Structure–activity relationship and biological evaluation of berberine derivatives as PCSK9 down-regulating agents

Article abstract of DOI:10.1016/j.bioorg.2021.104994

Proprotein convertase subtilisin/kexin type 9 (PCSK9) is a secreted protein and its deficiency markedly enhanced the survival rate of patient with cardiovascular diseases (CVDs). Forty berberine (BBR) derivatives were synthesized and evaluated for their activities on down-regulating the transcription of PCSK9 in HepG2 cells, taking BBR as the lead. Structure–activity relationship (SAR) analysis revealed that 2,3-dimethoxy moiety might be beneficial for activity. Among them, 9k displayed the most potent activity with IC₅₀ value of 9.5 ± 0.5 μM, better than that of BBR. Also, it significantly decreased PCSK9 protein level at cellular level, as well as in the liver and serum of mice in vivo. Furthermore, 9k markedly increased LDLR expression and LDL-C clearance via down-regulating PCSK9 protein. The mechanism of action of 9k is targeting HNF1α and/or Sp1 cluster modulation upstream of PCSK9, a different one from BBR. Therefore, 9k might have the potential to be a novel PCSK9 transcriptional inhibitor for the treatment of atherosclerosis, worthy for further investigation.

Full text of DOI:10.1016/j.bioorg.2021.104994

Bioorganic Chemistry 113 (2021) 104994
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Structure–activity relationship and biological evaluation of berberine
derivatives as PCSK9 down-regulating agents
Tian-Yun Fan1, Yu-Xin Yang1, Qing-Xuan Zeng, Xue-Lei Wang, Wei Wei, Xi-Xi Guo,
Li-Ping Zhao, Dan-Qing Song, Yan-Xiang Wang*, Li Wang*, Bin Hong*
Institute of Medicinal Biotechnology, Chinese Academy of Medical Science & Peking Union Medical College, Beijing 10050, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Proprotein convertase subtilisin/kexin type 9 (PCSK9) is a secreted protein and its deficiency markedly enhanced
the survival rate of patient with cardiovascular diseases (CVDs). Forty berberine (BBR) derivatives were syn-
thesized and evaluated for their activities on down-regulating the transcription of PCSK9 in HepG2 cells, taking
BBR as the lead. Structure–activity relationship (SAR) analysis revealed that 2,3-dimethoxy moiety might be
PCSK9
Berberine
Structure–activity relationship
Transcriptional inhibition
LDL-C
beneficial for activity. Among them, 9k displayed the most potent activity with IC50 value of 9.5 ± 0.5 μM, better
than that of BBR. Also, it significantly decreased PCSK9 protein level at cellular level, as well as in the liver and
serum of mice in vivo. Furthermore, 9k markedly increased LDLR expression and LDL-C clearance via down-
regulating PCSK9 protein. The mechanism of action of 9k is targeting HNF1α and/or Sp1 cluster modulation
upstream of PCSK9, a different one from BBR. Therefore, 9k might have the potential to be a novel PCSK9
transcriptional inhibitor for the treatment of atherosclerosis, worthy for further investigation.
1. Introduction
development of small-molecule PCSK9 inhibitors with lower cost and
ease of administration are conceptually attractive [12–14], and shed a
Cardiovascular diseases (CVDs) are still a major cause of death
worldwide [1]. Hepatic low-density lipoprotein receptor (LDLR) is a
major target responsible for the removal of circulating low-density li-
poprotein cholesterol (LDL-C), which is closely associated with the
development of atherosclerosis [2,3]. Proprotein convertase subtilisin/
kexin type 9 (PCSK9) is a secreted plasma protease that reduce hepatic
LDLR protein level and induce increased plasma LDL-C levels by binding
and shuttling LDLR to lysosomes for degradation. Clinically, deficiency
of PCSK9 significantly improves the survival rate of patient with CVDs
[4–6]. Thus, development of agents targeting PCSK9 expression or ac-
tivity inhibition shows benefit in reducing the risk of cardiovascular
disease. Two PCSK9-blocking monoclonal antibodies (alirocumab and
evolocumab) have been successfully used in clinic alone or in combi-
nation with statin [7–10]. Besides, inclisiran, a novel siRNA-based
PCSK9 silencer, has just been approved by the European Medicines
Agency (EMA), which effectively sustains low density lipoprotein
cholesterol (LDL-C) reduction by ~40% among patients with high risk of
CVDs [11]. Since both the high cost of PCSK9 monoclonal antibodies
and the subcutaneous administration limit their widespread use,
light on the novel anti-atherosclerosis drug discovery.
However, the shallow interaction surface between PCSK9 and low
density lipoprotein receptor (LDLR) makes it difficult to be targeted
directly by small-molecule inhibitors [15]. Therefore, we have suc-
cessfully established a cell-based high-throughput screening (HTS)
model for human PCSK9 transcriptional down-regulators and identified
an effective small-molecule PCSK9 inhibitor to profoundly reduce
atherosclerosis progression in vivo [16].
In the past few years, our group has been dedicated to the discovery
of innovate candidates from Traditional Chinese Medicine. The natural
alkaloid berberine (BBR, Fig. 1) could decrease PCSK9 expression in a
time- and dose-dependent manner through transcription factor HNF1
α
and SREBP, followed by increasing expression of LDLR as well as
clearance of LDL-C [17–21]. Based on this, taking BBR as the lead, we
carried out structural modification and optimization and constructed a
set of BBR analogues to elucidate the comprehensive structure–activity
relationship (SAR) for PCSK9 expression (Fig. 1) and discover novel
modulators targeting PCSK9 transcriptional expression.
All the target compounds were evaluated by luciferase reporter-
* Corresponding authors.
E-mail addresses: wangyanxiang@imb.pumc.edu.cn (Y.-X. Wang), wangli_imb@163.com (L. Wang), binhong69@hotmail.com (B. Hong).
1
These authors made equal contributions to the work.
https://doi.org/10.1016/j.bioorg.2021.104994
Received 15 February 2021; Received in revised form 19 April 2021; Accepted 12 May 2021
Available online 18 May 2021
0045-2068/© 2021 Elsevier Inc. All rights reserved.

T.-Y. Fan et al.
Bioorganic Chemistry 113 (2021) 104994
based HTS assay, the representative compound was further confirmed
on the regulation of hepatic PCSK9 in vitro and in vivo, followed by its
effect on LDLR protein expression and mediated LDL-C clearance in
cultured hepatic cells. Additionally, preliminary mechanism of repre-
sentative compound was elucidated in this study.
benzyl were attached, the PCSK9 transcriptional inhibition rates of 9e–g
went up to 83–98%, much better than that of BBR [16]. Instead of alkyl
substituent, methoxy, trifloromethoxy and trifluomethylthio (9h–k)
were attached on benzyl respectively. Among them, compound 9h with
one methoxy at meta-position of benzyl group gave a declined potency
with an inhibitory rate of 28%. While 3′,5′-dimethoxy substituent was
introduced, the activity on 9i was lost completely. Compounds 9j and 9k
displayed potent inhibitory activities with inhibition rates of 88% and
98%, which indicated that fluorine-containing substituent might be
beneficial for the improvement on potency. Besides, various halogens,
and electron withdrawing groups such as nitro, cyano and ester group
were employed on benzyl, by which compounds 9l–q were created.
However, most of them lost their downregulatory potencies on PCSK9.
Finally, derivatives with various saturated rings on position 9 were
synthesized and assessed. Both compounds 9r and 9u with six-member
ring could obviously down-regulate PCSK9 expression. These results
suggested that lipophilic substituent might be helpful for PCSK9 down-
regulating activity.
2. Results and discussion
2.1. Synthesis
The synthetic routes of all the target compounds were presented in
Schemes 1 and 2, taking commercially available substituted benzalde-
hydes or palmatine (PMT) as the starting materials. Among all target
products, compounds 7a–s, 9a, 9b, 9f–g, 9i and 9q have been previously
reported in our lab [22–27].
As presented in Scheme 2, demethylation on position 9 was first
conducted under the influence of quaternary ammonium cation on PMT
skeleton [25–28]. PMT was first heated at 195–210 ◦C for 30 min under
vacuum and then acidified in concentrated HCl/ethanol (5/95 by vol) to
obtain the key intermediate 8 with a free hydroxyl in a 78% yield after
purification. Afterwards, taking K2CO3 as the base and CH3CN as the
solvent, compound 8 was reacted with substituted benzyl halide or
halogenated hydrocarbons in the presence or absence of sodium iodide
(NaI) as the catalyst. The mixture was stirred at 71 ◦C for 4–10 h, and
target products 9a–u were acquired in yields of 43–68%. All of the final
compounds were purified by flash column chromatography on silica gel
with using CH2Cl2/CH3OH as eluent.
Meanwhile, when the dimethoxy moiety was moved to positions 3
and 4, the activity of 7h and 7i descended apparently. Furthermore,
larger volume substituents including ethoxy and benzyloxy groups, as
well as smaller volume hydroxyl were respectively attached to elucidate
the SAR on ring A. Still, weakened potencies were detected on com-
pounds 7j–q with inhibition rates of –2 to 40%. Finally, compounds 7r
and 7s with trimethoxy substituted on ring A displayed declined activ-
ity. These results suggested that 2,3-dimethoxy might also be beneficial
for PCSK9 down-regulatory activity, taking BBR as the positive control.
Therefore, four compounds 9e, 9k, 9r and 9u with inhibitory rates
above 90% were selected as active compounds for the further confir-
mation of PCSK9 inhibitory effects. The PCSK9p-Luc HepG2 cells were
treated with these active compounds under different concentrations for
24 h respectively, and the luciferase activities were detected. As repre-
sented in Fig. 2A, these four compounds dose-dependently suppressed
the PCSK9 promoter activity (structures shown in Fig. 2C). Especially,
compounds 9e and 9k showed the moderate activity with respective
2.2. Pharmacological evaluation
2.2.1. SAR analysis for PCSK9 transcriptional activity in PCSK9p-Luc
HepG2 cells
All the target derivatives (10 μg/mL) were initially examined for
their PCSK9 transcription inhibitory activities using the luciferase
reporter-based HTS assay in PCSK9p-Luc HepG2 cells established in our
lab [16]. Structures and PCSK9 down-regulatory activities of the 40
analogues are displayed in Table 1.
IC50 value of 7.6 ± 1.1 μM and 9.5 ± 0.5 μM, which were lower than that
of BBR (13.1 ± 0.6 μM) (Fig. 2B). Thus, the two derivatives were chosen
as key compounds for further evaluations.
Firstly, SAR was focused on the D ring, and compounds 7a–e with
various alkoxy and halogens on position 9, 10 or 11 were prepared and
tested. As shown in Table 1, most of them showed suppressed potency
with inhibition rates of no more than 35%, lower than that of BBR
(56%). Then, SAR was moved to ring A, and 7f and 7g with dimethoxy
on positions 2 and 3 instead of methylenedioxy exhibited comparable
PCSK9 down-regulatory activity to BBR, which hinted that 2,3-dime-
thoxy might be helpful for potency.
2.2.2. Preliminary safety evaluations for key compounds 9e and 9k in vitro
and in vivo
The effects of compounds 9e and 9k on cell viability were further
evaluated using MTT assay in HepG2 cells. The cells were treated with
9e or 9k under a series of concentrations for 24 h and the cell viability
was determined. As depicted in Fig. 3, compound 9k displayed a lower
cytotoxicity than 9e with median toxic concentration (TC50) value of
Thus, a set of PMT analogues with 2,3-dimethoxy moiety were
constructed and screened. It has been reported that substituent on po-
sition 9 might help maintain or improve potencies [28], structural
modification at this position was first carried out. Fat rings including
substituted benzyl groups and saturated carbocycles were introduced in
order to improve their lipophilic property and bioavailability. There-
fore, benzyl or various methylated benzyl groups were introduced on
position 9 respectively, and 9b–d displayed comparable potencies
against PCSK9 compared with that of BBR. While large-volume alkyls on
61.1 μM. Thus, compound 9k was selected as a representative com-
pound for next step. Furthermore, we examined the acute toxicity of 9k
on Kunming mice. Compound 9k was given orally in a single dose of 0,
250, 500, or 1,000 mg/kg, respectively. Then the body weight, survival
and behavioral characteristics were monitored for seven days. The mice
behaved well under the dose of up to 1,000 mg/kg, indicating a good in
vivo safety feature of compound 9k with the half-lethal dose (LD50) value
of over 1,000 mg/kg.
Fig. 1. Chemical structure of BBR and modification strategies.
2

T.-Y. Fan et al.
Bioorganic Chemistry 113 (2021) 104994
Scheme 1. Reagents and conditions: (a) CH3NO2, methanol, CH3NH2, 50 ◦C, 3.5 h; (b) NaBH4, methanol, reflux, 2 h; (c) Zn, HCl, ethanol, 3 h; (d) 1) substituted of
benzaldehyde, 100 ◦C, 8 h; 2) NaBH4, methanol, reflux, 5 h; (e) glyoxal, formic acid, CuSO4, HCl, 100 ◦C, 5 h; (f) 1) methanol/H2O, CaO, rt, 2 h; 2) ethanol, HCl, rt,
0.5 h.
Scheme 2. Reagents and conditions: (a) 195–210 ◦C, 30–40 mmHg, 30 min.; (b) Substituted benzyl halide or halogenated hydrocarbons, K2CO3 for 9a–q or K2CO3
and NaI for 9r–u, CH3CN, 71 ◦C, 4–10 h.
2.2.3. Effects of 9k on PCSK9 expression at protein level in vitro and in
vivo
tetramethylindocarbocyanine perchlorate-labeled LDL (DiI-LDL) uptake
was conducted using flow cytometry analysis. HepG2 cells were treated
with vehicle or different concentrations of 9k for 24 h, then incubated
To further validate the inhibitory activity of 9k on PCSK9 expression,
western blotting assay was applied to evaluate the effects on PCSK9
protein level in HepG2 cells firstly. The cells were treated with different
concentrations of 9k for 24 h, and PCSK9 protein level was detected. As
shown in Fig. 4A, compound 9k significantly decreased PCSK9 protein
with DiI-LDL (2
μ
g/mL) at 37 ◦C for 4 h before detection by flow
cytometry. The results showed that 9k treatment markedly promoted
DiI-LDL uptake in a dose-dependent manner with a scale of 2–3 folds,
which was much more effective than that of BBR (Fig. 5B, C), implying a
dominant role of 9k in regulating LDL-C metabolism. Besides, fluores-
cence visualization of DiI-LDL uptake by HepG2 cells was viewed by a
Confocal Laser Scanning Microscope, which revealed a similar result
that 9k increased DiI-LDL-induced fluorescence accumulation in HepG2
cells (Fig. 5D). Taken together, these results suggested that 9k is supe-
rior to BBR with better effect on LDL-C clearance.
level even at the concentration of 5 μM. We further investigated the
effect of 9k on hepatic and serum PCSK9 content in vivo. Male C57BL/6J
mice on standard chow diet in 2 groups (10 mice per group) were
administrated with 50 mg/kg intragastrically every day for 4 weeks.
Then the mice were sacrificed with 1% sodium pentobarbital, the liver
and serum samples were collected individually to determine PCSK9
expression by western blotting and ELISA assay respectively. As showed
in Fig. 4B, 9k significantly decreased hepatic and serum PCSK9 protein
level in C57BL/6J mice, which confirmed the effect of 9k on PCSK9
expression in vivo.
2.2.5. 9k transcriptionally reduced PCSK9 expression through HNF1α and
Sp1modulation
To investigate the mechanism involved in PCSK9 reduction mediated
by 9k, we analyzed the effect of 9k on the transcriptional activity of
PCSK9 promoter by transfection of HepG2 cells with the luciferase-
reporter plasmids containing different truncated promoter region
(D1–D7, Fig. 6A) of human PCSK9 gene [16]. The luciferase activity was
detected after the incubation of 9k with the HepG2 cells transfected with
different individual construct. The results revealed that the luciferase
activity of D1–D5 was significantly decreased by 9k, which was mark-
edly eliminated upon removal of the bases between ꢀ 392 bp and ꢀ 351
bp (D5 vs. D6, Fig. 6A). Besides, the inhibition effect of 9k on D7 was
abolished compared with pGL4 (D7 vs. pGL4, Fig. 6A). Taken together,
all these results indicated that the regions between ꢀ 392 bp and ꢀ 351
bp (D5 vs. D6), ꢀ 335 bp and ꢀ 1 bp (D7 vs. pGL4) were both important
for 9k mediated PCSK9 transcriptional reduction, possibly associating
2.2.4. Effects of 9k on LDLR expression and LDL cholesterol uptake in
HepG2 cells
Since PCSK9 could directly bind to LDLR and mediate its lysosomal
degradation in hepatic cells, agents which lower PCSK9 protein level
may also promote LDLR expression and increase LDL-C clearance
[29,30]. Therefore, the effect of compound 9k on LDLR expression was
conducted in HepG2 cells. Compound 9k with indicated concentrations
was incubated with HepG2 cells for 24 h before LDLR protein level was
detected by western blotting assay. As shown in Fig. 5A, 9k significantly
up-regulated LDLR protein levels in a dose-dependent manner, sug-
gesting a potential role of 9k in the regulation of LDLR mediated LDL-C
uptake.
To further confirm whether the reduction of PCSK9 expression and
the following LDLR increasement induced by 9k could promote LDL-C
with transcriptional factors HNF1
which is closely related to BBR [31]. There are two factors involved in
the regions between ꢀ 392 bp and ꢀ 351 bp including HNF1 and HINFP.
α, HINFP or Sp1, rather than SREBP
clearance
in
HepG2
cells,
1,1′-dioctadecyl-3,3,3′,3′-
α
3

T.-Y. Fan et al.
Bioorganic Chemistry 113 (2021) 104994
Table 1
The structures and PCSK9 transcriptional activities of all target compounds (10
μ
g/mL) in HepG2 cells.
R5
No.
R1
R2
R3
R4
R6
R7
Inhibition Rate (%)
BBR
7a
7b
7c
7d
7e
7f
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
OCH3
H
H
H
H
H
H
H
H
H
OCH2O
OCH2O
OCH2O
OCH2O
OCH2O
OCH2O
OCH3
OCH3
H
H
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OH
H
56
34
34
10
4
H
H
OCH3
OC2H5
H
H
H
H
Cl
H
Cl
OCH3
H
H
F
ꢀ 9
60
59
24
8
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OC2H5
OCH3
OH
H
H
OCH3
H
7g
7h
7i
H
H
OCH3
OCH3
H
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
H
H
H
OCH3
H
7j
OC2H5
OC2H5
OCH3
C6H5CH2O
H
OCH3
19
5
7k
7l
H
H
OCH3
H
H
OCH3
9
7m
7n
7o
7p
7q
7r
H
OCH3
H
16
32
ꢀ 2
40
36
13
22
38
ꢀ 11
55
67
87
98
83
86
H
OCH3
H
H
OH
H
H
OCH3
OCH3
OCH3
H
OCH3
OH
OH
H
H
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
H
H
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
H
OCH3
7s
OCH3
H
OCH3
H
PMT
9a
9b
9c
9d
9e
9f
OCH3
H
H
C6H5CH2O
H
H
p-CH3C6H4CH2O
o-CH3C6H4CH2O
3’,5’-(CH3)2C6H3CH2O
p-CH3(CH2)3C6H4CH2O
p-C(CH3)3C6H4CH2O
H
H
H
H
H
H
H
H
H
9g
H
H
9h
9i
H
H
H
H
H
H
H
H
H
H
H
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
H
H
H
H
H
H
H
H
H
H
H
m-CH3OC6H4CH2O
3′,4′-(CH3O)2C6H3CH2O
p-CF3OC6H4CH2O
p-SCF3C6H4CH2O
2′,4′-F2C6H3CH2O
3′,5′-F2C6H3CH2O
p-CH3OCOC6H4CH2O
p-CNC6H4CH2O
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
H
H
H
H
H
H
H
H
H
H
H
28
ꢀ 9
88
9j
9k
9l
98
26
9m
9n
9o
9p
9q
9r
37
ꢀ 43
ꢀ 34
ꢀ 5
19
m-NO2C6H4CH2O
o-NO2C6H4CH2O
96
9s
H
OCH3
OCH3
H
OCH3
H
68
9t
H
H
OCH3
OCH3
OCH3
OCH3
H
H
OCH3
OCH3
H
H
8
9u
98
Thus, to further identify which factor plays a key role in 9k mediated
PCSK9 modulation, respective binding site in PCSK9 promoter was
mutated. As shown in Fig. 6B, the inhibitory effect of 9k on D5 was
improve PCSK9 down-regulatory activity. Among them, 9k displayed
the most potent activity on expression of PCSK9 with IC50 value of 9.5 ±
0.5 μM, and exhibited good in vivo safety characters with LD50 value of
significantly abolished with missing HNF1
α
binding site, but not HINFP,
instead of
HINFP, which was similar to BBR [31]. Meanwhile, beside HNF1 , Sp1
over 1,000 mg/kg. Notably, 9k significantly decreased PCSK9 protein
level in vitro and in vivo. Most importantly, 9k markedly increased LDL-
C uptake in a dose-dependent manner through key PCSK9 protein.
Preliminary mechanism research indicated that 9k targeted transcrip-
implying that 9k regulated PCSK9 mainly through HNF1
α
α
cluster might also take a part in 9k induced PCSK9 reduction. In sum-
mary, 9k might exert its effect on PCSK9 expression through HNF1
α
tion factor HNF1α and/or Sp1 cluster modulation to reduce PCSK9
and/or Sp1 cluster, not exactly same as BBR.
expression, thereby enhance LDLR expression and LDL-C clearance, a
different mechanism from BBR, suggesting a change in the mechanism of
action after structural modifications. Therefore, 9k might have the po-
tential to be a novel PCSK9 down-regulatory agent for the treatment of
cardiovascular diseases, worthy for further investigations.
3. Conclusions
Taken together, 40 BBR derivatives, of which 15 were new, were
synthesized and screened for their activities on reduction of PCSK9
transcription in HepG2 cells. SAR study indicated that 2,3-dimethoxy
moiety and introduction of a suitable substituent at position 9 might
4

T.-Y. Fan et al.
Bioorganic Chemistry 113 (2021) 104994
Fig. 2. Identification of active compounds
through PCSK9 transcriptional inhibitor HTS
assay and structures of representative compounds
9e, 9k, 9r, 9u and BBR. (A) Structures of com-
pounds 9e, 9k, 9r, 9u (BBR: See Fig. 1.) (B) The
dose–effect curves were drawn and the respective
IC50 was calculated. Representative dose–effect
curves are shown. Each concentration was repli-
cated three times. (C) IC50 values of 9e, 9k and
BBR are shown with bars as mean ± SEM from
three independent experiments. *p < 0.05.
Fig. 3. Cell toxicity of active compounds 9e and 9k was evaluated by MTT assay in HepG2 cells. The dose–effect curves were drawn and the respective TC50 was
calculated. Each concentration was replicated three times.
4. Experimental section
performed on CombiflashRf 200 (Teledyne, Nebraska, USA), silica gel
particle size 0.038 mm.
4.1. General
4.2. Chemistry
Melting point (mp) was uncorrected using an MPA 100 OptiMelt
automated melting point system (Stanford Research Systems, California,
USA). The IR spectra were tested by Frontier FT-IR/NIR spectrometer
using Spectrum 10 software (PerkinElmer Frontier, Massachusetts,
USA). The 1H and 13C NMR spectra were performed on a Bruker Avance
600 MHz spectrometer (AV600-III, Bruker, Swiss) in DMSO‑d6 with
Me4Si as the internal standard. ESI high-resolution mass spectra (HRMS-
ESI) were recorded on an Autospec UItima-TOF mass spectrometer
(Micromass UK Ltd, Manchester, UK). Flash chromatography was
4.2.1. Synthesis procedure for compounds 7a–s, 9a, 9b, 9f–g, 9i and 9q
The synthetic methods and structural identification of compounds
7a–s, 9a, 9b, 9f–g, 9i and 9q have been previously reported in our lab
[19–23].
4.2.2. General synthesis procedure for compounds 9c–e, 9h, 9j–p, 9r–u
PMT (3.87 g, 10 mmol) was heated at 195–210 ◦C for 30 min under
vacuum (20–30 mmHg), and then a dark purple solid was obtained.
5

T.-Y. Fan et al.
Bioorganic Chemistry 113 (2021) 104994
Fig. 4. Effects of 9k on PCSK9 expression in
HepG2 cells and C57BL/6J mice. (A) West-
ern blotting assay for 9k-regulated PCSK9
protein level. Representative images are
shown. Gray scanning analysis was per-
formed from 3 independent experiments. (B)
Male C57BL/6 mice were intragastrically
administrated with vehicle (veh) or 9k (50
mg/kg). At the end of experiment, liver
samples and serum were collected and
PCSK9 protein level was detected. Data are
shown as mean ± SEM. *p < 0.05, **p <
0.01, ***p < 0.001 vs. vehicle group.
After the solid was cooled to room temperature, ethanol/concentrated
HCl (95/5) were added and the yellow solution was acquired. Then,
after the solvent was removed, the remaining residue was purified by
flash chromatography while using CH2Cl2/CH3OH as the gradient eluent
to obtain the intermediate 8 (3.43 g, 80%) as a brown solid.
1602, 1509, 1357, 1275, 1022, 873 cmꢀ 1; 1H NMR: δ 9.69 (s, 1H), 9.00
(s, 1H), 8.22 (d, J = 9.0 Hz, 1H), 8.02 (d, J = 9.0 Hz, 1H), 7.70 (s, 1H),
7.47 (d, J = 7.8 Hz, 2H), 7.20 (d, J = 7.8 Hz, 2H), 7.10 (s, 1H), 5.31 (s,
2H), 4.91 (t, J = 6.6 Hz, 2H), 4.10 (s, 3H), 3.93 (s, 3H), 3.87 (s, 3H), 3.21
(t, J = 6.6 Hz, 2H), 2.56 (t, J = 7.8 Hz, 2H), 1.54–1.45 (m, 2H),
1.32–1.10 (m, 2H), 0.83 (t, J = 7.2 Hz, 3H); 13C NMR: δ 151.5, 150.7,
148.7, 145.4, 142.7, 141.9, 137.5, 133.6, 133.0, 129.0 (2), 128.6, 128.2
(2), 126.6, 123.6, 121.9, 119.8, 118.9, 111.3, 108.7, 75.2, 57.0, 56.1,
55.9, 55.5, 34.5, 33.1, 26.0, 21.6, 13.7; HRMS: calcd for C31H34NO4Br
[Mꢀ Br]+: 484.2482, found: 484.2483.
Compound 8 (100 mg, 0.40 mmol), K2CO3 (1.60 mmol) with or
without NaI (1.60 mmol) were added in anhydrous DMF (6 mL), and the
mixture was stirred for 10 min at room temperature. Afterwards, cor-
responding benzyl halides (1.60 mmol) or halogenated hydrocarbons
(1.60 mmol) were added, and the mixture was heated at 71 ◦C for 4–10
h, and then filtered. The resulting residue was acidified with concen-
trated HCl/ethanol (5:95 by vol.), purified through flash chromatog-
raphy on silica gel using CH2Cl2/CH3OH as the gradient eluent to
acquire the desired compounds 9c–e, 9h, 9j–p, 9r–u.
4.2.2.4. 2,3,10-Trimethoxy-9-m-methoxylbenzyloxyprotopalmatine bro-
mide (9h). Faint yellow solid; yield: 33%; mp: 179–181 ◦C; IR: 3451,
2960, 1608, 1511, 1241, 1110, 870, 779 cmꢀ 1; 1H NMR: δ 9.77 (s, 1H),
9.05 (s, 1H), 8.23 (d, J = 9.0 Hz, 1H), 8.05 (d, J = 9.0 Hz, 1H), 7.72 (s,
1H), 7.30 (t, J = 7.8 Hz, 1H), 7.19 (s, 1H), 7.13 (d, J = 7.8 Hz, 1H), 7.10
(s, 1H), 6.91 (dd, J = 8.4, 2.4 Hz, 1H), 5.34 (s, 2H), 4.94 (t, J = 6.6 Hz,
2H), 4.10 (s, 3H), 3.94 (s, 3H), 3.88 (s, 3H), 3.77 (s, 3H), 3.22 (t, J = 6.6
Hz, 2H); 13C NMR: δ 159.3, 151.5, 150.5, 148.7, 145.3, 141.9, 137.9,
137.6, 133.0, 129.4, 128.6, 126.6, 123.6, 121.8, 120.8, 119.9, 118.8,
114.1, 114.0, 111.3, 108.8, 75.2, 57.0, 56.2, 55.6, 55.5, 55.1, 26.0;
HRMS: calcd for C28H28NO5Br [Mꢀ Br]+: 458.1962, found: 458.1966.
4.2.2.1. 2,3,10-Trimethoxy-9-o-methylbenzyloxyprotopalmatine bromide
(9c). Brown solid; yield: 24%; mp: 164–166 ◦C; IR: 3405, 2920, 1512,
1359, 1272, 1106, 1018, 744 cmꢀ 1; 1H NMR: δ 9.59 (s, 1H), 9.07 (s, 1H),
8.24 (dd, J = 9.0, 3.6 Hz, 1H), 8.08 (d, J = 9.0 Hz, 1H), 7.72 (s, 1H), 7.56
(d, J = 9.0 Hz, 1H), 7.49–7.17 (m, 3H), 7.10 (s, 1H), 5.35 (s, 2H), 4.90 (t,
J = 6.6 Hz, 2H), 4.07 (s, 3H), 3.94 (s, 3H), 3.87 (s, 3H), 3.21 (t, J = 6.6
Hz, 2H), 2.45 (s, 3H); 13C NMR δ 151.5, 150.8, 148.7, 145.2, 142.2,
137.6, 136.7, 134.7, 133.1, 130.1, 129.5, 128.6, 128.5, 126.6, 125.8,
123.8, 121.7, 112.0, 118.8, 111.3, 108.8, 73.5, 57.0, 56.2, 55.9, 55.7,
26.0, 18.6; HRMS: calcd for C28H28NO4Br [Mꢀ Br]+: 442.2013, found:
442.2012.
4.2.2.5. 2,3,10-Trimethoxy-9-p-trifluoromethoxybenzyloxyprotopalmatine
bromide (9j). Faint yellow solid; yield: 38%; mp: 187–189 ◦C; IR: 3400,
2982, 2610, 1607, 1359, 1276, 1110, 809 cmꢀ 1; 1H NMR: δ 9.83 (s, 1H),
9.07 (s, 1H), 8.24 (d, J = 9.0 Hz, 1H), 8.07 (d, J = 9.0 Hz, 1H), 7.86 (d, J
= 8.4 Hz, 2H), 7.81 (d, J = 8.4 Hz, 2H), 7.73 (s, 1H), 7.11 (s, 1H), 5.46 (s,
2H), 4.96 (t, J = 6.6 Hz, 2H), 4.08 (s, 3H), 3.94 (s, 3H), 3.87 (s, 3H), 3.23
(t, J = 6.6 Hz, 2H); 13C NMR: δ 151.5, 150.4, 148.7, 145.3, 141.8, 141.4,
137.7, 133.1, 128.9 (2), 128.7, 128.6, 126.6, 125.2 (2), 124.2, 123.8,
121.5, 119.9, 118.9, 111.3, 108.8, 74.4, 57.1, 56.2, 55.9, 55.5, 26.0;
HRMS: calcd for C28H25NO5F3Br [Mꢀ Br]+: 512.1679, found: 512.1685.
4.2.2.2. 2,3,10-Trimethoxy-9-3′,5′-dimethylbenzyloxyprotopalmatine bro-
mide (9d). Brown solid; yield: 46%; mp: 191–193 ◦C; IR: 3466, 2973,
1608, 1359, 1276, 1110, 822, 705 cmꢀ 1; 1H NMR: δ 9.72 (s, 1H), 9.06 (s,
1H), 8.24 (dd, J = 9.0, 2.4 Hz, 1H), 8.06 (d, J = 9.0 Hz, 1H), 7.72 (s, 1H),
7.19 (d, J = 10.8 Hz, 2H), 7.10 (s, 1H), 6.99 (s, 1H), 5.26 (s, 2H), 4.94 (t,
J = 6.6 Hz, 2H), 4.10 (d, J = 1.8 Hz, 3H), 3.94 (d, J = 1.2 Hz, 3H), 3.87
(d, J = 1.8 Hz, 3H), 3.22 (t, J = 6.6 Hz, 2H), 2.29 (s, 6H); 13C NMR: δ
151.5, 150.7, 148.7, 145.3, 142.1, 137.6, 137.4 (2), 136.3, 133.0, 130.8,
129.7, 128.6, 126.5 (2), 123.7, 121.8, 119.9, 118.9, 111.3, 108.7, 75.5,
57.0, 56.2, 55.9, 55.5, 26.0, 20.9 (2); HRMS: calcd for C29H30NO4Br
[Mꢀ Br]+: 456.2169, found: 456.2169.
4.2.2.6. 2,3,10-Trimethoxy-9-p-trifluomethylthiobenzyloxyprotopalmatine
bromide (9k). Brown solid; yield: 47%; mp: 182–184 ◦C; IR: 3382, 2980,
1606, 1360, 1275, 1108, 1018, 812 cmꢀ 1; 1H NMR: δ 9.84 (s, 1H), 9.07
(s, 1H), 8.24 (d, J = 9.0 Hz, 1H), 8.07 (d, J = 9.0 Hz, 1H), 7.79 (s, 4H),
7.72 (s, 1H), 7.11 (s, 1H), 5.42 (s, 2H), 4.95 (t, J = 6.6 Hz, 2H), 4.07 (s,
3H), 3.94 (s, 3H), 3.87 (s, 3H), 3.23 (t, J = 6.6 Hz, 2H); 13C NMR: δ
151.5, 150.4, 148.7, 145.3, 141.8, 140.3, 137.7, 136.1 (2), 133.1,
4.2.2.3. 2,3,10-Trimethoxy-9-p-(n-butyl)benzyloxyprotopalmatine
bro-
mide (9e). Yellow solid; yield: 57%; mp: 215–217 ◦C; IR: 3397, 2937,
6

T.-Y. Fan et al.
Bioorganic Chemistry 113 (2021) 104994
Fig. 5. Effects of 9k on LDLR expression and LDL-C uptake in HepG2 cells. (A) Western blotting assay for LDLR protein level regulated by 9k. Gray scanning analysis
was performed from 3 independent experiments. (B) Flow cytometric analysis was conducted to detect LDL-C uptake. Representative images are shown and statistical
analysis is performed from 3 independent experiments. (C) Overlay graph of blank (bln), vehicle (veh), 9k (20 μM) and BBR (20 μM). (D) Fluorescence microscopy
was used to analyze DiI uptake by HepG2 cells. Representative images are shown. Scale bar, 50 μm. The data are shown as mean ± SEM. *p < 0.05, **p < 0.01, ***p
< 0.001 vs. vehicle group.
130.10, 129.6 (3), 126.5, 123.9, 122.7, 121.5, 119.9, 118.8, 111.3,
108.8, 74.3, 57.0, 56.2, 55.8, 55.5, 25.9; HRMS: calcd for
3.94 (s, 3H), 3.87 (s, 3H), 3.22 (t, J = 6.6 Hz, 2H); 13C NMR: δ 162.5,
160.8, 151.5, 150.6, 148.7, 145.1, 141.5, 137.7, 133.1, 133.0, 128.6,
126.6, 123.9, 121.6, 120.0 (2), 118.9, 111.6, 111.3, 108.8, 104.0, 68.5,
57.0, 56.2, 55.9, 55.6, 26.0; HRMS: calcd for C27H24NO4F2Br [Mꢀ Br]+:
464.1668, found: 464.1665.
C
28H25NO4F3SBr [Mꢀ Br]+: 528.1451, found: 528.1448.
4.2.2.7. 2,3,10-Trimethoxy-9-2′,4′-difluorobenzyloxyprotopalmatine bro-
mide (9l). Heavy brown solid; yield: 41%; mp: 187–189 ◦C; IR: 3008,
1603, 1507, 1363, 1215, 1023, 865, 726 cmꢀ 1; 1H NMR: δ 9.69 (s, 1H),
9.06 (s, 1H), 8.23 (d, J = 9.0 Hz, 1H), 8.07 (d, J = 9.0 Hz, 1H), 7.83–7.76
(m, 1H), 7.72 (s, 1H), 7.30 (td, J = 9.6, 2.4 Hz, 1H), 7.16 (td, J = 8.4, 2.4
Hz, 1H), 7.11 (s, 1H), 5.41 (s, 2H), 4.93 (t, J = 6.6 Hz, 2H), 4.06 (s, 3H),
4.2.2.8. 2,3,10-Trimethoxy-9-3′,5′-difluorobenzyloxyprotopalmatine bro-
mide (9m). Yellow solid; yield: 42%; mp: 245–247 ◦C; IR: 2986, 1608,
1359, 1276, 1106, 1023, 866, 696 cmꢀ 1; 1H NMR: δ 9.85 (s, 1H), 9.06 (s,
1H), 8.24 (d, J = 9.0 Hz, 1H), 8.07 (d, J = 9.0 Hz, 1H), 7.72 (s, 1H),
7

T.-Y. Fan et al.
Bioorganic Chemistry 113 (2021) 104994
Fig. 6. 9k downregulates PCSK9 transcriptional expression through HNF1
α
and Sp1 cluster. (A) Effects of 9k on PCSK9 luciferase-reporter plasmids containing
different truncated promoter region D1–D7 in HepG2 cells. Position ꢀ 1 was the 3′ end of PCSK9 promoter inserts. HepG2 cells were transfected with the D1–D7 or
pGL4 plasmids for 24 h and treated with vehicle or 9k (10 μM) for 24 h. Then the luciferase activity was detected. (B) Mutation analysis of core nucleotide sequence
of HNF1α or HINFP binding site in PCSK9 promoter. Data are shown as mean ± SEM (at least 3 independent experiments). **p < 0.01 vs. vehicle group.
7.45–7.33 (m, 2H), 7.27 (td, J = 9.0, 2.4 Hz, 1H), 7.11 (s, 1H), 5.37 (s,
2H), 4.96 (t, J = 6.6 Hz, 2H), 4.08 (s, 3H), 3.94 (s, 3H), 3.88 (s, 3H), 3.23
(t, J = 6.6 Hz, 2H); 13C NMR: δ 162.3 (2), 151.5, 150.4, 148.7, 145.3,
141.6, 141.1, 137.8, 133.1, 128.7, 126.6, 123.9, 121.5, 119.9, 118.9,
111.3 (2), 111.1, 108.8, 103.5, 73.7, 57.1, 56.2, 55.9, 55.5, 26.0; HRMS:
calcd for C27H24NO4F2Br [Mꢀ Br]+: 464.1668, found: 464.1666.
1350, 1215, 1114, 1018, 735 cmꢀ 1; 1H NMR: δ 9.89 (s, 1H), 9.07 (s, 1H),
8.51 (t, J = 1.8 Hz, 1H), 8.25 (d, J = 9.0 Hz, 2H), 8.08 (t, J = 9.0 Hz, 2H),
7.78–7.66 (m, 2H), 7.11 (s, 1H), 5.49 (s, 2H), 4.96 (t, J = 6.6 Hz, 2H),
4.09 (s, 3H), 3.94 (s, 3H), 3.87 (s, 3H), 3.23 (t, J = 6.6 Hz, 2H); 13C NMR:
δ 152.0, 150.9, 149.2, 148.2, 145.8, 142.1, 139.3, 138.2, 135.4, 133.5,
130.5, 129.1, 126.9, 124.5, 123.6, 123.4, 122.0, 120.4, 119.3, 111.7,
109.2, 74.5, 57.5, 56.6, 56.3, 55.9, 26.4; HRMS: calcd for C27H25N2O6Br
[Mꢀ Br]+: 473.1707, found: 473.1705.
4.2.2.9. 2,3,10-Trimethoxy-9-p-carbomethoxybenzyloxyprotopalmatine
bromide (9n). Brown solid; yield: 39%; mp: 211–213 ◦C; IR: 3421, 2916,
1716, 1363, 1272, 1110, 866, 757 cmꢀ 1; 1H NMR: δ 9.82 (s, 1H), 9.05 (s,
1H), 8.23 (d, J = 9.0 Hz, 1H), 8.06 (d, J = 9.0 Hz, 1H), 8.03–7.98 (m,
2H), 7.76 (d, J = 7.8 Hz, 2H), 7.72 (s, 1H), 7.11 (s, 1H), 5.45 (s, 2H),
4.95 (t, J = 6.6 Hz, 2H), 4.08 (s, 3H), 3.94 (s, 3H), 3.87 (s, 3H), 3.86 (s,
3H), 3.22 (t, J = 6.6 Hz, 2H); 13C NMR: δ 166.4, 152.0, 150.8, 149.2,
145.7, 142.5, 142.2, 138.2, 133.5, 129.8, 129.7 (2), 129.1, 128.9 (2),
127.0, 124.2, 122.0, 120.4, 119.3, 111.7, 109.2, 75.0, 57.5, 56.6, 56.3,
56.0, 52.7, 26.4; HRMS: calcd for C29H28NO6Br [Mꢀ Br]+: 486.1911,
found: 486.1911.
4.2.2.12. 2,3,10-Trimethoxy-9-cyclohexylmethoxyprotopalmatine bromide
(9r). Yellow solid; yield: 45%; mp: 260–262 ◦C; IR: 2930, 1606, 1524,
1360, 1278, 1115, 1022, 879 cmꢀ 1; 1H NMR: δ 9.66 (d, J = 2.4 Hz, 1H),
9.04 (s, 1H), 8.20 (dd, J = 9.0, 2.4 Hz, 1H), 8.02 (d, J = 9.0 Hz, 1H), 7.71
(d, J = 1.8 Hz, 1H), 7.11 (d, J = 1.8 Hz, 1H), 4.96 (t, J = 6.6 Hz, 2H),
4.10 (d, J = 6.6 Hz, 2H), 4.05 (d, J = 1.8 Hz, 3H), 3.94 (d, J = 1.8 Hz,
3H), 3.87 (d, J = 1.8 Hz, 3H), 3.23 (t, J = 6.6 Hz, 2H), 2.05–1.84 (m,
1H), 1.81–1.66 (m, 4H), 1.42–1.01 (m, 6H); 13C NMR: δ 151.5, 150.2,
148.7, 145.1, 143.0, 137.6, 133.2, 128.6, 126.8, 123.1, 121.4, 119.9,
118.9, 111.3, 108.7, 79.4, 57.1, 56.2, 55.9, 55.7, 37.9, 29.2 (2), 26.1,
26.0, 25.3 (2); HRMS: calcd for C27H32NO4Br [Mꢀ Br]+: 434.2326,
found: 434.2326.
4.2.2.10. 2,3,10-Trimethoxy-9-p-cyanobenzyloxyprotopalmatine bromide
(9o). Yellow solid; yield: 51%; mp: 198–200 ◦C; IR: 3387, 2982, 2227,
1603, 1359, 1276, 1110, 822 cmꢀ 1; 1H NMR: δ 9.83 (s, 1H), 9.06 (s, 1H),
8.24 (d, J = 9.0 Hz, 1H), 8.07 (d, J = 9.0 Hz, 1H), 7.94–7.87 (m, 2H),
7.87–7.76 (m, 2H), 7.72 (s, 1H), 7.11 (s, 1H), 5.45 (s, 2H), 4.95 (t, J =
6.6 Hz, 2H), 4.07 (s, 3H), 3.94 (s, 3H), 3.87 (s, 3H), 3.23 (t, J = 6.6 Hz,
2H); 13C NMR: δ 151.5, 150.3, 148.7, 145.3, 142.3, 141.7, 137.8, 133.1,
132.4 (2), 128.9 (2), 128.6, 126.6, 123.9, 121.5, 119.9, 118.9, 118.7,
111.3, 110.9, 108.8, 74.3, 57.1, 56.2, 55.9, 55.5, 26.0; HRMS: calcd for
4.2.2.13. 2,3,10-Trimethoxy-9-cyclopentylmethoxylprotopalmatine bro-
mide (9s). Faint yellow solid; yield: 39%; mp: 234–236 ◦C; IR: 3392,
2920, 1603, 1359, 1241, 1114, 1018, 861 cmꢀ 1; 1H NMR: δ 9.69 (s, 1H),
9.04 (s, 1H), 8.20 (d, J = 9.0 Hz, 1H), 8.03 (d, J = 9.0 Hz, 1H), 7.72 (s,
1H), 7.10 (s, 1H), 4.97 (t, J = 6.6 Hz, 2H), 4.18 (d, J = 7.2 Hz, 2H), 4.05
(s, 3H), 3.94 (s, 3H), 3.87 (s, 3H), 3.23 (t, J = 6.6 Hz, 2H), 2.04–1.94 (m,
1H),1.90–1.79 (m, 2H), 1.69–1.53 (m, 4H), 1.50–1.38 (m, 2H); 13C
NMR: δ 151.5, 150.2, 148.7, 145.2, 142.8, 137.6, 133.1, 128.6, 126.7,
123.1, 121.5, 119.9, 118.9, 111.3, 108.7, 78.3, 57.0, 56.2, 55.9, 55.6,
39.6, 29.0 (2), 26.0, 25.1 (2); HRMS: calcd for C26H30NO4Br [Mꢀ Br]+:
420.2169, found: 420.2170.
C
28H25N2O4Br [Mꢀ Br]+: 453.1809, found: 453.1809.
4.2.2.11. 2,3,10-Trimethoxy-9-m-nitrobenzyloxyprotopalmatine bromide
(9p). Yellow solid; yield: 37%; mp: 230–232 ◦C; IR: 2920, 1607, 1525,
8

T.-Y. Fan et al.
Bioorganic Chemistry 113 (2021) 104994
4.2.2.14. 2,3,10-Trimethoxy-9-cyclobutylmethoxylprotopalmatine
bro-
using a Victor X5 multilabel plate reader (PerkinElmer, Waltham, MA,
USA).
mide (9t). Yellow solid; yield: 47%; mp: 237–239 ◦C; IR: 3392, 2977,
2231, 1608, 1511, 1359, 1110, 866 cmꢀ 1; 1H NMR: δ 9.69 (s, 1H), 9.03
(s, 1H), 8.20 (d, J = 9.0 Hz, 1H), 8.03 (d, J = 9.0 Hz, 1H), 7.71 (s, 1H),
7.10 (s, 1H), 4.96 (t, J = 6.6 Hz, 2H), 4.29 (d, J = 7.20 Hz, 2H), 4.06 (s,
3H), 3.94 (s, 3H), 3.87 (s, 3H), 3.23 (t, J = 6.6 Hz, 2H), 2.88 (p, J = 7.2
Hz, 1H), 2.16–2.02 (m, 2H), 1.99–1.88 (m, 4H); 13C NMR: δ 151.5,
150.3, 148.7, 145.1, 142.7, 137.7, 133.1, 128.6, 126.7, 123.2, 121.6,
119.9, 118.9, 111.3, 108.7, 78.0, 57.0, 56.2, 55.9, 55.6, 34.6, 26.0, 24.5
(2), 18.1; HRMS: calcd for C25H28NO4Br [Mꢀ Br]+: 406.2013, found:
406.2012.
4.3.4. Western blotting assay
HepG2 cells plated in 12-well plate were treated with compound 9k
in MEM for 24 h. Then the cells were lysed using RIPA lysis buffer
(Applygen, Beijing, China). The lysed protein samples were separated
using 10% SDS-polyacrylamide gel electrophoresis (PAGE), then elec-
troblotted onto a 0.45 μm polyvinylidene fluoride (PVDF) membrane
(Millipore, Bedford, USA). The PVDF membranes were then incubated
with primary antibodies against PCSK9 (R&D Systems, Minneapolis,
USA), LDLR (Abcam, Cambridge, UK) or GAPDH (ZSJQ-Bio, Beijing,
China) at 4 ◦C overnight. The membranes were washed and incubated
with horse radish peroxidase (HRP)-conjugated secondary antibodies
(ZSJQ-Bio) at room temperature for 1 h. All bands were visualized with
an enhanced chemiluminescence detection system (Millipore). Image J
software was used to quantify PCSK9 or LDLR protein amounts,
normalized to GAPDH.
4.2.2.15. 2,3,10-Trimethoxy-9–2′-cyclohexylethyoxylprotopalmatine bro-
mide (9u). Faint yellow solid; yield: 32%; mp: 204–206 ◦C; IR: 3390,
2920, 1606, 1360, 1275, 1111, 1019, 812 cmꢀ 1; 1H NMR: δ 9.74 (s, 1H),
9.04 (d, J = 3.0 Hz, 1H), 8.20 (dd, J = 9.0, 2.4 Hz, 1H), 8.03 (d, J = 9.0
Hz, 1H), 7.72 (s, 1H), 7.10 (d, J = 1.8 Hz, 1H), 4.97 (t, J = 6.6 Hz, 2H),
4.33 (td, J = 6.6, 1.8 Hz, 2H), 4.05 (d, J = 1.8 Hz, 3H), 3.94 (d, J = 1.2
Hz, 3H), 3.87 (d, J = 1.8 Hz, 3H), 3.23 (t, J = 6.6 Hz, 2H), 1.79–0.93 (m,
13H); 13C NMR: δ 151.5, 150.2, 148.7, 145.3, 142.8, 137.6, 133.1,
128.6, 126.7, 123.2, 121.6, 119.9, 118.9, 111.3, 108.7, 72.3, 57.0, 56.2,
55.9, 55.5, 36.8, 33.9, 32.8 (2), 26.1, 26.0, 25.8 (2); HRMS: calcd for
4.3.5. DiI-LDL uptake assay
HepG2 cells were treated with 9k as described above. One day later,
cells were added with 2 μg/mL DiI-LDL (ADI Alpha Diagnostic Inter-
national, San Antonio, USA) at 37 ◦C for 4 h. After cells were washed and
resuspended with PBS, flow cytometer (ACEA NovoCyte, USA) was
applied to measure the DiI fluorescence as described previously [32].
Additionally, DiI-LDL uptake by HepG2 cells treated as above was
viewed by a Confocal Laser Scanning Microscope (FV500, Olympus,
Tokyo, Japan).
C
28H34NO4Br [Mꢀ Br]+: 448.2482, found: 448.2478.
4.3. Biological methods
4.3.1. Cells and animals
The human hepatoma cell line HepG2 cells were cultured in Eagle
minimal essential medium (MEM, Invitrogen, Carlsbad, USA) supple-
mented with 10% fetal bovine serum (FBS, Gibco, Grand Island, USA),
1% nonessential amino acids and 1% sodium pyruvate (Invitrogen).
The stable pGL4-PCSK9-P transfected HepG2 cells, named as
PCSK9p-Luc HepG2 cells and used as PCSK9 transcriptional inhibitor
HTS assay [12], were cultured in MEM supplemented with 10% FBS, 1%
nonessential amino acids, 1% sodium pyruvate and additional G418
(700 mg/mL, Invitrogen). Cells were maintained at 37 ◦C in the presence
of 5% CO2.
4.3.6. ELISA assay
The serum PCSK9 levels in C57BL/6J mice were detected using the
mouse PCSK9 simple step ELISA kit (Abcam). Briefly, all samples or
standards were added to appropriate wells and subsequently the Anti-
body Cocktail was added. The plate was sealed and incubated for 1 h at
room temperature on a plate shaker. Each well was washed with 3 ×
350 μL 1× Wash Buffer PT. Then the TMB Development Solution was
added and incubated for 20 min in the dark on a plate shaker. Finally,
Stop Solution was added to each well and shaked plate on a plate shaker
for 1 min to mix. The OD value at 450 nm was measured with a Victor X5
multilabel plate reader (PerkinElmer).
All experimental procedures relating animals were approved by the
Institutional Laboratory Animal Care and Use Committee of Institute of
Medicinal Biotechnology. Kunming mice and C57BL/6J mice were
purchased from Vital River Laboratory Animal Technology (Beijing,
China). Kunming mice with body weights of 20.0 ± 1.0 g were fed
regular rodent chow diet and randomly divided into four groups with six
mice each. Compound 9k was given orally in a single dose of 0, 250,
500, or 1,000 mg/kg, respectively. Then the body weight, survival and
behavioral characteristics were monitored for seven days. C57BL/6J
mice (~8 weeks old) were used to investigate the effect of 9k on PCSK9
expression in vivo, which were divided into 2 groups (10 mice/group)
intragastric administrated with 9k (50 mg/kg) or vehicle every day for 4
weeks. Mice were sacrificed, and serum and liver samples were collected
for further determination of PCSK9 by western blotting assay.
4.3.7. Statistical analyses
All Biological experiments were duplicated at least 3 independent
times and the values were shown as the mean ± SEM. Student’s t-test
was applied to compare two groups and one-way ANOVA with Bonfer-
roni’s correction was used to compare the difference among multiple
groups by GraphPad Prism Software. Two-sided P-values of <0.05 were
considered a statistically significant difference.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
4.3.2. Screening
The PCSK9p-Luc HepG2 cells cultured in MEM medium were treated
with different compounds at a concentration of 10
μg/mL or 0.1% DMSO
Acknowledgments
for 24 h. The luciferase activities were measured by the Luciferase Assay
System (Promega, Madison, USA) and detected using a Victor X5 mul-
tilabel plate reader (PerkinElmer, Waltham, USA). Then the inhibitory
rates were calculated as percentage.
This work was supported by the CAMS initiative for innovative
medicine (2019-I2M-1-005, 2016-I2M-2-002), the National Natural
Science Foundation of China (81473214) and the Drug Innovation Major
Project of China (2018ZX09711001-003-006).
4.3.3. MTT assay
Compounds 9e and 9k were added to HepG2 cells for 24 h and then
MTT reagent (1 mg/mL) was added and incubated at 37 ◦C for 4 h, using
MTT assay. Cells were washed and added with DMSO to dissolve the
purple formazan crystals, and the absorbance at 550 nm was measured
Appendix A. Supplementary material
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bioorg.2021.104994.
9

T.-Y. Fan et al.
Bioorganic Chemistry 113 (2021) 104994
[17] J. Cameron, T. Ranheim, M.A. Kulseth, T.P. Leren, K.E. Berge, Berberine decreases
PCSK9 expression in HepG2 cells, Atherosclerosis 201 (2008) 266–273.
[18] B. Dong, H. Li, A.B. Singh, A. Cao, J. Liu, Inhibition of PCSK9 transcription by
berberine involves down-regulation of hepatic HNF1alpha protein expression
through the ubiquitin-proteasome degradation pathway, J. Biol. Chem. 290 (2015)
4047–4058.
References
[1] G.A. Mensah, G.A. Roth, V. Fuster, The Global Burden of Cardiovascular Diseases
and Risk Factors: 2020 and Beyond, J. Am. Coll. Cardiol. 74 (2019) 2529–2532.
[2] A.C. Burke, D.E. Telford, M.W. Huff, Bempedoic acid: effects on lipoprotein
metabolism and atherosclerosis, Curr. Opin. Lipidol. 30 (2019) 1–9.
[3] Z.H. Tang, T.H. Li, J. Peng, J. Zheng, T.T. Li, L.S. Liu, Z.S. Jiang, X.L. Zheng,
PCSK9: A novel inflammation modulator in atherosclerosis? J. Cell Physiol. 234
(2019) 2345–2355.
[19] H. Li, B. Dong, S.W. Park, H.S. Lee, W. Chen, J. Liu, Hepatocyte nuclear factor 1
alpha plays a critical role in PCSK9 gene transcription and regulation by the
natural hypocholesterolemic compound berberine, J. Biol. Chem. 284 (2009)
28885–28895.
[4] M. Benn, B.G. Nordestgaard, P. Grande, P. Schnohr, A. Tybjaerg-Hansen, PCSK9
R46L, low-density lipoprotein cholesterol levels, and risk of ischemic heart disease:
3 independent studies and meta-analyses, J. Am. Coll. Cardiol. 55 (2010)
2833–2842.
[20] S. Cao, P. Xu, J. Yan, H. Liu, L. Liu, L. Cheng, F. Qiu, N. Kang, Berberrubine and its
analog, hydroxypropyl-berberrubine, regulate LDLR and PCSK9 expression via the
ERK signal pathway to exert cholesterol-lowering effects in human hepatoma
HepG2 cells, Cell Biochem. 120 (2018) 1340–1349.
[5] J. Cohen, A. Pertsemlidis, I.K. Kotowski, R. Graham, C.K. Garcia, H.H. Hobbs, Low
LDL cholesterol in individuals of African descent resulting from frequent nonsense
mutations in PCSK9, Nat. Genet. 37 (2005) 161–165.
[21] C.L. Wu, C. Xi, J.H. Tong, J. Zhao, H.L. Jiang, J. Wang, Y.P. Wang, H. Liu, Design,
synthesis, and biological evaluation of novel tetrahydroprotoberberine derivatives
(THPBs) as proprotein convertase subtilisin/kexin type 9 (PCSK9) modulators for
the treatment of hyperlipidemia, Acta Pharm. Sin. B. 9 (2019) 1216–1230.
[22] W.J. Kong, J. Wei, P. Abidi, M.H. Lin, S. Inaba, C. Li, Y.L. Wang, Z.Z. Wang, S.Y. Si,
H.N. Pan, S.K. Wang, J.D. Wu, Y. Wang, Z.R. Li, J.W. Liu, J.D. Jiang, Berberine is a
novel cholesterol-lowering drug working through a unique mechanism distinct
from statins, Nat. Med. 10 (2004) 1344–1351.
[6] J.C. Cohen, E. Boerwinkle, T.H. Jr Mosley, H.H. Hobbs, Sequence variations in
PCSK9, low LDL, and protection against coronary heart disease, N. Engl. J. Med.
354 (2006) 1264–1272.
[7] D.L. Dixon, L.G. Pamulapati, J.D. Bucheit, E.M. Sisson, S.R. Smith, C.J. Kim, G.
F. Wohlford, J. Pozen, Recent updates on the use of PCSK9 inhibitors in patients
with atherosclerotic cardiovascular disease, Curr. Atheroscler. Rep. 21 (2019) 16.
[8] D.J. Kereiakes, J.G. Robinson, C.P. Cannon, C. Lorenzato, R. Pordy, U. Chaudhari,
H.M. Colhoun, Efficacy and safety of the proprotein convertase subtilisin/kexin
type 9 inhibitor alirocumab among high cardiovascular risk patients on maximally
tolerated statin therapy: The ODYSSEY COMBO I study, Am. Heart J. 169 (2015)
906–915.
[23] Y.X. Wang, W.J. Kong, Y.H. Li, S. Tang, Z. Li, Y.B. Li, Y.Q. Shan, C.W. Bi, J.D. Jiang,
D.Q. Song, Synthesis and structure–activity relationship of berberine analogues in
LDLR up-regulation and AMPK activation, Bioorg. Med. Chem. 20 (2012)
6552–6558.
[24] Y.X. Wang, Y.P. Wang, H. Zhang, W.J. Kong, Y.H. Li, F. Liu, R.M. Gao, T. Liu, J.
D. Jiang, D.Q. Song, Synthesis and biological evaluation of berberine analogues as
novel up-regulators for both low-density-lipoprotein receptor and insulin receptor,
Bioorg. Med. Chem. Let. 19 (2009) 6004–6008.
[9] M.S. Sabatine, R.P. Giugliano, S.D. Wiviott, F.J. Raal, D.J. Blom, J. Robinson, C.
M. Ballantyne, R. Somaratne, J. Legg, S.M. Wasserman, R. Scott, M.J. Koren, E.
A. Stein, Efficacy and safety of evolocumab in reducing lipids and cardiovascular
events, N. Engl. J. Med. 372 (2015) 1500–1509.
[25] T.Y. Fan, M.X. Ge, Z.H. Guo, H.W. He, N. Zhang, Y.H. Li, D.Q. Song, Discovery of
9O-substituted palmatine derivatives as a new class of anti-COL1A1 agents via
repressing TGF-β1/Smads and JAK1/STAT3 pathways, Molecules 25 (2020) 773.
[26] T.Y. Fan, X.X. Guo, Q.X. Zeng, W. Wei, X.F. You, Y.X. Wang, J. Pang, D.Q. Song,
Synthesis and structure-activity relationship of palmatine derivatives as a novel
class of antibacterial agents against Helicobacter pylori, Molecules 25 (2020) 1352.
[27] T.Y. Fan, J. Pang, Q.X. Zeng, Y.X. Wang, X.F. You, D.Q. Song, Anti-Helicobacter
pylori activities of 9-substituted palmatine derivatives, Acta Pharm. Sin. 55 (2020)
1237–1244.
[10] F.J. Raal, E.A. Stein, R. Dufour, T. Turner, F. Civeira, L. Burgess, G. Langslet,
R. Scott, A.G. Olsson, D. Sullivan, G.K. Hovingh, B. Cariou, I.G. Berthold,
R. Somaratne, I. Bridges, R. Scott, S.M. Wasserman, D. Gaudet, PCSK9 inhibition
with evolocumab (AMG 145) in heterozygous familial hypercholesterolaemia
(RUTHERFORD-2): a randomised, double-blind, placebo-controlled trial, Lancet
385 (2015) 331–340.
[11] K.K. Ray, U. Landmesser, L.A. Leiter, D. Kallend, R. Dufour, M. Karakas, T. Hall, R.
P. Troquay, T. Turner, F.L. Visseren, P. Wijngaard, R.S. Wright, J.J. Kastelein,
Inclisiran in patients at high cardiovascular risk with elevated LDL cholesterol,
N. Engl. J. Med. 376 (2017) 1430–1440.
[28] J. Chen, T. Wang, S. Xu, A. Lin, H. Yao, W. Xie, Z. Zhu, J. Xu, Design, synthesis and
biological evaluation of novel nitric oxide-donating protoberberine derivatives as
antitumor agents, Eur. J. Med. Chem. 132 (2017) 173–183.
[12] A.L. Catapano, A. Pirillo, G.D. Norata, New pharmacological approaches to target
PCSK9, Curr. Atheroscler. Rep. 22 (2020) 24.
[29] D.W. Zhang, T.A. Lagace, R. Garuti, Z.Z. Zhao, M. McDonald, J.D. Horton, J.
C. Cohen, H.H. Hobbs, Binding of proprotein convertase subtilisin/kexin type 9 to
epidermal growth factor-like repeat A of low density lipoprotein receptor decreases
receptor recycling and increases degradation, J. Biol. Chem. 282 (2007)
18602–18612.
[13] D.S. Kazi, A.E. Moran, P.G. Coxson, J. Penko, D.A. Ollendorf, S.D. Pearson, J.
A. Tice, D. Guzman, K. Bibbins-Domingo, Cost-effectiveness of PCSK9 inhibitor
therapy in patients with heterozygous familial hypercholesterolemia or
atherosclerotic cardiovascular disease, J. Am. Med. Assoc. 316 (2016) 743–753.
[14] S. Xu, S. Luo, Z. Zhu, J. Xu, Small molecules as inhibitors of PCSK9: Current status
and future challenges, Eur. J. Med. Chem. 162 (2019) 212–233.
[30] N. Nassoury, D.A. Blasiole, A. Tebon Oler, N. Nassoury, D.A. Blasiole, A.T. Oler,
S. Benjannet, J. Hamelin, V. Poupon, P.S. McPherson, A.D. Attie, A. Prat, N.
G. Seidah, The cellular trafficking of the secretory proprotein convertase PCSK9
and its dependence on the LDLR, Traffic 8 (2007) 718–732.
[15] P. Lo Surdo, M.J. Bottomley, A. Calzetta, E.C. Settembre, A. Cirillo, S. Pandit, Y.
G. Ni, B. Hubbard, A. Sitlani, A. Carfí, Mechanistic implications for LDL receptor
degradation from the PCSK9/LDLR structure at neutral pH, EMBO Rep. 12 (2011)
1300–1305.
[31] H. Li, B. Dong, S.W. Park, H.S. Lee, W. Chen, J.W. Liu, Hepatocyte nuclear factor 1
plays a critical role in PCSK9 gene transcription and regulation by the natural
hypocholesterolemic compound Berberine, J. Biol. Chem. 284 (2009)
28885–28895.
α
[16] X.L. Wang, X.F. Chen, X.M. Zhang, C.Y. Su, M.X. Yang, W. He, Y. Du, S.Y. Si,
L. Wang, B. Hong, A small-molecule inhibitor of PCSK9 transcription ameliorates
[32] L. Wang, X.J. Jia, H.J. Jiang, Y. Du, F. Yang, S.Y. Si, B. Hong, MicroRNAs 185, 96,
and 223 repress selective high-density lipoprotein cholesterol uptake through
posttranscriptional inhibition, Mol. Cell. Biol. 33 (2013) 1956–1964.
atherosclerosis through the modulation of FoxO1/3 and HNF1
(2020) 102650.
α, EBioMedicine 52
10