Berberine analogues as a novel class of the low-density-lipoprotein receptor up-regulators: Synthesis, structure-activity relationships, and cholesterol-lowering efficacy

Article abstract of DOI:10.1021/jm801157z

Twenty-nine derivatives of berberine (1) or pseudoberberine (2) were designed, semisynthesized, and evaluated for their up-regulatory activity on the low-density-lipoprotein receptor (LDLR) expression. SAR analysis revealed that (i) the methylenedioxy group at the 2- and 3-position is an essential element to keep the activity, (ii) the 7-position quaternary ammonium and planar structure of the compound are activity-required, and (iii) addition of electron-donating groups at the 7- or 13-position reduced the activity. Of the compound 1 analogues, compound 2 exhibited an increased activity on LDLR expression compared to 1.Inthe hyperlipidemic rats, compound 2 (100 (mg/kg)/day) reduced blood CHO and LDL-c by 42.6% and 49.4%, respectively, more efficient than 1 did (p < 0.01 for both). The results were confirmed in the hyperlipidemic mice. LD₅₀ of 2 in mice was over 5000 mg/kg (oral). We consider compound 2 a promising cholesterol- lowering drug candidate.

Full text of DOI:10.1021/jm801157z

492
J. Med. Chem. 2009, 52, 492–501
Berberine Analogues as a Novel Class of the Low-Density-Lipoprotein Receptor Up-Regulators:
Synthesis, Structure-Activity Relationships, and Cholesterol-Lowering Efficacy
Ying-Hong Li,†,‡ Peng Yang,†,‡ Wei-Jia Kong,†,‡ Yan-Xiang Wang,† Chang-Qin Hu,§ Zeng-Yan Zuo,† Yue-Ming Wang,†
Hong Gao,† Li-Mei Gao,† Yan-Chun Feng,§ Na-Na Du,† Ying Liu,† Dan-Qing Song,*,† and Jian-Dong Jiang*,†
Institute of Medicinal Biotechnology, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100050, China, and
National Institute for the Control of Pharmaceutical and Biological Products, Beijing 100050, China
ReceiVed September 16, 2008
Twenty-nine derivatives of berberine (1) or pseudoberberine (2) were designed, semisynthesized, and evaluated
for their up-regulatory activity on the low-density-lipoprotein receptor (LDLR) expression. SAR analysis
revealed that (i) the methylenedioxy group at the 2- and 3-position is an essential element to keep the
activity, (ii) the 7-position quaternary ammonium and planar structure of the compound are activity-required,
and (iii) addition of electron-donating groups at the 7- or 13-position reduced the activity. Of the compound
1 analogues, compound 2 exhibited an increased activity on LDLR expression compared to 1. In the
hyperlipidemic rats, compound 2 (100 (mg/kg)/day) reduced blood CHO and LDL-c by 42.6% and 49.4%,
respectively, more efficient than 1 did (p < 0.01 for both). The results were confirmed in the hyperlipidemic
mice. LD50 of 2 in mice was over 5000 mg/kg (oral). We consider compound 2 a promising cholesterol-
lowering drug candidate.
Introduction
Expression of the low-density-lipoprotein receptor (LDLRa)
on the liver cell surface regulates homeostasis of human blood
LDL cholesterol (LDL-c). Increased hepatic LDLR expression
improves the clearance of blood LDL-c through a receptor-
mediated endocytosis1,2 and is strongly associated with a
decreased risk of atherosclerosis and coronary heart disease.3,4
In the previous study, we have found that berberine (1, Figure
1), a natural product extracted from a traditional Chinese herb
Huanglian (Coptis chinensis), is a promising cholesterol-
lowering drug.5 Compound 1 increases LDLR expression at the
post-transcriptional level through stabilization of LDLR mRNA
in an extracellular signal-regulated kinase (ERK) dependent
manner. Its mode of action is different from that of statins.5
Human experiment showed that the T1/2 (kꢀ) of compound 1
(orally) was 2.94 ( 0.14 h.6 Compound 1 (oral administration)
showed a significant cholesterol-lowering effect in animals as
well as in hypercholesterolemic patients with almost no
complaints of side effects.5,7,8 Also, it elevated the therapeutic
efficacy against hyperlipidemia in combination with statins.8
Figure 1. Structures of the lead compounds 1 and 2.
dimethoxy (called pseudoberberine as an isomer of 1) showed
a LDLR up-regulatory activity greater than 1 did. In the
exploration of the chemical mechanism of this class of
compounds and in the search for potent lipid-modifying agents,
SAR analysis was continued by focusing our study on the
aromatic ring C and side chains of the aromatic rings A and C,
taking 1 or 2 as the lead compound. This report includes
compound design and synthesis, SAR analysis, efficacy evalu-
ation, and toxicity and mechanism study.
The structure-activity relationship (SAR) of 1 was recently
initiated with the D aromatic ring analysis first.9 We found the
two vicinal methoxy at the 9- and 10-position or 10- and 11-
position on the ring D to be crucial structures for LDLR up-
regulation; and the compound 2 (Figure 1) bearing 10, 11-
Chemistry
In the present study we retained the 9,10-dimethoxy of 1 or
10,11-dimethoxy of 2 at the aromatic ring D and focused SAR
analysis on the variations of the aromatic ring C as well as
substituents at the 2-, 3-, 7-, and 13-position of the aromatic
rings A and C, respectively. On the basis of this strategy, 15
analogues of 1 or 2 and 14 of tetrahydrogenated 1 or 2 were
designed and synthesized.
The 29 derivatives were semisynthesized with 1 or 2 as
starting material and were grouped as analogues of 1 or 2 (Table
1) and of tetrahydrogenated 1 or 2 (Table 2). The synthetic
routes are described in the Scheme 1.
Scheme 1 includes three synthetic methods. The first one
(route A) used commercially available 1 as the starting material,
which reacted with phloroglucinol in H2SO4 (60%) at 90-95
°C to yield the key intermediate 3.10 The desired products 4a-e
were obtained with a good yield with substitution reactions of
3 with different RX in methanol in the presence of NaOH.11-13
* To whom correspondence should be addressed. For D.-Q.S.: phone,
86-10-63165268; fax, 86-10-63165268; e-mail, songdanqingsdq@
hotmail.com. For J.-D.J.: phone, 86-10-63188423; fax, 86-10-63017302;
e-mail, jiang.jdong@163.com.
†
Chinese Academy of Medical Sciences and Peking Union Medical
College.
‡
These authors made equal contribution to this work.
National Institute for the Control of Pharmaceutical and Biological
§
Products.
a Abbreviations: SAR, structure-activity relationship; LDLR, low-
density-lipoprotein receptor; LDL-c, low-density-lipoprotein cholesterol;
CHO, total cholesterol; HDL-c, high-density-lipoprotein cholesterol; HFHC,
high fat and high cholesterol; ALT, alanine aminotransferase; AST, aspartate
aminotransferase; BUN, blood-urea-nitrogen; CRE, creatinine; LPDS,
lipoprotein-deficient serum.
10.1021/jm801157z CCC: $40.75
2009 American Chemical Society
Published on Web 12/17/2008

Berberine Analogues
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 493
Table 1. Effect on the LDLR Expression by 1 Analogues with
Variations at the 2-, 3-, and 13-Position
Table 2. Effect on LDLR Expression by 1 Analogues with Variations at
the 7-Position
compd R1
R2
R3
R4
R5
X
LDLR mRNAa
compd
R1
R2
R6
X
LDLR mRNAa
1
2
OCH3
H
H
OCH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
-OCH2O-
Cl
Cl
I
2.2 ( 0.3
3.5 ( 0.9
1.0 ( 0.1
0.9 ( 0.1
0.8 ( 0.1
0.9 ( 0.1
1.0 ( 0.1
1.1 ( 0.1
1.04 ( 0.1
0.9 ( 0.1
0.9 ( 0.1
1.0 ( 0.1
1.4 ( 0.1
0.9 ( 0.1
0.9 ( 0.1
1.1 ( 0.1
1.0 ( 0.1
1
2
2.2 ( 0.3
3.5 ( 0.9
1.0 ( 0.1
1.3 ( 0.1
1.1 ( 0.1
1.0 ( 0.1
1.2 ( 0.1
1.2 ( 0.1
1.1 ( 0.1
0.8 ( 0.1
1.1 ( 0.1
0.9 ( 0.1
0.8 ( 0.1
0.9 ( 0.1
1.0 ( 0.1
0.9 ( 0.1
-OCH2O-
4a
OCH3
OH
OCH3
OC2H5 Br
OCH3
7a
7b
8a
8b
8c
8d
8e
8f
8g
8h
8i
OCH3
H
H
OCH3
H
H
H
H
H
H
H
4b OCH3
4c OCH3
4d OCH3
OH
OCH3
n-OC3H7 n-OC3H7 Br
OCH2Ph OCH2Ph Br
I
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
H
H
H
H
H
CH3
I
CH2CH3
Br
Br
I
4e
6a
OCH3
OCH3
CH2CH2OH
CH2CH2CH3
CH2CH2 CH2CH3
CH3
C2H5
n-C3H7
n-C4H9
n-C6H13
PhCH2
-OCH2O-
-OCH2O-
-OCH2O-
-OCH2O-
-OCH2O-
-OCH2O-
-OCH2O-
-OCH2O-
-OCH2O-
-OCH2O-
Cl
Cl
Cl
Cl
Cl
Br
Cl
Cl
Cl
Br
6b OCH3
6c OCH3
6d OCH3
Br
CH2COOCH2CH3 Br
CH2Ph
Br
I
Br
I
6e
6f
6g
6h
6i
OCH3
OCH3
H
H
H
OCH3 CH3
OCH3 CH2CH3
OCH3 CH2CH2CH3
OCH3 CH2COOCH2CH3 Br
OCH3 CH2Ph Br
OCH3 CH3
8j
8k
8l
OCH3 n-C4H9
OCH3 n-C6H13
OCH3 PhCH2
6j
H
a Human liver HepG2 cells were cultured in the EMEM medium
containing 0.5% of LPDS and were then incubated with 1 or 2 or their
analogues (5.0 µg/mL) for 12 h at 37 °C. Up-regulation of LDLR expression
was determined by the real-time RT-PCR. Abundance of LDLR mRNA in
the untreated cells was defined as 1, and the levels of LDLR mRNA from
the study compounds treated cells were defined as fold of the untreated
control. The data shown are the mean of three separate experiments.
a Human liver HepG2 cells were cultured in the EMEM medium
containing 0.5% of LPDS and were then incubated with 1 or 2 or their
analogues (5.0 µg/mL) for 12 h at 37 °C. Up-regulation of LDLR expression
was determined by the real-time RT-PCR. Abundance of LDLR mRNA in
the untreated cells was defined as 1, and the levels of LDLR mRNA from
the study compounds treated cells were defined as fold of the untreated
control. The data shown are the mean of three separate experiments.
The second one (route B) used compound 1 or 29 as starting
material for a selective reduction reaction, in which NaBH4 was
used as a reducing agent and methanol as the solvent.14 The
key intermediates (5a,b), dihydrogenated product of 1 or 2, were
obtained at room temperature conditions, and the total amount
of NaBH4 was controlled strictly with an optimal level of 75%
of 1 or 2 (mol/mol). The resultant intermediate (5a,b) reacted
with a series of alkylaldehydes in the solvent mixture of EtOH
and HOAc and then acidified with 2 N HCl to yield the desired
compounds (6a-e,g-i).14,15 Compounds 6f and 6j were readily
obtained through a substitution reaction of benzyl bromide with
the intermediate 5a or 5b, in which benzyl bromide was first
treated with NaI in CH3CN to improve substituent activity,
followed by an oxidation reaction using N-bromosuccinimide
(NBS).16,17
In the third synthetic route (route C), the key compounds 7a
and 7b, the tetrahydrogenated product of 1 or 2, were obtained
through a complete reduction in which NaBH4 was used as a
reducing agent. The reaction was carried out at reflux conditions
with ethanol (80%) as the solvent, and the total amount of
NaBH4 was at 1:1 ratio (mol/mol) to the compound 1 or 2. Then
the resultant intermediate 7 underwent a quaternization reaction
with alkyl halides to achieve the corresponding quaternary
ammonium salts (8a-l).18
The SAR study was first focused on the effect of the side
chains at the 2- and 3-position of the aromatic ring A (Table
1). The oxacyclopentene was opened, and a methoxyl or ethoxyl
was attached at the 3-position of the compound 3. The obtained
compounds 4a (columbamine) and 4b showed no activity on
LDLR expression. The compound 4c (palmatine), 4d, or 4e,
bearing dimethoxy, dipropoxy, or dibenzyloxy at the 2- and
3-position, respectively, completely lost their activity on LDLR.
Therefore, all of the semisynthesized analogues with modifica-
tion at the 2- and 3-position partially or completely lost their
activity on LDLR, regardless of the size of the side chains. We
conclude that the methylenedioxy at the 2- and 3-position plays
a significant role in up-regulating LDLR expression.
Next, on the aromatic ring C of 1 or 2 the hydrogen atom at
the 13-position was replaced with electron-donating groups
(Table 1). Introduction of methyl, ethyl, propyl, butyl, hexyl,
or benzyl at the 13-position generated compounds 6a-j, all of
which had their activity much less than did compound 1 or 2.
As the quaternary ammonium at the 7-position plays an
important role in binding to speculated biological targets,9,19,20
we consider that the electron-donating groups at the 13-position
might reduce the electropositivity of quaternary ammonium and
subsequently interfere with the bioactivity of the analogues.
Meanwhile, the steric obstacle from the enlarged side chains
on the 13-position could be another possible explanation.
The SAR was further explored on the variation of the aromatic
ring C. The ring C was modified with hydrogenation. As shown
in Table 2, the tetrahydrogenated product 7a (canadine) or 7b
had much lower activity on the LDLR expression compared
with parent compound 1 or 2. There could be at least two
chemical factors that contribute to the activity reduction after
Results and Discussion
SAR Analysis. The 29 analogues were examined in human
liver cells for their up-regulatory activity on LDLR mRNA
expression using a specific real time RT-PCR assay. Structures
of the analogues and their regulatory activity on LDLR mRNA
expression are shown in Tables 1 and 2.

494 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
Scheme 1. Semisynthesis of analogues of 1 and 2a
Li et al.
a Reagents and conditions: (a) phloroglucinol, 60% H2SO4, 90-95 °C; (b) RX, NaOH, CH3OH, reflux; (c) NaBH4, K2CO3, CH3OH, room temperature;
(d) RCHO, HOAc, 80% ethanol, 85-95 °C; (e) PhCH2Br, NaI, CH3CN, room temperature, reflux, NBS; (f) NaBH4, K2CO3, 80% ethanol, reflux; (g) RX,
K2CO3, CH3CN, reflux.
Therefore, the quaternary ammonium and planar structure are
both crucial in keeping compounds with potency.
Modification at the 7-position of the aromatic ring C was
explored as well. All of the compounds (8a-l) with substituents
at the 7-position exhibited an activity much less than did its
parent compound 1 or 2. It appears that the substituents at the
7-position, regardless of the size of the side chains, form a
blockage that prevents the quaternary ammonium from conduct-
ing its action on the pathway molecules for LDLR expression
regulation. Another possible factor is the nonplanar structure
of the compounds (8a-l), which might cause the activity loss
as discussed in 7a.
Taken together, we have de novo synthesized and/or semi-
synthesized 48 compound 1 analogues in total, of which 19 were
synthesized in our previous study9 and 29 were in this study.
Through SAR analysis we found and confirmed that compound
2 possesses an optimal structure in up-regulating LDLR
expression. Its activity was higher than that of compound 1.
After overall evaluation, this compound was selected for further
biological evaluation in vitro and in vivo.
Figure 2. Modeled skeleton structures of 1 and 7 analogues. The steric
structures were established on the basis of the minimum energy
conformation principle, generated by ChemOffice Ultra Drawing,
Modeling and Information (version 4.5), applying the calculator in
molecular dynamics MOPAC component to fulfill the optimization.
Compound 2 Up-Regulated LDLR Expression in Hu-
man Hepatocytes. The LDLR up-regulatory activity of 2 was
first examined in a Caucasian liver cell line HepG2, in
comparison with that of 1. In the cultured HepG2 cells,
compound 2 significantly increased LDLR mRNA expression,
and the activity of 2 was about 1.5-fold of that of 1 in the real
time RT-PCR assay (Figure 3a). With respect to the reference
drug simvastatin, we found that switch of “-OCH3” from the
9-position (compound 1) to the 11-position (compound 2)
increases the effect equivalent to that of a doubled concentration
of simvastatin (Figure 3b). Accordingly, LDLR protein expres-
sion was also increased on the surface of the 2-treated
hepatocytes, with a degree higher than on those treated with
hydrogenation. First, the quaternary ammonium was converted
into the tertiary ammonium, resulting in a loss of the positive
ion at this position, and second, the planar structure was
converted into bended structure (Figure 2) at the pseudo-trans
ring junction between B and C rings, causing a significant
configuration change. These factors appeared to alter the
interaction between the compounds (7a and 7b) and speculated
target(s) and reduced their activity on LDLR expression.

Berberine Analogues
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 495
compound 1 (Figure 3c). Besides HepG2 cells, the enhanced
activity of 2 was also detected in the Bel-7402 cells (not shown),
a liver cell line derived from a Chinese hepatoma patient,
indicating the effectiveness of the compound in cells with either
Caucasian or oriental genetic background.
Lipid-Lowering Efficacy of Compound 2 in Hyperlipi-
demic Animals. The superior LDLR up-regulatory activity of
2 encouraged us to test its cholesterol-lowering efficacy in vivo.
The rats were fed with a high-fat and high-cholesterol (HFHC)
diet for 8 weeks before treatment. Compared to the normal diet
group, 2.4-fold increases in blood LDL-c and total cholesterol
(CHO) were detected in the HFHC-fed rats (P < 0.01, Figure
4a). No alteration of the high-density-lipoprotein cholesterol
(HDL-c) level was found in this model.
Drug interventions lowered blood lipids. The results are
shown in Figure 4a. After a 30-day treatment, compound 1 at
100 (mg/kg)/day reduced CHO and LDL-c by 27.4% and 31.6%,
respectively (P < 0.01, compared to the untreated HFHC
control). The 100 (mg/kg)/day of 2 reduced blood CHO and
LDL-c by 42.6% and 49.4%, respectively, significantly higher
than that of 1 (P < 0.01). The lipid reduction by 1 or 2 was
time-dependent and became statistically significant after 10 days
on therapy (Figure 4b). The LDL cholesterol-lowering efficacy
of 2 was higher than that of 1 at every time-point examined. In
addition, Figure 4c reveals the total cholesterol- and LDL-c-
lowering effect of 2 in comparison with that of the reference
drug simvastatin.
Livers were dissected from the animals at the end of
experiment. Hepatic RNA was isolated and followed by LDLR
mRNA expression analysis. As shown in Figure 4d, treatment
with 100 (mg/kg)/day of 2 increased liver LDLR mRNA
expression by 2.9-fold of the untreated HFHC control and 1.5-
fold of the compound 1 treatment (P < 0.01). This result was
consistent with that observed in cultured cells. As statins up-
regulate HMG-CoA reductase mRNA expression,8 the expres-
sion of this gene was also measured for 1 or 2. As shown in
Figure 4d, neither 1 nor 2 enhanced the expression of HMG-
CoA reductase mRNA in the rat livers, confirming the mech-
anism difference between statins and compound 1 or 2.
Compound 2 was safe and well tolerated in the rats. We did
not observe weight loss, food intake reduction, or liver/kidney
function abnormalities (Figure 4e,f) in the rats treated with
compound 2. The safety results of 2 agreed with that of
compound 1.5,8
Compound 2 lowered blood lipids also in a dose-dependent
manner. As shown in Figure 5a, after a 17-day treatment of the
hyperlipidemic rats, 2 at 100 or 300 (mg/kg)/day reduced CHO
by 28.5% or 46.3% and LDL-c by 31% or 53.6%, respectively.
Compound 2 at 300 (mg/kg)/day increased liver LDLR expres-
sion by 4.8-fold of the untreated control, significantly higher
than did the 100 (mg/kg)/day group (Figure 5b; P < 0.01).
The in vivo lipid-lowering activity of 2 was confirmed in
another animal model, the hyperlipidemic C57BL/6J mice, that
developed hyperlipidemia after 8-weeks of HFHC diet feeding
(Figure 6a,b). Treatment with compound 1 or 2 at 100 (mg/
kg)/day for 21 days lowered CHO by 16.7% or 29.7% and
LDL-c by 23.9% or 39%, respectively. Again, the cholesterol-
lowering activity of 2 was significantly higher than that of 1 in
the mice (P < 0.01). Furthermore, compound 2 increased the
liver LDLR mRNA expression by 3.2-fold of the control (Figure
6c) and 1.5-fold of 1 (P < 0.01). The results in mice fully agreed
with that in rats as well as in cell culture.
Figure 3. Up-regulation of LDLR expression by 2 in the HepG2 cells.
(a) LDLR mRNA levels. HepG2 cells cultured in the LPDS containing
media were left untreated or treated for 12 h with 1 or 2 at the
concentrations indicated. Total cellular RNAs were extracted and
reversely transcribed into cDNAs. LDLR mRNA levels were analyzed
by real-time RT-PCR using GAPDH as internal control. The normalized
LDLR mRNA levels were plotted as fold of the untreated control, which
was designated as 1. Values are the mean ( SE of at least three
independently experiments: ($$) P < 0.01 as compared to that of 1 at
the same concentration. (b) LDLR up-regulatory activity in 1, 2, and
simvastatin. HepG2 cells were left untreated or treated with 1, 2, or
simvastatin as indicated for 12 h. Cellular LDLR mRNA levels were
measured by real-time RT-PCR and plotted as fold of untreated control.
Values are the mean ( SEM of three separate experiments: (//) p <
0.01 as compared to that of untreated control; (##) p < 0.01 as compared
to that of 20 µM BBR or 1 µM simvastatin. (c) LDLR protein
expression on cell surface. HepG2 cells were untreated or treated with
10 µg/mL 1 or 2 for 12 h. Cells were analyzed by flow cytometry for
cell surface LDLR protein expression using a rabbit-polyclonal antibody
against LDLR. Normal rabbit IgG was used as control for the
background staining. Fluorescence intensities on cell surface were
analyzed in a FACSCalibur system. Count on the Y axis is the absolute
number of cells, and FL1 on the X axis represents the fluorescent
intensity of the LDLR protein expressed on the HepG2 cell surface.
Safety Evaluation of 2. While the pharmacokinetics of
compound 2 is still under investigation, we performed safety

496 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
Li et al.
Figure 4. Lipid-lowering effect of 1 and 2 in the hyperlipidemic rats. Male Sprague-Dawley rats were fed with HFHC diet for 8 weeks. Then the
animals were untreated (n ) 6) or treated with 100 (mg/kg)/day of 1 (n ) 8) or 2 (n ) 8) for 30 days, respectively. (a) End-point blood lipid levels.
At the end of the treatment course, blood samples were taken from the animals, and CHO, LDL-c and HDL-c levels were determined. (b) Time-
dependent reduction of blood LDL-c by 1 and 2. (c) Comparison of the cholesterol-lowering effects of compound 2 with simvastatin. Hyperlipidemic
rats were untreated (n ) 6) or treated with either 5 (mg/kg)/day of simvastatin (n ) 7) or 100 (mg/kg)/day of compound 2 (n ) 7). Four weeks
later, serum cholesterol and LDL-c levels were assayed. Values are the mean ( SEM: (//) p < 0.01, as compared to that of normal diet; (##) p
< 0.01, as compared to that of HFHC; ($$) p < 0.01, as compared to that of simvastatin. (d) LDLR up-regulatory activities in vivo. After the
30-day experiment, rats were sacrificed and their livers were dissected. Liver total RNAs were isolated for real-time RT-PCR assay. The expression
levels of LDLR and HMG-CoA reductase mRNAs were normalized to that of GAPDH and presented as fold of the HFHC group. (e) Aspartate
aminotransferase (AST) and alanine aminotransferase (ALT) levels in the treatment with compounds 1 and 2 in rats. Hyperlipidemic rat were
treated with the study compound (100 (mg/kg)/day, oral) for 30 days. Blood samples were taken before and after the treatment, and their AST and
ALT levels were measured. Presented are mean and sd values. (f) Blood urea nitrogen (BUN) and creatinine (CRE) levels in the treatment with
compounds 1 and 2 in rats. Hyperlipidemic rat were treated as mentioned above. Blood samples were assayed for their BUN and CRE levels.
Presented are mean and sd values. Values are the mean ( SE of all of the animals in each group: (//) P < 0.01 vs that of the normal diet group
(n ) 5); (#) P < 0.05; (##) P < 0.01 vs that of the HFHC diet group; ($) P < 0.05; ($$) P < 0.01 vs that of 1 by one-way ANOVA and multiple
comparisons.
evaluation and acute toxicity test in the Kunming mice.
Compound 2 was given orally in a single-dosing experiment at
250, 1000, or 5000 mg/kg. Then the animals were closely
monitored. No animal died in the 14-day treatment, indicating
that the LD50 of 2 was larger than 5000 mg/kg. Oral administra-
tion of 2 at 250 or 1000 mg/kg did not affect the body weights
(BW) of the mice. For the 5000 mg/kg group, transient BW
loss (-7%) was observed in the first 2-3 days, followed by a
rebound to the normal level (not shown). Furthermore, the
animals were healthy in behavior and had shining fur even at
the dose of 5000 mg/kg. The histological examination for liver,
kidney, heart, lung, brain, and colon of the mice was done 14
days after treatment. Compound 2 showed no toxicity to the
liver, kidney (Figure 7), as well as other organs (not shown),
even when the dose was up to 5000 mg/kg. These results suggest
compound 2 to be considerably safe in vivo.
Conclusions
In the continuation of SAR study for the compound 1
derivatives as a novel class of LDLR up-regulators, we have
semisynthesized 29 analogues defined through variations of the
aromatic ring C and the substituents at the 2-, 3-, 7-, and 13-
position of the parent compound 1 or 2. The SAR analysis
reveals that (i) the methylenedioxy at the 2- and 3-position is

Berberine Analogues
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 497
Figure 5. Dose-dependent activity of 2 in the hyperlipidemic rats.
Hyperlipidemic Sprague-Dawley rats were untreated (n ) 6) or treated
with 2 at 100 (mg/kg)/day (n ) 8) or 300 (mg/kg)/day (n ) 8) for 17
days. (a) End-point blood levels of CHO and LDL-c. (b) Liver LDLR
mRNA. At the end of treatment, animals were sacrificed and their livers
were dissected. Liver total RNAs were isolated for real-time RT-PCR
assay of LDLR mRNA expression levels. The normalized LDLR
mRNA levels were presented as fold of the HFHC group, which was
designated as 1. Values are the mean ( SE of all of the animals in
each group: (##) P < 0.01 vs that of the HFHC diet group; ($$) P <
0.01 vs that of the low dose group of 2 (100 (mg/kg)/day) by one-way
ANOVA and multiple comparisons.
Figure 6. Lipid-lowering effect of 1 and 2 in the hyperlipidemic mice.
Male C57BL/6J mice were fed with HFHC diet for 8 weeks. Then the
animals were untreated (n ) 6) or treated with 100 (mg/kg)/day of 1
(n ) 8) or 2 (n ) 8) for 21 days, respectively. Before and after
treatment, blood samples were taken for the assay of CHO (a) and
LDL-c (b) levels. At the end of treatment, livers were dissected from
the animals and total liver RNAs were isolated for the real-time RT-
PCR of the LDLR mRNA expression (c). The normalized LDLR
mRNA levels were presented as fold of the HFHC group, which was
designated as 1. Values are the mean ( SE of all of the animals in
each group: (//) P < 0.01 vs that of the normal diet group (n ) 5);
(##) P < 0.01 vs that of the HFHC diet group; ($$) P < 0.01 vs that
of 1 by one-way ANOVA and multiple comparisons.
an important element to keep the compound potent in up-
regulating LDLR expression, (ii) adding electron-donating
groups at the 7- or 13-position reduces the activity, and (iii)
the 7-position quaternary ammonium ion and planar structure
of the compound play crucial roles for its interaction with
speculated target molecule(s). Among the analogues, compound
2 afforded an up-regulatory activity on LDLR expression greater
than 1 did in vitro. Accordingly, 2 had a significantly enhanced
lipid-lowering efficacy in vivo compared to 1. The increased
activity of 2 might result from the higher electronpositivity at
the 7-position in respect to that of 1.9 As compound 2 shows
an improved lipid-lowering activity with a mechanism different
from that of statins and has a considerably good safety in vivo,
it merits further investigation.
Experimental Section
Chemistry. Melting points (mp) were obtained with YRT-3
1
melting point apparatus and were uncorrected. H NMR spectra
were performed on a Varian Inova 400 MHz spectrometer (Varian,
San Francisco, CA) in DMSO-d6 (as internal standard on a δ scale).
FAB and EI mass spectra were recorded on an Autospec UItima-
TOF mass spectrometer (Micromass UK Ltd., Manchester, U.K.).
Elemental analyses were from Flash 1112 series EA (Thermo
Finnigan) and were within (0.4 of the theoretical values.
Route A (for 4a-e). 2-Hydroxy-3-methoxy-9,10-dimethox-
yprotoberberine Iodide (4a). To a stirred mixture of 310 (360 mg,
1.0 mmol) and NaOH (100 mg, 2.5 mmol) in methanol (30 mL),
methyl iodide (0.07 mL, 1.2 mmol) was added. The reaction mixture
Figure 7. Histological examination of compound 2. Male Kunming
mice were divided into four groups with six mice in each. They were
untreated or orally treated with 2 at 250, 1000, or 5000 mg/kg in one
dose. Fourteen days later, animal organs were taken for histological
examination using H&E staining.
was stirred at room temperature overnight. The solvent was removed
by evaporation, and then the residue was purified by flash
chromatography over silica gel, affording the title compound (237

498 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
Li et al.
mg, 51%) as a yellow solid, mp 214-216 °C. MS m/z 338 (M -
2H), 3.35 (q, J ) 7.2 Hz, 2H), 4.09 (s, 3H), 4.09 (s, 3H), 4.81 (t,
J ) 5.6 Hz, 2H), 6.18 (s, 2H), 7.16 (s, 1H), 7.29 (s, 1H), 8.19 (d,
J ) 9.2 Hz, 1H), 8.24 (d, J ) 9.6 Hz, 1H), 9.92 (s, 1H). Anal.
(C22H22NO4Cl) C, H, N.
1
I)+. H NMR: δ 3.18 (t, J ) 6.0 Hz, 2H), 3.89 (s, 3H), 4.06 (s,
3H), 4.08 (s, 3H), 4.92 (t, J ) 6.0 Hz, 2H), 7.05 (s, 1H), 7.54 (s,
1H), 8.05 (d, J ) 9.2 Hz, 1H), 8.18 (d, J ) 9.2 Hz, 1H), 8.81 (s,
1H), 9.36 (br, 1H), 9.85 (s, 1H). Anal. (C20H20NO4I) C, H, N.
2-Hydroxy-3-ethoxy-9,10-dimethoxyprotoberberine Bromide
(4b). To a stirred mixture of 3 (360 mg, 1.0 mmol) and NaOH
(100 mg, 2.5 mmol) in methanol (30 mL), ethyl bromide (0.10 mL,
1.5 mmol) was added. The reaction mixture was stirred at room
temperature for 0.5 h and then heated to reflux for 5 h. The solvent
was removed by evaporation, and then the residue was purified by
flash chromatography over silica gel, affording the title compound
(155 mg, 36%) as a redlike solid, mp 172-174 °C. MS m/z 352
(M - Br)+. 1H NMR: δ 1.39 (t, J ) 9.2 Hz, 3H), 3.17 (t, J ) 8.4
Hz, 2H), 4.06 (s, 3H), 4.08 (s, 3H), 4.14 (q, J ) 9.2 Hz, 2H), 4.91
(t, J ) 8.4 Hz, 2H), 7.03 (s, 1H), 7.54 (s, 1H), 8.05 (d, J ) 12.0
Hz, 1H), 8.18 (d, J ) 12.0, 1H), 8.80 (s, 1H), 9.27 (br, 1H), 9.85
(s, 1H). Anal. (C21H22NO4Br) C, H, N.
2,3-Dimethoxy-9,10-dimethoxyprotoberberine Iodide (4c).
The title compound was obtained from 3 (1.0 mmol) and methyl
iodide (2.5 mmol) with a procedure similar to that for compound
4b. Yield: 78%. Yellow solid, mp 222-224 °C. MS m/z 352 (M
- I)+. 1H NMR: δ 3.18 (t, J ) 6.0 Hz, 2H), 3.87 (s, 3H), 3.93 (s,
3H), 4.07 (s, 3H), 4.09 (s, 3H), 4.94 (t, J ) 6.0 Hz, 2H), 7.09 (s,
1H), 7.70 (s, 1H), 8.02 (d, J ) 9.2 Hz, 1H), 8.20 (d, J ) 9.2 Hz,
1H), 9.00 (s, 1H), 9.87 (s, 1H). Anal. (C21H22NO4I) C, H, N.
2,3-Di-n-propoxy-9,10-dimethoxyprotoberberine Bromide
(4d). The title compound was obtained from 3 (1.0 mmol) and
propyl bromide (2.5 mmol) with a procedure similar to that for
compound 4b. Yield: 80%. Yellow solid, mp 218-220 °C. MS
m/z 408 (M - Br)+. 1H NMR: δ 1.00 (t, J ) 7.2 Hz, 3H), 1.03 (t,
J ) 7.2 Hz, 3H),1.78 (q, J ) 7.2 Hz, 4H), 3.20 (t, J ) 6.0 Hz,
2H), 4.02-4.11 (m, 4H), 4.06 (s, 3H), 4.09 (s, 3H), 4.93 (t, J )
6.0 Hz, 2H), 7.08 (s, 1H), 7.71 (s, 1H), 8.02 (d, J ) 9.2 Hz, 1H),
8.21 (d, J ) 9.2 Hz, 1H), 8.98 (s, 1H), 9.86 (s, 1H). Anal.
(C25H30NO4Br) C, H, N.
2,3-Dibenzyloxy-9,10-dimethoxyprotoberberine Bromide (4e).
The title compound was obtained from 3 (1.0 mmol) and benzyl
bromide (2.5 mmol) with a procedure similar to that for compound
4b. Yield: 89%. Yellow solid, mp 244-246 °C. MS m/z 504 (M
- Br)+. 1H NMR: δ 3.20 (t, J ) 6.0 Hz, 2H), 4.07 (s, 3H), 4.09 (s,
3H), 4.93 (t, J ) 6.0 Hz, 2H), 5.25 (s, 2H), 5.30 (s, 2H), 7.22 (s,
1H), 7.34-7.53 (m, 10H), 7.90 (s, 1H), 8.02 (d, J ) 9.2 Hz, 1H),
8.21 (d, J ) 9.2 Hz, 1H), 9.00 (s, 1H), 9.88 (s, 1H). Anal.
(C33H30NO4Br) C, H, N.
Route B (for 6a-j). 2,3-Methylenedioxy-9,10-dimethoxy-13-
methylprotoberberine Chloride (6a). To a stirred solution of 1
(3.71 g, 10 mmol) and K2CO3 (3.6 g, 30 mmol) in methanol (125
mL), 5% NaOH (5 mL) solution containing NaBH4 (0.30 g, 7.5
mmol) was added dropwise. The reaction mixture was stirred at
room temperature for 1 h and the precipitated product was filtered,
washed with 30% ethanol (20 mL) and 80% ethanol (20 mL), and
then recrystallized from absolute ethanol to provide 5a (2.5 g, 74%)
as a brownlike solid.
2,3-Methylenedioxy-9,10-dimethoxy-13-n-propylprotober-
berine Chloride (6c). The title compound was obtained from 5a
and propaldehyde with a procedure similar to that for compound
6a. Yield: 54%. Yellow solid, mp 201-202 °C. MS m/z 378 (M
- Cl)+. 1H NMR: δ 0.99 (t, J ) 7.2 Hz, 3H), 1.78 (q, J ) 7.2 Hz,
2H), 3.07 (t, J ) 5.6 Hz, 2H), 3.31 (m, 2H), 4.08 (s, 6H), 4.79 (t,
J ) 5.6 Hz, 2H), 6.18 (s, 2H), 7.15 (s, 1H), 7.25 (s, 1H), 8.17 (d,
J ) 9.6 Hz, 1H), 8.22 (d, J ) 9.6 Hz, 1H), 9.90 (s, 1H). Anal.
(C23H24NO4Cl) C, H, N.
2,3-Methylenedioxy-9,10-dimethoxy-13-n-butylprotober-
berine Chloride (6d). The title compound was obtained from 5a
and butaldehyde with a procedure similar to that for compound
6a. Yield: 39%. Yellow solid, mp 201-202 °C. MS m/z 392 (M
- Cl)+. 1H NMR: δ 0.91 (t, J ) 7.2 Hz, 3H), 1.41 (q, J ) 7.2 Hz,
2H), 1.75 (m, 2H), 3.07 (t, J ) 5.6 Hz, 2H), 3.35 (m, 2H), 4.08 (s,
3H), 4.09 (s, 3H), 4.79 (t, J ) 5.6 Hz, 2H), 6.18 (s, 2H), 7.15 (s,
1H), 7.28 (s, 1H), 8.19 (d, J ) 9.2 Hz, 1H), 8.20 (d, J ) 9.2 Hz,
1H), 9.90 (s, 1H). Anal. (C24H26NO4Cl) C, H, N.
2,3-Methylenedioxy-9,10-dimethoxy-13-n-hexylprotober-
berine Chloride (6e). The title compound was obtained from 5a
and hexaldehyde with a procedure similar to that for compound
6a. Yield: 40%. Yellow solid, mp 198-199 °C. MS m/z 420 (M
1
- Cl)+. H NMR: δ 0.84 (m, 3H), 1.25 (m, 4H), 1.37 (m 2H),
1.74 (m, 2H), 3.07 (t, J ) 5.6 Hz, 2H), 3.34 (m, 2H), 4.06 (s, 3H),
4.09 (s, 3H), 4.78 (t, J ) 5.6 Hz, 2H), 6.18 (s, 2H), 7.15 (s, 1H),
7.28 (s, 1H), 8.19 (s, 2H), 9.86 (s, 1H). Anal. (C26H30NO4Cl) C,
H, N.
2,3-Methylenedioxy-9,10-dimethoxy-13-benzylprotober-
berine Bromide (6f). To a stirred suspension of NaI (310 mg, 2.06
mmol) in acetonitrile (16 mL), benzyl bromide (350 mg, 2.06 mmol)
was added dropwise. The reaction mixture was stirred at room
temperature for 1 h. After 6a (580 mg, 1.72 mmol) was added, the
system was heated to reflux for 3 h, the solvent was removed by
evaporation, and then the residue was redissolved in CHCl3 (20
mL). The organic layer was then washed with 10% K2CO3 (25 mL),
dried over MgSO4, and filtrated. To the filtration was added NBS
(367 mg, 2.06 mmol), and the reaction mixture was stirred at room
temperature for 1 h. The solvent was removed by evaporation, and
the residue was purified by flash chromatography over silica gel,
affording the title compound (620 mg, 72%) as a yellow solid, mp
204-206 °C. MS m/z 426 (M - Br)+. 1H NMR: δ 3.14 (t, J ) 5.6
Hz, 2H), 4.02 (s, 3H), 4.11 (s, 3H), 4.74 (s, 2H), 4.87 (t, J ) 5.6
Hz, 2H), 6.07 (s, 2H), 6.96 (s, 1H), 7.14-7.17 (m, 3H), 7.26-7.37
(m, 3H), 7.77 (d, J ) 9.6 Hz, 1H), 8.08 (d, J ) 9.6 Hz, 1H), 10.03
(s, 1H). Anal. (C27H24NO4Br) C, H, N.
2,3-Methylenedioxy-10,11-dimethoxy-13-methylprotober-
berine Chloride (6g). Synthetic intermediate 5b was obtained from
2 and NaBH4 with a procedure similar to that for compound 5a.
Yield: 59%. Yellow-like solid.
To a stirred solution of 5a (170 mg, 5.0 mmol) in 80% ethanol
(8 mL) and HOAc (2 mL), formaldehyde (2 mL) was added. The
reaction mixture was heated to 85-95 °C for 5 h. The solvent was
removed by evaporation, and the residue was acidified by 2 N HCl
(5 mL), then stirred at room temperature for 1 h. The solid was
collected by filtration and then purified by flash chromatography
over silica gel, affording the title compound (74 mg, 38%) as a
yellow solid, mp 198-200 °C. MS m/z 350 (M - Cl)+. 1H NMR:
δ 2.92 (s, 3H), 3.10 (t, J ) 5.6 Hz, 2H), 4.08 (s, 3H), 4.09 (s, 3H),
4.81 (t, J ) 5.6 Hz, 2H), 6.17 (s, 2H), 7.14 (s, 1H), 7.46 (s, 1H),
8.18 (d, J ) 9.6 Hz, 1H), 8.19 (d, J ) 9.6 Hz, 1H), 9.89 (s, 1H).
Anal. (C21H20NO4Cl) C, H, N.
The title compound was obtained from 5b and formaldehyde
with a procedure similar to that for compound 6a. Yield: 39%.
Yellow solid, mp 243-244 °C. MS m/z 350 (M - Cl)+. 1H NMR:
δ 2.91 (s, 3H), 3.09 (t, J ) 6.0 Hz, 2H), 4.02 (s, 3H), 4.12 (s, 3H),
4.63 (t, J ) 6.0 Hz, 2H), 6.06 (s, 2H), 6.97(s, 1H), 7.30 (s, 1H),
7.50 (s, 1H), 7.60 (s, 1H), 9.29 (s, 1H). Anal. (C21H20NO4Cl) C,
H, N.
2,3-Methylenedioxy-10,11-dimethoxy-13-n-butylprotober-
berine Chloride (6h). The title compound was obtained from 5b
and butaldehyde with a procedure similar to that for compound
6a. Yield: 39%. Yellow solid, 222-224 °C. MS m/z 392 (M -
1
2,3-Methylenedioxy-9,10-dimethoxy-13-ethylprotoberberine
Chloride (6b). The title compound was obtained from 5a and
acetaldehyde with a procedure similar to that for compound 6a.
Yield: 42%. Yellow solid, mp 181-183 °C. MS m/z 364 (M -
Cl)+. H NMR: δ 0.92 (t, J ) 7.2 Hz, 3H), 1.41 (q, J ) 7.2 Hz,
2H), 1.78 (m, 2H), 3.07 (t, J ) 5.6 Hz, 2H), 3.35 (s, 2H), 4.00 (s,
3H), 4.13 (s, 3H), 4.63 (t, J ) 5.6 Hz, 2H), 6.18 (s, 2H), 7.16 (s,
1H), 7.29 (s, 1H), 7.54 (s, 1H), 7.75 (s, 1H), 9.55 (s, 1H). Anal.
(C24H26NO4Cl) C, H, N.
1
Cl)+. H NMR: δ 1.45 (t, J ) 7.2 Hz, 3H), 3.08 (t, J ) 5.6 Hz,

Berberine Analogues
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 499
2,3-Methylenedioxy-10,11-dimethoxy-13-n-hexylprotober-
berine Chloride (6i). The title compound was obtained from 5b
and hexaldehyde with a procedure similar to that for compound
6a. Yield: 40%. Yellow solid, mp 195-196 °C. MS m/z 420 (M
- Cl)+. 1H NMR: δ 0.84 (t, J ) 6.8 Hz, 3H), 1.22-1.27 (m, 4H),
1.39 (m, 2H), 1.78 (m, 2H), 3.06 (t, J ) 5.6 Hz, 2H), 3.34 (t, J )
6.8 Hz, 2H), 4.00 (s, 3H), 4.13 (s, 3H), 4.65 (t, J ) 5.6 Hz, 2H),
6.18 (s, 2H), 7.17 (s, 1H), 7.28 (s, 1H), 7.54 (s, 1H), 7.79 (s, 1H),
9.65 (s, 1H). Anal. (C26H30NO4Cl) C, H, N.
1H), 3.77 (s, 3H), 3.83 (s, 3H), 3.94-3.98 (m, 1H), 4.01 (q, J )
5.2 Hz, 1H), 4.24 (q, J ) 5.2 Hz, 1H), 4.55 (d, J ) 15.6 Hz, 1H),
5.07 (q, J ) 5.2 Hz, 1H), 5.35 (d, J ) 15.6 Hz, 1H), 5.37 (t, J )
4.8 Hz, 2H), 6.05 (d, J ) 7.2 Hz, 2H), 6.88 (s, 1H), 7.07 (s, 1H),
7.09 (d, J ) 8.4 Hz, 1H), 7.14 (d, J ) 8.4 Hz, 1H). Anal.
(C22H26NO5Br) C, H, N.
2,3-Methylenedioxy-9,10-dimethoxy-7-n-propyltetrahydro-
protoberberine Iodide (8d). The title compound was obtained from
7a and propyl iodide with a procedure similar to that for compound
8a. Yield: 60%. White-like solid, mp 165-166 °C. MS m/z 382
2,3-Methylenedioxy-10,11-dimethoxy-13-benzylprotober-
berine Bromide (6j). The title compound was obtained from 5b
and benzyl bromide with a procedure similar to that for compound
6e. Yield: 54%. Yellow solid, mp 218-220 °C. MS m/z 426 (M -
1
(M - I)+. H NMR: δ 0.76 (t, J ) 7.2 Hz, 3H), 1.55-1.81 (m,
1H), 1.79 (m, 1H), 2.89 (t, J ) 8.4 Hz, 2H), 3.05-3.10 (m, 3H),
3.68-3.79 (m, 1H), 3.78 (s, 3H), 3.83 (s, 3H), 4.01 (q, J ) 4.8
Hz, 1H), 4.13 (q, J ) 4.0 Hz, 1H), 4.55 (d, J ) 16.4 Hz, 1H), 4.88
(d, J ) 16.4 Hz, 1H), 5.14 (q, J ) 5.2 Hz, 1H), 6.05 (d, J ) 8.0
Hz, 2H), 6.88 (s, 1H), 7.07 (s, 1H), 7.09 (d, J ) 8.8 Hz, 1H), 7.15
(d, J ) 8.8 Hz, 1H). Anal. (C23H28NO4I) C, H, N.
1
Br)+. H NMR: δ 3.15 (t, J ) 5.6 Hz, 2H), 3.74 (s, 3H), 4.00 (s,
3H), 4.71 (t, J ) 5.6 Hz, 2H), 4.77 (s, 2H), 6.08 (s, 2H), 7.04 (s,
1H), 7.17 (s, 1H), 7.14-7.17 (m, 3H), 7.25-7.37 (m, 3H), 7.78
(s, 1H), 9.68 (s, 1H). Anal. (C27H24NO4Br) C, H, N.
Route C (for 7a,b, 8a-l). 2,3-Methylenedioxy-9,10-dimethox-
ytetrahydroprotoberberine (7a). To a stirred refluxing solution
of 2 (3.71 g, 10 mmol) and K2CO3 (3.6 g, 30 mmol) in 80% ethanol
(60 mL), NaBH4 (0.4 g, 10 mmol) was added portionwise. The
reaction mixture was allowed to react at this reflux temperature
for 20 min, then at room temperature for 4 h, and generated a lot
of white solid. The solid was collected by filtration and then
recrystallized from 95% ethanol twice to afford the title compound
(2 g, 59%) as a white needle, mp 174-176 °C. MS m/z 340 (M +
2,3-Methylenedioxy-9,10-dimethoxy-7-n-butyltetrahydropro-
toberberine Bromide (8e). The title compound was obtained from
7a and butyl bromide with a procedure similar to that for compound
8a. Yield: 22%. White-like solid, mp 130-131 °C. MS m/z 396
1
(M - Br)+. H NMR: δ 0.87 (t, J ) 7.2 Hz, 3H), 1.18-1.28 (m,
2H), 1.71 (m, 1H), 1.87 (m, 1H), 3.05-3.14 (m, 3H), 3.38-3.40
(m, 2H), 3.65-3.75 (m, 2H), 3.78-3.83 (m, 1H), 3.79 (s, 3H),
3.81 (s, 3H), 4.65 (d, J ) 12.4 Hz, 1H), 4.74 (q, J ) 5.6 Hz, 1H),
4.80 (d, J ) 12.4 Hz, 1H), 6.03 (d, J ) 8.0 Hz, 2H), 6.88 (s, 1H),
7.07 (s, 1H), 7.09 (d, J ) 9.2 Hz, 1H), 7.12 (d, J ) 8.8 Hz, 1H).
Anal. (C24H30NO4Br) C, H, N.
1
H)+. H NMR: δ 2.43 (q, J ) 3.2 Hz, 1H), 2.55 (q, J ) 4.0 Hz,
1H), 2.61 (m, 1H), 2.85-2.93 (m, 1H), 3.08 (q, J ) 3.6 Hz, 1H),
3.27-3.36 (m, 2H), 3.35 (d, J ) 16.0 Hz, 1H), 3.71 (s, 3H), 3.76
(s, 3H), 4.04 (d, J ) 16.0 Hz, 1H), 5.93 (d, J ) 2.4 Hz, 2H), 6.65
(s, 1H), 6.84 (d, J ) 8.4 Hz, 1H), 6.87 (d, J ) 8.4 Hz, 1H), 6.90
(s, 1H). Anal. (C20H21NO4) C, H, N.
2,3-Methylenedioxy-9,10-dimethoxy-7-ethylacetatetetrahydro-
protoberberine Bromide (8f). The title compound was obtained
from 7a and ethyl acetate bromide with a procedure similar to that
for compound 8a. Yield: 71%. White-like solid, mp 142-144 °C.
2,3-Methylenedioxy-10,11-dimethoxytetrahydroprotober-
berine (7b). The title compound was obtained from 2 and NaBH4
with a procedure similar to that for compound 7a. Yield: 63%.
1
MS m/z 426 (M - Br)+. H NMR: δ 1.21 (t, J ) 7.2 Hz, 3H),
3.10-3.25 (m, 2H), 3.42 (q, J ) 5.6 Hz, 1H), 3.75-3.94 (m, 2H),
3.76 (s, 3H), 3.86 (s, 3H), 4.14-4.26 (m, 2H), 4.22 (q, J ) 5.6
Hz, 1H), 4.40-4.49 (m, 2H), 4.90 (d, J ) 16.4 Hz, 1H), 4.99 (d,
J ) 16.4 Hz, 1H), 5.07 (q, J ) 5.6 Hz, 1H), 6.02 (d, J ) 6.8 Hz,
2H), 6.88 (s, 1H), 6.92 (s, 1H), 7.01 (d, J ) 8.4 Hz, 1H), 7.10 (d,
J ) 8.4 Hz, 1H). Anal. (C24H28NO6Br) C, H, N.
1
White solid, mp 176-177 °C. MS m/z 340 (M + H)+. H NMR:
δ 2.39-2.44 (m, 1H), 2.52-2.61 (m, 2H), 2.85-2.93 (m, 1H), 3.03
(q, J ) 3.6 Hz, 1H), 3.26 (d, J ) 10.8 Hz, 1H), 3.39 (d, J ) 10.8
Hz, 1H), 3.47 (d, J ) 14.8 Hz, 1H), 3.70 (s, 6H), 3.85 (d, J ) 14.8
Hz, 1H), 5.94 (s, 2H), 6.65 (s, 1H), 6.65 (s, 1H), 6.69 (s, 1H), 6.89
(s, 1H). Anal. (C20H21NO4) C, H, N.
2,3-Methylenedioxy-9,10-dimethoxy-7-benzyltetrahydropro-
toberberine Bromide (8g). The title compound was obtained from
7a and benzyl bromide with a procedure similar to that for
compound 8a. Yield: 70%. White powder, mp 180-181 °C. MS
2,3-Methylenedioxy-9,10-dimethoxy-7-methyltetrahydropro-
toberberine Iodide (8a). To a stirred solution of 3 (340 mg, 1.0
mmol) and K2CO3 (410 mg, 3.0 mmol) in acetonitrile (40 mL),
methyl iodide (3.0 mL, 5.0 mmol) was added. The reaction mixture
was stirred at 40 °C for 36 h and filtered to remove undissolved
substitutes. The filtrate was evaporated under vacuum, and the
residue was washed with ether, then recrystallized from methanol/
acetone, affording the title compound (328 mg, 68%) as a white-
1
m/z 430 (M - Br)+. H NMR: δ 3.23 (q, J ) 14.4 Hz, 1H), 3.38
(q, J ) 12.4 Hz, 1H), 3.64-3.71 (m, 3H), 3.74 (s, 3H), 3.87 (s,
3H), 4.10-4.24 (m, 1H), 4.19 (m, 2H), 4.36 (d, J ) 16.4 Hz, 1H),
4.48 (d, J ) 16.4 Hz, 1H), 5.34 (q, J ) 5.6 Hz, 1H), 6.08 (d, J )
5.6 Hz, 2H), 6.95 (s, 1H), 7.14 (s, 1H), 7.14-7.25 (m, 4H),
7.45-7.52 (m, 3H). Anal. (C27H28NO4Br) C, H, N.
1
like solid, mp 233-234 °C. MS m/z 354 (M - I)+. H NMR: δ
2.93 (q, J ) 12.0 Hz, 1H), 3.08 (q, J ) 9.2 Hz, 1H), 3.14-3.18
(m, 2H), 3.63-3.68 (m, 1H), 3.77 (s, 3H), 3.80 (s, 3H), 3.83 (s,
3H), 3.87-3.94 (m, 1H), 4.72 (d, J ) 16.4 Hz, 1H), 4.84 (d, J )
16.4 Hz, 1H), 5.00 (q, J ) 4.8 Hz, 1H), 6.05 (d, J ) 5.6 Hz, 2H),
6.89 (s, 1H), 6.97 (d, J ) 8.8 Hz, 1H), 7.09 (s, 1H), 7.15 (d, J )
8.8 Hz, 1H). Anal. (C21H24NO4I) C, H, N.
2,3-Methylenedioxy-10,11-dimethoxy-7-methyltetrahydropro-
toberberine Iodide (8h). The title compound was obtained from
7b and methyl iodide with a procedure similar to that for compound
8a. Yield: 40%. White-like solid, mp 197-198 °C. MS m/z 354
1
(M - I)+. H NMR: δ 2.90 (q, J ) 12.8 Hz, 1H), 3.00-3.08 (m,
2H), 3.38 (q, J ) 5.6 Hz, 1H), 3.63-3.72 (m, 1H), 3.74 (s, 3H),
3.77 (s, 3H), 3.79 (s, 3H), 3.90 (q, J ) 5.6 Hz, 1H), 4.55 (d, J )
12.4 Hz, 1H), 4.59 (d, J ) 12.4 Hz, 1H), 5.02 (q, J ) 5.6 Hz, 1H),
6.02 (s, 2H), 6.78 (s, 1H), 6.83 (s, 1H), 6.92 (s, 1H), 6.93 (s, 1H).
Anal. (C21H24NO4I) C, H, N.
2,3-Methylenedioxy-9,10-dimethoxy-7-ethyltetrahydroproto-
berberine Bromide (8b). The title compound was obtained from
7a and ethyl bromide with a procedure similar to that for compound
8a. Yield: 26%. Yellow-like powder, mp 218-219 °C. MS m/z
1
368 (M - Br)+. H NMR: δ 1.20 (t, J ) 7.2 Hz, 3H), 3.00-3.08
2,3-Methylenedioxy-10,11-dimethoxy-7-ethyltetrahyroproto-
berberine Bromide (8i). The title compound was obtained from
7b and ethyl bromide with a procedure similar to that for compound
8a. Yield: 30%. White-like solid, mp 226-227 °C. MS m/z 368
(m, 4H), 3.68-3.76 (m, 1H), 3.79 (s, 3H), 3.73 (s, 3H), 4.06-4.12
(m, 3H), 4.52 (d, J ) 16.4 Hz, 1H), 4.86 (d, J ) 16.4 Hz, 1H),
5.16 (q, J ) 5.6 Hz, 1H), 6.05 (d, J ) 6.4 Hz, 2H), 6.88 (s, 1H),
7.06 (s, 1H), 7.09 (d, J ) 8.4 Hz, 1H), 7.15 (d, J ) 8.4 Hz, 1H).
Anal. (C22H26NO4Br) C, H, N.
1
(M - Br)+. H NMR: δ 1.34 (t, J ) 7.2 Hz, 3H), 2.97-3.15 (m,
2H), 3.33-3.36 (m, 1H), 3.44-3.48 (m, 1H), 3.60-3.64 (m, 2H),
3.72 (s, 3H), 3.74 (s, 3H), 3.91-4.00 (m, 2H), 4.59 (d, J ) 13.2
Hz, 1H), 4.71 (d, J ) 13.2 Hz, 1H), 5.20 (q, J ) 5.6 Hz, 1H), 6.05
(d, J ) 6.8 Hz, 2H), 6.82 (s, 1H), 6.91 (s, 1H), 6.95 (s, 1H), 7.02
(s, 1H). Anal. (C22H26NO4Br) C, H, N.
2,3-Methylenedioxy-9,10-dimethoxy-7-ꢀ-ethanoltetrahydro-
protoberberine Bromide (8c). The title compound was obtained
from 7a and ethanol bromide with a procedure similar to that for
compound 8a. Yield: 29%. Yellow-green solid, mp 238-239 °C.
MS m/z 384 (M - Br)+. 1H NMR: δ 3.03-3.18 (m, 5H), 3.70 (m,

500 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2
Li et al.
2,3-Methylenedioxy-10,11-dimethoxy-7-n-propyltetrahydro-
protoberberine Iodide (8j). The title compound was obtained from
7b and propyl iodide with a procedure similar to that for compound
8a. Yield: 28%. White-like solid, mp 235-236 °C. MS m/z 382
of Medical Sciences. Animals were housed in an air-conditioned
room with 3-5 rats/mice per cage, and with a 12 h light and 12 h
dark cycle. After 1 week of accommodation period, all of the
animals were fed with HFHC diet containing 2% cholesterol, 10%
yolk powder, 15% lard, and 0.2% of sodium cholate for 8 weeks.
Five of the rats and mice were fed with regular rodent chow serving
as normal diet controls. The HFHC as well as the regular rodent
chow was from Institute of Laboratory Animal Science (Beijing,
China).
Hyperlipidemic animals (rats or mice) were untreated (n ) 6)
or treated with 100 (mg/kg)/day of 1 (n ) 8) or 2 (n ) 8) orally
for 30 days for rat experiment or 21 days for mouse experiment.
Study compounds were suspended in saline (1 mL/rat, 0.2 mL/
mice) prior to use. Compound 1 or 2 was orally administered to
the animals using gavage twice a day at 8 a.m. and 5 p.m.,
respectively, and simvastatin was given once a day at 5 p.m. Before,
after, and during the treatment, blood samples were taken by tail
snip after fast. CHO, LDL-c, HDL-c, alanine aminotransferase
(ALT), aspartate aminotransferase (AST), blood urea nitrogen
(BUN), and creatinine (CRE) levels were assayed using com-
mercially available kits. In another set of experiments, in order to
observe the dose-dependent effect of 2, the hyperlipidemic rats were
untreated (n ) 6) or treated with 2 at 100 (mg/kg)/day (n ) 8) or
300 (mg/kg)/day (n ) 8), respectively, for 17 days. Before and
after treatment, blood lipid levels were assayed.
1
(M - Br)+. H NMR: δ 0.76 (t, J ) 7.2 Hz, 3H), 1.66-1.72 (m,
1H), 1.74-1.80 (m, 1H), 2.84-2.94 (m, 2H), 3.05-3.10 (m, 3H),
3.70-3.74 (m, 1H), 3.76 (s, 3H), 3.77 (s, 3H), 3.94-4.00 (m, 2H),
4.58 (d, J ) 15.6 Hz, 1H), 4.71 (d, J ) 15.6 Hz, 1H), 5.17 (q, J )
5.6 Hz, 1H), 6.05 (d, J ) 6.0 Hz, 2H), 6.88 (s, 1H), 6.89 (s, 1H),
6.92 (s, 1H), 7.02 (s, 1H). Anal. (C23H28NO4I) C, H, N.
2,3-Methylenedioxy-10,11-dimethoxy-7-ethylacetatetetrahy-
droprotoberberine Bromide (8k). The title compound was
obtained from 7b and ethyl acetate bromide with a procedure similar
to that for compound 8a. Yield: 60%. White solid, mp 140-141
°C. MS m/z 426 (M - Br)+. 1H NMR: δ 1.23 (t, J ) 7.2 Hz, 3H),
3.07-3.26 (m, 2H), 3.44 (q, J ) 5.6 Hz, 1H), 3.74 (s, 6H), 3.82
(s, 2H), 3.96 (q, J ) 7.2 Hz, 1H), 4.17 (m, 2H), 4.42 (s, 2H), 4.80
(d, J ) 15.6 Hz, 1H), 4.96 (d, J ) 15.6 Hz, 1H), 5.04 (q, J ) 5.6
Hz, 1H), 6.05 (d, J ) 8.0 Hz, 2H), 6.88 (s, 1H), 6.92 (s, 1H), 6.94
(s, 1H), 7.03 (s, 1H). Anal. (C24H28NO6Br) C, H, N.
2,3-Methylenedioxy-10,11-dimethoxy-7-benzyltetrahydropro-
toberberine Bromide (8l). The title compound was obtained from
7b and benzyl bromide with a procedure similar to that for
compound 8a. Yield: 50%. White solid, mp 178-179 °C. MS m/z
1
430 (M - Br)+. H NMR: δ 3.19 (m, 1H), 3.30-3.36 (m, 2H),
At the end of the treatment course, animals were sacrificed; their
livers were dissected and quickly frozen at -80 °C. Liver tissues
were homogenized with the Ultraspec RNA lysis solution, and total
RNAs were isolated. After reversely transcribed into cDNAs, the
liver mRNA levels of LDLR and HMG-CoA reductase were
analyzed with real-time RT- PCR.
Acute Toxicity. Male Kunming mice (19 ( 1 g) were purchased
from the Institute of Laboratory Animal Science. They were fed
with regular rodent chow and housed in an air-conditioned room.
Then they were randomly divided into four groups with six mice
each, and received saline (control) or 2 (at 250, 1000, or 5000 mg/
kg). 2 was given orally in a single-dosing. After treatment, animals
were weighed every day during the experiment. Fourteen days later,
animals were sacrificed; their livers, kidneys, hearts, lungs, brains,
and colons were fixed in 10% formaldehyde at room temperature
for histological examination using hematoxylin and eosin (H&E)
staining.
3.53 (q, J ) 5.6 Hz, 1H), 3.56-3.66 (m, 2H), 3.74 (s, 3H), 3.78
(s, 3H), 4.06-4.22 (m, 3H), 4.53 (d, J ) 15.6 Hz, 1H), 5.38 (q, J
) 5.6 Hz, 1H), 6.08 (d, J ) 3.6 Hz, 2H), 6.83 (s, 1H), 6.96 (s,
1H), 7.01 (s, 1H), 7.09 (s, 1H), 7.21 (s, 1H), 7.23 (s, 1H), 7.46-7.55
(m, 3H). Anal. (C27H28NO4Br) C, H, N.
Biological Methods. Cell Culture. HepG2 and Bel-7402 cells
were maintained in RPMI-1640 medium containing 10% fetal
bovine serum (FBS) and 1% antibiotics at 37 °C with 5% CO2. At
24 h before drug treatment, cells were trypsinized and grown in
RPMI-1640 containing 10% lipoprotein-deficient serum (LPDS,
Sigma). When cells reached about 80% confluence, media were
switched to RPMI-1640 containing 0.5% LPDS and supplemented
with 1 (Sigma) or 2 or simvastatin (Merck, Rahway, NJ) at the
concentrations indicated. Cells were treated for 12 h before harvest
for RNA extraction.
RNA Extraction and Real-Time RT-PCR. Total RNAs were
isolated from cells or animal livers using the Ultraspec RNA lysis
solution (Biotecxs Laboratory, Houston, TX) according to the
supplier’s protocol. For reverse transcription, 1 µg of total cellular
or liver RNA from each sample was used as template in a 20 µL
reaction system (Promega, Madison, WI). The reverse transcription
reactions were conducted at 42 °C for 30 min and then inactivated
at 95 °C for 5 min. Real-time RT-PCR was performed with these
cDNAs using the Applied Biosystems 7500 real-time RT-PCR
system and the 2 × TaqMan Universal PCR Master Mix (Applied
Biosystems, Foster City, CA). In each of the 20 µL real-time RT-
PCR reaction, 2 µL of cDNA, 10 µL of the Universal PCR Master
Mix, and 1 µL of the 20 × TaqMan gene expression assay reagent
were used. Each experiment was repeated at least 3 times. The
comparative threshold cycle (CT) method was used in relative gene
expression quantification with the GAPDH as the endogenous
control.
Statistical Analysis. The in vitro gene expression data were
analyzed by the Student’s t-test. For animal experiments, after
validation of the test for homogeneity of variance, results were
examined by one-way ANOVA followed by the Newman-Keuls
test for multiple comparisons. P < 0.05 was considered as
statistically significant.
Acknowledgment. This work was supported by the IMB
Research Foundation.
Supporting Information Available: Element analysis data of
the studied analogues. This material is available free of charge via
the Internet at http://pubs.acs.org.
References
Flow Cytometry. Cells were detached and fixed in 2% paraform-
aldehyde at 4 °C overnight. After blocking, cells were incubated
on ice for 1 h with a rabbit-polyclonal antibody against LDLR
(Santa Cruz Biotechnology, Inc., Santa Cruz, CA) with a dilution
of 1:25. Normal rabbit IgG was used as a control for background
staining. Then cells were washed and stained on ice for 30 min
with a FITC-conjugated goat-antirabbit IgG (Santa Cruz) with a
dilution of 1:50. After washing, the fluorescent intensities on the
cell surface were analyzed in a FACSCalibur system (BD Bio-
sciences, San Jose, CA).
Animal Experiments. Male Sprague-Dawley rats (180 ( 10 g)
and C57BL/6J mice (21 ( 2 g) were obtained from the Institute of
Laboratory Animal Science (Beijing, China). Animals were cared
according to the institutional guidelines of the Chinese Academy
(1) Stamler, J.; Daviglus, M. L.; Garside, D. B.; Dyer, A. R.; Greenland,
P.; Neaton, J. D. Relationship of baseline serum cholesterol levels in
3 large cohorts of younger men to long-term coronary, cardiovascular,
and all-cause mortality and to longevity. J. Am. Med. Assoc. 2000,
284, 311–318.
(2) Anderson, K. M.; Castelli, W. P.; Levy, D. Cholesterol and mortality:
30 years of follow-up from the Framingham study. J. Am. Med. Assoc.
1987, 257, 2176–2180.
(3) Grundy, S. M.; Cleeman, J. I.; Merz, C. N., Jr.; Clark, L. T.;
Hunninghake, D. B.; Pasternak, R. C., Jr.; Stone, N. J. Implications
of recent clinical trials for the National Cholesterol Education Program
Adult Treatment Panel III Guidelines. Circulation 2004, 110, 227–
239.
(4) Kong, W. J.; Liu, J.; Jiang, J. D. Human low-density lipoprotein
receptor gene and its regulation. J. Mol. Med. 2006, 84, 29–36.

Berberine Analogues
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 2 501
(5) Kong, W. J.; Abidi, P.; Lin, M.; Inaba, S.; Li, C.; Wang, Y.; Wang,
Z.; Si, S.; Pan, H.; Wang, S.; Wu, J.; Wang, Y.; Li, Z.; Liu, J.; Jiang,
J. D. Berberine is a novel cholesterol-lowering drug working through
a unique mechanism distinct from statins. Nat. Med. 2004, 10, 1344–
1351.
(6) Li, B. X.; Zhang, M. S.; Bao, L. H. Study on the pharmacokinetics of
berberine after oral administration in human being. J. Harbin Med.
UniV. 1995, 29, 382–384.
(7) Zhao, W.; Xue, R.; Zhou, Z. X.; Kong, W. J.; Jiang, J. D. Reduction
of blood lipid by berberine in hyperlipidemic patients with chronic
hepatitis or liver cirrhosis. Biomed. Pharmacother. 2008, 62, 730–
731.
(8) Kong, W. J.; Wei, J.; Zuo, Z. Y.; Wang, Y. M.; Song, D. Q.; You,
X. F.; Zhao, L. X.; Pan, H. N.; Jiang, J. D. Combination of simvastatin
with berberine improves the lipid-lowering efficacy. Metabolism 2008,
57, 1029–1037.
activity of 3-alkoxyjatrorrhizine derivatives. Planta Med. 2008, 74,
290–292.
(13) Zhang, W. J.; Ou, T. M.; Lu, Y. J.; Huang, Y. Y.; Wu, W. B.; Huang,
Z. S.; Zhou, J. L.; Wong, K. Y.; Gu, L. Q. 9-Substituted berberine
derivatives as G-quadruplex stabilizing ligands in telomeric DNA.
Bioorg. Med. Chem. 2007, 15, 5493–5501.
(14) Iwasa, K.; Kamigauchi, M.; Sugiura, M.; Nanba, H. Antimicrobial
activity of some 13-alkyl substituted protoberberinium salts. Planta
Med. 1997, 63, 196–198.
(15) Ikekawa, T.; Ikeda, Y. Antitumor activity of 13-methylberberrubine
derivatives. J. Pharmacobio-Dyn. 1982, 5, 469–474.
(16) Park, K. D.; Lee, J. H.; Kim, S. H.; Kang, T. H.; Moon, J. S.; Kim,
S. U. Synthesis of 13-(substituted benzyl)berberine and berberrubine
derivatives as antifungal agents. Bioorg. Med. Chem. Lett. 2006, 16,
3913–3916.
(17) Ikekawa, T.; Shimada, F.; Jong-Chol, C.; Uebaba, K. 13-Propylber-
(9) Yang, P.; Song, D. Q.; Li, Y. H.; Kong, W. J.; Wang, Y. X.; Gao,
L. M.; Liu, S. Y.; Cao, R. Q.; Jiang, J. D. Synthesis and structure-
activity relationships of berberine analogues as a novel class of low-
density lipoprotein receptor up-regulators. Bioorg. Med. Chem. Lett.
2008, 18, 4675–4677.
(10) Feng, Z.; Norio, N.; Teruaki, A.; Masao, H. Pharmacokinetics of
berberine and its main metabolites in conventional and pseudo germ-
free rats determined by liquid chromatography ion trap mass spec-
trometry. Drug Metab. Dispos. 2006, 34, 2064–2072.
(11) Pan, J. F.; Yu, C.; Zhu, D. Y.; Zhang, H.; Ren, J. Y. Preparation of
Six Isoquinoline Alkaloid Salt. CN 1314347, 2001.
(12) Wang, L. J.; Ye, X. L.; Li, X. G.; Sun, Q. L.; Yu, G.; Cao, X. G.;
Liang, Y. T.; Zhang, H. S.; Zhou, J. Z. Synthesis and antimicrobial
berine Salts. EP 184118 A1, 1986.
(18) Wang, R. F.; Xu, G. Y.; Hua, W. Y.; Peng, S. X. Synthesis and anti-
arrhythmic activity of some tetrahydroberberine quaternary ammonium
compounds. Chin. J. Med. Chem. 1996, 4, 235–248.
(19) Qin, Y.; Pang, J. Y.; Chen, W. H.; Cai, Z.; Jiang, Z. H. Synthesis,
DNA-binding affinities, and binding mode of berberine dimers. Bioorg.
Med. Chem. Lett. 2006, 14, 25–32.
(20) Abidi, P.; Chen, W.; Kraemer, F. B.; Li, H.; Liu, J. W. Identification
of medicinal plant goldenseal as a natural cholesterol-lowering agent:
mechanisms of actions and new modulators of LDL receptor expres-
sion. J. Lipid Res. 2006, 47, 2134–2147.
JM801157Z