+:
558.2179, found: 558.2174.
4.2.2.4. 2’-Ethylamino-12N-p-methlybenzenesulfonyl matrinic thiadiazole (8d). Yield:
20
39%; white solid; mp 177 °C (dec.); [α] = -4.06; 1H NMR (500 MHz, DMSO-d6) δ
589
7.68 (d, J = 8.1 Hz, 2H), 7.53 (t, J = 5.3 Hz, 1H), 7.38 (d, J = 8.1 Hz, 2H), 3.43–3.39
(m, 2H), 3.26–3.21 (m, 2H), 3.15–3.09 (m, 1H), 2.75–2.56 (m, 3H), 2.38 (s, 3H), 1.93
(t, J = 3.0 Hz, 1H), 1.87–1.80 (m, 2H), 1.78–1.71 (m, 2H), 1.71–1.60 (m, 5H),
1.50–1.45 (m, 2H), 1.41–1.33 (m, 2H), 1.33–1.21 (m, 4H), 1.15 (t, J = 7.2 Hz, 3H).
23

13C NMR (126 MHz) δ 168.2, 157.4, 142.9, 136.7, 129.5(2), 127.2(2), 62.3, 56.6,
56.2, 56.1, 47.0, 39.4, 38.6, 34.0, 29.6, 29.5, 27.6, 27.5, 24.8, 21.0, 20.4, 20.3, 14.3.
HRMS: calcd for C25H38O2N5S2[M+H]+: 504.2461, found: 504.2462.
4.2.3. General procedures for the synthesis of 9a and 9b. A mixture of 7 (5.0 mmol),
aminourea (5.0 mmol), methanesulfonic acid (25.0 mmol) and phosphorus pentoxide
(5.0 mmol) were heated at 70 °C for 9 h. The reaction mixture was poured into ice
water (60 mL) and neutralized with 20% NaOH to pH 8. The solution was extracted
with CH2Cl2 (60 mL) and dried over Na2SO4, and condensed, which was purified by
flash column chromatography with CH2Cl2/CH3OH as the eluent.
4.2.3.1. 2’-Amino-12N-m-trifluoromethlybenzenesulfonyl matrinicoxadiazol (9a).
20
Yield: 63%; white solid; mp 146–147 °C; [α] = -6.25; 1H NMR (600 MHz, CDCl3) δ
589
8.12 (s, 1H), 8.05 (d, J = 7.8 Hz, 1H), 7.78 (d, J = 7.8 Hz, 1H), 7.62 (t, J = 7.9 Hz,
1H), 5.37 (s, 2H), 3.70–3.59 (m, 1H), 3.50–3.43 (m, 1H), 3.20 (t, J = 11.6 Hz, 1H),
2.69–2.60 (m, 2H), 2.49 (d, J = 11.1 Hz, 1H), 2.40 (d, J = 11.1 Hz, 1H), 1.98–1.63 (m,
11H), 1.53–1.31 (m, 6H). 13C NMR (126 MHz) δ 163.0, 160.7, 141.3, 130.9, 130.8,
129.2 (q, JCF = 33.2 Hz), 128.7 (q, JCF = 3.02 Hz), 124.6 (q, JCF = 3.02 Hz), 123.3 (q,
JCF = 273.3 Hz), 62.2, 56.8, 56.3, 56.2, 45.6, 39.5, 33.6, 32.6, 28.5, 27.8, 25.2, 22.5,
20.7, 20.3. HRMS: calcd for C23H31O3N5F3S [M+H]+: 514.2094, found: 514.2100.
4.2.3.2. 2’-Amino-12N-p-trifluoromethlybenzenesulfonyl matrinicoxadiazol (9b).
20
Yield: 60%; white solid; mp 213 °C (dec.); [α] = -5.81; 1H NMR (600 MHz, CDCl3)
589
δ 7.96 (d, J = 8.2 Hz, 2H), 7.72 (d, J = 8.2 Hz, 2H), 5.36 (s, 2H), 3.69–3.57 (m, 1H),
3.58–3.40 (m, 1H), 3.21 (t, J = 11.5 Hz, 1H), 2.90–2.75 (m, 2H), 2.54 (d, J = 11.5 Hz,
1H), 2.45 (d, J = 11.5 Hz, 1H), 1.96 (d, J = 12.5 Hz, 2H), 1.90–1.65 (m, 8H),
1.49–1.25 (m, 7H). 13C NMR (126 MHz) δ 168.2, 161.2, 143.9, 133.9 (q, JCF = 32.8
Hz), 128.1(2), 125.9 (q, JCF = 3.78 Hz, 2), 123.4 (q, JCF = 273.4 Hz), 62.7, 57.5, 56.6,
24

56.5, 46.9, 39.7, 34.4, 31.6, 30.3, 28.3, 27.9, 25.7, 20.9, 20.6. HRMS: calcd for
C23H31O3N5F3S[M+H]+: 514.2094, found: 514.2092.
4.2.4.
2’-Acetylamino-12N-m-trifluoromethlybenzenesulfonyl
matrinicthiazole
hydrochloride (10). To a solution of 8a (0.8 mmol) in CH2Cl2 (15 mL), TEA (5.0
mmol) and acetyl chloride (4.0 mmol) were added at 0 °C and stirred at rt for 1 h. The
reaction mixture was washed by brine and dried over Na2SO4 and condensed, which
was purified by flash column chromatography with CH2Cl2/CH3OH to give 10. Yield:
20
87%; white solid; mp 109–110 °C ; [α] = 8.93; 1H NMR (600 MHz, CDCl3) δ 13.20
589
(br, 1H), 8.14 (s, 1H), 8.06 (d, J = 7.9 Hz, 1H), 7.79 (d, J = 7.9 Hz, 1H), 7.63 (t, J =
7.9 Hz, 1H), 3.74–3.65 (m, 1H), 3.49 (dd, J = 12.7, 6.7 Hz, 1H), 3.25 (t, J = 11.5 Hz,
1H), 3.03–2.91 (m, 2H), 2.53–2.40 (m, 5H), 1.91–1.66 (m, 10H), 1.51–1.26 (m, 8H).
13C NMR (126 MHz) δ 168.8, 164.5, 160.4, 141.7, 131.2 (q, JCF = 32.8 Hz), 130.9,
129.3, 128.8 (q, JCF = 3.78 Hz), 124.6 (q, JCF = 3.78 Hz), 123.5 (q, JCF = 273.4 Hz),
62.4, 57.0, 56.4, 56.3, 45.9, 39.7, 33.9, 32.7, 29.7, 28.7, 27.9, 26.0, 23.2, 20.8, 20.4.
HRMS: calcd for C25H33O3N5F3S2 [M-HCl+H]+: 572.1971, found: 572.1966.
4.2.5. 12N-p-Trifluoromethlybenzenesulfonyl matrinicbutyl phthalimide (11). To
a
solution of phthalimide (2.5 mmol) and triphenyl phosphine (2.5 mmol) in THF (30
mL), a solution of 6a (2.0 mmol) in THF (2 mL) was added dropwise, and the mixture
was stirred for 15 min. DIAD (2.5 mmol) was added dropwise and the mixture was
stirred for an additional 12 h. The reaction was quenched with hydrogen peroxide (2
mL) and saturated ZnCl2 (1 mL). The precipitation was filtrated off, and the filtration
was concentrated to afford 11, which was purified by flash column chromatography
with ethyl acetate/ petroleum ether to afford 11 in a 47% yield. white solid; mp
97–98 °C ;1H NMR (600 MHz, CDCl3) δ 7.99 (d, J = 8.2 Hz, 2H), 7.83 (dd, J = 5.4,
3.1 Hz, 2H), 7.75 (d, J = 8.3 Hz, 2H), 7.71 (dd, J = 5.4, 3.0 Hz, 2H), 3.67–3.64 (m,
25

1H), 3.58 (t, J = 7.2 Hz, 2H), 3.50 – 3.47 (m, 1H), 3.32–3.28 (m, 1H), 2.56 (d, J =
11.5 Hz, 1H), 2.51 (d, J = 11.6 Hz, 1H), 2.03–1.98 (m, 2H), 1.83–1.72 (m, 6H),
13
1.67–1.62 (m, 1H), 1.58–1.51 (m, 2H), 1.47–1.31 (m, 7H), 1.28–1.24 (m, 1H); C
NMR (151 MHz) δ 168.4(2), 144.7, 133.9(2), 133.7(q, JCF = 33.22 Hz), 132.2(2),
127.9(2), 125.7(q, JCF = 4.53 Hz, 2), 123.4(q, JCF = 273.31 Hz), 123.2(2), 62.9, 58.1,
56.6(2), 47.0, 39.8, 37.7, 34.5, 31.7, 28.5, 28.4, 27.9, 23.4, 20.8, 20.6.
4.2.6. 12N-p-Trifluoromethlybenzenesulfonyl matrinicbutyl isothiocyanate (13). To a
solution of 11 (1.0 mmol) in ethanol (25 mL), hydrazine hydrate (1 mL) was added,
and the mixture was refluxed for 2 h. The residue was dissolved in CH2Cl2 (50 mL),
and the organic layer was dried over Na2SO4 and concentrated to afford 12, which
was dissolved in CH2Cl2 (40 mL), then TCDI (1.5 mmol) and DMAP (0.05 mmol)
were added. The mixture was stirred for 2 h, and concentrated to afford 13, which was
purified by flash column chromatography to afford 13. Yield: 49%; white solid; mp
20
72–73 °C ; [α] = -5.16; 1H NMR (600 MHz, CDCl3) δ 7.98 (d, J = 8.1 Hz, 2H), 7.75
589
(d, J = 8.1 Hz, 2H), 3.64 (s, 1H), 3.53–3.40 (m, 3H), 3.27 (t, J = 11.3 Hz, 1H), 2.57 (d,
J = 11.3 Hz, 1H), 2.49 (d, J = 11.3 Hz, 1H), 2.04–1.95 (m, 2H), 1.84–1.72 (m, 6H),
1.58–1.28 (m, 11H). 13C NMR (151 MHz) δ 144.1, 133.9 (q, JCF = 33.2 Hz), 130.0,
128.1(2), 125.8(2), 123.4 (q, JCF = 273.3 Hz), 65.6, 62.7, 57.6, 56.6, 46.9, 45.0, 39.7,
34.4, 31.6, 29.9, 28.4, 27.9, 22.9, 20.9, 20.6. HRMS: calcd for C23H31O3N2F3S2
[M+H]+: 502.1804, found: 502.1808.
4.2.7. General procedures for the synthesis of compounds 17c–l
The solution of 15 (8.0 mmol) in THF (15 mL) was added to a solution of
sodium naphthalenide (16.0 mmol) in THF and stirred for 1 h in an ice bath. Then the
pH of solution was adjusted to 8–9 with 2 N HCl, and the residue was suspended in
methanol, and the filtration was condensed to give 16. To a solution of 16 (2.0 mmol)
26

in CH2Cl2 (50 mL), substituted benzenesulfonyl chloride or benzene isothiocyanate
(3.0 mmol) and K2CO3 (6.0 mmol) were added and stirred for 4 h. Water (30 mL) was
added, and the organic phase was separated and dried with Na2SO4 and concentrated.
The residue was purified by flash column chromatography with CH2Cl2/CH3OH and
acidized with 2 N HCl/ether to afford title compounds.
4.2.7.1. 12N-p-Methlybenzenesulfonyl matrinane hydrochloride (17c). Yield: 88%;
20
white solid; mp 166–167 °C ; [α] = 2.06; 1H NMR (600 MHz, DMSO-d6) δ 10.51 (s,
589
1H), 7.70 (d, J = 8.2 Hz, 2H), 7.40 (d, J = 8.2 Hz, 2H), 4.08–4.01 (m, 1H), 3.77 (dd, J
= 13.9, 12.2 Hz, 1H), 3.71 (dd, J = 13.9, 5.1 Hz, 1H), 3.56 (dt, J = 10.6, 3.4 Hz, 1H),
3.22–3.16 (m, 2H), 2.92–2.83 (m, 2H), 2.39 (s, 3H), 2.28–2.21 (m, 2H), 1.90–1.83 (m,
2H), 1.69–1.50 (m, 8H), 1.16–1.09 (m, 1H), 1.06–0.98 (m, 1H), 0.97–0.87 (m, 1H),
0.83–0.73 (m, 1H), 0.62 (t, J = 7.2 Hz, 3H). 13C NMR (151 MHz,) δ 143.0, 139.2,
129.6(2), 126.6(2), 63.1, 57.8, 54.6, 54.5, 48.3, 37.9, 34.5, 28.0, 27.6, 24.9, 24.5, 22.0,
20.9, 18.3, 18.1, 13.7. HRMS: calcd for C22H35O2N2S [M-HCl+H]+: 391.2414, found:
391.2413.
4.2.7.2. 12N-o-Trifluoromethylphenylcarbamoyl matrinane (17d). Yield: 60%; white
20
589
1
solid; mp 190–191 °C ; [α] = -8.15; H NMR (600 MHz, CDCl3) δ 8.04 (d, J = 8.4
Hz, 1H), 7.53 (d, J = 7.9 Hz, 1H), 7.48 (t, J = 7.9 Hz, 1H), 7.07 (t, J = 7.9 Hz, 1H),
6.84 (s, 1H), 3.81–3.73 (m, 2H), 3.29–3.22 (m, 1H), 2.74 (t, J = 14.1 Hz, 2H), 2.06 (s,
1H), 1.92–1.81 (m, 4H), 1.74–1.66 (m, 2H), 1.61–1.56 (m, 1H), 1.54–1.27 (m, 11H),
0.88 (t, J = 6.9 Hz, 3H). 13C NMR (151 MHz) δ 155.6, 137.9, 132.8, 125.9 (q, JCF
=
4.53 Hz), 124.7 (q, JCF = 273.3 Hz), 123.5, 122.2, 118.6 (q, JCF = 28.7 Hz), 63.3, 56.9,
56.7, 55.8, 45.7, 39.7, 35.2, 32.5, 28.9, 28.3, 22.9, 21.4, 21.3, 18.6, 14.2. HRMS:
calcd for C23H33ON3F3[M+H]+: 424.2570, found: 424.2574.
27

4.2.7.3. 12N-m-Trifluoromethylphenylcarbamoyl matrinane hydrochloride (17e).
20
589
1
Yield: 72%; white solid; mp 194–195 °C; [α] = -6.43; H NMR (600 MHz,
DMSO-d6) δ 9.91 (br, 1H), 9.19 (s, 1H), 7.95 (s, 1H), 7.71 (d, J = 7.9 Hz, 1H), 7.45 (t,
J = 7.9 Hz, 1H), 7.25 (d, J = 7.9 Hz, 1H), 4.00–3.97 (m, 1H), 3.82 (dd, J = 14.3, 11.5
Hz, 1H), 3.65 (dd, J = 14.3, 5.8 Hz, 1H), 3.50–3.44 (m, 1H), 3.19 (t, J = 13.6 Hz, 2H),
2.93–2.84 (m, 2H), 2.31–2.25 (m, 1H), 2.12–2.06 (m, 1H), 2.02–1.88 (m, 2H), 1.83 (d,
J = 13.8 Hz, 1H), 1.81–1.71 (m, 1H), 1.68–1.53 (m, 5H), 1.49–1.40 (m, 1H),
1.38–1.29 (m, 1H), 1.29–1.16 (m, 3H), 0.83 (t, J = 7.0 Hz, 3H). 13C NMR (151 MHz)
δ 156.6, 141.5, 129.4, 129.1(q, JCF = 31.7 Hz), 124.3 (q, JCF = 273.3 Hz), 122.6, 117.7
(q, JCF = 3.02 Hz), 115.1117.7 (q, JCF = 3.02 Hz), 63.0, 54.6, 54.3, 54.2, 47.0, 36.1,
33.7, 30.1, 28.1, 25.2, 25.0, 22.1, 18.4, 18.3, 14.0. HRMS: calcd for
C23H33ON3F3[M-HCl+H]+: 424.2570, found: 424.2570.
4.2.7.4. 12N-p-Trifluoromethylphenylcarbamoyl matrinane hydrochloride (17f).
20
Yield: 64%; white solid; mp 209 °C(dec.) ; [α] = -12.16; 1H NMR (600 MHz, CDCl3)
589
δ 9.93 (br, 1H), 9.25 (s, 1H), 7.67 (d, J = 8.6 Hz, 2H), 7.57 (d, J = 8.6 Hz, 2H),
3.99–3.88 (m, 1H), 3.82 (dd, J = 14.3, 11.4 Hz, 1H), 3.64 (dd, J = 14.3, 5.7 Hz, 1H),
3.51–3.44 (m, 1H), 3.23–3.15 (m, 2H), 2.93–2.85 (m, 2H), 2.31–2.25 (m, 1H),
2.11–1.87 (m, 3H), 1.87–1.71 (m, 2H), 1.70–1.58 (m, 4H), 1.58–1.53 (m, 1H),
1.49–1.41 (m, 1H), 1.39–1.14 (m, 4H), 0.83 (t, J = 7.0 Hz, 3H). 13C NMR (151 MHz)
δ 156.4, 144.5, 125.5 (q, JCF = 3.02 Hz, 2), 124.6 (q, JCF = 271.8 Hz), 121.4 (q, JCF
=
31.7 Hz), 118.7(2), 63.0, 54.6, 54.3, 54.2, 47.1, 36.2, 33.8, 30.0, 28.1, 25.2, 25.0, 22.1,
18.4, 18.3, 14.0. HRMS: calcd for C24H33O3N3F3[M-HCl+H]+: 424.2570, found:
424.2569.
4.2.7.5. 12N-p-Trifluoromethoxyphenylcarbamoyl matrinane hydrochloride (17g).
20
589
1
Yield: 43%; white solid; mp 127–128 °C; [α] = -4.05; H NMR (600 MHz,
28